diff --git a/AE/01-AE-with-MNIST.ipynb b/AE/00-AE-with-MNIST.ipynb
similarity index 100%
rename from AE/01-AE-with-MNIST.ipynb
rename to AE/00-AE-with-MNIST.ipynb
diff --git a/AE/01-AE-with-MNIST==done==.ipynb b/AE/01-AE-with-MNIST==done==.ipynb
deleted file mode 100644
index 169469542fd0989cde52e7d5f551b1f9d3835585..0000000000000000000000000000000000000000
--- a/AE/01-AE-with-MNIST==done==.ipynb
+++ /dev/null
@@ -1,9887 +0,0 @@
-{
- "cells": [
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "<img width=\"800px\" src=\"../fidle/img/00-Fidle-header-01.svg\"></img>\n",
- "\n",
- "# <!-- TITLE --> [AE1] - Building and training an AE denoiser model\n",
- "<!-- DESC --> Episode 1 : After construction, the model is trained with noisy data from the MNIST dataset.\n",
- "\n",
- "<!-- AUTHOR : Jean-Luc Parouty (CNRS/SIMaP) -->\n",
- "\n",
- "## Objectives :\n",
- " - Understanding and implementing a denoizing **autoencoder** neurals network (AE)\n",
- " - Understanding a more **advanced programming model**\n",
- "\n",
- "The calculation needs being important, it is preferable to use a very simple dataset such as MNIST. \n",
- "The use of a GPU is often indispensable.\n",
- "\n",
- "## What we're going to do :\n",
- "\n",
- " - Defining a VAE model\n",
- " - Build the model\n",
- " - Train it\n",
- " - Follow the learning process with Tensorboard\n",
- "\n",
- "Thanks to **François Chollet** whose VAE example has greatly inspired this example. \n",
- "See : https://keras.io/examples/generative/vae\n"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "## Step 1 - Init python stuff"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 1,
- "metadata": {
- "execution": {
- "iopub.execute_input": "2021-03-01T19:32:38.031692Z",
- "iopub.status.busy": "2021-03-01T19:32:38.031219Z",
- "iopub.status.idle": "2021-03-01T19:32:41.792505Z",
- "shell.execute_reply": "2021-03-01T19:32:41.793007Z"
- }
- },
- "outputs": [
- {
- "data": {
- "text/html": [
- "<style>\n",
- "\n",
- "div.warn { \n",
- " background-color: #fcf2f2;\n",
- " border-color: #dFb5b4;\n",
- " border-left: 5px solid #dfb5b4;\n",
- " padding: 0.5em;\n",
- " font-weight: bold;\n",
- " font-size: 1.1em;;\n",
- " }\n",
- "\n",
- "\n",
- "\n",
- "div.nota { \n",
- " background-color: #DAFFDE;\n",
- " border-left: 5px solid #92CC99;\n",
- " padding: 0.5em;\n",
- " }\n",
- "\n",
- "div.todo:before { content:url(data:image/svg+xml;base64,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);\n",
- " float:left;\n",
- " margin-right:20px;\n",
- " margin-top:-20px;\n",
- " margin-bottom:20px;\n",
- "}\n",
- "div.todo{\n",
- " font-weight: bold;\n",
- " font-size: 1.1em;\n",
- " margin-top:40px;\n",
- "}\n",
- "div.todo ul{\n",
- " margin: 0.2em;\n",
- "}\n",
- "div.todo li{\n",
- " margin-left:60px;\n",
- " margin-top:0;\n",
- " margin-bottom:0;\n",
- "}\n",
- "\n",
- "div .comment{\n",
- " font-size:0.8em;\n",
- " color:#696969;\n",
- "}\n",
- "\n",
- "\n",
- "\n",
- "</style>\n",
- "\n"
- ],
- "text/plain": [
- "<IPython.core.display.HTML object>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "text/markdown": [
- "<br>**FIDLE 2020 - Practical Work Module**"
- ],
- "text/plain": [
- "<IPython.core.display.Markdown object>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Version : 2.0.17\n",
- "Notebook id : AE1\n",
- "Run time : Monday 01 March 2021, 20:32:41\n",
- "TensorFlow version : 2.4.0\n",
- "Keras version : 2.4.0\n",
- "Datasets dir : /gpfswork/rech/mlh/uja62cb/datasets\n",
- "Run dir : ./run\n",
- "Update keras cache : False\n",
- "Save figs : True\n",
- "Path figs : ./run/figs\n"
- ]
- }
- ],
- "source": [
- "import numpy as np\n",
- "\n",
- "import tensorflow as tf\n",
- "from tensorflow import keras\n",
- "from tensorflow.keras import layers\n",
- "\n",
- "from modules.callbacks import ImagesCallback\n",
- "from tensorflow.keras.callbacks import ModelCheckpoint, TensorBoard\n",
- "\n",
- "import os,sys,json,time,datetime\n",
- "from importlib import reload\n",
- "from IPython.display import display,Image,Markdown,HTML\n",
- "import h5py\n",
- "\n",
- "from skimage import io\n",
- "from skimage.util import random_noise\n",
- "\n",
- "import modules.AE\n",
- "from modules.AE import AE\n",
- "from modules.loader_MNIST import Loader_MNIST\n",
- "from modules.callbacks import ImagesCallback, BestModelCallback\n",
- "\n",
- "sys.path.append('..')\n",
- "import fidle.pwk as pwk\n",
- "\n",
- "datasets_dir = pwk.init('AE1')"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "## Step 2 - Prepare data\n",
- "### 2.1 - Get MNIST dataset"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 2,
- "metadata": {
- "execution": {
- "iopub.execute_input": "2021-03-01T19:32:41.796320Z",
- "iopub.status.busy": "2021-03-01T19:32:41.795847Z",
- "iopub.status.idle": "2021-03-01T19:32:42.315216Z",
- "shell.execute_reply": "2021-03-01T19:32:42.315702Z"
- }
- },
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Dataset loaded.\n",
- "Normalized.\n",
- "Expanded.\n",
- "Concatenate.\n",
- "x shape : (70000, 28, 28, 1)\n",
- "y shape : (70000,)\n"
- ]
- }
- ],
- "source": [
- "x_data, y_data = Loader_MNIST.get()"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "### 2.2 - Add noise to dataset\n",
- "We will add noise to our input dataset (x_data) \n",
- "Our goal is to predict noiselessly data (y_data)"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 3,
- "metadata": {
- "execution": {
- "iopub.execute_input": "2021-03-01T19:32:42.321261Z",
- "iopub.status.busy": "2021-03-01T19:32:42.320773Z",
- "iopub.status.idle": "2021-03-01T19:33:30.882862Z",
- "shell.execute_reply": "2021-03-01T19:33:30.883361Z"
- }
- },
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [#---------------------------------------] 2.5% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [##--------------------------------------] 5.0% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [###-------------------------------------] 7.5% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [####------------------------------------] 10.0% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [#####-----------------------------------] 12.5% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [######----------------------------------] 15.0% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [#######---------------------------------] 17.5% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [########--------------------------------] 20.0% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [#########-------------------------------] 22.5% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [##########------------------------------] 25.0% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [###########-----------------------------] 27.5% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [############----------------------------] 30.0% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [#############---------------------------] 32.5% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [##############--------------------------] 35.0% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [###############-------------------------] 37.5% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [################------------------------] 40.0% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [#################-----------------------] 42.5% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [##################----------------------] 45.0% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [###################---------------------] 47.5% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [####################--------------------] 50.0% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [#####################-------------------] 52.5% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [######################------------------] 55.0% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [#######################-----------------] 57.5% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [########################----------------] 60.0% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [#########################---------------] 62.5% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [##########################--------------] 65.0% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [###########################-------------] 67.5% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [############################------------] 70.0% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [#############################-----------] 72.5% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [##############################----------] 75.0% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [###############################---------] 77.5% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [################################--------] 80.0% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [#################################-------] 82.5% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [##################################------] 85.0% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [###################################-----] 87.5% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [####################################----] 90.0% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [#####################################---] 92.5% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [######################################--] 95.0% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [#######################################-] 97.5% of 70000\r"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Add noise to x_train : [########################################] 100.0% of 70000\n"
- ]
- }
- ],
- "source": [
- "def noise_it(x_data, about):\n",
- " for i,image in enumerate(x_data):\n",
- " pwk.update_progress(about,i+1,len(x_data))\n",
- " image=random_noise(image, mode='gaussian', mean=0, var=0.3)\n",
- " image=random_noise(image, mode='s&p', amount=0.2, salt_vs_pepper=0.5)\n",
- " image=random_noise(image, mode='poisson') \n",
- " image=random_noise(image, mode='speckle', mean=0, var=0.1)\n",
- " x_data[i]=image\n",
- " return x_data\n",
- "\n",
- "\n",
- "# ---- What we want to predict\n",
- "#\n",
- "y_data = np.copy(x_data)\n",
- "\n",
- "# ---- Add noise to input data\n",
- "#\n",
- "x_data = noise_it(x_data, 'Add noise to x_train :')\n"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "### 2.3 - Have a look"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 4,
- "metadata": {
- "execution": {
- "iopub.execute_input": "2021-03-01T19:33:30.887881Z",
- "iopub.status.busy": "2021-03-01T19:33:30.887408Z",
- "iopub.status.idle": "2021-03-01T19:33:32.448127Z",
- "shell.execute_reply": "2021-03-01T19:33:32.447612Z"
- }
- },
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Noisy dataset (x_data) : (70000, 28, 28, 1)\n",
- "Clean dataset (y_data) : (70000, 28, 28, 1)\n"
- ]
- },
- {
- "data": {
- "text/markdown": [
- "<br>**Noisy images (input data)**"
- ],
- "text/plain": [
- "<IPython.core.display.Markdown object>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "text/html": [
- "<div class=\"comment\">Saved: ./run/figs/AE1-01-noisy</div>"
- ],
- "text/plain": [
- "<IPython.core.display.HTML object>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "image/png": "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\n",
- "text/plain": [
- "<Figure size 1080x241.2 with 5 Axes>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "text/markdown": [
- "<br>**Original images we want to obtain (output data)**"
- ],
- "text/plain": [
- "<IPython.core.display.Markdown object>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "text/html": [
- "<div class=\"comment\">Saved: ./run/figs/AE1-02-original</div>"
- ],
- "text/plain": [
- "<IPython.core.display.HTML object>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "image/png": "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\n",
- "text/plain": [
- "<Figure size 1080x241.2 with 5 Axes>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- }
- ],
- "source": [
- "print('Noisy dataset (x_data) : ',x_data.shape)\n",
- "print('Clean dataset (y_data) : ',y_data.shape)\n",
- "\n",
- "pwk.subtitle('Noisy images (input data)')\n",
- "pwk.plot_images(x_data[:5], None, indices='all', columns=5, x_size=3,y_size=3, interpolation=None, save_as='01-noisy')\n",
- "pwk.subtitle('Original images we want to obtain (output data)')\n",
- "pwk.plot_images(y_data[:5], None, indices='all', columns=5, x_size=3,y_size=3, interpolation=None, save_as='02-original')\n"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "### 2.4 - Shuffle and split dataset"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 5,
- "metadata": {
- "execution": {
- "iopub.execute_input": "2021-03-01T19:33:32.453035Z",
- "iopub.status.busy": "2021-03-01T19:33:32.451708Z",
- "iopub.status.idle": "2021-03-01T19:33:32.600666Z",
- "shell.execute_reply": "2021-03-01T19:33:32.601162Z"
- }
- },
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "x_train:(60000, 28, 28, 1) y_train:(60000, 28, 28, 1)\n",
- "x_test :(10000, 28, 28, 1) y_test :(10000, 28, 28, 1)\n"
- ]
- }
- ],
- "source": [
- "p = np.random.permutation(len(x_data))\n",
- "x_data, y_data = x_data[p], y_data[p]\n",
- "x_train, y_train = x_data[:60000], y_data[:60000]\n",
- "x_test, y_test = x_data[60000:], y_data[60000:]\n",
- "\n",
- "print(f'x_train:{x_train.shape} y_train:{y_train.shape}')\n",
- "print(f'x_test :{x_test.shape} y_test :{y_test.shape}')"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "### 2.5 - Save it"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 6,
- "metadata": {
- "execution": {
- "iopub.execute_input": "2021-03-01T19:33:32.605215Z",
- "iopub.status.busy": "2021-03-01T19:33:32.604741Z",
- "iopub.status.idle": "2021-03-01T19:33:32.915077Z",
- "shell.execute_reply": "2021-03-01T19:33:32.914568Z"
- }
- },
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Saved.\n"
- ]
- }
- ],
- "source": [
- "pwk.mkdir('./data')\n",
- "\n",
- "with h5py.File('./data/mnist-noisy.h5', \"w\") as f:\n",
- " f.create_dataset(\"x_train\", data=x_train)\n",
- " f.create_dataset(\"y_train\", data=y_train)\n",
- " f.create_dataset(\"x_test\", data=x_test)\n",
- " f.create_dataset(\"y_test\", data=y_test)\n",
- " print('Saved.')"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "## Step 3 - Build model"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 7,
- "metadata": {
- "execution": {
- "iopub.execute_input": "2021-03-01T19:33:32.918147Z",
- "iopub.status.busy": "2021-03-01T19:33:32.917681Z",
- "iopub.status.idle": "2021-03-01T19:33:32.919271Z",
- "shell.execute_reply": "2021-03-01T19:33:32.919737Z"
- }
- },
- "outputs": [],
- "source": [
- "latent_dim = 10"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "#### Encoder"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 8,
- "metadata": {
- "execution": {
- "iopub.execute_input": "2021-03-01T19:33:32.924600Z",
- "iopub.status.busy": "2021-03-01T19:33:32.924129Z",
- "iopub.status.idle": "2021-03-01T19:33:34.048441Z",
- "shell.execute_reply": "2021-03-01T19:33:34.047897Z"
- }
- },
- "outputs": [],
- "source": [
- "reload(modules.AE)\n",
- "from modules.AE import AE\n",
- "\n",
- "inputs = keras.Input(shape=(28, 28, 1))\n",
- "x = layers.Conv2D(32, 3, activation=\"relu\", strides=2, padding=\"same\")(inputs)\n",
- "x = layers.Conv2D(64, 3, activation=\"relu\", strides=2, padding=\"same\")(x)\n",
- "x = layers.Flatten()(x)\n",
- "x = layers.Dense(16, activation=\"relu\")(x)\n",
- "z = layers.Dense(latent_dim)(x)\n",
- "\n",
- "encoder = keras.Model(inputs, z, name=\"encoder\")\n",
- "encoder.compile()\n",
- "# encoder.summary()"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "#### Decoder"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 9,
- "metadata": {
- "execution": {
- "iopub.execute_input": "2021-03-01T19:33:34.055219Z",
- "iopub.status.busy": "2021-03-01T19:33:34.054739Z",
- "iopub.status.idle": "2021-03-01T19:33:34.112626Z",
- "shell.execute_reply": "2021-03-01T19:33:34.113105Z"
- }
- },
- "outputs": [],
- "source": [
- "inputs = keras.Input(shape=(latent_dim,))\n",
- "x = layers.Dense(7 * 7 * 64, activation=\"relu\")(inputs)\n",
- "x = layers.Reshape((7, 7, 64))(x)\n",
- "x = layers.Conv2DTranspose(64, 3, activation=\"relu\", strides=2, padding=\"same\")(x)\n",
- "x = layers.Conv2DTranspose(32, 3, activation=\"relu\", strides=2, padding=\"same\")(x)\n",
- "outputs = layers.Conv2DTranspose(1, 3, activation=\"sigmoid\", padding=\"same\")(x)\n",
- "\n",
- "decoder = keras.Model(inputs, outputs, name=\"decoder\")\n",
- "decoder.compile()\n",
- "# decoder.summary()"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "#### AE\n"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 10,
- "metadata": {
- "execution": {
- "iopub.execute_input": "2021-03-01T19:33:34.117669Z",
- "iopub.status.busy": "2021-03-01T19:33:34.117208Z",
- "iopub.status.idle": "2021-03-01T19:33:34.127075Z",
- "shell.execute_reply": "2021-03-01T19:33:34.127546Z"
- }
- },
- "outputs": [],
- "source": [
- "ae = AE(encoder, decoder)\n",
- "\n",
- "ae.compile(optimizer=keras.optimizers.Adam(), loss='binary_crossentropy')"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "## Step 4 - Train\n",
- "20' on a CPU \n",
- "1'12 on a GPU (V100, IDRIS)"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 11,
- "metadata": {
- "execution": {
- "iopub.execute_input": "2021-03-01T19:33:34.131585Z",
- "iopub.status.busy": "2021-03-01T19:33:34.131119Z",
- "iopub.status.idle": "2021-03-01T19:33:34.329762Z",
- "shell.execute_reply": "2021-03-01T19:33:34.330261Z"
- }
- },
- "outputs": [],
- "source": [
- "# ---- Callback : Images\n",
- "pwk.mkdir('./run/images')\n",
- "filename = './run/images/image-{epoch:03d}-{i:02d}.jpg'\n",
- "callback_images = ImagesCallback(filename, x=x_test[:5], encoder=encoder,decoder=decoder)\n",
- "\n",
- "# ---- Callback : Best model\n",
- "pwk.mkdir('./run/models')\n",
- "filename = './run/models/best_model'\n",
- "callback_bestmodel = BestModelCallback(filename)\n",
- "\n",
- "# ---- Callback tensorboard\n",
- "dirname = './run/logs'\n",
- "callback_tensorboard = TensorBoard(log_dir=dirname, histogram_freq=1)\n",
- "\n",
- "callbacks_list = [callback_images, callback_bestmodel, callback_tensorboard]"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 12,
- "metadata": {
- "execution": {
- "iopub.execute_input": "2021-03-01T19:33:34.334554Z",
- "iopub.status.busy": "2021-03-01T19:33:34.334087Z",
- "iopub.status.idle": "2021-03-01T19:34:39.883997Z",
- "shell.execute_reply": "2021-03-01T19:34:39.884500Z"
- }
- },
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Epoch 1/30\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\r",
- " 1/469 [..............................] - ETA: 27:52 - loss: 0.6932"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 3/469 [..............................] - ETA: 20s - loss: 0.6922 "
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "WARNING:tensorflow:Callback method `on_train_batch_end` is slow compared to the batch time (batch time: 0.0045s vs `on_train_batch_end` time: 0.0130s). Check your callbacks.\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 15/469 [..............................] - ETA: 4s - loss: 0.6629 "
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 28/469 [>.............................] - ETA: 3s - loss: 0.5991"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 40/469 [=>............................] - ETA: 2s - loss: 0.5513"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 53/469 [==>...........................] - ETA: 2s - loss: 0.5131"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 66/469 [===>..........................] - ETA: 2s - loss: 0.4846"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 79/469 [====>.........................] - ETA: 2s - loss: 0.4624"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 92/469 [====>.........................] - ETA: 1s - loss: 0.4448"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "106/469 [=====>........................] - ETA: 1s - loss: 0.4293"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "120/469 [======>.......................] - ETA: 1s - loss: 0.4165"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "134/469 [=======>......................] - ETA: 1s - loss: 0.4058"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "148/469 [========>.....................] - ETA: 1s - loss: 0.3966"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "162/469 [=========>....................] - ETA: 1s - loss: 0.3886"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "176/469 [==========>...................] - ETA: 1s - loss: 0.3816"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "190/469 [===========>..................] - ETA: 1s - loss: 0.3755"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "204/469 [============>.................] - ETA: 1s - loss: 0.3699"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "218/469 [============>.................] - ETA: 1s - loss: 0.3649"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "232/469 [=============>................] - ETA: 0s - loss: 0.3603"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "246/469 [==============>...............] - ETA: 0s - loss: 0.3561"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "255/469 [===============>..............] - ETA: 1s - loss: 0.3536"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "269/469 [================>.............] - ETA: 0s - loss: 0.3499"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "283/469 [=================>............] - ETA: 0s - loss: 0.3465"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "297/469 [=================>............] - ETA: 0s - loss: 0.3434"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "311/469 [==================>...........] - ETA: 0s - loss: 0.3404"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "325/469 [===================>..........] - ETA: 0s - loss: 0.3375"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "339/469 [====================>.........] - ETA: 0s - loss: 0.3348"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "353/469 [=====================>........] - ETA: 0s - loss: 0.3323"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "367/469 [======================>.......] - ETA: 0s - loss: 0.3299"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "381/469 [=======================>......] - ETA: 0s - loss: 0.3276"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "395/469 [========================>.....] - ETA: 0s - loss: 0.3253"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "409/469 [=========================>....] - ETA: 0s - loss: 0.3232"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "423/469 [==========================>...] - ETA: 0s - loss: 0.3212"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "437/469 [==========================>...] - ETA: 0s - loss: 0.3192"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "451/469 [===========================>..] - ETA: 0s - loss: 0.3173"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "465/469 [============================>.] - ETA: 0s - loss: 0.3154"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "469/469 [==============================] - ETA: 0s - loss: 0.3149"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "469/469 [==============================] - 6s 4ms/step - loss: 0.3148\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- " (saved) Epoch 2/30\n",
- "\r",
- " 1/469 [..............................] - ETA: 1s - loss: 0.1967"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 15/469 [..............................] - ETA: 1s - loss: 0.1966"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 29/469 [>.............................] - ETA: 1s - loss: 0.1958"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 43/469 [=>............................] - ETA: 1s - loss: 0.1955"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 57/469 [==>...........................] - ETA: 1s - loss: 0.1950"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 71/469 [===>..........................] - ETA: 1s - loss: 0.1946"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 85/469 [====>.........................] - ETA: 1s - loss: 0.1942"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 99/469 [=====>........................] - ETA: 1s - loss: 0.1938"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "113/469 [======>.......................] - ETA: 1s - loss: 0.1934"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "127/469 [=======>......................] - ETA: 1s - loss: 0.1931"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "141/469 [========>.....................] - ETA: 1s - loss: 0.1928"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "155/469 [========>.....................] - ETA: 1s - loss: 0.1925"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "169/469 [=========>....................] - ETA: 1s - loss: 0.1922"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "183/469 [==========>...................] - ETA: 1s - loss: 0.1919"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "197/469 [===========>..................] - ETA: 1s - loss: 0.1916"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "211/469 [============>.................] - ETA: 0s - loss: 0.1914"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "225/469 [=============>................] - ETA: 0s - loss: 0.1911"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "239/469 [==============>...............] - ETA: 0s - loss: 0.1909"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "253/469 [===============>..............] - ETA: 0s - loss: 0.1906"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "267/469 [================>.............] - ETA: 0s - loss: 0.1904"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "281/469 [================>.............] - ETA: 0s - loss: 0.1902"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "295/469 [=================>............] - ETA: 0s - loss: 0.1900"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "309/469 [==================>...........] - ETA: 0s - loss: 0.1897"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "323/469 [===================>..........] - ETA: 0s - loss: 0.1895"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "337/469 [====================>.........] - ETA: 0s - loss: 0.1893"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "351/469 [=====================>........] - ETA: 0s - loss: 0.1891"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "365/469 [======================>.......] - ETA: 0s - loss: 0.1888"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "379/469 [=======================>......] - ETA: 0s - loss: 0.1886"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "393/469 [========================>.....] - ETA: 0s - loss: 0.1884"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "407/469 [=========================>....] - ETA: 0s - loss: 0.1882"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "421/469 [=========================>....] - ETA: 0s - loss: 0.1880"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "435/469 [==========================>...] - ETA: 0s - loss: 0.1878"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "449/469 [===========================>..] - ETA: 0s - loss: 0.1876"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "463/469 [============================>.] - ETA: 0s - loss: 0.1873"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "469/469 [==============================] - 2s 4ms/step - loss: 0.1872\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- " (saved) "
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Epoch 3/30\n",
- "\r",
- " 1/469 [..............................] - ETA: 1s - loss: 0.1689"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 15/469 [..............................] - ETA: 1s - loss: 0.1686"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 29/469 [>.............................] - ETA: 1s - loss: 0.1681"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 43/469 [=>............................] - ETA: 1s - loss: 0.1683"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 57/469 [==>...........................] - ETA: 1s - loss: 0.1684"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 71/469 [===>..........................] - ETA: 1s - loss: 0.1686"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 85/469 [====>.........................] - ETA: 1s - loss: 0.1686"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 99/469 [=====>........................] - ETA: 1s - loss: 0.1687"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "113/469 [======>.......................] - ETA: 1s - loss: 0.1687"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "127/469 [=======>......................] - ETA: 1s - loss: 0.1686"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "141/469 [========>.....................] - ETA: 1s - loss: 0.1685"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "155/469 [========>.....................] - ETA: 1s - loss: 0.1684"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "169/469 [=========>....................] - ETA: 1s - loss: 0.1683"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "183/469 [==========>...................] - ETA: 1s - loss: 0.1682"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "197/469 [===========>..................] - ETA: 1s - loss: 0.1681"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "211/469 [============>.................] - ETA: 0s - loss: 0.1681"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "225/469 [=============>................] - ETA: 0s - loss: 0.1680"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "239/469 [==============>...............] - ETA: 0s - loss: 0.1679"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "253/469 [===============>..............] - ETA: 0s - loss: 0.1678"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "267/469 [================>.............] - ETA: 0s - loss: 0.1678"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "281/469 [================>.............] - ETA: 0s - loss: 0.1677"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "295/469 [=================>............] - ETA: 0s - loss: 0.1676"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "309/469 [==================>...........] - ETA: 0s - loss: 0.1675"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "323/469 [===================>..........] - ETA: 0s - loss: 0.1675"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "337/469 [====================>.........] - ETA: 0s - loss: 0.1674"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "351/469 [=====================>........] - ETA: 0s - loss: 0.1673"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "365/469 [======================>.......] - ETA: 0s - loss: 0.1673"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "379/469 [=======================>......] - ETA: 0s - loss: 0.1672"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "393/469 [========================>.....] - ETA: 0s - loss: 0.1672"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "407/469 [=========================>....] - ETA: 0s - loss: 0.1671"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "421/469 [=========================>....] - ETA: 0s - loss: 0.1670"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "435/469 [==========================>...] - ETA: 0s - loss: 0.1670"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "449/469 [===========================>..] - ETA: 0s - loss: 0.1669"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "463/469 [============================>.] - ETA: 0s - loss: 0.1668"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "469/469 [==============================] - 2s 4ms/step - loss: 0.1668\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- " (saved) "
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Epoch 4/30\n",
- "\r",
- " 1/469 [..............................] - ETA: 2s - loss: 0.1541"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 15/469 [..............................] - ETA: 1s - loss: 0.1602"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 29/469 [>.............................] - ETA: 1s - loss: 0.1610"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 43/469 [=>............................] - ETA: 1s - loss: 0.1610"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 57/469 [==>...........................] - ETA: 1s - loss: 0.1610"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 71/469 [===>..........................] - ETA: 1s - loss: 0.1609"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 85/469 [====>.........................] - ETA: 1s - loss: 0.1608"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 99/469 [=====>........................] - ETA: 1s - loss: 0.1607"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "113/469 [======>.......................] - ETA: 1s - loss: 0.1606"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "127/469 [=======>......................] - ETA: 1s - loss: 0.1605"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "141/469 [========>.....................] - ETA: 1s - loss: 0.1604"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "155/469 [========>.....................] - ETA: 1s - loss: 0.1604"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "169/469 [=========>....................] - ETA: 1s - loss: 0.1603"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "183/469 [==========>...................] - ETA: 1s - loss: 0.1602"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "197/469 [===========>..................] - ETA: 1s - loss: 0.1602"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "211/469 [============>.................] - ETA: 0s - loss: 0.1602"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "225/469 [=============>................] - ETA: 0s - loss: 0.1601"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "239/469 [==============>...............] - ETA: 0s - loss: 0.1601"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "253/469 [===============>..............] - ETA: 0s - loss: 0.1600"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "267/469 [================>.............] - ETA: 0s - loss: 0.1600"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "281/469 [================>.............] - ETA: 0s - loss: 0.1600"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "295/469 [=================>............] - ETA: 0s - loss: 0.1599"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "309/469 [==================>...........] - ETA: 0s - loss: 0.1599"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "323/469 [===================>..........] - ETA: 0s - loss: 0.1598"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "337/469 [====================>.........] - ETA: 0s - loss: 0.1598"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "351/469 [=====================>........] - ETA: 0s - loss: 0.1597"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "365/469 [======================>.......] - ETA: 0s - loss: 0.1597"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "379/469 [=======================>......] - ETA: 0s - loss: 0.1597"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "393/469 [========================>.....] - ETA: 0s - loss: 0.1596"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "407/469 [=========================>....] - ETA: 0s - loss: 0.1596"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "421/469 [=========================>....] - ETA: 0s - loss: 0.1595"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "435/469 [==========================>...] - ETA: 0s - loss: 0.1595"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "449/469 [===========================>..] - ETA: 0s - loss: 0.1595"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "463/469 [============================>.] - ETA: 0s - loss: 0.1594"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "469/469 [==============================] - 2s 4ms/step - loss: 0.1594\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- " (saved) Epoch 5/30\n",
- "\r",
- " 1/469 [..............................] - ETA: 1s - loss: 0.1538"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 15/469 [..............................] - ETA: 1s - loss: 0.1554"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 29/469 [>.............................] - ETA: 1s - loss: 0.1553"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 43/469 [=>............................] - ETA: 1s - loss: 0.1553"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 57/469 [==>...........................] - ETA: 1s - loss: 0.1553"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 71/469 [===>..........................] - ETA: 1s - loss: 0.1553"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 85/469 [====>.........................] - ETA: 1s - loss: 0.1552"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 99/469 [=====>........................] - ETA: 1s - loss: 0.1552"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "113/469 [======>.......................] - ETA: 1s - loss: 0.1552"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "127/469 [=======>......................] - ETA: 1s - loss: 0.1552"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "141/469 [========>.....................] - ETA: 1s - loss: 0.1552"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "155/469 [========>.....................] - ETA: 1s - loss: 0.1551"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "169/469 [=========>....................] - ETA: 1s - loss: 0.1551"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "183/469 [==========>...................] - ETA: 1s - loss: 0.1551"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "197/469 [===========>..................] - ETA: 1s - loss: 0.1551"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "211/469 [============>.................] - ETA: 0s - loss: 0.1551"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "225/469 [=============>................] - ETA: 0s - loss: 0.1551"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "239/469 [==============>...............] - ETA: 0s - loss: 0.1551"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "253/469 [===============>..............] - ETA: 0s - loss: 0.1550"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "267/469 [================>.............] - ETA: 0s - loss: 0.1550"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "281/469 [================>.............] - ETA: 0s - loss: 0.1550"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "295/469 [=================>............] - ETA: 0s - loss: 0.1550"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "309/469 [==================>...........] - ETA: 0s - loss: 0.1550"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "323/469 [===================>..........] - ETA: 0s - loss: 0.1550"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "337/469 [====================>.........] - ETA: 0s - loss: 0.1550"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "351/469 [=====================>........] - ETA: 0s - loss: 0.1550"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "365/469 [======================>.......] - ETA: 0s - loss: 0.1549"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "379/469 [=======================>......] - ETA: 0s - loss: 0.1549"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "393/469 [========================>.....] - ETA: 0s - loss: 0.1549"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "407/469 [=========================>....] - ETA: 0s - loss: 0.1549"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "421/469 [=========================>....] - ETA: 0s - loss: 0.1549"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "435/469 [==========================>...] - ETA: 0s - loss: 0.1549"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "449/469 [===========================>..] - ETA: 0s - loss: 0.1549"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "463/469 [============================>.] - ETA: 0s - loss: 0.1549"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "469/469 [==============================] - 2s 4ms/step - loss: 0.1549\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- " (saved) "
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Epoch 6/30\n",
- "\r",
- " 1/469 [..............................] - ETA: 2s - loss: 0.1533"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 15/469 [..............................] - ETA: 1s - loss: 0.1520"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 29/469 [>.............................] - ETA: 1s - loss: 0.1522"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 43/469 [=>............................] - ETA: 1s - loss: 0.1522"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 57/469 [==>...........................] - ETA: 1s - loss: 0.1523"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 71/469 [===>..........................] - ETA: 1s - loss: 0.1524"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 85/469 [====>.........................] - ETA: 1s - loss: 0.1524"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 99/469 [=====>........................] - ETA: 1s - loss: 0.1524"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "113/469 [======>.......................] - ETA: 1s - loss: 0.1525"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "127/469 [=======>......................] - ETA: 1s - loss: 0.1525"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "141/469 [========>.....................] - ETA: 1s - loss: 0.1525"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "155/469 [========>.....................] - ETA: 1s - loss: 0.1525"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "169/469 [=========>....................] - ETA: 1s - loss: 0.1525"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "183/469 [==========>...................] - ETA: 1s - loss: 0.1525"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "197/469 [===========>..................] - ETA: 1s - loss: 0.1525"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "211/469 [============>.................] - ETA: 0s - loss: 0.1526"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "225/469 [=============>................] - ETA: 0s - loss: 0.1526"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "239/469 [==============>...............] - ETA: 0s - loss: 0.1526"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "253/469 [===============>..............] - ETA: 0s - loss: 0.1526"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "267/469 [================>.............] - ETA: 0s - loss: 0.1526"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "281/469 [================>.............] - ETA: 0s - loss: 0.1526"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "295/469 [=================>............] - ETA: 0s - loss: 0.1526"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "309/469 [==================>...........] - ETA: 0s - loss: 0.1526"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "323/469 [===================>..........] - ETA: 0s - loss: 0.1526"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "337/469 [====================>.........] - ETA: 0s - loss: 0.1526"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "351/469 [=====================>........] - ETA: 0s - loss: 0.1526"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "365/469 [======================>.......] - ETA: 0s - loss: 0.1526"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "379/469 [=======================>......] - ETA: 0s - loss: 0.1526"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "393/469 [========================>.....] - ETA: 0s - loss: 0.1526"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "407/469 [=========================>....] - ETA: 0s - loss: 0.1526"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "421/469 [=========================>....] - ETA: 0s - loss: 0.1526"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "435/469 [==========================>...] - ETA: 0s - loss: 0.1526"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "449/469 [===========================>..] - ETA: 0s - loss: 0.1526"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "463/469 [============================>.] - ETA: 0s - loss: 0.1526"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "469/469 [==============================] - 2s 4ms/step - loss: 0.1526\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- " (saved) "
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Epoch 7/30\n",
- "\r",
- " 1/469 [..............................] - ETA: 1s - loss: 0.1485"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 15/469 [..............................] - ETA: 1s - loss: 0.1490"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 29/469 [>.............................] - ETA: 1s - loss: 0.1497"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 43/469 [=>............................] - ETA: 1s - loss: 0.1500"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 57/469 [==>...........................] - ETA: 1s - loss: 0.1502"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 71/469 [===>..........................] - ETA: 1s - loss: 0.1504"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 85/469 [====>.........................] - ETA: 1s - loss: 0.1504"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 99/469 [=====>........................] - ETA: 1s - loss: 0.1505"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "113/469 [======>.......................] - ETA: 1s - loss: 0.1505"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "127/469 [=======>......................] - ETA: 1s - loss: 0.1505"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "141/469 [========>.....................] - ETA: 1s - loss: 0.1505"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "155/469 [========>.....................] - ETA: 1s - loss: 0.1505"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "169/469 [=========>....................] - ETA: 1s - loss: 0.1505"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "183/469 [==========>...................] - ETA: 1s - loss: 0.1505"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "197/469 [===========>..................] - ETA: 1s - loss: 0.1505"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "211/469 [============>.................] - ETA: 0s - loss: 0.1505"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "225/469 [=============>................] - ETA: 0s - loss: 0.1505"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "239/469 [==============>...............] - ETA: 0s - loss: 0.1506"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "253/469 [===============>..............] - ETA: 0s - loss: 0.1506"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "267/469 [================>.............] - ETA: 0s - loss: 0.1506"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "281/469 [================>.............] - ETA: 0s - loss: 0.1506"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "295/469 [=================>............] - ETA: 0s - loss: 0.1506"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "309/469 [==================>...........] - ETA: 0s - loss: 0.1506"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "323/469 [===================>..........] - ETA: 0s - loss: 0.1506"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "337/469 [====================>.........] - ETA: 0s - loss: 0.1506"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "351/469 [=====================>........] - ETA: 0s - loss: 0.1506"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "365/469 [======================>.......] - ETA: 0s - loss: 0.1506"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "379/469 [=======================>......] - ETA: 0s - loss: 0.1506"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "393/469 [========================>.....] - ETA: 0s - loss: 0.1506"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "407/469 [=========================>....] - ETA: 0s - loss: 0.1506"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "421/469 [=========================>....] - ETA: 0s - loss: 0.1506"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "435/469 [==========================>...] - ETA: 0s - loss: 0.1506"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "449/469 [===========================>..] - ETA: 0s - loss: 0.1506"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "463/469 [============================>.] - ETA: 0s - loss: 0.1506"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "469/469 [==============================] - 2s 4ms/step - loss: 0.1506\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- " (saved) "
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Epoch 8/30\n",
- "\r",
- " 1/469 [..............................] - ETA: 1s - loss: 0.1517"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 15/469 [..............................] - ETA: 1s - loss: 0.1514"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 29/469 [>.............................] - ETA: 1s - loss: 0.1510"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 43/469 [=>............................] - ETA: 1s - loss: 0.1507"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 57/469 [==>...........................] - ETA: 1s - loss: 0.1506"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 71/469 [===>..........................] - ETA: 1s - loss: 0.1505"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 85/469 [====>.........................] - ETA: 1s - loss: 0.1505"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 99/469 [=====>........................] - ETA: 1s - loss: 0.1504"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "113/469 [======>.......................] - ETA: 1s - loss: 0.1504"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "127/469 [=======>......................] - ETA: 1s - loss: 0.1503"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "141/469 [========>.....................] - ETA: 1s - loss: 0.1502"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "155/469 [========>.....................] - ETA: 1s - loss: 0.1502"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "169/469 [=========>....................] - ETA: 1s - loss: 0.1501"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "183/469 [==========>...................] - ETA: 1s - loss: 0.1501"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "197/469 [===========>..................] - ETA: 1s - loss: 0.1500"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "211/469 [============>.................] - ETA: 0s - loss: 0.1500"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "225/469 [=============>................] - ETA: 0s - loss: 0.1500"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "239/469 [==============>...............] - ETA: 0s - loss: 0.1499"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "253/469 [===============>..............] - ETA: 0s - loss: 0.1499"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "267/469 [================>.............] - ETA: 0s - loss: 0.1499"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "281/469 [================>.............] - ETA: 0s - loss: 0.1498"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "295/469 [=================>............] - ETA: 0s - loss: 0.1498"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "309/469 [==================>...........] - ETA: 0s - loss: 0.1498"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "323/469 [===================>..........] - ETA: 0s - loss: 0.1498"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "337/469 [====================>.........] - ETA: 0s - loss: 0.1498"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "351/469 [=====================>........] - ETA: 0s - loss: 0.1498"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "365/469 [======================>.......] - ETA: 0s - loss: 0.1497"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "379/469 [=======================>......] - ETA: 0s - loss: 0.1497"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "393/469 [========================>.....] - ETA: 0s - loss: 0.1497"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "407/469 [=========================>....] - ETA: 0s - loss: 0.1497"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "421/469 [=========================>....] - ETA: 0s - loss: 0.1497"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "435/469 [==========================>...] - ETA: 0s - loss: 0.1497"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "449/469 [===========================>..] - ETA: 0s - loss: 0.1497"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "463/469 [============================>.] - ETA: 0s - loss: 0.1497"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "469/469 [==============================] - 2s 4ms/step - loss: 0.1496\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- " (saved) "
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Epoch 9/30\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\r",
- " 1/469 [..............................] - ETA: 1s - loss: 0.1500"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 15/469 [..............................] - ETA: 1s - loss: 0.1474"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 29/469 [>.............................] - ETA: 1s - loss: 0.1474"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 43/469 [=>............................] - ETA: 1s - loss: 0.1475"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 57/469 [==>...........................] - ETA: 1s - loss: 0.1476"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 71/469 [===>..........................] - ETA: 1s - loss: 0.1477"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 85/469 [====>.........................] - ETA: 1s - loss: 0.1477"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 99/469 [=====>........................] - ETA: 1s - loss: 0.1478"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "113/469 [======>.......................] - ETA: 1s - loss: 0.1478"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "127/469 [=======>......................] - ETA: 1s - loss: 0.1479"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "141/469 [========>.....................] - ETA: 1s - loss: 0.1479"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "155/469 [========>.....................] - ETA: 1s - loss: 0.1480"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "169/469 [=========>....................] - ETA: 1s - loss: 0.1480"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "183/469 [==========>...................] - ETA: 1s - loss: 0.1480"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "197/469 [===========>..................] - ETA: 1s - loss: 0.1480"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "211/469 [============>.................] - ETA: 0s - loss: 0.1481"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "225/469 [=============>................] - ETA: 0s - loss: 0.1481"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "239/469 [==============>...............] - ETA: 0s - loss: 0.1481"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "253/469 [===============>..............] - ETA: 0s - loss: 0.1481"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "267/469 [================>.............] - ETA: 0s - loss: 0.1481"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "281/469 [================>.............] - ETA: 0s - loss: 0.1481"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "295/469 [=================>............] - ETA: 0s - loss: 0.1481"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "309/469 [==================>...........] - ETA: 0s - loss: 0.1481"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "323/469 [===================>..........] - ETA: 0s - loss: 0.1481"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "337/469 [====================>.........] - ETA: 0s - loss: 0.1481"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "351/469 [=====================>........] - ETA: 0s - loss: 0.1482"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "365/469 [======================>.......] - ETA: 0s - loss: 0.1482"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "379/469 [=======================>......] - ETA: 0s - loss: 0.1482"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "393/469 [========================>.....] - ETA: 0s - loss: 0.1482"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "407/469 [=========================>....] - ETA: 0s - loss: 0.1482"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "421/469 [=========================>....] - ETA: 0s - loss: 0.1482"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "435/469 [==========================>...] - ETA: 0s - loss: 0.1482"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "449/469 [===========================>..] - ETA: 0s - loss: 0.1482"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "463/469 [============================>.] - ETA: 0s - loss: 0.1482"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "469/469 [==============================] - 2s 4ms/step - loss: 0.1482\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- " (saved) Epoch 10/30\n",
- "\r",
- " 1/469 [..............................] - ETA: 1s - loss: 0.1467"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 15/469 [..............................] - ETA: 1s - loss: 0.1463"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 29/469 [>.............................] - ETA: 1s - loss: 0.1465"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 43/469 [=>............................] - ETA: 1s - loss: 0.1467"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 57/469 [==>...........................] - ETA: 1s - loss: 0.1469"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 71/469 [===>..........................] - ETA: 1s - loss: 0.1470"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 85/469 [====>.........................] - ETA: 1s - loss: 0.1470"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 99/469 [=====>........................] - ETA: 1s - loss: 0.1469"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "113/469 [======>.......................] - ETA: 1s - loss: 0.1469"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "127/469 [=======>......................] - ETA: 1s - loss: 0.1469"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "141/469 [========>.....................] - ETA: 1s - loss: 0.1468"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "155/469 [========>.....................] - ETA: 1s - loss: 0.1468"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "169/469 [=========>....................] - ETA: 1s - loss: 0.1468"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "183/469 [==========>...................] - ETA: 1s - loss: 0.1468"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "197/469 [===========>..................] - ETA: 1s - loss: 0.1468"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "211/469 [============>.................] - ETA: 0s - loss: 0.1468"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "225/469 [=============>................] - ETA: 0s - loss: 0.1468"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "239/469 [==============>...............] - ETA: 0s - loss: 0.1468"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "253/469 [===============>..............] - ETA: 0s - loss: 0.1469"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "267/469 [================>.............] - ETA: 0s - loss: 0.1469"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "281/469 [================>.............] - ETA: 0s - loss: 0.1469"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "295/469 [=================>............] - ETA: 0s - loss: 0.1469"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "309/469 [==================>...........] - ETA: 0s - loss: 0.1469"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "323/469 [===================>..........] - ETA: 0s - loss: 0.1469"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "337/469 [====================>.........] - ETA: 0s - loss: 0.1469"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "351/469 [=====================>........] - ETA: 0s - loss: 0.1470"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "365/469 [======================>.......] - ETA: 0s - loss: 0.1470"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "379/469 [=======================>......] - ETA: 0s - loss: 0.1470"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "393/469 [========================>.....] - ETA: 0s - loss: 0.1470"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "407/469 [=========================>....] - ETA: 0s - loss: 0.1470"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "421/469 [=========================>....] - ETA: 0s - loss: 0.1470"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "435/469 [==========================>...] - ETA: 0s - loss: 0.1470"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "449/469 [===========================>..] - ETA: 0s - loss: 0.1470"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "463/469 [============================>.] - ETA: 0s - loss: 0.1470"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "469/469 [==============================] - 2s 4ms/step - loss: 0.1470\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- " (saved) "
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Epoch 11/30\n",
- "\r",
- " 1/469 [..............................] - ETA: 1s - loss: 0.1478"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 15/469 [..............................] - ETA: 1s - loss: 0.1477"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 29/469 [>.............................] - ETA: 1s - loss: 0.1472"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 43/469 [=>............................] - ETA: 1s - loss: 0.1471"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 57/469 [==>...........................] - ETA: 1s - loss: 0.1469"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 71/469 [===>..........................] - ETA: 1s - loss: 0.1469"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 85/469 [====>.........................] - ETA: 1s - loss: 0.1468"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 98/469 [=====>........................] - ETA: 1s - loss: 0.1468"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "111/469 [======>.......................] - ETA: 1s - loss: 0.1468"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "124/469 [======>.......................] - ETA: 1s - loss: 0.1467"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "137/469 [=======>......................] - ETA: 1s - loss: 0.1467"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "150/469 [========>.....................] - ETA: 1s - loss: 0.1467"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "163/469 [=========>....................] - ETA: 1s - loss: 0.1467"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "176/469 [==========>...................] - ETA: 1s - loss: 0.1466"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "189/469 [===========>..................] - ETA: 1s - loss: 0.1466"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "202/469 [===========>..................] - ETA: 1s - loss: 0.1466"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "215/469 [============>.................] - ETA: 0s - loss: 0.1466"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "228/469 [=============>................] - ETA: 0s - loss: 0.1466"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "241/469 [==============>...............] - ETA: 0s - loss: 0.1465"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "254/469 [===============>..............] - ETA: 0s - loss: 0.1465"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "267/469 [================>.............] - ETA: 0s - loss: 0.1465"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "280/469 [================>.............] - ETA: 0s - loss: 0.1465"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "293/469 [=================>............] - ETA: 0s - loss: 0.1465"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "306/469 [==================>...........] - ETA: 0s - loss: 0.1465"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "319/469 [===================>..........] - ETA: 0s - loss: 0.1465"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "332/469 [====================>.........] - ETA: 0s - loss: 0.1465"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "345/469 [=====================>........] - ETA: 0s - loss: 0.1465"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "358/469 [=====================>........] - ETA: 0s - loss: 0.1465"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "371/469 [======================>.......] - ETA: 0s - loss: 0.1465"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "384/469 [=======================>......] - ETA: 0s - loss: 0.1465"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "397/469 [========================>.....] - ETA: 0s - loss: 0.1465"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "410/469 [=========================>....] - ETA: 0s - loss: 0.1465"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "423/469 [==========================>...] - ETA: 0s - loss: 0.1465"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "436/469 [==========================>...] - ETA: 0s - loss: 0.1465"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "449/469 [===========================>..] - ETA: 0s - loss: 0.1465"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "462/469 [============================>.] - ETA: 0s - loss: 0.1465"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "469/469 [==============================] - 2s 4ms/step - loss: 0.1465\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- " (saved) "
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Epoch 12/30\n",
- "\r",
- " 1/469 [..............................] - ETA: 2s - loss: 0.1400"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 14/469 [..............................] - ETA: 1s - loss: 0.1431"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 27/469 [>.............................] - ETA: 1s - loss: 0.1435"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 40/469 [=>............................] - ETA: 1s - loss: 0.1440"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 53/469 [==>...........................] - ETA: 1s - loss: 0.1444"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 66/469 [===>..........................] - ETA: 1s - loss: 0.1446"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 79/469 [====>.........................] - ETA: 1s - loss: 0.1448"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 92/469 [====>.........................] - ETA: 1s - loss: 0.1449"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "105/469 [=====>........................] - ETA: 1s - loss: 0.1450"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "118/469 [======>.......................] - ETA: 1s - loss: 0.1451"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "131/469 [=======>......................] - ETA: 1s - loss: 0.1452"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "144/469 [========>.....................] - ETA: 1s - loss: 0.1453"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "157/469 [=========>....................] - ETA: 1s - loss: 0.1453"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "170/469 [=========>....................] - ETA: 1s - loss: 0.1454"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "183/469 [==========>...................] - ETA: 1s - loss: 0.1454"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "196/469 [===========>..................] - ETA: 1s - loss: 0.1454"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "209/469 [============>.................] - ETA: 1s - loss: 0.1455"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "222/469 [=============>................] - ETA: 0s - loss: 0.1455"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "235/469 [==============>...............] - ETA: 0s - loss: 0.1455"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "248/469 [==============>...............] - ETA: 0s - loss: 0.1455"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "261/469 [===============>..............] - ETA: 0s - loss: 0.1455"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "274/469 [================>.............] - ETA: 0s - loss: 0.1455"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "287/469 [=================>............] - ETA: 0s - loss: 0.1456"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "300/469 [==================>...........] - ETA: 0s - loss: 0.1456"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "313/469 [===================>..........] - ETA: 0s - loss: 0.1456"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "326/469 [===================>..........] - ETA: 0s - loss: 0.1456"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "339/469 [====================>.........] - ETA: 0s - loss: 0.1456"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "352/469 [=====================>........] - ETA: 0s - loss: 0.1456"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "365/469 [======================>.......] - ETA: 0s - loss: 0.1456"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "378/469 [=======================>......] - ETA: 0s - loss: 0.1457"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "391/469 [========================>.....] - ETA: 0s - loss: 0.1457"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "404/469 [========================>.....] - ETA: 0s - loss: 0.1457"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "417/469 [=========================>....] - ETA: 0s - loss: 0.1457"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "430/469 [==========================>...] - ETA: 0s - loss: 0.1457"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "443/469 [===========================>..] - ETA: 0s - loss: 0.1457"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "456/469 [============================>.] - ETA: 0s - loss: 0.1457"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "469/469 [==============================] - ETA: 0s - loss: 0.1457"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "469/469 [==============================] - 2s 4ms/step - loss: 0.1457\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- " (saved) "
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Epoch 13/30\n",
- "\r",
- " 1/469 [..............................] - ETA: 2s - loss: 0.1427"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 14/469 [..............................] - ETA: 1s - loss: 0.1447"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 27/469 [>.............................] - ETA: 1s - loss: 0.1447"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 40/469 [=>............................] - ETA: 1s - loss: 0.1446"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 53/469 [==>...........................] - ETA: 1s - loss: 0.1447"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 66/469 [===>..........................] - ETA: 1s - loss: 0.1447"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 79/469 [====>.........................] - ETA: 1s - loss: 0.1447"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 92/469 [====>.........................] - ETA: 1s - loss: 0.1447"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "105/469 [=====>........................] - ETA: 1s - loss: 0.1447"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "118/469 [======>.......................] - ETA: 1s - loss: 0.1446"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "131/469 [=======>......................] - ETA: 1s - loss: 0.1446"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "144/469 [========>.....................] - ETA: 1s - loss: 0.1446"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "157/469 [=========>....................] - ETA: 1s - loss: 0.1446"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "170/469 [=========>....................] - ETA: 1s - loss: 0.1446"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "183/469 [==========>...................] - ETA: 1s - loss: 0.1447"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "196/469 [===========>..................] - ETA: 1s - loss: 0.1447"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "209/469 [============>.................] - ETA: 1s - loss: 0.1447"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "222/469 [=============>................] - ETA: 0s - loss: 0.1447"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "235/469 [==============>...............] - ETA: 0s - loss: 0.1447"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "248/469 [==============>...............] - ETA: 0s - loss: 0.1447"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "261/469 [===============>..............] - ETA: 0s - loss: 0.1447"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "274/469 [================>.............] - ETA: 0s - loss: 0.1447"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "287/469 [=================>............] - ETA: 0s - loss: 0.1448"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "300/469 [==================>...........] - ETA: 0s - loss: 0.1448"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "313/469 [===================>..........] - ETA: 0s - loss: 0.1448"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "326/469 [===================>..........] - ETA: 0s - loss: 0.1448"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "339/469 [====================>.........] - ETA: 0s - loss: 0.1448"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "352/469 [=====================>........] - ETA: 0s - loss: 0.1448"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "365/469 [======================>.......] - ETA: 0s - loss: 0.1448"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "378/469 [=======================>......] - ETA: 0s - loss: 0.1448"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "391/469 [========================>.....] - ETA: 0s - loss: 0.1448"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "404/469 [========================>.....] - ETA: 0s - loss: 0.1449"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "417/469 [=========================>....] - ETA: 0s - loss: 0.1449"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "430/469 [==========================>...] - ETA: 0s - loss: 0.1449"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "443/469 [===========================>..] - ETA: 0s - loss: 0.1449"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "456/469 [============================>.] - ETA: 0s - loss: 0.1449"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "469/469 [==============================] - ETA: 0s - loss: 0.1449"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "469/469 [==============================] - 2s 4ms/step - loss: 0.1449\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- " (saved) "
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Epoch 14/30\n",
- "\r",
- " 1/469 [..............................] - ETA: 2s - loss: 0.1461"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 14/469 [..............................] - ETA: 1s - loss: 0.1447"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 27/469 [>.............................] - ETA: 1s - loss: 0.1446"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 40/469 [=>............................] - ETA: 1s - loss: 0.1445"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 53/469 [==>...........................] - ETA: 1s - loss: 0.1444"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 66/469 [===>..........................] - ETA: 1s - loss: 0.1443"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 79/469 [====>.........................] - ETA: 1s - loss: 0.1443"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 92/469 [====>.........................] - ETA: 1s - loss: 0.1442"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "105/469 [=====>........................] - ETA: 1s - loss: 0.1442"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "118/469 [======>.......................] - ETA: 1s - loss: 0.1442"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "131/469 [=======>......................] - ETA: 1s - loss: 0.1442"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "144/469 [========>.....................] - ETA: 1s - loss: 0.1442"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "157/469 [=========>....................] - ETA: 1s - loss: 0.1442"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "170/469 [=========>....................] - ETA: 1s - loss: 0.1442"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "183/469 [==========>...................] - ETA: 1s - loss: 0.1442"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "196/469 [===========>..................] - ETA: 1s - loss: 0.1442"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "209/469 [============>.................] - ETA: 1s - loss: 0.1442"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "222/469 [=============>................] - ETA: 0s - loss: 0.1442"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "235/469 [==============>...............] - ETA: 0s - loss: 0.1442"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "248/469 [==============>...............] - ETA: 0s - loss: 0.1443"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "261/469 [===============>..............] - ETA: 0s - loss: 0.1443"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "274/469 [================>.............] - ETA: 0s - loss: 0.1443"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "287/469 [=================>............] - ETA: 0s - loss: 0.1443"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "300/469 [==================>...........] - ETA: 0s - loss: 0.1443"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "313/469 [===================>..........] - ETA: 0s - loss: 0.1443"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "326/469 [===================>..........] - ETA: 0s - loss: 0.1443"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "339/469 [====================>.........] - ETA: 0s - loss: 0.1443"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "352/469 [=====================>........] - ETA: 0s - loss: 0.1443"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "365/469 [======================>.......] - ETA: 0s - loss: 0.1443"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "378/469 [=======================>......] - ETA: 0s - loss: 0.1443"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "391/469 [========================>.....] - ETA: 0s - loss: 0.1443"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "404/469 [========================>.....] - ETA: 0s - loss: 0.1444"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "417/469 [=========================>....] - ETA: 0s - loss: 0.1444"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "430/469 [==========================>...] - ETA: 0s - loss: 0.1444"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "443/469 [===========================>..] - ETA: 0s - loss: 0.1444"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "456/469 [============================>.] - ETA: 0s - loss: 0.1444"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "469/469 [==============================] - ETA: 0s - loss: 0.1444"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "469/469 [==============================] - 2s 4ms/step - loss: 0.1444\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- " (saved) Epoch 15/30\n",
- "\r",
- " 1/469 [..............................] - ETA: 2s - loss: 0.1392"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 14/469 [..............................] - ETA: 1s - loss: 0.1410"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 27/469 [>.............................] - ETA: 1s - loss: 0.1421"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 40/469 [=>............................] - ETA: 1s - loss: 0.1427"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 53/469 [==>...........................] - ETA: 1s - loss: 0.1429"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 66/469 [===>..........................] - ETA: 1s - loss: 0.1431"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 79/469 [====>.........................] - ETA: 1s - loss: 0.1432"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 92/469 [====>.........................] - ETA: 1s - loss: 0.1433"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "105/469 [=====>........................] - ETA: 1s - loss: 0.1434"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "118/469 [======>.......................] - ETA: 1s - loss: 0.1434"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "131/469 [=======>......................] - ETA: 1s - loss: 0.1435"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "144/469 [========>.....................] - ETA: 1s - loss: 0.1435"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "157/469 [=========>....................] - ETA: 1s - loss: 0.1436"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "170/469 [=========>....................] - ETA: 1s - loss: 0.1436"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "183/469 [==========>...................] - ETA: 1s - loss: 0.1436"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "196/469 [===========>..................] - ETA: 1s - loss: 0.1437"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "209/469 [============>.................] - ETA: 1s - loss: 0.1437"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "222/469 [=============>................] - ETA: 0s - loss: 0.1437"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "235/469 [==============>...............] - ETA: 0s - loss: 0.1438"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "248/469 [==============>...............] - ETA: 0s - loss: 0.1438"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "261/469 [===============>..............] - ETA: 0s - loss: 0.1438"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "274/469 [================>.............] - ETA: 0s - loss: 0.1438"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "287/469 [=================>............] - ETA: 0s - loss: 0.1438"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "300/469 [==================>...........] - ETA: 0s - loss: 0.1439"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "313/469 [===================>..........] - ETA: 0s - loss: 0.1439"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "326/469 [===================>..........] - ETA: 0s - loss: 0.1439"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "339/469 [====================>.........] - ETA: 0s - loss: 0.1439"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "352/469 [=====================>........] - ETA: 0s - loss: 0.1439"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "365/469 [======================>.......] - ETA: 0s - loss: 0.1439"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "378/469 [=======================>......] - ETA: 0s - loss: 0.1439"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "391/469 [========================>.....] - ETA: 0s - loss: 0.1439"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "404/469 [========================>.....] - ETA: 0s - loss: 0.1439"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "417/469 [=========================>....] - ETA: 0s - loss: 0.1439"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "430/469 [==========================>...] - ETA: 0s - loss: 0.1439"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "443/469 [===========================>..] - ETA: 0s - loss: 0.1440"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "456/469 [============================>.] - ETA: 0s - loss: 0.1440"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "469/469 [==============================] - ETA: 0s - loss: 0.1440"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "469/469 [==============================] - 2s 4ms/step - loss: 0.1440\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- " (saved) Epoch 16/30\n",
- "\r",
- " 1/469 [..............................] - ETA: 2s - loss: 0.1450"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 13/469 [..............................] - ETA: 1s - loss: 0.1426"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 26/469 [>.............................] - ETA: 1s - loss: 0.1427"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 39/469 [=>............................] - ETA: 1s - loss: 0.1431"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 52/469 [==>...........................] - ETA: 1s - loss: 0.1433"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 65/469 [===>..........................] - ETA: 1s - loss: 0.1434"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 79/469 [====>.........................] - ETA: 1s - loss: 0.1435"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 93/469 [====>.........................] - ETA: 1s - loss: 0.1436"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "107/469 [=====>........................] - ETA: 1s - loss: 0.1436"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "121/469 [======>.......................] - ETA: 1s - loss: 0.1437"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "135/469 [=======>......................] - ETA: 1s - loss: 0.1437"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "149/469 [========>.....................] - ETA: 1s - loss: 0.1437"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "163/469 [=========>....................] - ETA: 1s - loss: 0.1437"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "177/469 [==========>...................] - ETA: 1s - loss: 0.1437"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "191/469 [===========>..................] - ETA: 1s - loss: 0.1437"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "205/469 [============>.................] - ETA: 1s - loss: 0.1437"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "219/469 [=============>................] - ETA: 0s - loss: 0.1437"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "233/469 [=============>................] - ETA: 0s - loss: 0.1437"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "247/469 [==============>...............] - ETA: 0s - loss: 0.1437"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "261/469 [===============>..............] - ETA: 0s - loss: 0.1437"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "275/469 [================>.............] - ETA: 0s - loss: 0.1437"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "289/469 [=================>............] - ETA: 0s - loss: 0.1437"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "303/469 [==================>...........] - ETA: 0s - loss: 0.1437"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "317/469 [===================>..........] - ETA: 0s - loss: 0.1437"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "331/469 [====================>.........] - ETA: 0s - loss: 0.1437"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "345/469 [=====================>........] - ETA: 0s - loss: 0.1437"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "359/469 [=====================>........] - ETA: 0s - loss: 0.1437"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "373/469 [======================>.......] - ETA: 0s - loss: 0.1437"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "387/469 [=======================>......] - ETA: 0s - loss: 0.1437"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "401/469 [========================>.....] - ETA: 0s - loss: 0.1437"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "415/469 [=========================>....] - ETA: 0s - loss: 0.1437"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "429/469 [==========================>...] - ETA: 0s - loss: 0.1437"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "443/469 [===========================>..] - ETA: 0s - loss: 0.1437"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "457/469 [============================>.] - ETA: 0s - loss: 0.1437"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "469/469 [==============================] - 2s 4ms/step - loss: 0.1437\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- " (saved) Epoch 17/30\n",
- "\r",
- " 1/469 [..............................] - ETA: 1s - loss: 0.1364"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 15/469 [..............................] - ETA: 1s - loss: 0.1426"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 29/469 [>.............................] - ETA: 1s - loss: 0.1427"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 43/469 [=>............................] - ETA: 1s - loss: 0.1427"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 57/469 [==>...........................] - ETA: 1s - loss: 0.1428"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 71/469 [===>..........................] - ETA: 1s - loss: 0.1428"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 85/469 [====>.........................] - ETA: 1s - loss: 0.1429"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 99/469 [=====>........................] - ETA: 1s - loss: 0.1429"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "113/469 [======>.......................] - ETA: 1s - loss: 0.1429"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "127/469 [=======>......................] - ETA: 1s - loss: 0.1429"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "141/469 [========>.....................] - ETA: 1s - loss: 0.1429"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "155/469 [========>.....................] - ETA: 1s - loss: 0.1429"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "169/469 [=========>....................] - ETA: 1s - loss: 0.1429"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "183/469 [==========>...................] - ETA: 1s - loss: 0.1429"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "197/469 [===========>..................] - ETA: 1s - loss: 0.1429"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "211/469 [============>.................] - ETA: 0s - loss: 0.1429"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "225/469 [=============>................] - ETA: 0s - loss: 0.1429"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "239/469 [==============>...............] - ETA: 0s - loss: 0.1430"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "253/469 [===============>..............] - ETA: 0s - loss: 0.1430"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "267/469 [================>.............] - ETA: 0s - loss: 0.1430"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "281/469 [================>.............] - ETA: 0s - loss: 0.1430"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "295/469 [=================>............] - ETA: 0s - loss: 0.1430"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "309/469 [==================>...........] - ETA: 0s - loss: 0.1430"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "323/469 [===================>..........] - ETA: 0s - loss: 0.1430"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "337/469 [====================>.........] - ETA: 0s - loss: 0.1430"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "351/469 [=====================>........] - ETA: 0s - loss: 0.1431"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "365/469 [======================>.......] - ETA: 0s - loss: 0.1431"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "379/469 [=======================>......] - ETA: 0s - loss: 0.1431"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "393/469 [========================>.....] - ETA: 0s - loss: 0.1431"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "407/469 [=========================>....] - ETA: 0s - loss: 0.1431"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "421/469 [=========================>....] - ETA: 0s - loss: 0.1431"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "435/469 [==========================>...] - ETA: 0s - loss: 0.1431"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "449/469 [===========================>..] - ETA: 0s - loss: 0.1431"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "463/469 [============================>.] - ETA: 0s - loss: 0.1431"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "469/469 [==============================] - 2s 4ms/step - loss: 0.1431\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- " (saved) Epoch 18/30\n",
- "\r",
- " 1/469 [..............................] - ETA: 1s - loss: 0.1416"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 15/469 [..............................] - ETA: 1s - loss: 0.1431"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 29/469 [>.............................] - ETA: 1s - loss: 0.1428"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 43/469 [=>............................] - ETA: 1s - loss: 0.1427"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 57/469 [==>...........................] - ETA: 1s - loss: 0.1427"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 71/469 [===>..........................] - ETA: 1s - loss: 0.1427"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 85/469 [====>.........................] - ETA: 1s - loss: 0.1427"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 99/469 [=====>........................] - ETA: 1s - loss: 0.1426"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "113/469 [======>.......................] - ETA: 1s - loss: 0.1426"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "127/469 [=======>......................] - ETA: 1s - loss: 0.1426"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "141/469 [========>.....................] - ETA: 1s - loss: 0.1426"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "155/469 [========>.....................] - ETA: 1s - loss: 0.1426"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "169/469 [=========>....................] - ETA: 1s - loss: 0.1425"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "183/469 [==========>...................] - ETA: 1s - loss: 0.1425"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "197/469 [===========>..................] - ETA: 1s - loss: 0.1425"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "211/469 [============>.................] - ETA: 0s - loss: 0.1425"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "225/469 [=============>................] - ETA: 0s - loss: 0.1425"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "239/469 [==============>...............] - ETA: 0s - loss: 0.1425"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "253/469 [===============>..............] - ETA: 0s - loss: 0.1425"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "267/469 [================>.............] - ETA: 0s - loss: 0.1425"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "281/469 [================>.............] - ETA: 0s - loss: 0.1425"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "295/469 [=================>............] - ETA: 0s - loss: 0.1425"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "309/469 [==================>...........] - ETA: 0s - loss: 0.1426"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "323/469 [===================>..........] - ETA: 0s - loss: 0.1426"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "337/469 [====================>.........] - ETA: 0s - loss: 0.1426"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "351/469 [=====================>........] - ETA: 0s - loss: 0.1426"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "365/469 [======================>.......] - ETA: 0s - loss: 0.1426"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "379/469 [=======================>......] - ETA: 0s - loss: 0.1426"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "393/469 [========================>.....] - ETA: 0s - loss: 0.1426"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "407/469 [=========================>....] - ETA: 0s - loss: 0.1426"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "421/469 [=========================>....] - ETA: 0s - loss: 0.1426"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "435/469 [==========================>...] - ETA: 0s - loss: 0.1426"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "449/469 [===========================>..] - ETA: 0s - loss: 0.1427"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "463/469 [============================>.] - ETA: 0s - loss: 0.1427"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "469/469 [==============================] - 2s 4ms/step - loss: 0.1427\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- " (saved) "
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Epoch 19/30\n",
- "\r",
- " 1/469 [..............................] - ETA: 1s - loss: 0.1371"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 15/469 [..............................] - ETA: 1s - loss: 0.1409"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 29/469 [>.............................] - ETA: 1s - loss: 0.1415"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 43/469 [=>............................] - ETA: 1s - loss: 0.1418"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 57/469 [==>...........................] - ETA: 1s - loss: 0.1419"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 71/469 [===>..........................] - ETA: 1s - loss: 0.1419"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 85/469 [====>.........................] - ETA: 1s - loss: 0.1419"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 99/469 [=====>........................] - ETA: 1s - loss: 0.1419"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "113/469 [======>.......................] - ETA: 1s - loss: 0.1419"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "127/469 [=======>......................] - ETA: 1s - loss: 0.1420"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "141/469 [========>.....................] - ETA: 1s - loss: 0.1420"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "155/469 [========>.....................] - ETA: 1s - loss: 0.1420"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "169/469 [=========>....................] - ETA: 1s - loss: 0.1421"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "183/469 [==========>...................] - ETA: 1s - loss: 0.1421"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "197/469 [===========>..................] - ETA: 1s - loss: 0.1421"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "211/469 [============>.................] - ETA: 0s - loss: 0.1421"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "225/469 [=============>................] - ETA: 0s - loss: 0.1421"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "239/469 [==============>...............] - ETA: 0s - loss: 0.1422"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "253/469 [===============>..............] - ETA: 0s - loss: 0.1422"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "267/469 [================>.............] - ETA: 0s - loss: 0.1422"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "281/469 [================>.............] - ETA: 0s - loss: 0.1422"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "295/469 [=================>............] - ETA: 0s - loss: 0.1422"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "309/469 [==================>...........] - ETA: 0s - loss: 0.1422"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "323/469 [===================>..........] - ETA: 0s - loss: 0.1423"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "337/469 [====================>.........] - ETA: 0s - loss: 0.1423"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "351/469 [=====================>........] - ETA: 0s - loss: 0.1423"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "365/469 [======================>.......] - ETA: 0s - loss: 0.1423"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "379/469 [=======================>......] - ETA: 0s - loss: 0.1423"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "393/469 [========================>.....] - ETA: 0s - loss: 0.1423"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "407/469 [=========================>....] - ETA: 0s - loss: 0.1423"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "421/469 [=========================>....] - ETA: 0s - loss: 0.1423"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "435/469 [==========================>...] - ETA: 0s - loss: 0.1423"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "449/469 [===========================>..] - ETA: 0s - loss: 0.1424"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "463/469 [============================>.] - ETA: 0s - loss: 0.1424"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "469/469 [==============================] - 2s 4ms/step - loss: 0.1424\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- " (saved) Epoch 20/30\n",
- "\r",
- " 1/469 [..............................] - ETA: 1s - loss: 0.1399"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 15/469 [..............................] - ETA: 1s - loss: 0.1417"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 29/469 [>.............................] - ETA: 1s - loss: 0.1420"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 43/469 [=>............................] - ETA: 1s - loss: 0.1419"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 57/469 [==>...........................] - ETA: 1s - loss: 0.1419"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 71/469 [===>..........................] - ETA: 1s - loss: 0.1418"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 85/469 [====>.........................] - ETA: 1s - loss: 0.1418"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 99/469 [=====>........................] - ETA: 1s - loss: 0.1418"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "113/469 [======>.......................] - ETA: 1s - loss: 0.1419"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "127/469 [=======>......................] - ETA: 1s - loss: 0.1419"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "141/469 [========>.....................] - ETA: 1s - loss: 0.1419"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "155/469 [========>.....................] - ETA: 1s - loss: 0.1419"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "169/469 [=========>....................] - ETA: 1s - loss: 0.1419"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "183/469 [==========>...................] - ETA: 1s - loss: 0.1419"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "197/469 [===========>..................] - ETA: 1s - loss: 0.1419"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "211/469 [============>.................] - ETA: 0s - loss: 0.1419"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "225/469 [=============>................] - ETA: 0s - loss: 0.1419"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "239/469 [==============>...............] - ETA: 0s - loss: 0.1419"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "253/469 [===============>..............] - ETA: 0s - loss: 0.1419"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "267/469 [================>.............] - ETA: 0s - loss: 0.1420"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "281/469 [================>.............] - ETA: 0s - loss: 0.1420"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "295/469 [=================>............] - ETA: 0s - loss: 0.1420"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "309/469 [==================>...........] - ETA: 0s - loss: 0.1420"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "323/469 [===================>..........] - ETA: 0s - loss: 0.1420"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "337/469 [====================>.........] - ETA: 0s - loss: 0.1420"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "351/469 [=====================>........] - ETA: 0s - loss: 0.1420"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "365/469 [======================>.......] - ETA: 0s - loss: 0.1420"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "379/469 [=======================>......] - ETA: 0s - loss: 0.1420"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "393/469 [========================>.....] - ETA: 0s - loss: 0.1420"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "407/469 [=========================>....] - ETA: 0s - loss: 0.1420"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "421/469 [=========================>....] - ETA: 0s - loss: 0.1420"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "435/469 [==========================>...] - ETA: 0s - loss: 0.1420"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "449/469 [===========================>..] - ETA: 0s - loss: 0.1421"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "463/469 [============================>.] - ETA: 0s - loss: 0.1421"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "469/469 [==============================] - 2s 4ms/step - loss: 0.1421\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- " (saved) "
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Epoch 21/30\n",
- "\r",
- " 1/469 [..............................] - ETA: 2s - loss: 0.1399"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 15/469 [..............................] - ETA: 1s - loss: 0.1410"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 29/469 [>.............................] - ETA: 1s - loss: 0.1407"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 43/469 [=>............................] - ETA: 1s - loss: 0.1406"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 57/469 [==>...........................] - ETA: 1s - loss: 0.1407"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 71/469 [===>..........................] - ETA: 1s - loss: 0.1408"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 85/469 [====>.........................] - ETA: 1s - loss: 0.1409"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 99/469 [=====>........................] - ETA: 1s - loss: 0.1411"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "113/469 [======>.......................] - ETA: 1s - loss: 0.1412"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "127/469 [=======>......................] - ETA: 1s - loss: 0.1412"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "141/469 [========>.....................] - ETA: 1s - loss: 0.1413"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "155/469 [========>.....................] - ETA: 1s - loss: 0.1413"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "169/469 [=========>....................] - ETA: 1s - loss: 0.1414"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "183/469 [==========>...................] - ETA: 1s - loss: 0.1414"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "197/469 [===========>..................] - ETA: 0s - loss: 0.1415"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "211/469 [============>.................] - ETA: 0s - loss: 0.1415"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "225/469 [=============>................] - ETA: 0s - loss: 0.1415"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "239/469 [==============>...............] - ETA: 0s - loss: 0.1415"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "253/469 [===============>..............] - ETA: 0s - loss: 0.1415"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "267/469 [================>.............] - ETA: 0s - loss: 0.1416"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "281/469 [================>.............] - ETA: 0s - loss: 0.1416"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "295/469 [=================>............] - ETA: 0s - loss: 0.1416"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "309/469 [==================>...........] - ETA: 0s - loss: 0.1416"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "323/469 [===================>..........] - ETA: 0s - loss: 0.1416"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "337/469 [====================>.........] - ETA: 0s - loss: 0.1416"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "351/469 [=====================>........] - ETA: 0s - loss: 0.1416"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "365/469 [======================>.......] - ETA: 0s - loss: 0.1417"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "379/469 [=======================>......] - ETA: 0s - loss: 0.1417"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "393/469 [========================>.....] - ETA: 0s - loss: 0.1417"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "407/469 [=========================>....] - ETA: 0s - loss: 0.1417"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "421/469 [=========================>....] - ETA: 0s - loss: 0.1417"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "435/469 [==========================>...] - ETA: 0s - loss: 0.1417"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "449/469 [===========================>..] - ETA: 0s - loss: 0.1417"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "463/469 [============================>.] - ETA: 0s - loss: 0.1417"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "469/469 [==============================] - 2s 4ms/step - loss: 0.1417\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- " (saved) "
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Epoch 22/30\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\r",
- " 1/469 [..............................] - ETA: 1s - loss: 0.1359"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 15/469 [..............................] - ETA: 1s - loss: 0.1393"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 29/469 [>.............................] - ETA: 1s - loss: 0.1399"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 43/469 [=>............................] - ETA: 1s - loss: 0.1403"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 57/469 [==>...........................] - ETA: 1s - loss: 0.1407"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 71/469 [===>..........................] - ETA: 1s - loss: 0.1408"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 85/469 [====>.........................] - ETA: 1s - loss: 0.1409"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 99/469 [=====>........................] - ETA: 1s - loss: 0.1409"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "113/469 [======>.......................] - ETA: 1s - loss: 0.1409"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "127/469 [=======>......................] - ETA: 1s - loss: 0.1410"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "141/469 [========>.....................] - ETA: 1s - loss: 0.1410"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "155/469 [========>.....................] - ETA: 1s - loss: 0.1410"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "169/469 [=========>....................] - ETA: 1s - loss: 0.1411"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "183/469 [==========>...................] - ETA: 1s - loss: 0.1411"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "197/469 [===========>..................] - ETA: 0s - loss: 0.1412"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "211/469 [============>.................] - ETA: 0s - loss: 0.1412"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "225/469 [=============>................] - ETA: 0s - loss: 0.1412"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "239/469 [==============>...............] - ETA: 0s - loss: 0.1412"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "253/469 [===============>..............] - ETA: 0s - loss: 0.1413"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "267/469 [================>.............] - ETA: 0s - loss: 0.1413"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "281/469 [================>.............] - ETA: 0s - loss: 0.1413"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "295/469 [=================>............] - ETA: 0s - loss: 0.1413"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "309/469 [==================>...........] - ETA: 0s - loss: 0.1414"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "323/469 [===================>..........] - ETA: 0s - loss: 0.1414"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "337/469 [====================>.........] - ETA: 0s - loss: 0.1414"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "351/469 [=====================>........] - ETA: 0s - loss: 0.1414"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "365/469 [======================>.......] - ETA: 0s - loss: 0.1414"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "379/469 [=======================>......] - ETA: 0s - loss: 0.1414"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "393/469 [========================>.....] - ETA: 0s - loss: 0.1414"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "407/469 [=========================>....] - ETA: 0s - loss: 0.1415"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "421/469 [=========================>....] - ETA: 0s - loss: 0.1415"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "435/469 [==========================>...] - ETA: 0s - loss: 0.1415"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "449/469 [===========================>..] - ETA: 0s - loss: 0.1415"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "463/469 [============================>.] - ETA: 0s - loss: 0.1415"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "469/469 [==============================] - 2s 4ms/step - loss: 0.1415\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- " (saved) "
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Epoch 23/30\n",
- "\r",
- " 1/469 [..............................] - ETA: 1s - loss: 0.1422"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 15/469 [..............................] - ETA: 1s - loss: 0.1428"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 29/469 [>.............................] - ETA: 1s - loss: 0.1426"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 43/469 [=>............................] - ETA: 1s - loss: 0.1425"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 57/469 [==>...........................] - ETA: 1s - loss: 0.1424"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 71/469 [===>..........................] - ETA: 1s - loss: 0.1423"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 85/469 [====>.........................] - ETA: 1s - loss: 0.1423"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 99/469 [=====>........................] - ETA: 1s - loss: 0.1422"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "113/469 [======>.......................] - ETA: 1s - loss: 0.1421"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "127/469 [=======>......................] - ETA: 1s - loss: 0.1420"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "141/469 [========>.....................] - ETA: 1s - loss: 0.1420"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "155/469 [========>.....................] - ETA: 1s - loss: 0.1419"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "169/469 [=========>....................] - ETA: 1s - loss: 0.1419"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "183/469 [==========>...................] - ETA: 1s - loss: 0.1419"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "197/469 [===========>..................] - ETA: 1s - loss: 0.1418"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "211/469 [============>.................] - ETA: 0s - loss: 0.1418"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "225/469 [=============>................] - ETA: 0s - loss: 0.1418"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "239/469 [==============>...............] - ETA: 0s - loss: 0.1418"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "253/469 [===============>..............] - ETA: 0s - loss: 0.1417"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "267/469 [================>.............] - ETA: 0s - loss: 0.1417"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "281/469 [================>.............] - ETA: 0s - loss: 0.1417"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "295/469 [=================>............] - ETA: 0s - loss: 0.1417"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "309/469 [==================>...........] - ETA: 0s - loss: 0.1417"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "323/469 [===================>..........] - ETA: 0s - loss: 0.1417"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "337/469 [====================>.........] - ETA: 0s - loss: 0.1417"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "351/469 [=====================>........] - ETA: 0s - loss: 0.1417"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "365/469 [======================>.......] - ETA: 0s - loss: 0.1417"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "379/469 [=======================>......] - ETA: 0s - loss: 0.1417"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "393/469 [========================>.....] - ETA: 0s - loss: 0.1417"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "407/469 [=========================>....] - ETA: 0s - loss: 0.1416"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "421/469 [=========================>....] - ETA: 0s - loss: 0.1416"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "435/469 [==========================>...] - ETA: 0s - loss: 0.1416"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "449/469 [===========================>..] - ETA: 0s - loss: 0.1416"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "463/469 [============================>.] - ETA: 0s - loss: 0.1416"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "469/469 [==============================] - 2s 4ms/step - loss: 0.1416\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- " (saved) Epoch 24/30\n",
- "\r",
- " 1/469 [..............................] - ETA: 1s - loss: 0.1476"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 15/469 [..............................] - ETA: 1s - loss: 0.1422"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 29/469 [>.............................] - ETA: 1s - loss: 0.1414"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 43/469 [=>............................] - ETA: 1s - loss: 0.1412"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 57/469 [==>...........................] - ETA: 1s - loss: 0.1411"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 71/469 [===>..........................] - ETA: 1s - loss: 0.1411"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 85/469 [====>.........................] - ETA: 1s - loss: 0.1411"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 99/469 [=====>........................] - ETA: 1s - loss: 0.1411"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "113/469 [======>.......................] - ETA: 1s - loss: 0.1411"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "127/469 [=======>......................] - ETA: 1s - loss: 0.1410"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "141/469 [========>.....................] - ETA: 1s - loss: 0.1410"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "155/469 [========>.....................] - ETA: 1s - loss: 0.1410"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "169/469 [=========>....................] - ETA: 1s - loss: 0.1410"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "183/469 [==========>...................] - ETA: 1s - loss: 0.1410"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "197/469 [===========>..................] - ETA: 1s - loss: 0.1410"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "211/469 [============>.................] - ETA: 0s - loss: 0.1410"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "225/469 [=============>................] - ETA: 0s - loss: 0.1410"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "239/469 [==============>...............] - ETA: 0s - loss: 0.1410"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "253/469 [===============>..............] - ETA: 0s - loss: 0.1410"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "267/469 [================>.............] - ETA: 0s - loss: 0.1410"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "281/469 [================>.............] - ETA: 0s - loss: 0.1410"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "295/469 [=================>............] - ETA: 0s - loss: 0.1410"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "309/469 [==================>...........] - ETA: 0s - loss: 0.1410"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "323/469 [===================>..........] - ETA: 0s - loss: 0.1410"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "337/469 [====================>.........] - ETA: 0s - loss: 0.1410"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "351/469 [=====================>........] - ETA: 0s - loss: 0.1411"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "365/469 [======================>.......] - ETA: 0s - loss: 0.1411"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "379/469 [=======================>......] - ETA: 0s - loss: 0.1411"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "393/469 [========================>.....] - ETA: 0s - loss: 0.1411"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "407/469 [=========================>....] - ETA: 0s - loss: 0.1411"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "421/469 [=========================>....] - ETA: 0s - loss: 0.1411"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "435/469 [==========================>...] - ETA: 0s - loss: 0.1411"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "449/469 [===========================>..] - ETA: 0s - loss: 0.1411"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "463/469 [============================>.] - ETA: 0s - loss: 0.1411"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "469/469 [==============================] - 2s 4ms/step - loss: 0.1411\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- " (saved) Epoch 25/30\n",
- "\r",
- " 1/469 [..............................] - ETA: 1s - loss: 0.1347"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 15/469 [..............................] - ETA: 1s - loss: 0.1409"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 29/469 [>.............................] - ETA: 1s - loss: 0.1409"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 43/469 [=>............................] - ETA: 1s - loss: 0.1406"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 57/469 [==>...........................] - ETA: 1s - loss: 0.1404"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 71/469 [===>..........................] - ETA: 1s - loss: 0.1404"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 85/469 [====>.........................] - ETA: 1s - loss: 0.1404"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 99/469 [=====>........................] - ETA: 1s - loss: 0.1404"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "113/469 [======>.......................] - ETA: 1s - loss: 0.1404"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "127/469 [=======>......................] - ETA: 1s - loss: 0.1404"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "141/469 [========>.....................] - ETA: 1s - loss: 0.1404"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "155/469 [========>.....................] - ETA: 1s - loss: 0.1404"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "169/469 [=========>....................] - ETA: 1s - loss: 0.1403"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "183/469 [==========>...................] - ETA: 1s - loss: 0.1403"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "197/469 [===========>..................] - ETA: 1s - loss: 0.1403"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "211/469 [============>.................] - ETA: 0s - loss: 0.1403"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "225/469 [=============>................] - ETA: 0s - loss: 0.1403"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "239/469 [==============>...............] - ETA: 0s - loss: 0.1404"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "253/469 [===============>..............] - ETA: 0s - loss: 0.1404"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "267/469 [================>.............] - ETA: 0s - loss: 0.1404"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "281/469 [================>.............] - ETA: 0s - loss: 0.1404"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "295/469 [=================>............] - ETA: 0s - loss: 0.1404"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "309/469 [==================>...........] - ETA: 0s - loss: 0.1405"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "323/469 [===================>..........] - ETA: 0s - loss: 0.1405"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "337/469 [====================>.........] - ETA: 0s - loss: 0.1405"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "351/469 [=====================>........] - ETA: 0s - loss: 0.1405"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "365/469 [======================>.......] - ETA: 0s - loss: 0.1405"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "379/469 [=======================>......] - ETA: 0s - loss: 0.1406"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "393/469 [========================>.....] - ETA: 0s - loss: 0.1406"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "407/469 [=========================>....] - ETA: 0s - loss: 0.1406"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "421/469 [=========================>....] - ETA: 0s - loss: 0.1406"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "435/469 [==========================>...] - ETA: 0s - loss: 0.1406"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "449/469 [===========================>..] - ETA: 0s - loss: 0.1406"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "463/469 [============================>.] - ETA: 0s - loss: 0.1407"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "469/469 [==============================] - 2s 4ms/step - loss: 0.1407\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- " (saved) Epoch 26/30\n",
- "\r",
- " 1/469 [..............................] - ETA: 1s - loss: 0.1416"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 15/469 [..............................] - ETA: 1s - loss: 0.1398"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 29/469 [>.............................] - ETA: 1s - loss: 0.1396"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 43/469 [=>............................] - ETA: 1s - loss: 0.1396"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 57/469 [==>...........................] - ETA: 1s - loss: 0.1396"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 71/469 [===>..........................] - ETA: 1s - loss: 0.1396"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 85/469 [====>.........................] - ETA: 1s - loss: 0.1397"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 99/469 [=====>........................] - ETA: 1s - loss: 0.1397"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "113/469 [======>.......................] - ETA: 1s - loss: 0.1397"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "127/469 [=======>......................] - ETA: 1s - loss: 0.1397"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "141/469 [========>.....................] - ETA: 1s - loss: 0.1397"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "155/469 [========>.....................] - ETA: 1s - loss: 0.1397"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "169/469 [=========>....................] - ETA: 1s - loss: 0.1397"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "183/469 [==========>...................] - ETA: 1s - loss: 0.1398"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "197/469 [===========>..................] - ETA: 1s - loss: 0.1398"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "211/469 [============>.................] - ETA: 0s - loss: 0.1398"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "225/469 [=============>................] - ETA: 0s - loss: 0.1398"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "239/469 [==============>...............] - ETA: 0s - loss: 0.1398"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "253/469 [===============>..............] - ETA: 0s - loss: 0.1398"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "267/469 [================>.............] - ETA: 0s - loss: 0.1398"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "281/469 [================>.............] - ETA: 0s - loss: 0.1399"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "295/469 [=================>............] - ETA: 0s - loss: 0.1399"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "309/469 [==================>...........] - ETA: 0s - loss: 0.1399"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "323/469 [===================>..........] - ETA: 0s - loss: 0.1399"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "337/469 [====================>.........] - ETA: 0s - loss: 0.1399"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "351/469 [=====================>........] - ETA: 0s - loss: 0.1399"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "365/469 [======================>.......] - ETA: 0s - loss: 0.1399"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "379/469 [=======================>......] - ETA: 0s - loss: 0.1400"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "393/469 [========================>.....] - ETA: 0s - loss: 0.1400"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "407/469 [=========================>....] - ETA: 0s - loss: 0.1400"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "421/469 [=========================>....] - ETA: 0s - loss: 0.1400"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "435/469 [==========================>...] - ETA: 0s - loss: 0.1400"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "449/469 [===========================>..] - ETA: 0s - loss: 0.1401"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "463/469 [============================>.] - ETA: 0s - loss: 0.1401"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "469/469 [==============================] - 2s 4ms/step - loss: 0.1401\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- " (saved) Epoch 27/30\n",
- "\r",
- " 1/469 [..............................] - ETA: 1s - loss: 0.1414"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 15/469 [..............................] - ETA: 1s - loss: 0.1416"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 29/469 [>.............................] - ETA: 1s - loss: 0.1408"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 43/469 [=>............................] - ETA: 1s - loss: 0.1405"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 57/469 [==>...........................] - ETA: 1s - loss: 0.1405"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 71/469 [===>..........................] - ETA: 1s - loss: 0.1404"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 85/469 [====>.........................] - ETA: 1s - loss: 0.1404"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 99/469 [=====>........................] - ETA: 1s - loss: 0.1403"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "113/469 [======>.......................] - ETA: 1s - loss: 0.1403"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "127/469 [=======>......................] - ETA: 1s - loss: 0.1402"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "141/469 [========>.....................] - ETA: 1s - loss: 0.1402"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "155/469 [========>.....................] - ETA: 1s - loss: 0.1402"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "169/469 [=========>....................] - ETA: 1s - loss: 0.1402"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "183/469 [==========>...................] - ETA: 1s - loss: 0.1401"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "197/469 [===========>..................] - ETA: 1s - loss: 0.1401"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "211/469 [============>.................] - ETA: 0s - loss: 0.1401"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "225/469 [=============>................] - ETA: 0s - loss: 0.1401"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "239/469 [==============>...............] - ETA: 0s - loss: 0.1401"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "253/469 [===============>..............] - ETA: 0s - loss: 0.1401"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "267/469 [================>.............] - ETA: 0s - loss: 0.1401"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "281/469 [================>.............] - ETA: 0s - loss: 0.1402"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "295/469 [=================>............] - ETA: 0s - loss: 0.1402"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "309/469 [==================>...........] - ETA: 0s - loss: 0.1402"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "323/469 [===================>..........] - ETA: 0s - loss: 0.1402"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "337/469 [====================>.........] - ETA: 0s - loss: 0.1402"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "351/469 [=====================>........] - ETA: 0s - loss: 0.1402"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "365/469 [======================>.......] - ETA: 0s - loss: 0.1402"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "379/469 [=======================>......] - ETA: 0s - loss: 0.1403"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "393/469 [========================>.....] - ETA: 0s - loss: 0.1403"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "407/469 [=========================>....] - ETA: 0s - loss: 0.1403"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "421/469 [=========================>....] - ETA: 0s - loss: 0.1403"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "435/469 [==========================>...] - ETA: 0s - loss: 0.1403"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "449/469 [===========================>..] - ETA: 0s - loss: 0.1403"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "463/469 [============================>.] - ETA: 0s - loss: 0.1403"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "469/469 [==============================] - 2s 4ms/step - loss: 0.1403\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- " (saved) "
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Epoch 28/30\n",
- "\r",
- " 1/469 [..............................] - ETA: 1s - loss: 0.1403"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 15/469 [..............................] - ETA: 1s - loss: 0.1410"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 29/469 [>.............................] - ETA: 1s - loss: 0.1410"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 43/469 [=>............................] - ETA: 1s - loss: 0.1407"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 57/469 [==>...........................] - ETA: 1s - loss: 0.1406"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 71/469 [===>..........................] - ETA: 1s - loss: 0.1405"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 85/469 [====>.........................] - ETA: 1s - loss: 0.1405"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 99/469 [=====>........................] - ETA: 1s - loss: 0.1405"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "113/469 [======>.......................] - ETA: 1s - loss: 0.1404"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "127/469 [=======>......................] - ETA: 1s - loss: 0.1404"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "141/469 [========>.....................] - ETA: 1s - loss: 0.1403"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "155/469 [========>.....................] - ETA: 1s - loss: 0.1403"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "169/469 [=========>....................] - ETA: 1s - loss: 0.1403"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "183/469 [==========>...................] - ETA: 1s - loss: 0.1403"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "197/469 [===========>..................] - ETA: 1s - loss: 0.1402"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "211/469 [============>.................] - ETA: 0s - loss: 0.1402"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "225/469 [=============>................] - ETA: 0s - loss: 0.1402"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "239/469 [==============>...............] - ETA: 0s - loss: 0.1402"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "253/469 [===============>..............] - ETA: 0s - loss: 0.1402"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "267/469 [================>.............] - ETA: 0s - loss: 0.1402"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "281/469 [================>.............] - ETA: 0s - loss: 0.1402"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "295/469 [=================>............] - ETA: 0s - loss: 0.1401"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "309/469 [==================>...........] - ETA: 0s - loss: 0.1401"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "323/469 [===================>..........] - ETA: 0s - loss: 0.1401"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "337/469 [====================>.........] - ETA: 0s - loss: 0.1401"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "351/469 [=====================>........] - ETA: 0s - loss: 0.1401"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "365/469 [======================>.......] - ETA: 0s - loss: 0.1401"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "379/469 [=======================>......] - ETA: 0s - loss: 0.1401"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "393/469 [========================>.....] - ETA: 0s - loss: 0.1401"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "407/469 [=========================>....] - ETA: 0s - loss: 0.1402"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "421/469 [=========================>....] - ETA: 0s - loss: 0.1402"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "435/469 [==========================>...] - ETA: 0s - loss: 0.1402"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "449/469 [===========================>..] - ETA: 0s - loss: 0.1402"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "463/469 [============================>.] - ETA: 0s - loss: 0.1402"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "469/469 [==============================] - 2s 4ms/step - loss: 0.1402\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- " (saved) Epoch 29/30\n",
- "\r",
- " 1/469 [..............................] - ETA: 1s - loss: 0.1374"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 15/469 [..............................] - ETA: 1s - loss: 0.1378"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 29/469 [>.............................] - ETA: 1s - loss: 0.1383"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 43/469 [=>............................] - ETA: 1s - loss: 0.1385"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 57/469 [==>...........................] - ETA: 1s - loss: 0.1386"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 71/469 [===>..........................] - ETA: 1s - loss: 0.1388"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 85/469 [====>.........................] - ETA: 1s - loss: 0.1389"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 99/469 [=====>........................] - ETA: 1s - loss: 0.1390"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "113/469 [======>.......................] - ETA: 1s - loss: 0.1390"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "127/469 [=======>......................] - ETA: 1s - loss: 0.1390"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "141/469 [========>.....................] - ETA: 1s - loss: 0.1391"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "155/469 [========>.....................] - ETA: 1s - loss: 0.1391"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "169/469 [=========>....................] - ETA: 1s - loss: 0.1391"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "183/469 [==========>...................] - ETA: 1s - loss: 0.1391"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "197/469 [===========>..................] - ETA: 1s - loss: 0.1391"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "211/469 [============>.................] - ETA: 0s - loss: 0.1391"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "225/469 [=============>................] - ETA: 0s - loss: 0.1392"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "239/469 [==============>...............] - ETA: 0s - loss: 0.1392"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "253/469 [===============>..............] - ETA: 0s - loss: 0.1392"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "267/469 [================>.............] - ETA: 0s - loss: 0.1392"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "281/469 [================>.............] - ETA: 0s - loss: 0.1393"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "295/469 [=================>............] - ETA: 0s - loss: 0.1393"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "309/469 [==================>...........] - ETA: 0s - loss: 0.1393"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "323/469 [===================>..........] - ETA: 0s - loss: 0.1394"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "337/469 [====================>.........] - ETA: 0s - loss: 0.1394"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "351/469 [=====================>........] - ETA: 0s - loss: 0.1394"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "365/469 [======================>.......] - ETA: 0s - loss: 0.1394"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "379/469 [=======================>......] - ETA: 0s - loss: 0.1395"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "393/469 [========================>.....] - ETA: 0s - loss: 0.1395"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "407/469 [=========================>....] - ETA: 0s - loss: 0.1395"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "421/469 [=========================>....] - ETA: 0s - loss: 0.1395"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "435/469 [==========================>...] - ETA: 0s - loss: 0.1395"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "449/469 [===========================>..] - ETA: 0s - loss: 0.1396"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "463/469 [============================>.] - ETA: 0s - loss: 0.1396"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "469/469 [==============================] - 2s 4ms/step - loss: 0.1396\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- " (saved) Epoch 30/30\n",
- "\r",
- " 1/469 [..............................] - ETA: 1s - loss: 0.1456"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 15/469 [..............................] - ETA: 1s - loss: 0.1407"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 29/469 [>.............................] - ETA: 1s - loss: 0.1398"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 43/469 [=>............................] - ETA: 1s - loss: 0.1397"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 57/469 [==>...........................] - ETA: 1s - loss: 0.1396"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 71/469 [===>..........................] - ETA: 1s - loss: 0.1394"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 85/469 [====>.........................] - ETA: 1s - loss: 0.1393"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- " 99/469 [=====>........................] - ETA: 1s - loss: 0.1393"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "113/469 [======>.......................] - ETA: 1s - loss: 0.1393"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "127/469 [=======>......................] - ETA: 1s - loss: 0.1393"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "141/469 [========>.....................] - ETA: 1s - loss: 0.1393"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "155/469 [========>.....................] - ETA: 1s - loss: 0.1393"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "169/469 [=========>....................] - ETA: 1s - loss: 0.1393"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "183/469 [==========>...................] - ETA: 1s - loss: 0.1393"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "197/469 [===========>..................] - ETA: 1s - loss: 0.1393"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "211/469 [============>.................] - ETA: 0s - loss: 0.1394"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "225/469 [=============>................] - ETA: 0s - loss: 0.1394"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "239/469 [==============>...............] - ETA: 0s - loss: 0.1394"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "253/469 [===============>..............] - ETA: 0s - loss: 0.1394"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "267/469 [================>.............] - ETA: 0s - loss: 0.1394"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "281/469 [================>.............] - ETA: 0s - loss: 0.1394"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "295/469 [=================>............] - ETA: 0s - loss: 0.1394"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "309/469 [==================>...........] - ETA: 0s - loss: 0.1394"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "323/469 [===================>..........] - ETA: 0s - loss: 0.1395"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "337/469 [====================>.........] - ETA: 0s - loss: 0.1395"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "351/469 [=====================>........] - ETA: 0s - loss: 0.1395"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "365/469 [======================>.......] - ETA: 0s - loss: 0.1395"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "379/469 [=======================>......] - ETA: 0s - loss: 0.1395"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "393/469 [========================>.....] - ETA: 0s - loss: 0.1396"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "407/469 [=========================>....] - ETA: 0s - loss: 0.1396"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "421/469 [=========================>....] - ETA: 0s - loss: 0.1396"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "435/469 [==========================>...] - ETA: 0s - loss: 0.1396"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "449/469 [===========================>..] - ETA: 0s - loss: 0.1396"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "463/469 [============================>.] - ETA: 0s - loss: 0.1396"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
- "469/469 [==============================] - 2s 4ms/step - loss: 0.1396\n"
- ]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- " (saved) CPU times: user 1min 13s, sys: 5.48 s, total: 1min 18s\n",
- "Wall time: 1min 5s\n"
- ]
- }
- ],
- "source": [
- "%%time\n",
- "\n",
- "n = 60000\n",
- "batch_size = 128\n",
- "epochs = 30\n",
- "\n",
- "history = ae.fit(x_train[:n],y_train[:n], epochs=epochs, batch_size=batch_size, callbacks=callbacks_list)"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "## Step 5 - History"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 13,
- "metadata": {
- "execution": {
- "iopub.execute_input": "2021-03-01T19:34:39.899720Z",
- "iopub.status.busy": "2021-03-01T19:34:39.899253Z",
- "iopub.status.idle": "2021-03-01T19:34:40.482800Z",
- "shell.execute_reply": "2021-03-01T19:34:40.483298Z"
- }
- },
- "outputs": [
- {
- "data": {
- "text/html": [
- "<div class=\"comment\">Saved: ./run/figs/AE1-03-history_0</div>"
- ],
- "text/plain": [
- "<IPython.core.display.HTML object>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "image/png": "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\n",
- "text/plain": [
- "<Figure size 576x432 with 1 Axes>"
- ]
- },
- "metadata": {
- "needs_background": "light"
- },
- "output_type": "display_data"
- }
- ],
- "source": [
- "pwk.plot_history(history, plot={\"Loss\":['loss']}, save_as='03-history')"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "## Step 7 - Learning progression"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 14,
- "metadata": {
- "execution": {
- "iopub.execute_input": "2021-03-01T19:34:40.490796Z",
- "iopub.status.busy": "2021-03-01T19:34:40.490319Z",
- "iopub.status.idle": "2021-03-01T19:34:53.017653Z",
- "shell.execute_reply": "2021-03-01T19:34:53.018155Z"
- }
- },
- "outputs": [
- {
- "data": {
- "text/markdown": [
- "<br>**Real images :**"
- ],
- "text/plain": [
- "<IPython.core.display.Markdown object>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "text/html": [
- "<div class=\"comment\">Saved: ./run/figs/AE1-04-original-real</div>"
- ],
- "text/plain": [
- "<IPython.core.display.HTML object>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "image/png": "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\n",
- "text/plain": [
- "<Figure size 720x169.2 with 5 Axes>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "text/markdown": [
- "<br>**Noisy images :**"
- ],
- "text/plain": [
- "<IPython.core.display.Markdown object>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "text/html": [
- "<div class=\"comment\">Saved: ./run/figs/AE1-05-original-noisy</div>"
- ],
- "text/plain": [
- "<IPython.core.display.HTML object>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "image/png": "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\n",
- "text/plain": [
- "<Figure size 720x169.2 with 5 Axes>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "text/markdown": [
- "<br>**Learning...**"
- ],
- "text/plain": [
- "<IPython.core.display.Markdown object>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "text/html": [
- "<div class=\"comment\">Saved: ./run/figs/AE1-06-learning</div>"
- ],
- "text/plain": [
- "<IPython.core.display.HTML object>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "image/png": "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\n",
- "text/plain": [
- "<Figure size 720x5076 with 150 Axes>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "text/markdown": [
- "<br>**Noisy images :**"
- ],
- "text/plain": [
- "<IPython.core.display.Markdown object>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "image/png": "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\n",
- "text/plain": [
- "<Figure size 720x169.2 with 5 Axes>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "text/markdown": [
- "<br>**Real images :**"
- ],
- "text/plain": [
- "<IPython.core.display.Markdown object>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "image/png": "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\n",
- "text/plain": [
- "<Figure size 720x169.2 with 5 Axes>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- }
- ],
- "source": [
- "imgs=[]\n",
- "labels=[]\n",
- "for epoch in range(epochs):\n",
- " for i in range(5):\n",
- " filename = './run/images/image-{epoch:03d}-{i:02d}.jpg'.format(epoch=epoch, i=i)\n",
- " img = io.imread(filename)\n",
- " imgs.append(img)\n",
- " \n",
- "\n",
- "pwk.subtitle('Real images :')\n",
- "pwk.plot_images(y_test[:5], None, indices='all', columns=5, x_size=2,y_size=2, interpolation=None, save_as='04-original-real')\n",
- "\n",
- "pwk.subtitle('Noisy images :')\n",
- "pwk.plot_images(x_test[:5], None, indices='all', columns=5, x_size=2,y_size=2, interpolation=None, save_as='05-original-noisy')\n",
- "\n",
- "pwk.subtitle('Learning...')\n",
- "pwk.plot_images(imgs, None, indices='all', columns=5, x_size=2,y_size=2, interpolation=None, save_as='06-learning')\n",
- "\n",
- "pwk.subtitle('Noisy images :')\n",
- "pwk.plot_images(x_test[:5], None, indices='all', columns=5, x_size=2,y_size=2, interpolation=None, save_as=None)\n",
- "\n",
- "pwk.subtitle('Real images :')\n",
- "pwk.plot_images(y_test[:5], None, indices='all', columns=5, x_size=2,y_size=2, interpolation=None, save_as=None)\n"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 15,
- "metadata": {
- "execution": {
- "iopub.execute_input": "2021-03-01T19:34:53.021474Z",
- "iopub.status.busy": "2021-03-01T19:34:53.021005Z",
- "iopub.status.idle": "2021-03-01T19:34:53.023315Z",
- "shell.execute_reply": "2021-03-01T19:34:53.023790Z"
- }
- },
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "End time is : Monday 01 March 2021, 20:34:53\n",
- "Duration is : 00:02:11 233ms\n",
- "This notebook ends here\n"
- ]
- }
- ],
- "source": [
- "pwk.end()"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "---\n",
- "<img width=\"80px\" src=\"../fidle/img/00-Fidle-logo-01.svg\"></img>"
- ]
- }
- ],
- "metadata": {
- "kernelspec": {
- "display_name": "Python 3",
- "language": "python",
- "name": "python3"
- },
- "language_info": {
- "codemirror_mode": {
- "name": "ipython",
- "version": 3
- },
- "file_extension": ".py",
- "mimetype": "text/x-python",
- "name": "python",
- "nbconvert_exporter": "python",
- "pygments_lexer": "ipython3",
- "version": "3.7.9"
- }
- },
- "nbformat": 4,
- "nbformat_minor": 4
-}
diff --git a/AE/01-Prepare_MNIST_dataset.ipynb b/AE/01-Prepare_MNIST_dataset.ipynb
new file mode 100644
index 0000000000000000000000000000000000000000..413926ad402bc2dd9b0892b656db1cc191ed391b
--- /dev/null
+++ b/AE/01-Prepare_MNIST_dataset.ipynb
@@ -0,0 +1,219 @@
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "<img width=\"800px\" src=\"../fidle/img/00-Fidle-header-01.svg\"></img>\n",
+ "\n",
+ "# <!-- TITLE --> [AE1] - Prepare a noisy MNIST dataset\n",
+ "<!-- DESC --> Episode 1: Preparation of a specially noisy MNIST dataset\n",
+ "\n",
+ "<!-- AUTHOR : Jean-Luc Parouty (CNRS/SIMaP) -->\n",
+ "\n",
+ "## Objectives :\n",
+ " - Prepare a MNIST dataset, usable with our denoiser autoencoder\n",
+ "\n",
+ "## What we're going to do :\n",
+ "\n",
+ " - Load original MNIST dataset\n",
+ " - Adding noise, a lot !\n",
+ " - Save it :-)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Step 1 - Init and set parameters\n",
+ "### 1.1 - Init python"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "import numpy as np\n",
+ "import sys\n",
+ "\n",
+ "from skimage import io\n",
+ "from skimage.util import random_noise\n",
+ "\n",
+ "import modules.MNIST\n",
+ "from modules.MNIST import MNIST\n",
+ "\n",
+ "sys.path.append('..')\n",
+ "import fidle.pwk as pwk\n",
+ "\n",
+ "run_dir='./run/AE1'\n",
+ "datasets_dir = pwk.init('AE1', run_dir)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### 1.2 - Parameters\n",
+ "`prepared_dataset` : Filename of the future prepared dataset (Can be in ./data) "
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "prepared_dataset = './data/mnist-noisy.h5'"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Step 2 - Get MNIST original dataset\n",
+ "We will have: \n",
+ "`clean_data` : Original and clean images - This is what we will want to get at the **output** \n",
+ "`noisy_data` : Noisy images - These are the images we will have as **input** \n",
+ "`class_data` : Image classes - Inutiles, because the training will be unsupervised"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "clean_data, class_data = MNIST.get_origine()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Step 3 - Add noise\n",
+ "We will add noise to our input dataset : **x_data** \n",
+ "Our goal is to predict noiselessly data : **y_data** \n",
+ "Need 30-40 seconds"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "def noise_it(data):\n",
+ " new_data = np.copy(data)\n",
+ " for i,image in enumerate(new_data):\n",
+ " pwk.update_progress('Add noise : ',i+1,len(new_data))\n",
+ " image=random_noise(image, mode='gaussian', mean=0, var=0.3)\n",
+ " image=random_noise(image, mode='s&p', amount=0.2, salt_vs_pepper=0.5)\n",
+ " image=random_noise(image, mode='poisson') \n",
+ " image=random_noise(image, mode='speckle', mean=0, var=0.1)\n",
+ " new_data[i]=image\n",
+ " print('Done.')\n",
+ " return new_data\n",
+ "\n",
+ "# ---- Add noise to input data : x_data\n",
+ "#\n",
+ "noisy_data = noise_it(clean_data)\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Step 4 - Have a look"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "print('Clean dataset (clean_data) : ',clean_data.shape)\n",
+ "print('Noisy dataset (noisy_data) : ',noisy_data.shape)\n",
+ "\n",
+ "pwk.subtitle(\"Noisy images we'll have in input (or x)\")\n",
+ "pwk.plot_images(noisy_data[:5], None, indices='all', columns=5, x_size=3,y_size=3, interpolation=None, save_as='01-noisy')\n",
+ "pwk.subtitle('Clean images we want to obtain (or y)')\n",
+ "pwk.plot_images(clean_data[:5], None, indices='all', columns=5, x_size=3,y_size=3, interpolation=None, save_as='02-original')\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Step 5 - Shuffle dataset"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "p = np.random.permutation(len(clean_data))\n",
+ "clean_data, noisy_data, class_data = clean_data[p], noisy_data[p], class_data[p]\n",
+ "print('Shuffled.')"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Step 6 - Save our prepared dataset"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "MNIST.save_prepared_dataset( clean_data, noisy_data, class_data, filename=prepared_dataset )"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "pwk.end()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "---\n",
+ "<img width=\"80px\" src=\"../fidle/img/00-Fidle-logo-01.svg\"></img>"
+ ]
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "Python 3",
+ "language": "python",
+ "name": "python3"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.8.5"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}
diff --git a/AE/02-AE-with-MNIST-post.ipynb b/AE/02-AE-with-MNIST-post.ipynb
deleted file mode 100644
index c8c87c71b3877c4785d3d2922789fd4cfe6c7a54..0000000000000000000000000000000000000000
--- a/AE/02-AE-with-MNIST-post.ipynb
+++ /dev/null
@@ -1,379 +0,0 @@
-{
- "cells": [
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "<img width=\"800px\" src=\"../fidle/img/00-Fidle-header-01.svg\"></img>\n",
- "\n",
- "# <!-- TITLE --> [AE2] - Exploring our denoiser model\n",
- "<!-- DESC --> Episode 2 : Using the previously trained autoencoder to denoise data\n",
- "<!-- AUTHOR : Jean-Luc Parouty (CNRS/SIMaP) -->\n",
- "\n",
- "## Objectives :\n",
- " - Retrieve and use our denoiser model\n",
- "\n",
- "\n",
- "## What we're going to do :\n",
- "\n",
- " - Load a saved model\n",
- " - Encode/decode some test images (neved used, never seen by the model)\n"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "## Step 1 - Init python stuff\n",
- "### 1.1 - Init"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 2,
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/html": [
- "<style>\n",
- "\n",
- "div.warn { \n",
- " background-color: #fcf2f2;\n",
- " border-color: #dFb5b4;\n",
- " border-left: 5px solid #dfb5b4;\n",
- " padding: 0.5em;\n",
- " font-weight: bold;\n",
- " font-size: 1.1em;;\n",
- " }\n",
- "\n",
- "\n",
- "\n",
- "div.nota { \n",
- " background-color: #DAFFDE;\n",
- " border-left: 5px solid #92CC99;\n",
- " padding: 0.5em;\n",
- " }\n",
- "\n",
- "div.todo:before { content:url(data:image/svg+xml;base64,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);\n",
- " float:left;\n",
- " margin-right:20px;\n",
- " margin-top:-20px;\n",
- " margin-bottom:20px;\n",
- "}\n",
- "div.todo{\n",
- " font-weight: bold;\n",
- " font-size: 1.1em;\n",
- " margin-top:40px;\n",
- "}\n",
- "div.todo ul{\n",
- " margin: 0.2em;\n",
- "}\n",
- "div.todo li{\n",
- " margin-left:60px;\n",
- " margin-top:0;\n",
- " margin-bottom:0;\n",
- "}\n",
- "\n",
- "div .comment{\n",
- " font-size:0.8em;\n",
- " color:#696969;\n",
- "}\n",
- "\n",
- "\n",
- "\n",
- "</style>\n",
- "\n"
- ],
- "text/plain": [
- "<IPython.core.display.HTML object>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "text/markdown": [
- "<br>**FIDLE 2020 - Practical Work Module**"
- ],
- "text/plain": [
- "<IPython.core.display.Markdown object>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Version : 2.0.19\n",
- "Notebook id : AE2\n",
- "Run time : Friday 12 March 2021, 22:06:34\n",
- "TensorFlow version : 2.2.0\n",
- "Keras version : 2.3.0-tf\n",
- "Datasets dir : /home/pjluc/datasets/fidle\n",
- "Run dir : ./run\n",
- "Update keras cache : False\n"
- ]
- }
- ],
- "source": [
- "import numpy as np\n",
- "import sys\n",
- "import h5py\n",
- "import random\n",
- "\n",
- "from modules.AE import AE\n",
- "\n",
- "sys.path.append('..')\n",
- "import fidle.pwk as pwk\n",
- "\n",
- "datasets_dir = pwk.init('AE2')"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "### 1.2 - Parameters\n",
- "`output_dir` : Place where prepared dataset was saved(need 420 Mo) \n",
- "`output_filename` : Filename of the prepared dataset "
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 3,
- "metadata": {},
- "outputs": [],
- "source": [
- "output_dir = './data'\n",
- "output_filename = 'mnist-noisy.h5'"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "## Step 2 - Reload prepared dataset"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 5,
- "metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Loaded.\n",
- "x_test shape is : (10000, 28, 28, 1)\n",
- "y_test shape is : (10000, 28, 28, 1)\n"
- ]
- }
- ],
- "source": [
- "with h5py.File(f'{output_dir}/{output_filename}','r') as f:\n",
- " x_test = f['x_test'][:]\n",
- " y_test = f['y_test'][:]\n",
- " print('Loaded.')\n",
- " print('x_test shape is : ',x_test.shape)\n",
- " print('y_test shape is : ',y_test.shape)"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "## Step 3 - Reload best model"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 6,
- "metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Reloaded.\n"
- ]
- }
- ],
- "source": [
- "ae=AE()\n",
- "ae.reload('./run/models/best_model')"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "## Step 4 - Image reconstruction"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 7,
- "metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "y_pred shape : (10000, 28, 28, 1)\n"
- ]
- }
- ],
- "source": [
- "y_pred = ae.predict(x_test)\n",
- "print('y_pred shape : ',y_pred.shape)"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "## Step 5 - Have a look"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 22,
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/markdown": [
- "<br>**Noisy test images (input):**"
- ],
- "text/plain": [
- "<IPython.core.display.Markdown object>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "image/png": "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\n",
- "text/plain": [
- "<Figure size 1152x169.2 with 8 Axes>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "text/markdown": [
- "<br>**Predict test images (output):**"
- ],
- "text/plain": [
- "<IPython.core.display.Markdown object>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "image/png": "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\n",
- "text/plain": [
- "<Figure size 1152x169.2 with 8 Axes>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "text/markdown": [
- "<br>**Real test images :**"
- ],
- "text/plain": [
- "<IPython.core.display.Markdown object>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "image/png": "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\n",
- "text/plain": [
- "<Figure size 1152x169.2 with 8 Axes>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- }
- ],
- "source": [
- "i=random.randint(0,len(y_pred)-8)\n",
- "j=i+8\n",
- "\n",
- "pwk.subtitle('Noisy test images (input):')\n",
- "pwk.plot_images(x_test[i:j], None, indices='all', columns=8, x_size=2,y_size=2, interpolation=None, save_as='07-test-noisy')\n",
- "\n",
- "pwk.subtitle('Predict test images (output):')\n",
- "pwk.plot_images(y_pred[i:j], None, indices='all', columns=8, x_size=2,y_size=2, interpolation=None, save_as='08-test-predict')\n",
- "\n",
- "pwk.subtitle('Real test images :')\n",
- "pwk.plot_images(y_test[i:j], None, indices='all', columns=8, x_size=2,y_size=2, interpolation=None, save_as='09-test-real')\n",
- "\n"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 12,
- "metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "End time is : Monday 28 December 2020, 23:59:16\n",
- "Duration is : 00:02:31 652ms\n",
- "This notebook ends here\n"
- ]
- }
- ],
- "source": [
- "pwk.end()"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "---\n",
- "<img width=\"80px\" src=\"../fidle/img/00-Fidle-logo-01.svg\"></img>"
- ]
- }
- ],
- "metadata": {
- "kernelspec": {
- "display_name": "Python 3",
- "language": "python",
- "name": "python3"
- },
- "language_info": {
- "codemirror_mode": {
- "name": "ipython",
- "version": 3
- },
- "file_extension": ".py",
- "mimetype": "text/x-python",
- "name": "python",
- "nbconvert_exporter": "python",
- "pygments_lexer": "ipython3",
- "version": "3.8.5"
- }
- },
- "nbformat": 4,
- "nbformat_minor": 4
-}
diff --git a/AE/02-AE-with-MNIST-post==done==.ipynb b/AE/02-AE-with-MNIST-post==done==.ipynb
deleted file mode 100644
index 63fcd60cb57a9b912fddd1a11c9a5ad977a1d81b..0000000000000000000000000000000000000000
--- a/AE/02-AE-with-MNIST-post==done==.ipynb
+++ /dev/null
@@ -1,439 +0,0 @@
-{
- "cells": [
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "<img width=\"800px\" src=\"../fidle/img/00-Fidle-header-01.svg\"></img>\n",
- "\n",
- "# <!-- TITLE --> [AE2] - Exploring our denoiser model\n",
- "<!-- DESC --> Episode 2 : Using the previously trained autoencoder to denoise data\n",
- "<!-- AUTHOR : Jean-Luc Parouty (CNRS/SIMaP) -->\n",
- "\n",
- "## Objectives :\n",
- " - Retrieve and use our denoiser model\n",
- "\n",
- "\n",
- "## What we're going to do :\n",
- "\n",
- " - Load a saved model\n",
- " - Encode/decode some test images (neved used, never seen by the model)\n"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "## Step 1 - Init python stuff"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 1,
- "metadata": {
- "execution": {
- "iopub.execute_input": "2021-03-01T19:34:56.536505Z",
- "iopub.status.busy": "2021-03-01T19:34:56.536033Z",
- "iopub.status.idle": "2021-03-01T19:34:59.294122Z",
- "shell.execute_reply": "2021-03-01T19:34:59.294609Z"
- }
- },
- "outputs": [
- {
- "data": {
- "text/html": [
- "<style>\n",
- "\n",
- "div.warn { \n",
- " background-color: #fcf2f2;\n",
- " border-color: #dFb5b4;\n",
- " border-left: 5px solid #dfb5b4;\n",
- " padding: 0.5em;\n",
- " font-weight: bold;\n",
- " font-size: 1.1em;;\n",
- " }\n",
- "\n",
- "\n",
- "\n",
- "div.nota { \n",
- " background-color: #DAFFDE;\n",
- " border-left: 5px solid #92CC99;\n",
- " padding: 0.5em;\n",
- " }\n",
- "\n",
- "div.todo:before { content:url(data:image/svg+xml;base64,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);\n",
- " float:left;\n",
- " margin-right:20px;\n",
- " margin-top:-20px;\n",
- " margin-bottom:20px;\n",
- "}\n",
- "div.todo{\n",
- " font-weight: bold;\n",
- " font-size: 1.1em;\n",
- " margin-top:40px;\n",
- "}\n",
- "div.todo ul{\n",
- " margin: 0.2em;\n",
- "}\n",
- "div.todo li{\n",
- " margin-left:60px;\n",
- " margin-top:0;\n",
- " margin-bottom:0;\n",
- "}\n",
- "\n",
- "div .comment{\n",
- " font-size:0.8em;\n",
- " color:#696969;\n",
- "}\n",
- "\n",
- "\n",
- "\n",
- "</style>\n",
- "\n"
- ],
- "text/plain": [
- "<IPython.core.display.HTML object>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "text/markdown": [
- "<br>**FIDLE 2020 - Practical Work Module**"
- ],
- "text/plain": [
- "<IPython.core.display.Markdown object>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Version : 2.0.17\n",
- "Notebook id : AE2\n",
- "Run time : Monday 01 March 2021, 20:34:59\n",
- "TensorFlow version : 2.4.0\n",
- "Keras version : 2.4.0\n",
- "Datasets dir : /gpfswork/rech/mlh/uja62cb/datasets\n",
- "Run dir : ./run\n",
- "Update keras cache : False\n",
- "Save figs : True\n",
- "Path figs : ./run/figs\n"
- ]
- }
- ],
- "source": [
- "import numpy as np\n",
- "import sys\n",
- "import h5py\n",
- "import random\n",
- "\n",
- "from modules.AE import AE\n",
- "\n",
- "sys.path.append('..')\n",
- "import fidle.pwk as pwk\n",
- "\n",
- "datasets_dir = pwk.init('AE2')"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "## Step 2 - Get data"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 2,
- "metadata": {
- "execution": {
- "iopub.execute_input": "2021-03-01T19:34:59.329848Z",
- "iopub.status.busy": "2021-03-01T19:34:59.329358Z",
- "iopub.status.idle": "2021-03-01T19:34:59.332329Z",
- "shell.execute_reply": "2021-03-01T19:34:59.332814Z"
- }
- },
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Loaded.\n",
- "x_test shape is : (10000, 28, 28, 1)\n",
- "y_test shape is : (10000, 28, 28, 1)\n"
- ]
- }
- ],
- "source": [
- "with h5py.File('./data/mnist-noisy.h5','r') as f:\n",
- " x_test = f['x_test'][:]\n",
- " y_test = f['y_test'][:]\n",
- " print('Loaded.')\n",
- " print('x_test shape is : ',x_test.shape)\n",
- " print('y_test shape is : ',y_test.shape)"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "## Step 3 - Reload best model"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 3,
- "metadata": {
- "execution": {
- "iopub.execute_input": "2021-03-01T19:34:59.336075Z",
- "iopub.status.busy": "2021-03-01T19:34:59.335603Z",
- "iopub.status.idle": "2021-03-01T19:35:00.572803Z",
- "shell.execute_reply": "2021-03-01T19:35:00.573307Z"
- }
- },
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Reloaded.\n"
- ]
- }
- ],
- "source": [
- "ae=AE()\n",
- "ae.reload('./run/models/best_model')"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "## Step 4 - Image reconstruction"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 4,
- "metadata": {
- "execution": {
- "iopub.execute_input": "2021-03-01T19:35:00.587871Z",
- "iopub.status.busy": "2021-03-01T19:35:00.587287Z",
- "iopub.status.idle": "2021-03-01T19:35:03.411996Z",
- "shell.execute_reply": "2021-03-01T19:35:03.412489Z"
- }
- },
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "y_pred shape : (10000, 28, 28, 1)\n"
- ]
- }
- ],
- "source": [
- "y_pred = ae.predict(x_test)\n",
- "print('y_pred shape : ',y_pred.shape)"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "## Step 5 - Latent space"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 5,
- "metadata": {
- "execution": {
- "iopub.execute_input": "2021-03-01T19:35:03.418629Z",
- "iopub.status.busy": "2021-03-01T19:35:03.418081Z",
- "iopub.status.idle": "2021-03-01T19:35:05.520943Z",
- "shell.execute_reply": "2021-03-01T19:35:05.521439Z"
- }
- },
- "outputs": [
- {
- "data": {
- "text/markdown": [
- "<br>**Noisy test images (input):**"
- ],
- "text/plain": [
- "<IPython.core.display.Markdown object>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "text/html": [
- "<div class=\"comment\">Saved: ./run/figs/AE2-07-test-noisy</div>"
- ],
- "text/plain": [
- "<IPython.core.display.HTML object>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "image/png": "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\n",
- "text/plain": [
- "<Figure size 1152x169.2 with 8 Axes>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "text/markdown": [
- "<br>**Predict test images (output):**"
- ],
- "text/plain": [
- "<IPython.core.display.Markdown object>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "text/html": [
- "<div class=\"comment\">Saved: ./run/figs/AE2-08-test-predict</div>"
- ],
- "text/plain": [
- "<IPython.core.display.HTML object>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "image/png": "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\n",
- "text/plain": [
- "<Figure size 1152x169.2 with 8 Axes>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "text/markdown": [
- "<br>**Real test images :**"
- ],
- "text/plain": [
- "<IPython.core.display.Markdown object>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "text/html": [
- "<div class=\"comment\">Saved: ./run/figs/AE2-09-test-real</div>"
- ],
- "text/plain": [
- "<IPython.core.display.HTML object>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- },
- {
- "data": {
- "image/png": "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\n",
- "text/plain": [
- "<Figure size 1152x169.2 with 8 Axes>"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- }
- ],
- "source": [
- "i=random.randint(0,len(y_pred)-8)\n",
- "j=i+8\n",
- "\n",
- "pwk.subtitle('Noisy test images (input):')\n",
- "pwk.plot_images(x_test[i:j], None, indices='all', columns=8, x_size=2,y_size=2, interpolation=None, save_as='07-test-noisy')\n",
- "\n",
- "pwk.subtitle('Predict test images (output):')\n",
- "pwk.plot_images(y_pred[i:j], None, indices='all', columns=8, x_size=2,y_size=2, interpolation=None, save_as='08-test-predict')\n",
- "\n",
- "pwk.subtitle('Real test images :')\n",
- "pwk.plot_images(y_test[i:j], None, indices='all', columns=8, x_size=2,y_size=2, interpolation=None, save_as='09-test-real')\n",
- "\n"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 6,
- "metadata": {
- "execution": {
- "iopub.execute_input": "2021-03-01T19:35:05.525013Z",
- "iopub.status.busy": "2021-03-01T19:35:05.524542Z",
- "iopub.status.idle": "2021-03-01T19:35:05.526936Z",
- "shell.execute_reply": "2021-03-01T19:35:05.527413Z"
- }
- },
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "End time is : Monday 01 March 2021, 20:35:05\n",
- "Duration is : 00:00:06 235ms\n",
- "This notebook ends here\n"
- ]
- }
- ],
- "source": [
- "pwk.end()"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "---\n",
- "<img width=\"80px\" src=\"../fidle/img/00-Fidle-logo-01.svg\"></img>"
- ]
- }
- ],
- "metadata": {
- "kernelspec": {
- "display_name": "Python 3",
- "language": "python",
- "name": "python3"
- },
- "language_info": {
- "codemirror_mode": {
- "name": "ipython",
- "version": 3
- },
- "file_extension": ".py",
- "mimetype": "text/x-python",
- "name": "python",
- "nbconvert_exporter": "python",
- "pygments_lexer": "ipython3",
- "version": "3.7.9"
- }
- },
- "nbformat": 4,
- "nbformat_minor": 4
-}
diff --git a/AE/02-AE-with-MNIST.ipynb b/AE/02-AE-with-MNIST.ipynb
new file mode 100644
index 0000000000000000000000000000000000000000..02939991551812877edc285fa4a1cd1cfe4b5186
--- /dev/null
+++ b/AE/02-AE-with-MNIST.ipynb
@@ -0,0 +1,362 @@
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "<img width=\"800px\" src=\"../fidle/img/00-Fidle-header-01.svg\"></img>\n",
+ "\n",
+ "# <!-- TITLE --> [AE2] - Building and training an AE denoiser model\n",
+ "<!-- DESC --> Episode 1 : After construction, the model is trained with noisy data from the MNIST dataset.\n",
+ "\n",
+ "<!-- AUTHOR : Jean-Luc Parouty (CNRS/SIMaP) -->\n",
+ "\n",
+ "## Objectives :\n",
+ " - Understanding and implementing a denoizing **autoencoder** neurals network (AE)\n",
+ " - Understanding a more **advanced programming model**\n",
+ "\n",
+ "The calculation needs being important, it is preferable to use a very simple dataset such as MNIST. \n",
+ "The use of a GPU is often indispensable.\n",
+ "\n",
+ "## What we're going to do :\n",
+ "\n",
+ " - Defining a VAE model\n",
+ " - Build the model\n",
+ " - Train it\n",
+ " - Follow the learning process with Tensorboard"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Step 1 - Init python stuff\n",
+ "### 1.1 - Init"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "import numpy as np\n",
+ "from skimage import io\n",
+ "\n",
+ "import tensorflow as tf\n",
+ "from tensorflow import keras\n",
+ "from tensorflow.keras import layers\n",
+ "\n",
+ "from modules.callbacks import ImagesCallback\n",
+ "from tensorflow.keras.callbacks import ModelCheckpoint, TensorBoard\n",
+ "\n",
+ "import os,sys\n",
+ "from importlib import reload\n",
+ "import h5py\n",
+ "\n",
+ "from modules.MNIST import MNIST\n",
+ "from modules.callbacks import ImagesCallback, BestModelCallback\n",
+ "\n",
+ "sys.path.append('..')\n",
+ "import fidle.pwk as pwk\n",
+ "\n",
+ "run_dir = './run/AE2'\n",
+ "datasets_dir = pwk.init('AE2', run_dir)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### 1.2 - Parameters\n",
+ "`prepared_dataset` : Filename of the prepared dataset (Need 400 Mo, but can be in ./data) \n",
+ "`dataset_seed` : Random seed for shuffling dataset \n",
+ "`scale` : % of the dataset to use (1. for 100%) \n",
+ "`latent_dim` : Dimension of the latent space \n",
+ "`train_prop` : Percentage for train (the rest being for the test)\n",
+ "`batch_size` : Batch size \n",
+ "`epochs` : Nb of epochs for training\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "prepared_dataset = './data/mnist-noisy.h5'\n",
+ "dataset_seed = 123\n",
+ "\n",
+ "scale = .1\n",
+ "\n",
+ "latent_dim = 10\n",
+ "\n",
+ "train_prop = .8\n",
+ "batch_size = 128\n",
+ "epochs = 30"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Override parameters (batch mode) - Just forget this cell"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "pwk.override('prepared_dataset', 'dataset_seed', 'scale', 'latent_dim')\n",
+ "pwk.override('train_prop', 'batch_size', 'epochs')"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Step 2 - Retrieve dataset\n",
+ "With our MNIST class, in one call, we can reload, rescale, shuffle and split our previously saved dataset :-)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "clean_train,clean_test, noisy_train,noisy_test, _,_ = MNIST.reload_prepared_dataset(scale = scale, \n",
+ " train_prop = train_prop,\n",
+ " seed = dataset_seed,\n",
+ " shuffle = True,\n",
+ " filename=prepared_dataset )"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Step 3 - Build model"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### Encoder"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "inputs = keras.Input(shape=(28, 28, 1))\n",
+ "x = layers.Conv2D(32, 3, activation=\"relu\", strides=2, padding=\"same\")(inputs)\n",
+ "x = layers.Conv2D(64, 3, activation=\"relu\", strides=2, padding=\"same\")(x)\n",
+ "x = layers.Flatten()(x)\n",
+ "x = layers.Dense(16, activation=\"relu\")(x)\n",
+ "z = layers.Dense(latent_dim)(x)\n",
+ "\n",
+ "encoder = keras.Model(inputs, z, name=\"encoder\")\n",
+ "# encoder.summary()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### Decoder"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "inputs = keras.Input(shape=(latent_dim,))\n",
+ "x = layers.Dense(7 * 7 * 64, activation=\"relu\")(inputs)\n",
+ "x = layers.Reshape((7, 7, 64))(x)\n",
+ "x = layers.Conv2DTranspose(64, 3, activation=\"relu\", strides=2, padding=\"same\")(x)\n",
+ "x = layers.Conv2DTranspose(32, 3, activation=\"relu\", strides=2, padding=\"same\")(x)\n",
+ "outputs = layers.Conv2DTranspose(1, 3, activation=\"sigmoid\", padding=\"same\")(x)\n",
+ "\n",
+ "decoder = keras.Model(inputs, outputs, name=\"decoder\")\n",
+ "# decoder.summary()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### AE\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "inputs = keras.Input(shape=(28, 28, 1))\n",
+ "\n",
+ "latents = encoder(inputs)\n",
+ "outputs = decoder(latents)\n",
+ "\n",
+ "ae = keras.Model(inputs,outputs, name=\"ae\")\n",
+ "\n",
+ "ae.compile(optimizer=keras.optimizers.Adam(), loss='binary_crossentropy')"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Step 4 - Train\n",
+ "20' on a CPU \n",
+ "1'12 on a GPU (V100, IDRIS)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "# ---- Callback : Images\n",
+ "#\n",
+ "pwk.mkdir( run_dir + '/images')\n",
+ "filename = run_dir + '/images/image-{epoch:03d}-{i:02d}.jpg'\n",
+ "callback_images = ImagesCallback(filename, x=clean_test[:5], encoder=encoder,decoder=decoder)\n",
+ "\n",
+ "# ---- Callback : Best model\n",
+ "#\n",
+ "pwk.mkdir( run_dir + '/models')\n",
+ "filename = run_dir + '/models/best_model.h5'\n",
+ "callback_bestmodel = tf.keras.callbacks.ModelCheckpoint(filepath=filename, verbose=0, save_best_only=True)\n",
+ "\n",
+ "# ---- Callback tensorboard\n",
+ "#\n",
+ "logdir = run_dir + '/logs'\n",
+ "callback_tensorboard = TensorBoard(log_dir=logdir, histogram_freq=1)\n",
+ "\n",
+ "# callbacks_list = [callback_images, callback_bestmodel, callback_tensorboard]\n",
+ "callbacks_list = [callback_images, callback_bestmodel]"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "pwk.chrono_start()\n",
+ "\n",
+ "history = ae.fit(noisy_train, clean_train,\n",
+ " batch_size = batch_size,\n",
+ " epochs = epochs,\n",
+ " verbose = 1,\n",
+ " validation_data = (noisy_test, clean_test),\n",
+ " callbacks = callbacks_list )\n",
+ "\n",
+ "pwk.chrono_show()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Step 5 - History"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "pwk.plot_history(history, plot={'loss':['loss','val_loss']}, save_as='01-history')"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Step 7 - Learning progression"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "imgs=[]\n",
+ "labels=[]\n",
+ "for epoch in range(epochs):\n",
+ " for i in range(5):\n",
+ " filename = run_dir + '/images/image-{epoch:03d}-{i:02d}.jpg'.format(epoch=epoch, i=i)\n",
+ " img = io.imread(filename)\n",
+ " imgs.append(img)\n",
+ " \n",
+ "\n",
+ "pwk.subtitle('Real images :')\n",
+ "pwk.plot_images(clean_test[:5], None, indices='all', columns=5, x_size=2,y_size=2, interpolation=None, save_as='02-original-real')\n",
+ "\n",
+ "pwk.subtitle('Noisy images :')\n",
+ "pwk.plot_images(noisy_test[:5], None, indices='all', columns=5, x_size=2,y_size=2, interpolation=None, save_as='03-original-noisy')\n",
+ "\n",
+ "pwk.subtitle('Learning...')\n",
+ "pwk.plot_images(imgs, None, indices='all', columns=5, x_size=2,y_size=2, interpolation=None, save_as='04-learning')\n",
+ "\n",
+ "pwk.subtitle('Noisy images :')\n",
+ "pwk.plot_images(noisy_test[:5], None, indices='all', columns=5, x_size=2,y_size=2, interpolation=None, save_as=None)\n",
+ "\n",
+ "pwk.subtitle('Real images :')\n",
+ "pwk.plot_images(clean_test[:5], None, indices='all', columns=5, x_size=2,y_size=2, interpolation=None, save_as=None)\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "pwk.end()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "---\n",
+ "<img width=\"80px\" src=\"../fidle/img/00-Fidle-logo-01.svg\"></img>"
+ ]
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "Python 3",
+ "language": "python",
+ "name": "python3"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.8.5"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}
diff --git a/AE/03-AE-with-MNIST-post.ipynb b/AE/03-AE-with-MNIST-post.ipynb
new file mode 100644
index 0000000000000000000000000000000000000000..6a1d5fb3301c666015a4bc5a5eb40563d5e6a90e
--- /dev/null
+++ b/AE/03-AE-with-MNIST-post.ipynb
@@ -0,0 +1,198 @@
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "<img width=\"800px\" src=\"../fidle/img/00-Fidle-header-01.svg\"></img>\n",
+ "\n",
+ "# <!-- TITLE --> [AE3] - Playing with our denoiser model\n",
+ "<!-- DESC --> Episode 2 : Using the previously trained autoencoder to denoise data\n",
+ "<!-- AUTHOR : Jean-Luc Parouty (CNRS/SIMaP) -->\n",
+ "\n",
+ "## Objectives :\n",
+ " - Retrieve and use our denoiser model\n",
+ "\n",
+ "\n",
+ "## What we're going to do :\n",
+ "\n",
+ " - Reload our dataset and saved best model\n",
+ " - Encode/decode some test images (neved used, never seen by the model)\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Step 1 - Init python stuff\n",
+ "### 1.1 - Init"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "import numpy as np\n",
+ "import sys\n",
+ "import h5py\n",
+ "import random\n",
+ "\n",
+ "import tensorflow as tf\n",
+ "from tensorflow import keras\n",
+ "\n",
+ "from modules.MNIST import MNIST\n",
+ "\n",
+ "sys.path.append('..')\n",
+ "import fidle.pwk as pwk\n",
+ "\n",
+ "run_dir = './run/AE2'\n",
+ "datasets_dir = pwk.init('AE3', run_dir)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### 1.2 - Parameters\n",
+ "`prepared_dataset` : Filename of the prepared dataset (Need 400 Mo, but can be in ./data) \n",
+ "`dataset_seed` : Random seed for shuffling dataset \n",
+ "`scale` : % of the dataset to use (1. for 100%) \n",
+ "`train_prop` : Percentage for train (the rest being for the test)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "prepared_dataset = './data/mnist-noisy.h5'\n",
+ "dataset_seed = 123\n",
+ "scale = .1\n",
+ "train_prop = .8"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Step 2 - Retrieve dataset\n",
+ "With our MNIST class, in one call, we can reload, rescale, shuffle and split our previously saved dataset :-) \n",
+ "**Important :** Make sure that the hash is identical to the one used during the training!"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "clean_train,clean_test, noisy_train,noisy_test, _,_ = MNIST.reload_prepared_dataset(scale = scale, \n",
+ " train_prop = train_prop,\n",
+ " seed = dataset_seed,\n",
+ " shuffle = True,\n",
+ " filename=prepared_dataset )"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Step 3 - Reload best model"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "ae = keras.models.load_model(f'{run_dir}/models/best_model.h5')"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Step 4 - Image reconstruction"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "predi_test = ae.predict(noisy_test)\n",
+ "print('predi_test shape : ',predi_test.shape)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Step 5 - Have a look"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "i=random.randint(0,len(predi_test)-8)\n",
+ "j=i+8\n",
+ "\n",
+ "pwk.subtitle('Noisy test images (input):')\n",
+ "pwk.plot_images(noisy_test[i:j], None, indices='all', columns=8, x_size=2,y_size=2, interpolation=None, save_as='05-test-noisy')\n",
+ "\n",
+ "pwk.subtitle('Predict test images (output):')\n",
+ "pwk.plot_images(predi_test[i:j], None, indices='all', columns=8, x_size=2,y_size=2, interpolation=None, save_as='06-test-predict')\n",
+ "\n",
+ "pwk.subtitle('Real test images :')\n",
+ "pwk.plot_images(clean_test[i:j], None, indices='all', columns=8, x_size=2,y_size=2, interpolation=None, save_as='07-test-real')\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "pwk.end()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "---\n",
+ "<img width=\"80px\" src=\"../fidle/img/00-Fidle-logo-01.svg\"></img>"
+ ]
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "Python 3",
+ "language": "python",
+ "name": "python3"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.8.5"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}
diff --git a/AE/modules/MNIST.py b/AE/modules/MNIST.py
new file mode 100644
index 0000000000000000000000000000000000000000..d81b3584070ca19983aeecc16602d0789cfe30a7
--- /dev/null
+++ b/AE/modules/MNIST.py
@@ -0,0 +1,161 @@
+
+# ------------------------------------------------------------------
+# _____ _ _ _
+# | ___(_) __| | | ___
+# | |_ | |/ _` | |/ _ \
+# | _| | | (_| | | __/
+# |_| |_|\__,_|_|\___|
+# ------------------------------------------------------------------
+# Formation Introduction au Deep Learning (FIDLE)
+# CNRS/SARI/DEVLOG 2020 - S. Arias, E. Maldonado, JL. Parouty
+# ------------------------------------------------------------------
+# 2.0 version by JL Parouty, feb 2021
+
+import h5py
+import os
+import numpy as np
+from hashlib import blake2b
+import tensorflow as tf
+import tensorflow.keras.datasets.mnist as mnist
+
+class MNIST():
+
+ version = '0.1'
+
+ def __init__(self):
+ pass
+
+ @classmethod
+ def get_origine(cls, normalize=True, expand=True, concatenate=True):
+ """
+ Return original MNIST dataset
+ args:
+ normalize : Normalize dataset or not (True)
+ expand : Reshape images as (28,28,1) instead (28,28) (True)
+ concatenate : Concatenate train and test sets (True)
+ returns:
+ x_data,y_data if concatenate is False
+ x_train,y_train,x_test,y_test if concatenate is True
+ """
+
+ # ---- Get data
+ #
+ (x_train, y_train), (x_test, y_test) = mnist.load_data()
+ print('Dataset loaded.')
+
+ # ---- Normalization
+ #
+ if normalize:
+ x_train = x_train.astype('float32') / 255.
+ x_test = x_test.astype( 'float32') / 255.
+ print('Normalized.')
+
+ # ---- Reshape : (28,28) -> (28,28,1)
+ #
+ if expand:
+ x_train = np.expand_dims(x_train, axis=-1)
+ x_test = np.expand_dims(x_test, axis=-1)
+ print('Reshaped.')
+
+ # ---- Concatenate
+ #
+ if concatenate:
+ x_data = np.concatenate([x_train, x_test], axis=0)
+ y_data = np.concatenate([y_train, y_test])
+ print('Concatenate.')
+ print('x shape :', x_data.shape)
+ print('y shape :', y_data.shape)
+ return x_data,y_data
+ else:
+ print('x_train shape :', x_train.shape)
+ print('y_train shape :', y_train.shape)
+ print('x_test shape :', x_test.shape)
+ print('y_test shape :', y_test.shape)
+ return x_train,y_train,x_test,y_test
+
+
+ @classmethod
+ def save_prepared_dataset(cls, clean_data, noisy_data, class_data, filename='./data/mnist-noisy.h5'):
+ """
+ Save a prepared dataset in a h5 file
+ args:
+ clean_data, noisy_data, class_data : clean, noisy and class dataset
+ filename : filename
+ return:
+ None
+ """
+ path=os.path.dirname(filename)
+ os.makedirs(path, mode=0o750, exist_ok=True)
+
+ with h5py.File(filename, "w") as f:
+ f.create_dataset("clean_data", data=clean_data)
+ f.create_dataset("noisy_data", data=noisy_data)
+ f.create_dataset("class_data", data=class_data)
+ print('Saved.')
+ print('clean_data shape is : ',clean_data.shape)
+ print('noisy_data shape is : ',noisy_data.shape)
+ print('class_data shape is : ',class_data.shape)
+
+
+ @classmethod
+ def reload_prepared_dataset(cls, scale=1., train_prop=0.8, shuffle=True, seed=None, filename='./data/mnist-noisy.h5'):
+ """
+ Reload a saved dataset
+ args:
+ scale : Scale of dataset to use. 1. mean 100% (1.)
+ train_prop : Ratio of train/test
+ shuffle : Shuffle data if True
+ seed : Random seed value (no seed if None) (None)
+ filename : filename of the prepared dataset
+ returns:
+ clean_train,clean_test, noisy_train,noisy_test, class_train,class_test
+ """
+ # ---- Load saved dataset
+ #
+ with h5py.File(filename,'r') as f:
+ clean_data = f['clean_data'][:]
+ noisy_data = f['noisy_data'][:]
+ class_data = f['class_data'][:]
+ print('Loaded.')
+
+ # ---- Rescale
+ #
+ n = int(scale*len(clean_data))
+ clean_data, noisy_data, class_data = clean_data[:n], noisy_data[:n], class_data[:n]
+ print(f'rescaled ({scale}).')
+
+ # ---- Seed
+ #
+ if seed is not None:
+ np.random.seed(seed)
+ print(f'Seeded ({seed})')
+
+ # ---- Shuffle
+ #
+ if shuffle:
+ p = np.random.permutation(len(clean_data))
+ clean_data, noisy_data, class_data = clean_data[p], noisy_data[p], class_data[p]
+ print('Shuffled.')
+
+ # ---- Split
+ #
+ n=int(len(clean_data)*train_prop)
+ clean_train, clean_test = clean_data[:n], clean_data[n:]
+ noisy_train, noisy_test = noisy_data[:n], noisy_data[n:]
+ class_train, class_test = class_data[:n], class_data[n:]
+ print(f'splited ({train_prop}).')
+
+ # ---- Hash
+ #
+ h = blake2b(digest_size=10)
+ for a in [clean_train,clean_test, noisy_train,noisy_test, class_train,class_test]:
+ h.update(a)
+
+ print('clean_train shape is : ', clean_train.shape)
+ print('clean_test shape is : ', clean_test.shape)
+ print('noisy_train shape is : ', noisy_train.shape)
+ print('noisy_test shape is : ', noisy_test.shape)
+ print('class_train shape is : ', class_train.shape)
+ print('class_test shape is : ', class_test.shape)
+ print('Blake2b digest is : ', h.hexdigest())
+ return clean_train,clean_test, noisy_train,noisy_test, class_train,class_test
\ No newline at end of file
diff --git a/AE/modules/loader_MNIST.py b/AE/modules/loader_MNIST.py
deleted file mode 100644
index 115dbfd8a8e5b0bc33a3c82c31725a480b02a835..0000000000000000000000000000000000000000
--- a/AE/modules/loader_MNIST.py
+++ /dev/null
@@ -1,61 +0,0 @@
-
-# ------------------------------------------------------------------
-# _____ _ _ _
-# | ___(_) __| | | ___
-# | |_ | |/ _` | |/ _ \
-# | _| | | (_| | | __/
-# |_| |_|\__,_|_|\___|
-# ------------------------------------------------------------------
-# Formation Introduction au Deep Learning (FIDLE)
-# CNRS/SARI/DEVLOG 2020 - S. Arias, E. Maldonado, JL. Parouty
-# ------------------------------------------------------------------
-# Initial version by JL Parouty, feb 2020
-
-import numpy as np
-import tensorflow as tf
-import tensorflow.keras.datasets.mnist as mnist
-
-class Loader_MNIST():
-
- version = '0.1'
-
- def __init__(self):
- pass
-
- @classmethod
- def get(normalize=True, expand=True, concatenate=True):
-
- # ---- Get data
- #
- (x_train, y_train), (x_test, y_test) = mnist.load_data()
- print('Dataset loaded.')
-
- # ---- Normalization
- #
- if normalize:
- x_train = x_train.astype('float32') / 255.
- x_test = x_test.astype( 'float32') / 255.
- print('Normalized.')
-
- # ---- Reshape : (28,28) -> (28,28,1)
- #
- if expand:
- x_train = np.expand_dims(x_train, axis=-1)
- x_test = np.expand_dims(x_test, axis=-1)
- print('Expanded.')
-
- # ---- Concatenate
- #
- if concatenate:
- x_data = np.concatenate([x_train, x_test], axis=0)
- y_data = np.concatenate([y_train, y_test])
- print('Concatenate.')
- print('x shape :', x_data.shape)
- print('y shape :', y_data.shape)
- return x_data,y_data
- else:
- print('x_train shape :', x_train.shape)
- print('y_train shape :', y_train.shape)
- print('x_test shape :', x_test.shape)
- print('y_test shape :', y_test.shape)
- return x_train,y_train,x_test,y_test
\ No newline at end of file