-
Notifications
You must be signed in to change notification settings - Fork 2
/
food_xception_dp0707_2345p.py
237 lines (203 loc) · 9.18 KB
/
food_xception_dp0707_2345p.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
import numpy as np
from os import listdir
from os.path import isfile, join
import h5py
from sklearn.model_selection import train_test_split
from keras.preprocessing import image
from keras.layers import Input
from keras.preprocessing.image import ImageDataGenerator
from keras.models import Sequential, Model
from keras.layers import Dense, Dropout, Activation, Flatten
from keras.layers import Convolution2D, MaxPooling2D, ZeroPadding2D, GlobalAveragePooling2D, AveragePooling2D
from keras.layers.normalization import BatchNormalization
from keras.preprocessing.image import ImageDataGenerator
from keras.callbacks import ModelCheckpoint, TensorBoard, CSVLogger, EarlyStopping
import keras.backend as K
from keras.optimizers import SGD, RMSprop, Adam
from keras.models import model_from_json
from keras.utils.np_utils import to_categorical
from keras.applications.inception_v3 import InceptionV3
from keras.applications.resnet50 import ResNet50
from keras.applications.vgg19 import VGG19
from keras.applications.xception import Xception
from keras.applications.inception_v3 import preprocess_input, decode_predictions
from random import sample
import tensorflow as tf
import glob
from PIL import Image
import matplotlib.pyplot as plt
class_to_ix = {}
ix_to_class = {}
with open('meta/classes.txt', 'r') as txt:
classes = [l.strip() for l in txt.readlines()]
class_to_ix = dict(zip(classes, range(len(classes))))
ix_to_class = dict(zip(range(len(classes)), classes))
class_to_ix = {v: k for k, v in ix_to_class.items()}
print("Loading Data")
h = h5py.File('all_data_300515.hdf5', 'r')
h.keys()
print("Load Training Data")
X_train = np.array(h.get('X_train')) # Size (m, n_h = 299 , n_w = 299, n_c = 3)
y_train = np.array(h.get('y_train')) # Size (m, 101)
index_train = sample(range(X_train.shape[0]),X_train.shape[0])
X_train = X_train[index_train,:,:,:]
y_train = y_train[index_train,:]
# print("Load Dev Data")
# X_dev = np.array(h.get('X_dev')) # Size (m, n_h = 299 , n_w = 299, n_c = 3)
# y_dev = np.array(h.get('y_dev')) # Size (m, 101)
# index_dev = sample(range(X_dev.shape[0]),X_dev.shape[0])
# X_dev = X_dev[index_dev,:,:,:]
# y_dev = y_dev[index_dev,:]
print("Load Test Data")
X_test = np.array(h.get('X_test')) # Size (m, n_h = 299 , n_w = 299, n_c = 3)
y_test = np.array(h.get('y_test')) # Size (m, 101)
h.close()
index_test = sample(range(X_test.shape[0]),X_test.shape[0])
X_test = X_test[index_test,:,:,:]
y_test = y_test[index_test,:]
######## Set up Image Augmentation
print("Setting up ImageDataGenerator")
datagen = ImageDataGenerator(
featurewise_center=False, # set input mean to 0 over the dataset
samplewise_center=False, # set each sample mean to 0
featurewise_std_normalization=False, # divide inputs by std of the dataset
samplewise_std_normalization=False, # divide each input by its std
zca_whitening=False, # apply ZCA whitening
# rotation_range=45, # randomly rotate images in the range (degrees, 0 to 180)
# width_shift_range=0.125, # randomly shift images horizontally (fraction of total width)
# height_shift_range=0.125, # randomly shift images vertically (fraction of total height)
# horizontal_flip=True, # randomly flip images
vertical_flip=False, # randomly flip images
rescale=1./255,
fill_mode='nearest')
# datagen.fit(X_train)
generator = datagen.flow(X_train, y_train, batch_size=9)
val_generator = datagen.flow(X_test, y_test, batch_size=9)
# generator = datagen.flow(X_train, y_train_cat, batch_size=32)
# val_generator = datagen.flow(X_val, y_val_cat, batch_size=32)
## Fine tuning. 70% with image augmentation.
## 83% with pre processing (14 mins).
## 84.5% with rmsprop/img.aug/dropout
## 86.09% with batchnorm/dropout/img.aug/adam(10)/rmsprop(140)
## InceptionV3
print("Load Model")
K.clear_session()
# base_model = InceptionV3(weights='imagenet', include_top=False, input_tensor=Input(shape=(299, 299, 3)))
# base_model = ResNet50(weights='imagenet', include_top=False, input_tensor=Input(shape=(299, 299, 3)))
# base_model = VGG19(weights='imagenet', include_top=False, input_tensor=Input(shape=(299, 299, 3)))
base_model = Xception(weights='imagenet', include_top=False, input_tensor=Input(shape=(299, 299, 3)))
x = base_model.output
x = GlobalAveragePooling2D()(x)
x = Dropout(0.7)(x)
x = Dense(4096)(x)
x = BatchNormalization()(x)
x = Activation('relu')(x)
x = Dropout(0.7)(x)
predictions = Dense(101, activation='softmax')(x)
model = Model(inputs=base_model.input, outputs=predictions)
import time
filename = time.strftime("%Y%m%d_%H%M") + "_xception_dp0707_2p3p4p"
# serialize model to JSON
model_json = model.to_json()
with open(filename + "_model.json", "w") as json_file:
json_file.write(model_json)
model.load_weights("20180607_1810_xception_dp0608_first.20-2.39.hdf5")
# print("First pass")
# for layer in base_model.layers:
# layer.trainable = False
# model.compile(optimizer=Adam(lr=0.001, beta_1=0.9, beta_2=0.999), loss='categorical_crossentropy', metrics=['accuracy'])
# checkpointer = ModelCheckpoint(filepath=filename + '_first.{epoch:02d}-{val_loss:.2f}.hdf5', verbose=1, save_best_only=False)
# csv_logger = CSVLogger(filename + '_first.log')
# model.fit_generator(generator,
# validation_data=val_generator,
# epochs=40,
# verbose=1,
# callbacks=[csv_logger, checkpointer])
# # serialize weights to HDF5
# model.save_weights(filename + "_finalweights.hdf5")
# print("Saved model to disk")
nh = 4
print("Second pass 4 layers")
for layer in model.layers[:132-nh]:
layer.trainable = False
for layer in model.layers[132-nh:]:
layer.trainable = True
model.compile(optimizer=Adam(lr=0.001, beta_1=0.9, beta_2=0.999), loss='categorical_crossentropy', metrics=['accuracy'])
checkpointer = ModelCheckpoint(filepath=filename + '_second.{epoch:02d}-{val_loss:.2f}.hdf5', verbose=1, save_best_only=False)
csv_logger = CSVLogger(filename + '_second.log')
model.fit_generator(generator,
validation_data=val_generator,
epochs=20,
verbose=1,
callbacks=[csv_logger, checkpointer])
# serialize weights to HDF5
model.save_weights(filename + "_2p_weights.hdf5")
print("Saved model to disk")
nh = 8
print("Third pass 8 layers")
for layer in model.layers[:132-nh]:
layer.trainable = False
for layer in model.layers[132-nh:]:
layer.trainable = True
model.compile(optimizer=Adam(lr=0.001, beta_1=0.9, beta_2=0.999), loss='categorical_crossentropy', metrics=['accuracy'])
checkpointer = ModelCheckpoint(filepath=filename + '_third.{epoch:02d}-{val_loss:.2f}.hdf5', verbose=1, save_best_only=False)
csv_logger = CSVLogger(filename + 'third.log')
model.fit_generator(generator,
validation_data=val_generator,
epochs=20,
verbose=1,
callbacks=[csv_logger, checkpointer])
# serialize weights to HDF5
model.save_weights(filename + "_3p_weights.hdf5")
print("Saved model to disk")
nh = 12
print("Fourth pass 12 layers")
for layer in model.layers[:132-nh]:
layer.trainable = False
for layer in model.layers[132-nh:]:
layer.trainable = True
model.compile(optimizer=Adam(lr=0.001, beta_1=0.9, beta_2=0.999), loss='categorical_crossentropy', metrics=['accuracy'])
checkpointer = ModelCheckpoint(filepath=filename + '_fourth.{epoch:02d}-{val_loss:.2f}.hdf5', verbose=1, save_best_only=False)
csv_logger = CSVLogger(filename + '_fourth.log')
model.fit_generator(generator,
validation_data=val_generator,
epochs=20,
verbose=1,
callbacks=[csv_logger, checkpointer])
# serialize weights to HDF5
model.save_weights(filename + "_4p_weights.hdf5")
print("Saved model to disk")
nh = 16
print("Fifth pass 16 layers")
for layer in model.layers[:132-nh]:
layer.trainable = False
for layer in model.layers[132-nh:]:
layer.trainable = True
model.compile(optimizer=Adam(lr=0.001, beta_1=0.9, beta_2=0.999), loss='categorical_crossentropy', metrics=['accuracy'])
checkpointer = ModelCheckpoint(filepath=filename + '_fifth.{epoch:02d}-{val_loss:.2f}.hdf5', verbose=1, save_best_only=False)
csv_logger = CSVLogger(filename + 'fifth.log')
model.fit_generator(generator,
validation_data=val_generator,
epochs=20,
verbose=1,
callbacks=[csv_logger, checkpointer])
# serialize weights to HDF5
model.save_weights(filename + "_5p_weights.hdf5")
print("Saved model to disk")
nh = 20
print("Sixth pass 20 layers")
for layer in model.layers[:132-nh]:
layer.trainable = False
for layer in model.layers[132-nh:]:
layer.trainable = True
model.compile(optimizer=Adam(lr=0.001, beta_1=0.9, beta_2=0.999), loss='categorical_crossentropy', metrics=['accuracy'])
checkpointer = ModelCheckpoint(filepath=filename + '_sixth.{epoch:02d}-{val_loss:.2f}.hdf5', verbose=1, save_best_only=False)
csv_logger = CSVLogger(filename + 'sixth.log')
model.fit_generator(generator,
validation_data=val_generator,
epochs=20,
verbose=1,
callbacks=[csv_logger, checkpointer])
# serialize weights to HDF5
model.save_weights(filename + "_6p_weights.hdf5")
print("Saved model to disk")