Translate in EN French lines

Former-commit-id: b7c6033c

Translate in EN French lines
Former-commit-id: b7c6033c
d0f35a35 · Soraya Arias · aa6460cb · d0f35a35
Commit d0f35a35 authored 5 years ago by Soraya Arias
--- a/BHPD/02-DNN-Regression-Premium.ipynb
+++ b/BHPD/02-DNN-Regression-Premium.ipynb
@@ -273,7 +273,7 @@
     "name": "stdout",
     "output_type": "stream",
     "text": [
-      "Données manquantes :  0   Shape is :  (506, 14)\n"
+      "Missing Data :  0   Shape is :  (506, 14)\n"
     ]
    }
   ],
@@ -281,7 +281,7 @@
    "data = pd.read_csv('./data/BostonHousing.csv', header=0)\n",
    "\n",
    "display(data.head(5).style.format(\"{0:.2f}\"))\n",
-    "print('Données manquantes : ',data.isna().sum().sum(), '  Shape is : ', data.shape)"
+    "print('Missing Data : ',data.isna().sum().sum(), '  Shape is : ', data.shape)"
   ]
  },
  {
@@ -1195,7 +1195,7 @@
   ],
   "source": [
    "predictions = loaded_model.predict( my_data )\n",
-    "print(\"Prédiction : {:.2f} K$   Reality : {:.2f} K$\".format(predictions[0][0], real_price))"
+    "print(\"Prediction : {:.2f} K$   Reality : {:.2f} K$\".format(predictions[0][0], real_price))"
   ]
  },
  {

 %% Cell type:markdown id: tags:
 <img width="800px" src="../fidle/img/00-Fidle-header-01.svg"></img>
 # <!-- TITLE --> [BHP2] - Regression with a Dense Network (DNN) - Advanced code
  <!-- DESC -->  More advanced example of DNN network code - BHPD dataset
  <!-- AUTHOR : Jean-Luc Parouty (CNRS/SIMaP) -->
 ## Objectives :
 - Predicts **housing prices** from a set of house features.
 - Understanding the principle and the architecture of a regression with a dense neural network with backup and restore of the trained model.
 The **[Boston Housing Dataset](https://www.cs.toronto.edu/~delve/data/boston/bostonDetail.html)** consists of price of houses in various places in Boston.
 Alongside with price, the dataset also provide these information :
 - CRIM: This is the per capita crime rate by town
 - ZN: This is the proportion of residential land zoned for lots larger than 25,000 sq.ft
 - INDUS: This is the proportion of non-retail business acres per town
 - CHAS: This is the Charles River dummy variable (this is equal to 1 if tract bounds river; 0 otherwise)
 - NOX: This is the nitric oxides concentration (parts per 10 million)
 - RM: This is the average number of rooms per dwelling
 - AGE: This is the proportion of owner-occupied units built prior to 1940
 - DIS: This is the weighted distances to five Boston employment centers
 - RAD: This is the index of accessibility to radial highways
 - TAX: This is the full-value property-tax rate per 10,000 dollars
 - PTRATIO: This is the pupil-teacher ratio by town
 - B: This is calculated as 1000(Bk — 0.63)^2, where Bk is the proportion of people of African American descent by town
 - LSTAT: This is the percentage lower status of the population
 - MEDV: This is the median value of owner-occupied homes in 1000 dollars
 ## What we're going to do :
 - (Retrieve data)
 - (Preparing the data)
 - (Build a model)
 - Train and save the model
 - Restore saved model
 - Evaluate the model
 - Make some predictions
 %% Cell type:markdown id: tags:
 ## Step 1 - Import and init
 %% Cell type:code id: tags:
 ``` python
 import tensorflow as tf
 from tensorflow import keras
 import numpy as np
 import matplotlib.pyplot as plt
 import pandas as pd
 import os,sys
 from IPython.display import Markdown
 from importlib import reload
 sys.path.append('..')
 import fidle.pwk as ooo
 ooo.init()
 os.makedirs('./run/models',   mode=0o750, exist_ok=True)
 ```
 %% Output
    FIDLE 2020 - Practical Work Module
    Version              : 0.4.3
    Run time             : Friday 28 February 2020, 10:23:12
    TensorFlow version   : 2.0.0
    Keras version        : 2.2.4-tf
 %% Cell type:markdown id: tags:
 ## Step 2 - Retrieve data
 ### 2.1 - Option 1  : From Keras
 Boston housing is a famous historic dataset, so we can get it directly from [Keras datasets](https://www.tensorflow.org/api_docs/python/tf/keras/datasets)
 %% Cell type:raw id: tags:
 (x_train, y_train), (x_test, y_test) = keras.datasets.boston_housing.load_data(test_split=0.2, seed=113)
 %% Cell type:markdown id: tags:
 ### 2.2 - Option 2 : From a csv file
 More fun !
 %% Cell type:code id: tags:
 ``` python
 data = pd.read_csv('./data/BostonHousing.csv', header=0)
 display(data.head(5).style.format("{0:.2f}"))
-print('Données manquantes : ',data.isna().sum().sum(), '  Shape is : ', data.shape)
+print('Missing Data : ',data.isna().sum().sum(), '  Shape is : ', data.shape)
 ```
 %% Output
-    Données manquantes :  0   Shape is :  (506, 14)
+    Missing Data :  0   Shape is :  (506, 14)
 %% Cell type:markdown id: tags:
 ## Step 3 - Preparing the data
 ### 3.1 - Split data
 We will use 80% of the data for training and 20% for validation.
 x will be input data and y the expected output
 %% Cell type:code id: tags:
 ``` python
 # ---- Split => train, test
 #
 data_train = data.sample(frac=0.7, axis=0)
 data_test  = data.drop(data_train.index)
 # ---- Split => x,y (medv is price)
 #
 x_train = data_train.drop('medv',  axis=1)
 y_train = data_train['medv']
 x_test  = data_test.drop('medv',   axis=1)
 y_test  = data_test['medv']
 print('Original data shape was : ',data.shape)
 print('x_train : ',x_train.shape, 'y_train : ',y_train.shape)
 print('x_test  : ',x_test.shape,  'y_test  : ',y_test.shape)
 ```
 %% Output
    Original data shape was :  (506, 14)
    x_train :  (354, 13) y_train :  (354,)
    x_test  :  (152, 13) y_test  :  (152,)
 %% Cell type:markdown id: tags:
 ### 3.2 - Data normalization
 **Note :**
 - All input data must be normalized, train and test.
 - To do this we will subtract the mean and divide by the standard deviation.
 - But test data should not be used in any way, even for normalization.
 - The mean and the standard deviation will therefore only be calculated with the train data.
 %% Cell type:code id: tags:
 ``` python
 display(x_train.describe().style.format("{0:.2f}").set_caption("Before normalization :"))
 mean = x_train.mean()
 std  = x_train.std()
 x_train = (x_train - mean) / std
 x_test  = (x_test  - mean) / std
 display(x_train.describe().style.format("{0:.2f}").set_caption("After normalization :"))
 x_train, y_train = np.array(x_train), np.array(y_train)
 x_test,  y_test  = np.array(x_test),  np.array(y_test)
 ```
 %% Output
 %% Cell type:markdown id: tags:
 ## Step 4 - Build a model
 More informations about :
 - [Optimizer](https://www.tensorflow.org/api_docs/python/tf/keras/optimizers)
 - [Activation](https://www.tensorflow.org/api_docs/python/tf/keras/activations)
 - [Loss](https://www.tensorflow.org/api_docs/python/tf/keras/losses)
 - [Metrics](https://www.tensorflow.org/api_docs/python/tf/keras/metrics)
 %% Cell type:code id: tags:
 ``` python
  def get_model_v1(shape):
    model = keras.models.Sequential()
    model.add(keras.layers.Input(shape, name="InputLayer"))
    model.add(keras.layers.Dense(64, activation='relu', name='Dense_n1'))
    model.add(keras.layers.Dense(64, activation='relu', name='Dense_n2'))
    model.add(keras.layers.Dense(1, name='Output'))
    model.compile(optimizer = 'rmsprop',
                  loss      = 'mse',
                  metrics   = ['mae', 'mse'] )
    return model
 ```
 %% Cell type:markdown id: tags:
 ## 5 - Train the model
 ### 5.1 - Get it
 %% Cell type:code id: tags:
 ``` python
 model=get_model_v1( (13,) )
 model.summary()
 img=keras.utils.plot_model( model, to_file='./run/model.png', show_shapes=True, show_layer_names=True, dpi=96)
 display(img)
 ```
 %% Output
    Model: "sequential"
    _________________________________________________________________
    Layer (type)                 Output Shape              Param #
    =================================================================
    Dense_n1 (Dense)             (None, 64)                896
    _________________________________________________________________
    Dense_n2 (Dense)             (None, 64)                4160
    _________________________________________________________________
    Output (Dense)               (None, 1)                 65
    =================================================================
    Total params: 5,121
    Trainable params: 5,121
    Non-trainable params: 0
    _________________________________________________________________
 %% Cell type:markdown id: tags:
 ### 5.2 - Add callback
 %% Cell type:code id: tags:
 ``` python
 os.makedirs('./run/models',   mode=0o750, exist_ok=True)
 save_dir = "./run/models/best_model.h5"
 savemodel_callback = tf.keras.callbacks.ModelCheckpoint(filepath=save_dir, verbose=0, save_best_only=True)
 ```
 %% Cell type:markdown id: tags:
 ### 5.3 - Train it
 %% Cell type:code id: tags:
 ``` python
 history = model.fit(x_train,
                    y_train,
                    epochs          = 100,
                    batch_size      = 10,
                    verbose         = 1,
                    validation_data = (x_test, y_test),
                    callbacks       = [savemodel_callback])
 ```
 %% Output
    Train on 354 samples, validate on 152 samples
    Epoch 1/100
    354/354 [==============================] - 1s 3ms/sample - loss: 452.0200 - mae: 19.1483 - mse: 452.0200 - val_loss: 295.4043 - val_mae: 15.2476 - val_mse: 295.4043
    Epoch 2/100
    354/354 [==============================] - 0s 318us/sample - loss: 203.8159 - mae: 11.9110 - mse: 203.8159 - val_loss: 102.2654 - val_mae: 8.2041 - val_mse: 102.2654
    Epoch 3/100
    354/354 [==============================] - 0s 322us/sample - loss: 82.2725 - mae: 6.6720 - mse: 82.2725 - val_loss: 48.9591 - val_mae: 5.3325 - val_mse: 48.9591
    Epoch 4/100
    354/354 [==============================] - 0s 330us/sample - loss: 47.3532 - mae: 4.9323 - mse: 47.3532 - val_loss: 30.3381 - val_mae: 4.0806 - val_mse: 30.3381
    Epoch 5/100
    354/354 [==============================] - 0s 332us/sample - loss: 34.2380 - mae: 4.1989 - mse: 34.2380 - val_loss: 25.3102 - val_mae: 3.6149 - val_mse: 25.3102
    Epoch 6/100
    354/354 [==============================] - 0s 330us/sample - loss: 27.7203 - mae: 3.7333 - mse: 27.7203 - val_loss: 24.1136 - val_mae: 3.4089 - val_mse: 24.1136
    Epoch 7/100
    354/354 [==============================] - 0s 336us/sample - loss: 23.4702 - mae: 3.4503 - mse: 23.4702 - val_loss: 21.9095 - val_mae: 3.1906 - val_mse: 21.9095
    Epoch 8/100
    354/354 [==============================] - 0s 322us/sample - loss: 19.8215 - mae: 3.1687 - mse: 19.8215 - val_loss: 21.9063 - val_mae: 3.2564 - val_mse: 21.9063
    Epoch 9/100
    354/354 [==============================] - 0s 326us/sample - loss: 17.6146 - mae: 2.9640 - mse: 17.6146 - val_loss: 19.1573 - val_mae: 2.9280 - val_mse: 19.1573
    Epoch 10/100
    354/354 [==============================] - 0s 288us/sample - loss: 15.9631 - mae: 2.8267 - mse: 15.9631 - val_loss: 19.1600 - val_mae: 2.8806 - val_mse: 19.1600
    Epoch 11/100
    354/354 [==============================] - 0s 326us/sample - loss: 14.4344 - mae: 2.6588 - mse: 14.4344 - val_loss: 18.0972 - val_mae: 2.7704 - val_mse: 18.0972
    Epoch 12/100
    354/354 [==============================] - 0s 330us/sample - loss: 13.3890 - mae: 2.5821 - mse: 13.3890 - val_loss: 18.0529 - val_mae: 2.7683 - val_mse: 18.0529
    Epoch 13/100
    354/354 [==============================] - 0s 326us/sample - loss: 12.7002 - mae: 2.5117 - mse: 12.7002 - val_loss: 17.7848 - val_mae: 2.6781 - val_mse: 17.7848
    Epoch 14/100
    354/354 [==============================] - 0s 279us/sample - loss: 11.8030 - mae: 2.4625 - mse: 11.8030 - val_loss: 18.4840 - val_mae: 2.7626 - val_mse: 18.4840
    Epoch 15/100
    354/354 [==============================] - 0s 331us/sample - loss: 11.4627 - mae: 2.3904 - mse: 11.4627 - val_loss: 17.1289 - val_mae: 2.6199 - val_mse: 17.1289
    Epoch 16/100
    354/354 [==============================] - 0s 284us/sample - loss: 11.1781 - mae: 2.3387 - mse: 11.1781 - val_loss: 17.9369 - val_mae: 2.6804 - val_mse: 17.9369
    Epoch 17/100
    354/354 [==============================] - 0s 326us/sample - loss: 10.7485 - mae: 2.3250 - mse: 10.7485 - val_loss: 16.6649 - val_mae: 2.5390 - val_mse: 16.6649
    Epoch 18/100
    354/354 [==============================] - 0s 294us/sample - loss: 10.5149 - mae: 2.2548 - mse: 10.5149 - val_loss: 18.1112 - val_mae: 2.6858 - val_mse: 18.1112
    Epoch 19/100
    354/354 [==============================] - 0s 295us/sample - loss: 10.4495 - mae: 2.2872 - mse: 10.4495 - val_loss: 17.9377 - val_mae: 2.6937 - val_mse: 17.9377
    Epoch 20/100
    354/354 [==============================] - 0s 277us/sample - loss: 10.2075 - mae: 2.2586 - mse: 10.2075 - val_loss: 17.5565 - val_mae: 2.5374 - val_mse: 17.5565
    Epoch 21/100
    354/354 [==============================] - 0s 325us/sample - loss: 10.0869 - mae: 2.2396 - mse: 10.0869 - val_loss: 16.2770 - val_mae: 2.4551 - val_mse: 16.2770
    Epoch 22/100
    354/354 [==============================] - 0s 278us/sample - loss: 9.5957 - mae: 2.1407 - mse: 9.5957 - val_loss: 17.8160 - val_mae: 2.6874 - val_mse: 17.8160
    Epoch 23/100
    354/354 [==============================] - 0s 304us/sample - loss: 9.7569 - mae: 2.2057 - mse: 9.7569 - val_loss: 15.5761 - val_mae: 2.4537 - val_mse: 15.5761
    Epoch 24/100
    354/354 [==============================] - 0s 286us/sample - loss: 9.4873 - mae: 2.1643 - mse: 9.4873 - val_loss: 16.8661 - val_mae: 2.5735 - val_mse: 16.8661
    Epoch 25/100
    354/354 [==============================] - 0s 287us/sample - loss: 9.0956 - mae: 2.1422 - mse: 9.0956 - val_loss: 16.5815 - val_mae: 2.4812 - val_mse: 16.5815
    Epoch 26/100
    354/354 [==============================] - 0s 288us/sample - loss: 9.3352 - mae: 2.1471 - mse: 9.3352 - val_loss: 16.0146 - val_mae: 2.4404 - val_mse: 16.0146
    Epoch 27/100
    354/354 [==============================] - 0s 286us/sample - loss: 8.6794 - mae: 2.0948 - mse: 8.6794 - val_loss: 18.2565 - val_mae: 2.7272 - val_mse: 18.2565
    Epoch 28/100
    354/354 [==============================] - 0s 294us/sample - loss: 8.9854 - mae: 2.1159 - mse: 8.9854 - val_loss: 16.4515 - val_mae: 2.5282 - val_mse: 16.4515
    Epoch 29/100
    354/354 [==============================] - 0s 286us/sample - loss: 8.8348 - mae: 2.1032 - mse: 8.8348 - val_loss: 17.2604 - val_mae: 2.5932 - val_mse: 17.2604
    Epoch 30/100
    354/354 [==============================] - 0s 244us/sample - loss: 8.7365 - mae: 2.0970 - mse: 8.7365 - val_loss: 16.1155 - val_mae: 2.4509 - val_mse: 16.1155
    Epoch 31/100
    354/354 [==============================] - 0s 282us/sample - loss: 8.6290 - mae: 2.0487 - mse: 8.6290 - val_loss: 16.9125 - val_mae: 2.5010 - val_mse: 16.9125
    Epoch 32/100
    354/354 [==============================] - 0s 280us/sample - loss: 8.6531 - mae: 2.0411 - mse: 8.6531 - val_loss: 15.7585 - val_mae: 2.4288 - val_mse: 15.7585
    Epoch 33/100
    354/354 [==============================] - 0s 330us/sample - loss: 8.6551 - mae: 2.0516 - mse: 8.6551 - val_loss: 15.4765 - val_mae: 2.4073 - val_mse: 15.4765
    Epoch 34/100
    354/354 [==============================] - 0s 292us/sample - loss: 8.4218 - mae: 2.0072 - mse: 8.4218 - val_loss: 16.2900 - val_mae: 2.5081 - val_mse: 16.2900
    Epoch 35/100
    354/354 [==============================] - 0s 293us/sample - loss: 8.3149 - mae: 1.9851 - mse: 8.3149 - val_loss: 15.7184 - val_mae: 2.4337 - val_mse: 15.7184
    Epoch 36/100
    354/354 [==============================] - 0s 289us/sample - loss: 8.4496 - mae: 2.0142 - mse: 8.4496 - val_loss: 16.2760 - val_mae: 2.5489 - val_mse: 16.2760
    Epoch 37/100
    354/354 [==============================] - 0s 334us/sample - loss: 8.0962 - mae: 1.9872 - mse: 8.0962 - val_loss: 15.3895 - val_mae: 2.3543 - val_mse: 15.3895
    Epoch 38/100
    354/354 [==============================] - 0s 298us/sample - loss: 8.1599 - mae: 1.9882 - mse: 8.1599 - val_loss: 16.0081 - val_mae: 2.3977 - val_mse: 16.0081
    Epoch 39/100
    354/354 [==============================] - 0s 286us/sample - loss: 7.9958 - mae: 1.9817 - mse: 7.9958 - val_loss: 15.8999 - val_mae: 2.4583 - val_mse: 15.8999
    Epoch 40/100
    354/354 [==============================] - 0s 282us/sample - loss: 7.8666 - mae: 1.9441 - mse: 7.8666 - val_loss: 16.8131 - val_mae: 2.6931 - val_mse: 16.8131
    Epoch 41/100
    354/354 [==============================] - 0s 293us/sample - loss: 7.9312 - mae: 1.9216 - mse: 7.9312 - val_loss: 15.4608 - val_mae: 2.3995 - val_mse: 15.4608
    Epoch 42/100
    354/354 [==============================] - 0s 286us/sample - loss: 7.6752 - mae: 1.9127 - mse: 7.6752 - val_loss: 15.8675 - val_mae: 2.5118 - val_mse: 15.8675
    Epoch 43/100
    354/354 [==============================] - 0s 332us/sample - loss: 7.7535 - mae: 1.9296 - mse: 7.7535 - val_loss: 15.2040 - val_mae: 2.3731 - val_mse: 15.2040
    Epoch 44/100
    354/354 [==============================] - 0s 331us/sample - loss: 7.6188 - mae: 1.9150 - mse: 7.6188 - val_loss: 15.0409 - val_mae: 2.3680 - val_mse: 15.0409
    Epoch 45/100
    354/354 [==============================] - 0s 284us/sample - loss: 7.6286 - mae: 1.8755 - mse: 7.6286 - val_loss: 15.1650 - val_mae: 2.3595 - val_mse: 15.1650
    Epoch 46/100
    354/354 [==============================] - 0s 278us/sample - loss: 7.7937 - mae: 1.9318 - mse: 7.7937 - val_loss: 15.7196 - val_mae: 2.4218 - val_mse: 15.7196
    Epoch 47/100
    354/354 [==============================] - 0s 275us/sample - loss: 7.4244 - mae: 1.9022 - mse: 7.4244 - val_loss: 15.5651 - val_mae: 2.4811 - val_mse: 15.5651
    Epoch 48/100
    354/354 [==============================] - 0s 330us/sample - loss: 7.4042 - mae: 1.9083 - mse: 7.4042 - val_loss: 14.7377 - val_mae: 2.3598 - val_mse: 14.7377
    Epoch 49/100
    354/354 [==============================] - 0s 270us/sample - loss: 7.3230 - mae: 1.8741 - mse: 7.3230 - val_loss: 15.2313 - val_mae: 2.4210 - val_mse: 15.2313
    Epoch 50/100
    354/354 [==============================] - 0s 276us/sample - loss: 7.3075 - mae: 1.8614 - mse: 7.3075 - val_loss: 14.7584 - val_mae: 2.3305 - val_mse: 14.7584
    Epoch 51/100
    354/354 [==============================] - 0s 270us/sample - loss: 7.4376 - mae: 1.8956 - mse: 7.4376 - val_loss: 15.3226 - val_mae: 2.3742 - val_mse: 15.3226
    Epoch 52/100
    354/354 [==============================] - 0s 287us/sample - loss: 7.1467 - mae: 1.8380 - mse: 7.1467 - val_loss: 15.5150 - val_mae: 2.4291 - val_mse: 15.5150
    Epoch 53/100
    354/354 [==============================] - 0s 271us/sample - loss: 6.9376 - mae: 1.8018 - mse: 6.9376 - val_loss: 16.2807 - val_mae: 2.5440 - val_mse: 16.2807
    Epoch 54/100
    354/354 [==============================] - 0s 289us/sample - loss: 7.0885 - mae: 1.8170 - mse: 7.0885 - val_loss: 15.2975 - val_mae: 2.4258 - val_mse: 15.2975
    Epoch 55/100
    354/354 [==============================] - 0s 284us/sample - loss: 7.0596 - mae: 1.8107 - mse: 7.0596 - val_loss: 15.7460 - val_mae: 2.4825 - val_mse: 15.7460
    Epoch 56/100
    354/354 [==============================] - 0s 288us/sample - loss: 6.7812 - mae: 1.8335 - mse: 6.7812 - val_loss: 14.7849 - val_mae: 2.3684 - val_mse: 14.7849
    Epoch 57/100
    354/354 [==============================] - 0s 290us/sample - loss: 6.9172 - mae: 1.8140 - mse: 6.9172 - val_loss: 15.1139 - val_mae: 2.4500 - val_mse: 15.1139
    Epoch 58/100
    354/354 [==============================] - 0s 284us/sample - loss: 6.8010 - mae: 1.7637 - mse: 6.8010 - val_loss: 16.7211 - val_mae: 2.5769 - val_mse: 16.7211
    Epoch 59/100
    354/354 [==============================] - 0s 271us/sample - loss: 6.8954 - mae: 1.8046 - mse: 6.8954 - val_loss: 15.1101 - val_mae: 2.4743 - val_mse: 15.1101
    Epoch 60/100
    354/354 [==============================] - 0s 280us/sample - loss: 6.7740 - mae: 1.7866 - mse: 6.7740 - val_loss: 15.0811 - val_mae: 2.3838 - val_mse: 15.0811
    Epoch 61/100
    354/354 [==============================] - 0s 323us/sample - loss: 6.8996 - mae: 1.7872 - mse: 6.8996 - val_loss: 14.4203 - val_mae: 2.3199 - val_mse: 14.4203
    Epoch 62/100
    354/354 [==============================] - 0s 277us/sample - loss: 6.7188 - mae: 1.7858 - mse: 6.7188 - val_loss: 14.5972 - val_mae: 2.3610 - val_mse: 14.5972
    Epoch 63/100
    354/354 [==============================] - 0s 266us/sample - loss: 6.5708 - mae: 1.7710 - mse: 6.5708 - val_loss: 14.5145 - val_mae: 2.3563 - val_mse: 14.5145
    Epoch 64/100
    354/354 [==============================] - 0s 273us/sample - loss: 6.1133 - mae: 1.6706 - mse: 6.1133 - val_loss: 14.9870 - val_mae: 2.4017 - val_mse: 14.9870
    Epoch 65/100
    354/354 [==============================] - 0s 313us/sample - loss: 6.4980 - mae: 1.7295 - mse: 6.4980 - val_loss: 14.0636 - val_mae: 2.3661 - val_mse: 14.0636
    Epoch 66/100
    354/354 [==============================] - 0s 262us/sample - loss: 6.5237 - mae: 1.7277 - mse: 6.5237 - val_loss: 14.2366 - val_mae: 2.3318 - val_mse: 14.2366
    Epoch 67/100
    354/354 [==============================] - 0s 326us/sample - loss: 6.3067 - mae: 1.7433 - mse: 6.3067 - val_loss: 14.0032 - val_mae: 2.3350 - val_mse: 14.0032
    Epoch 68/100
    354/354 [==============================] - 0s 286us/sample - loss: 6.4447 - mae: 1.7336 - mse: 6.4447 - val_loss: 14.4271 - val_mae: 2.3149 - val_mse: 14.4271
    Epoch 69/100
    354/354 [==============================] - 0s 332us/sample - loss: 6.3821 - mae: 1.7012 - mse: 6.3821 - val_loss: 13.9716 - val_mae: 2.3141 - val_mse: 13.9716
    Epoch 70/100
    354/354 [==============================] - 0s 251us/sample - loss: 6.3734 - mae: 1.7080 - mse: 6.3734 - val_loss: 14.9184 - val_mae: 2.4716 - val_mse: 14.9184
    Epoch 71/100
    354/354 [==============================] - 0s 321us/sample - loss: 6.4273 - mae: 1.7281 - mse: 6.4273 - val_loss: 13.8686 - val_mae: 2.3176 - val_mse: 13.8686
    Epoch 72/100
    354/354 [==============================] - 0s 285us/sample - loss: 6.2473 - mae: 1.6967 - mse: 6.2473 - val_loss: 14.2249 - val_mae: 2.3450 - val_mse: 14.2249
    Epoch 73/100
    354/354 [==============================] - 0s 286us/sample - loss: 6.3427 - mae: 1.7034 - mse: 6.3427 - val_loss: 14.3159 - val_mae: 2.3431 - val_mse: 14.3159
    Epoch 74/100
    354/354 [==============================] - 0s 287us/sample - loss: 6.0929 - mae: 1.6752 - mse: 6.0929 - val_loss: 14.2151 - val_mae: 2.3644 - val_mse: 14.2151
    Epoch 75/100
    354/354 [==============================] - 0s 289us/sample - loss: 6.1445 - mae: 1.6985 - mse: 6.1445 - val_loss: 14.8251 - val_mae: 2.4202 - val_mse: 14.8251
    Epoch 76/100
    354/354 [==============================] - 0s 311us/sample - loss: 6.2184 - mae: 1.6867 - mse: 6.2184 - val_loss: 14.0596 - val_mae: 2.3274 - val_mse: 14.0596
    Epoch 77/100
    354/354 [==============================] - 0s 340us/sample - loss: 6.1201 - mae: 1.6785 - mse: 6.1201 - val_loss: 13.4886 - val_mae: 2.2769 - val_mse: 13.4886
    Epoch 78/100
    354/354 [==============================] - 0s 286us/sample - loss: 5.9001 - mae: 1.6716 - mse: 5.9001 - val_loss: 14.0295 - val_mae: 2.3214 - val_mse: 14.0295
    Epoch 79/100
    354/354 [==============================] - 0s 284us/sample - loss: 6.0389 - mae: 1.6783 - mse: 6.0389 - val_loss: 14.0250 - val_mae: 2.3245 - val_mse: 14.0250
    Epoch 80/100
    354/354 [==============================] - 0s 288us/sample - loss: 5.8268 - mae: 1.6458 - mse: 5.8268 - val_loss: 15.2746 - val_mae: 2.4834 - val_mse: 15.2746
    Epoch 81/100
    354/354 [==============================] - 0s 284us/sample - loss: 5.8671 - mae: 1.6680 - mse: 5.8671 - val_loss: 14.4935 - val_mae: 2.4353 - val_mse: 14.4935
    Epoch 82/100
    354/354 [==============================] - 0s 274us/sample - loss: 5.8115 - mae: 1.6742 - mse: 5.8115 - val_loss: 14.6922 - val_mae: 2.3631 - val_mse: 14.6922
    Epoch 83/100
    354/354 [==============================] - 0s 331us/sample - loss: 5.8561 - mae: 1.6727 - mse: 5.8561 - val_loss: 13.4236 - val_mae: 2.2982 - val_mse: 13.4236
    Epoch 84/100
    354/354 [==============================] - 0s 290us/sample - loss: 5.7500 - mae: 1.5833 - mse: 5.7500 - val_loss: 14.4867 - val_mae: 2.4330 - val_mse: 14.4867
    Epoch 85/100
    354/354 [==============================] - 0s 286us/sample - loss: 5.6700 - mae: 1.6435 - mse: 5.6700 - val_loss: 13.9873 - val_mae: 2.3614 - val_mse: 13.9873
    Epoch 86/100
    354/354 [==============================] - 0s 268us/sample - loss: 5.6816 - mae: 1.6524 - mse: 5.6816 - val_loss: 13.4864 - val_mae: 2.3505 - val_mse: 13.4864
    Epoch 87/100
    354/354 [==============================] - 0s 267us/sample - loss: 5.5838 - mae: 1.6220 - mse: 5.5838 - val_loss: 15.4727 - val_mae: 2.5215 - val_mse: 15.4727
    Epoch 88/100
    354/354 [==============================] - 0s 284us/sample - loss: 5.6117 - mae: 1.6208 - mse: 5.6117 - val_loss: 13.6392 - val_mae: 2.3150 - val_mse: 13.6392
    Epoch 89/100
    354/354 [==============================] - 0s 324us/sample - loss: 5.5648 - mae: 1.6051 - mse: 5.5648 - val_loss: 13.2082 - val_mae: 2.2858 - val_mse: 13.2082
    Epoch 90/100
    354/354 [==============================] - 0s 288us/sample - loss: 5.6019 - mae: 1.5946 - mse: 5.6019 - val_loss: 13.7882 - val_mae: 2.3677 - val_mse: 13.7882
    Epoch 91/100
    354/354 [==============================] - 0s 336us/sample - loss: 5.4979 - mae: 1.6000 - mse: 5.4979 - val_loss: 12.9619 - val_mae: 2.2898 - val_mse: 12.9619
    Epoch 92/100
    354/354 [==============================] - 0s 294us/sample - loss: 5.4595 - mae: 1.5815 - mse: 5.4595 - val_loss: 13.8617 - val_mae: 2.3735 - val_mse: 13.8617
    Epoch 93/100
    354/354 [==============================] - 0s 321us/sample - loss: 5.1999 - mae: 1.6043 - mse: 5.1999 - val_loss: 12.9011 - val_mae: 2.3053 - val_mse: 12.9011
    Epoch 94/100
    354/354 [==============================] - 0s 268us/sample - loss: 5.2630 - mae: 1.5463 - mse: 5.2630 - val_loss: 14.3610 - val_mae: 2.4348 - val_mse: 14.3610
    Epoch 95/100
    354/354 [==============================] - 0s 300us/sample - loss: 5.3272 - mae: 1.6047 - mse: 5.3272 - val_loss: 12.9650 - val_mae: 2.2989 - val_mse: 12.9650
    Epoch 96/100
    354/354 [==============================] - 0s 283us/sample - loss: 5.3137 - mae: 1.5796 - mse: 5.3137 - val_loss: 13.9091 - val_mae: 2.3669 - val_mse: 13.9091
    Epoch 97/100
    354/354 [==============================] - 0s 280us/sample - loss: 5.2891 - mae: 1.5773 - mse: 5.2891 - val_loss: 13.2578 - val_mae: 2.3012 - val_mse: 13.2578
    Epoch 98/100
    354/354 [==============================] - 0s 264us/sample - loss: 5.3977 - mae: 1.5920 - mse: 5.3977 - val_loss: 13.8690 - val_mae: 2.4075 - val_mse: 13.8690
    Epoch 99/100
    354/354 [==============================] - 0s 273us/sample - loss: 5.3071 - mae: 1.5391 - mse: 5.3071 - val_loss: 12.9043 - val_mae: 2.2816 - val_mse: 12.9043
    Epoch 100/100
    354/354 [==============================] - 0s 318us/sample - loss: 5.2748 - mae: 1.5458 - mse: 5.2748 - val_loss: 12.6915 - val_mae: 2.2683 - val_mse: 12.6915
 %% Cell type:markdown id: tags:
 ## Step 6 - Evaluate
 ### 6.1 - Model evaluation
 MAE =  Mean Absolute Error (between the labels and predictions)
 A mae equal to 3 represents an average error in prediction of $3k.
 %% Cell type:code id: tags:
 ``` python
 score = model.evaluate(x_test, y_test, verbose=0)
 print('x_test / loss      : {:5.4f}'.format(score[0]))
 print('x_test / mae       : {:5.4f}'.format(score[1]))
 print('x_test / mse       : {:5.4f}'.format(score[2]))
 ```
 %% Output
    x_test / loss      : 12.6915
    x_test / mae       : 2.2683
    x_test / mse       : 12.6915
 %% Cell type:markdown id: tags:
 ### 6.2 - Training history
 What was the best result during our training ?
 %% Cell type:code id: tags:
 ``` python
 print("min( val_mae ) : {:.4f}".format( min(history.history["val_mae"]) ) )
 ```
 %% Output
    min( val_mae ) : 2.2683
 %% Cell type:code id: tags:
 ``` python
 ooo.plot_history(history, plot={'MSE' :['mse', 'val_mse'],
                                'MAE' :['mae', 'val_mae'],
                                'LOSS':['loss','val_loss']})
 ```
 %% Output
 %% Cell type:markdown id: tags:
 ## Step 7 - Restore a model :
 %% Cell type:markdown id: tags:
 ### 7.1 - Reload model
 %% Cell type:code id: tags:
 ``` python
 loaded_model = tf.keras.models.load_model('./run/models/best_model.h5')
 loaded_model.summary()
 print("Loaded.")
 ```
 %% Output
    Model: "sequential"
    _________________________________________________________________
    Layer (type)                 Output Shape              Param #
    =================================================================
    Dense_n1 (Dense)             (None, 64)                896
    _________________________________________________________________
    Dense_n2 (Dense)             (None, 64)                4160
    _________________________________________________________________
    Output (Dense)               (None, 1)                 65
    =================================================================
    Total params: 5,121
    Trainable params: 5,121
    Non-trainable params: 0
    _________________________________________________________________
    Loaded.
 %% Cell type:markdown id: tags:
 ### 7.2 - Evaluate it :
 %% Cell type:code id: tags:
 ``` python
 score = loaded_model.evaluate(x_test, y_test, verbose=0)
 print('x_test / loss      : {:5.4f}'.format(score[0]))
 print('x_test / mae       : {:5.4f}'.format(score[1]))
 print('x_test / mse       : {:5.4f}'.format(score[2]))
 ```
 %% Output
    x_test / loss      : 12.6915
    x_test / mae       : 2.2683
    x_test / mse       : 12.6915
 %% Cell type:markdown id: tags:
 ### 7.3 - Make a prediction
 %% Cell type:code id: tags:
 ``` python
 my_data = [ 1.26425925, -0.48522739,  1.0436489 , -0.23112788,  1.37120745,
       -2.14308942,  1.13489104, -1.06802005,  1.71189006,  1.57042287,
        0.77859951,  0.14769795,  2.7585581 ]
 real_price = 10.4
 my_data=np.array(my_data).reshape(1,13)
 ```
 %% Cell type:code id: tags:
 ``` python
 predictions = loaded_model.predict( my_data )
-print("Prédiction : {:.2f} K$   Reality : {:.2f} K$".format(predictions[0][0], real_price))
+print("Prediction : {:.2f} K$   Reality : {:.2f} K$".format(predictions[0][0], real_price))
 ```
 %% Output
    Prédiction : 10.75 K$   Reality : 10.40 K$
 %% Cell type:markdown id: tags:
 ---
 <img width="80px" src="../fidle/img/00-Fidle-logo-01.svg"></img>