Contenu du didacticiel
Preparation1. Vagrant et Virtual Box.
2. Apache
3. MySQL
4. Python
5. Python Hello World
Apprentissage profond
1. Installer Keras et tester l'apprentissage profond (Deep learning)
2. Enregistrer et charger les paramètres
3. Enregistrer et charger les paramètres en même temps
4. Utiliser son propre le jeu de données
Enregistrer et charger les paramètres
Code original
On va créer un nouveau fichier qui apprend comment catégoriser les reuters news. On va utiliser MNIST dataset.
Creér un fichier "reuters.py" dans le dossier partagée comme ça:
Écrire ce code dans le "reuters.py" (source: fchollet/keras):
#!/usr/bin/env python
# -*- coding: UTF-8 -*-
'''Trains and evaluate a simple MLP
on the Reuters newswire topic classification task.
'''
from __future__ import print_function
import numpy as np
import keras
from keras.datasets import reuters
from keras.models import Sequential
from keras.layers import Dense, Dropout, Activation
from keras.preprocessing.text import Tokenizer
max_words = 1000
batch_size = 32
epochs = 5
print('Loading data...')
(x_train, y_train), (x_test, y_test) = reuters.load_data(num_words=max_words,
test_split=0.2)
print(len(x_train), 'train sequences')
print(len(x_test), 'test sequences')
num_classes = np.max(y_train) + 1
print(num_classes, 'classes')
print('Vectorizing sequence data...')
tokenizer = Tokenizer(num_words=max_words)
x_train = tokenizer.sequences_to_matrix(x_train, mode='binary')
x_test = tokenizer.sequences_to_matrix(x_test, mode='binary')
print('x_train shape:', x_train.shape)
print('x_test shape:', x_test.shape)
print('Convert class vector to binary class matrix '
'(for use with categorical_crossentropy)')
y_train = keras.utils.to_categorical(y_train, num_classes)
y_test = keras.utils.to_categorical(y_test, num_classes)
print('y_train shape:', y_train.shape)
print('y_test shape:', y_test.shape)
print('Building model...')
model = Sequential()
model.add(Dense(512, input_shape=(max_words,)))
model.add(Activation('relu'))
model.add(Dropout(0.5))
model.add(Dense(num_classes))
model.add(Activation('softmax'))
model.compile(loss='categorical_crossentropy',
optimizer='adam',
metrics=['accuracy'])
history = model.fit(x_train, y_train,
batch_size=batch_size,
epochs=epochs,
verbose=1,
validation_split=0.1)
score = model.evaluate(x_test, y_test,
batch_size=batch_size, verbose=1)
print('Test score:', score[0])
print('Test accuracy:', score[1])
# -*- coding: UTF-8 -*-
'''Trains and evaluate a simple MLP
on the Reuters newswire topic classification task.
'''
from __future__ import print_function
import numpy as np
import keras
from keras.datasets import reuters
from keras.models import Sequential
from keras.layers import Dense, Dropout, Activation
from keras.preprocessing.text import Tokenizer
max_words = 1000
batch_size = 32
epochs = 5
print('Loading data...')
(x_train, y_train), (x_test, y_test) = reuters.load_data(num_words=max_words,
test_split=0.2)
print(len(x_train), 'train sequences')
print(len(x_test), 'test sequences')
num_classes = np.max(y_train) + 1
print(num_classes, 'classes')
print('Vectorizing sequence data...')
tokenizer = Tokenizer(num_words=max_words)
x_train = tokenizer.sequences_to_matrix(x_train, mode='binary')
x_test = tokenizer.sequences_to_matrix(x_test, mode='binary')
print('x_train shape:', x_train.shape)
print('x_test shape:', x_test.shape)
print('Convert class vector to binary class matrix '
'(for use with categorical_crossentropy)')
y_train = keras.utils.to_categorical(y_train, num_classes)
y_test = keras.utils.to_categorical(y_test, num_classes)
print('y_train shape:', y_train.shape)
print('y_test shape:', y_test.shape)
print('Building model...')
model = Sequential()
model.add(Dense(512, input_shape=(max_words,)))
model.add(Activation('relu'))
model.add(Dropout(0.5))
model.add(Dense(num_classes))
model.add(Activation('softmax'))
model.compile(loss='categorical_crossentropy',
optimizer='adam',
metrics=['accuracy'])
history = model.fit(x_train, y_train,
batch_size=batch_size,
epochs=epochs,
verbose=1,
validation_split=0.1)
score = model.evaluate(x_test, y_test,
batch_size=batch_size, verbose=1)
print('Test score:', score[0])
print('Test accuracy:', score[1])
Et faire ces commandes dans Teraterm:
$ cd /vagrant
$ python3 reuters.py
$ python3 reuters.py
D'abord, la précision n'est pas bonne.
L'apprentissage est fini. La précision est 79%.
Enregistrer les paramètres
Pour enregistrer les paramètres, on va utiliser save_weights/load_weights comme ça:
model.save_weights('param.hdf5')
model.load_weights('param.hdf5')
model.load_weights('param.hdf5')
Pour enregistrer les paramètres pendant l'exécution, on va utiliser les fonctions de rappel (ça s'appelle "ModelCheckpoint" en Keras). Les fonctions de rappel sont appelé à la fin de chaque époque.
Arguments
filepath : le chemin d'accès d'un fichier où les paramètres sons enregistré.
monitor: quelle valeur doit être surveillée.
verbose : si les commentaires doivent être imprimés ou non.
save_best_only: enregistrer uniquement lorsque la précision est meilleure.
mode: comment les valeurs sont enregistrés.
filepath : le chemin d'accès d'un fichier où les paramètres sons enregistré.
monitor: quelle valeur doit être surveillée.
verbose : si les commentaires doivent être imprimés ou non.
save_best_only: enregistrer uniquement lorsque la précision est meilleure.
mode: comment les valeurs sont enregistrés.
On va créer un fichier "reuters_save.py" est deux dossiers "model" et "log".
Écrire ce code dans le "reuters.py":
#!/usr/bin/env python
# -*- coding: UTF-8 -*-
'''Trains and evaluate a simple MLP
on the Reuters newswire topic classification task.
'''
from __future__ import print_function
import numpy as np
import keras
from keras.datasets import reuters
from keras.models import Sequential
from keras.layers import Dense, Dropout, Activation
from keras.layers.convolutional import Convolution2D, MaxPooling2D
from keras.layers.normalization import BatchNormalization
from keras.utils import np_utils
from keras.optimizers import Adam
import keras.callbacks
from keras.preprocessing.text import Tokenizer
import keras.backend.tensorflow_backend as KTF
import tensorflow as tf
import os.path
f_log = './'
f_model = './'
max_words = 1000
batch_size = 32
epochs = 5
print('Loading data...')
(x_train, y_train), (x_test, y_test) = reuters.load_data(num_words=max_words,
test_split=0.2)
print(len(x_train), 'train sequences')
print(len(x_test), 'test sequences')
num_classes = np.max(y_train) + 1
print(num_classes, 'classes')
print('Vectorizing sequence data...')
tokenizer = Tokenizer(num_words=max_words)
x_train = tokenizer.sequences_to_matrix(x_train, mode='binary')
x_test = tokenizer.sequences_to_matrix(x_test, mode='binary')
print('x_train shape:', x_train.shape)
print('x_test shape:', x_test.shape)
print('Convert class vector to binary class matrix '
'(for use with categorical_crossentropy)')
y_train = keras.utils.to_categorical(y_train, num_classes)
y_test = keras.utils.to_categorical(y_test, num_classes)
old_session = KTF.get_session()
print('y_train shape:', y_train.shape)
print('y_test shape:', y_test.shape)
print('Building model...')
model = Sequential()
model.add(Dense(512, input_shape=(max_words,)))
model.add(Activation('relu'))
model.add(Dropout(0.5))
model.add(Dense(num_classes))
model.add(Activation('softmax'))
model.compile(loss='categorical_crossentropy',
optimizer='adam',
metrics=['accuracy'])
tb_cb = keras.callbacks.TensorBoard(log_dir=f_log, histogram_freq=0)
cp_cb = keras.callbacks.ModelCheckpoint(filepath = os.path.join(f_model,'reuters_model{epoch:02d}-loss{loss:.2f}-acc{acc:.2f}-vloss{val_loss:.2f}-vacc{val_acc:.2f}.hdf5'), monitor='val_loss', verbose=1, save_best_only=True, mode='auto')
cbks = [tb_cb, cp_cb]
history = model.fit(x_train, y_train,
batch_size=batch_size,
epochs=epochs,
verbose=1,
validation_split=0.1,
callbacks=cbks)
score = model.evaluate(x_test, y_test,
batch_size=batch_size, verbose=1)
print('')
print('Test score:', score[0])
print('Test accuracy:', score[1])
print('save the architecture of a model')
json_string = model.to_json()
open(os.path.join(f_model,'reuters_model.json'), 'w').write(json_string)
yaml_string = model.to_yaml()
open(os.path.join(f_model,'reuters_model.yaml'), 'w').write(yaml_string)
print('save weights')
model.save_weights(os.path.join(f_model,'reuters_model_weights.hdf5'))
KTF.set_session(old_session)
# -*- coding: UTF-8 -*-
'''Trains and evaluate a simple MLP
on the Reuters newswire topic classification task.
'''
from __future__ import print_function
import numpy as np
import keras
from keras.datasets import reuters
from keras.models import Sequential
from keras.layers import Dense, Dropout, Activation
from keras.layers.convolutional import Convolution2D, MaxPooling2D
from keras.layers.normalization import BatchNormalization
from keras.utils import np_utils
from keras.optimizers import Adam
import keras.callbacks
from keras.preprocessing.text import Tokenizer
import keras.backend.tensorflow_backend as KTF
import tensorflow as tf
import os.path
f_log = './'
f_model = './'
max_words = 1000
batch_size = 32
epochs = 5
print('Loading data...')
(x_train, y_train), (x_test, y_test) = reuters.load_data(num_words=max_words,
test_split=0.2)
print(len(x_train), 'train sequences')
print(len(x_test), 'test sequences')
num_classes = np.max(y_train) + 1
print(num_classes, 'classes')
print('Vectorizing sequence data...')
tokenizer = Tokenizer(num_words=max_words)
x_train = tokenizer.sequences_to_matrix(x_train, mode='binary')
x_test = tokenizer.sequences_to_matrix(x_test, mode='binary')
print('x_train shape:', x_train.shape)
print('x_test shape:', x_test.shape)
print('Convert class vector to binary class matrix '
'(for use with categorical_crossentropy)')
y_train = keras.utils.to_categorical(y_train, num_classes)
y_test = keras.utils.to_categorical(y_test, num_classes)
old_session = KTF.get_session()
print('y_train shape:', y_train.shape)
print('y_test shape:', y_test.shape)
print('Building model...')
model = Sequential()
model.add(Dense(512, input_shape=(max_words,)))
model.add(Activation('relu'))
model.add(Dropout(0.5))
model.add(Dense(num_classes))
model.add(Activation('softmax'))
model.compile(loss='categorical_crossentropy',
optimizer='adam',
metrics=['accuracy'])
tb_cb = keras.callbacks.TensorBoard(log_dir=f_log, histogram_freq=0)
cp_cb = keras.callbacks.ModelCheckpoint(filepath = os.path.join(f_model,'reuters_model{epoch:02d}-loss{loss:.2f}-acc{acc:.2f}-vloss{val_loss:.2f}-vacc{val_acc:.2f}.hdf5'), monitor='val_loss', verbose=1, save_best_only=True, mode='auto')
cbks = [tb_cb, cp_cb]
history = model.fit(x_train, y_train,
batch_size=batch_size,
epochs=epochs,
verbose=1,
validation_split=0.1,
callbacks=cbks)
score = model.evaluate(x_test, y_test,
batch_size=batch_size, verbose=1)
print('')
print('Test score:', score[0])
print('Test accuracy:', score[1])
print('save the architecture of a model')
json_string = model.to_json()
open(os.path.join(f_model,'reuters_model.json'), 'w').write(json_string)
yaml_string = model.to_yaml()
open(os.path.join(f_model,'reuters_model.yaml'), 'w').write(yaml_string)
print('save weights')
model.save_weights(os.path.join(f_model,'reuters_model_weights.hdf5'))
KTF.set_session(old_session)
Et faire cette commande:
$ python3 reuters_tosave.py
Les paramètres sont enregistrés.
Les paramètres sont enregistrés dans le dossier "Model":
Charger les paramètres
Maintenant, on va charger les paramètres de les fichiers. Créer "reuters_toload.py" dans le dossier partagée:
Écrire ce code dans le "reuters.py":
#!/usr/bin/env python
# -*- coding: UTF-8 -*-
'''Trains and evaluate a simple MLP
on the Reuters newswire topic classification task.
'''
from __future__ import print_function
import numpy as np
import keras
from keras.datasets import reuters
from keras.models import Sequential
from keras.models import model_from_json
from keras.layers import Dense, Dropout, Activation
from keras.preprocessing.text import Tokenizer
from keras.utils import np_utils
from keras.optimizers import Adam
import keras.callbacks
import keras.backend.tensorflow_backend as KTF
import tensorflow as tf
import os.path
f_log = './'
f_model = './'
model_filename = 'reuters_model.json'
weights_filename = 'reuters_model_weights.hdf5'
max_words = 1000
batch_size = 32
epochs = 5
print('Loading data...')
(x_train, y_train), (x_test, y_test) = reuters.load_data(num_words=max_words,
test_split=0.2)
print(len(x_train), 'train sequences')
print(len(x_test), 'test sequences')
num_classes = np.max(y_train) + 1
print(num_classes, 'classes')
print('Vectorizing sequence data...')
tokenizer = Tokenizer(num_words=max_words)
x_train = tokenizer.sequences_to_matrix(x_train, mode='binary')
x_test = tokenizer.sequences_to_matrix(x_test, mode='binary')
print('x_train shape:', x_train.shape)
print('x_test shape:', x_test.shape)
print('Convert class vector to binary class matrix '
'(for use with categorical_crossentropy)')
y_train = keras.utils.to_categorical(y_train, num_classes)
y_test = keras.utils.to_categorical(y_test, num_classes)
old_session = KTF.get_session()
print('y_train shape:', y_train.shape)
print('y_test shape:', y_test.shape)
print('Building model...')
session = tf.Session('')
KTF.set_session(session)
json_string = open(os.path.join(f_model, model_filename)).read()
model = model_from_json(json_string)
model.summary()
model.compile(loss='categorical_crossentropy',
optimizer='adam',
metrics=['accuracy'])
model.load_weights(os.path.join(f_model,weights_filename))
cbks = []
history = model.fit(x_train, y_train,
batch_size=batch_size,
epochs=epochs,
verbose=1,
validation_split=0.1,
callbacks=cbks)
score = model.evaluate(x_test, y_test,
batch_size=batch_size, verbose=1)
print('')
print('Test score:', score[0])
print('Test accuracy:', score[1])
# -*- coding: UTF-8 -*-
'''Trains and evaluate a simple MLP
on the Reuters newswire topic classification task.
'''
from __future__ import print_function
import numpy as np
import keras
from keras.datasets import reuters
from keras.models import Sequential
from keras.models import model_from_json
from keras.layers import Dense, Dropout, Activation
from keras.preprocessing.text import Tokenizer
from keras.utils import np_utils
from keras.optimizers import Adam
import keras.callbacks
import keras.backend.tensorflow_backend as KTF
import tensorflow as tf
import os.path
f_log = './'
f_model = './'
model_filename = 'reuters_model.json'
weights_filename = 'reuters_model_weights.hdf5'
max_words = 1000
batch_size = 32
epochs = 5
print('Loading data...')
(x_train, y_train), (x_test, y_test) = reuters.load_data(num_words=max_words,
test_split=0.2)
print(len(x_train), 'train sequences')
print(len(x_test), 'test sequences')
num_classes = np.max(y_train) + 1
print(num_classes, 'classes')
print('Vectorizing sequence data...')
tokenizer = Tokenizer(num_words=max_words)
x_train = tokenizer.sequences_to_matrix(x_train, mode='binary')
x_test = tokenizer.sequences_to_matrix(x_test, mode='binary')
print('x_train shape:', x_train.shape)
print('x_test shape:', x_test.shape)
print('Convert class vector to binary class matrix '
'(for use with categorical_crossentropy)')
y_train = keras.utils.to_categorical(y_train, num_classes)
y_test = keras.utils.to_categorical(y_test, num_classes)
old_session = KTF.get_session()
print('y_train shape:', y_train.shape)
print('y_test shape:', y_test.shape)
print('Building model...')
session = tf.Session('')
KTF.set_session(session)
json_string = open(os.path.join(f_model, model_filename)).read()
model = model_from_json(json_string)
model.summary()
model.compile(loss='categorical_crossentropy',
optimizer='adam',
metrics=['accuracy'])
model.load_weights(os.path.join(f_model,weights_filename))
cbks = []
history = model.fit(x_train, y_train,
batch_size=batch_size,
epochs=epochs,
verbose=1,
validation_split=0.1,
callbacks=cbks)
score = model.evaluate(x_test, y_test,
batch_size=batch_size, verbose=1)
print('')
print('Test score:', score[0])
print('Test accuracy:', score[1])
Faire cette commande:
$ python3 reuters_toload.py
La précision est bon toujours parce que ça utilise les paramètres enregistré.
Les résultats.
Aucun commentaire:
Enregistrer un commentaire