-
Notifications
You must be signed in to change notification settings - Fork 3
/
training.py
178 lines (142 loc) · 4.54 KB
/
training.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
import os
import matplotlib.pyplot as plt
from keras.models import Model
from tensorflow.keras import layers
from tensorflow.keras.applications import VGG19
from tensorflow.keras.utils import image_dataset_from_directory
from tensorflow.keras.preprocessing.image import ImageDataGenerator
from tensorflow.keras import optimizers
from tensorflow.keras.applications.resnet50 import preprocess_input, decode_predictions
from tensorflow.keras.layers import Dense, GlobalAveragePooling2D, Dropout
import numpy as np
import deploy
base_dir = "chihuahua_vs_muffin"
train_dir = os.path.join(base_dir, "train")
validation_dir = os.path.join(base_dir, "validation")
test_dir = os.path.join(base_dir, "test")
# Loading datasets into tensors
train_dataset = image_dataset_from_directory(
directory=train_dir,
image_size=(112, 112),
batch_size=10)
validation_dataset = image_dataset_from_directory(
directory=validation_dir,
image_size=(112, 112),
batch_size=10)
test_dataset = image_dataset_from_directory(
directory=test_dir,
image_size=(112, 112),
batch_size=10)
# Data augmentation generation
train_datagen = ImageDataGenerator(
rescale=1./255,
rotation_range=40,
width_shift_range=0.2,
height_shift_range=0.2,
shear_range=0.2,
zoom_range=0.2,
horizontal_flip=True)
val_datagen = ImageDataGenerator(1./255)
train_generator = train_datagen.flow_from_directory(
train_dir,
target_size=(112, 112),
batch_size=10,
class_mode="binary")
validation_generator = val_datagen.flow_from_directory(
validation_dir,
target_size=(112, 112),
batch_size=10,
class_mode="binary")
# Setup network
conv_base = VGG19(include_top=False,
weights="imagenet",
input_shape=(112, 112, 3))
z = conv_base.output
z = Dense(128)(z)
z = GlobalAveragePooling2D()(z)
z = Dropout(0.3)(z)
predictions = Dense(2, activation="softmax")(z)
model = Model(inputs=conv_base.input, outputs=predictions)
# print(model.summary())
conv_base.trainable = False
model.compile(
loss="sparse_categorical_crossentropy",
optimizer=optimizers.Adagrad(),
metrics=['acc']
)
n_training_images = 250
n_validation_images = 150
batch_size = 10
n_steps_epoch = n_training_images / batch_size
n_validation_steps = n_validation_images / batch_size
history = model.fit(
train_generator,
steps_per_epoch=n_steps_epoch,
epochs=20,
validation_data=validation_generator,
validation_steps=n_validation_steps)
model.save("vgg19_chihuahua_vs_muffin.h5")
acc = history.history['acc']
val_acc = history.history['val_acc']
loss = history.history['loss']
val_loss = history.history['val_loss']
epochs = range(1, len(acc)+1)
plt.plot(epochs, acc, 'b', label='train accuracy')
plt.plot(epochs, val_acc, 'orange', label='validation accuracy')
plt.title('train acc vs val acc')
plt.legend()
plt.figure()
plt.plot(epochs, loss, 'b', label='training loss')
plt.plot(epochs, val_loss, 'orange', label='validation loss')
plt.title('train loss vs val loss')
plt.legend()
plt.show(block=False)
test_datagen = ImageDataGenerator(1./255)
test_generator = test_datagen.flow_from_directory(
test_dir,
target_size=(112, 112),
batch_size=10,
class_mode="binary")
test_loss, test_acc = model.evaluate(test_generator, steps=n_steps_epoch)
print("\ntest acc :", test_acc)
class_labels_map = train_generator.class_indices
print("\nClasses : ", class_labels_map)
folder_id = 2
folder_name = ""
instance_id = 0
# 1 - 900 chihuahua
# 1 - 500 muffin
while folder_id != 0:
print("Choose an option: ")
print("2) Muffin")
print("1) Chihuahua")
print("0) Exit")
folder_id = int(input())
if folder_id == 2:
folder_name = "muffin"
print("Valid id's for muffins 1-500")
instance_id = input()
casted_instance_id = int(instance_id)
if 1 <= casted_instance_id <= 900:
deploy.predict_image_class(model, class_labels_map, folder_name, instance_id)
else:
print("Choose a valid instance id!")
elif folder_id == 1:
folder_name = "chihuahua"
print("Valid id's for chihuahuas 1-900")
instance_id = input()
casted_instance_id = int(instance_id)
if 1 <= casted_instance_id <= 500:
deploy.predict_image_class(model, class_labels_map, folder_name, instance_id)
else:
print("Choose a valid instance id!")
elif folder_id == 0:
break
else:
print("Choose a valid option!")
plt.show()
# Suggested queries:
# chihuahua_232
# muffin_138
# chihuahua 241
# muffin 433