{
"cells": [
{
"cell_type": "markdown",
"id": "alpha-bahrain",
"metadata": {},
"source": [
"<img width=\"800px\" src=\"../fidle/img/header.svg\"></img>\n",
"\n",
"# <!-- TITLE --> [SCRATCH1] - Scratchbook\n",
"<!-- DESC --> A scratchbook for small examples\n",
"<!-- AUTHOR : Jean-Luc Parouty (CNRS/SIMaP) -->\n",
"\n",
"## Objectives :\n",
" - Take a quick look at thousands of little things\n",
"\n",
"## Inside this scratchbook :\n",
"\n",
"[1 - LSTM Keras layer](#1---LSTM-Keras-layer) \n",
"[2 - TimeseriesGenerator](#2---TimeseriesGenerator) \n",
"[3 - Upsampling](#3---Upsampling) \n",
"[4 - Reduce mean](#4---Reduce-mean) \n",
"[5 - Gradient tape](#5---Gradient-tape)\n"
]
},
{
"cell_type": "markdown",
"id": "accessory-church",
"metadata": {},
"source": [
"# One init to rule them all"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "floppy-organic",
"metadata": {},
"outputs": [],
"source": [
"import tensorflow as tf\n",
"from tensorflow import keras\n",
"from tensorflow.keras.callbacks import TensorBoard\n",
"from tensorflow.keras.preprocessing.sequence import TimeseriesGenerator\n",
"\n",
"import numpy as np\n",
"import math, random"
]
},
{
"cell_type": "markdown",
"id": "danish-rebound",
"metadata": {},
"source": [
"## 1 - LSTM Keras layer\n",
"[Back to home](#)"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "opposite-plasma",
"metadata": {},
"outputs": [],
"source": [
"inputs = tf.random.normal([32, 20, 8])\n",
"\n",
"lstm = tf.keras.layers.LSTM(16)\n",
"output = lstm(inputs)\n",
"\n",
"print('Inputs shape is : ', inputs.shape)\n",
"print('Output shape is : ', output.shape)"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "forbidden-murray",
"metadata": {},
"outputs": [],
"source": [
"lstm = tf.keras.layers.LSTM(18, return_sequences=True, return_state=True)\n",
"\n",
"output, memory_state, carry_state = lstm(inputs)\n",
"\n",
"print('Output shape : ', output.shape)\n",
"print('Memory state : ', memory_state.shape)\n",
"print('Carry state : ', carry_state.shape)\n"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "verified-fruit",
"metadata": {},
"outputs": [],
"source": [
"# --- See the last vector of the output\n",
"output[-1,-1]"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "homeless-library",
"metadata": {},
"outputs": [],
"source": [
"# ---- Memory state is the last output\n",
"memory_state[-1]"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "preliminary-psychiatry",
"metadata": {},
"outputs": [],
"source": [
"carry_state[-1]"
]
},
{
"cell_type": "markdown",
"id": "41d326b2-376e-49d6-9429-07016d98dc09",
"metadata": {},
"source": [
"## 2 - TimeseriesGenerator\n",
"[Back to home](#)"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "42276389-4ea6-42d1-93bc-6650062ef86a",
"metadata": {},
"outputs": [],
"source": [
"from keras.preprocessing.sequence import TimeseriesGenerator\n",
"\n",
"# ---- Define a dataset\n",
"\n",
"series = np.array([1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15])\n",
"\n",
"# ---- Generator\n",
"\n",
"generator = TimeseriesGenerator(series, series, length=5, batch_size=1)\n",
"\n",
"# ---- Samples\n",
"\n",
"nb_batch = len(generator)\n",
"\n",
"print(f'Number of batch : {nb_batch}\\n')\n",
"for i in range(nb_batch):\n",
" x, y = generator[i]\n",
" print(f'#{i} : {x} => {y}')"
]
},
{
"cell_type": "markdown",
"id": "67e3c888-aaa4-4166-90a1-cdb63920fd7d",
"metadata": {},
"source": [
"## 3 - Upsampling\n",
"[Back to home](#)"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "20f12cf0-1fdb-4b53-92c6-d03b140e42d1",
"metadata": {},
"outputs": [],
"source": [
"x = np.array([1,2,3,4])\n",
"x = x.reshape(2,2)\n",
"print('\\nInitial : ', x.shape)\n",
"print(x)\n",
"\n",
"x = x.reshape((1,2,2,1))\n",
"print('\\nReshape as a batch of (2,2) vectors : ', x.shape)\n",
"print(x)\n",
"\n",
"y = tf.keras.layers.UpSampling2D( size=(2, 2), interpolation=\"nearest\" )(x)\n",
"\n",
"y = np.array(y)\n",
"print('\\ny shape : ',y.shape)\n",
"\n",
"y = y.reshape(4,4)\n",
"print('\\n After a (4,4) reshape :')\n",
"print(y)\n",
"\n"
]
},
{
"cell_type": "markdown",
"id": "e8fb1472",
"metadata": {},
"source": [
"## 4 - Reduce mean\n",
"[Back to home](#)"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "09ac4e52-8953-41d9-b712-e6a83a9ae860",
"metadata": {},
"outputs": [],
"source": [
"import numpy as np\n",
"import tensorflow as tf\n",
"\n",
"c = np.array([[3.,4], [5.,6], [7.,8]])\n",
"print(np.mean(c,1))\n",
"\n",
"print(tf.reduce_mean(c,1))\n"
]
},
{
"cell_type": "markdown",
"id": "72be9368",
"metadata": {},
"source": [
"## 5 - Gradient tape\n",
"[Back to home](#)"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "13fcc722",
"metadata": {},
"outputs": [],
"source": [
"import numpy as np\n",
"import tensorflow as tf\n",
"\n",
"# ---- My function f\n",
"#\n",
"def f(x):\n",
" y = x**2 + 4*x - 5\n",
" return y\n",
"\n",
"# ---- Examples :\n",
"#\n",
"print('f(1) is : ', f(1))\n",
"print('f(2) is : ', f(2))\n",
"\n",
"# ---- With a tensor :\n",
"#\n",
"x = tf.Variable(2.0)\n",
"\n",
"print('f(x) is : ', f(x))"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "6fab93ce",
"metadata": {},
"outputs": [],
"source": [
"# ---- Derivative function of f(x)\n",
"#\n",
"def g(x):\n",
" y = 2*x + 4\n",
" return y\n",
"\n",
"print('Derivative of f(x) for x=3 : ', g(3))\n"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "3f3f0c4c",
"metadata": {},
"outputs": [],
"source": [
"# ---- Derivative using tf\n",
"#\n",
"with tf.GradientTape() as tape:\n",
" x = tf.Variable(3.0)\n",
" y = f(x)\n",
"\n",
"dy = tape.gradient(y, x) # dy/dx\n",
"\n",
"print(dy)\n"
]
},
{
"cell_type": "markdown",
"id": "43f9a625",
"metadata": {},
"source": [
"---\n",
"<img width=\"80px\" src=\"../fidle/img/logo-paysage.svg\"></img>"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3.9.2 ('fidle-env')",
"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": "3.9.2"
},
"vscode": {
"interpreter": {
"hash": "b3929042cc22c1274d74e3e946c52b845b57cb6d84f2d591ffe0519b38e4896d"
}
}
},
"nbformat": 4,
"nbformat_minor": 5
}