-
Notifications
You must be signed in to change notification settings - Fork 0
/
main.py
103 lines (89 loc) · 2.67 KB
/
main.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
import os
import cv2
import numpy as np
import keras
from sklearn.model_selection import train_test_split
from keras import models
from keras import layers
from keras.preprocessing.image import ImageDataGenerator
from keras.utils.np_utils import to_categorical
images = []
classes = []
path = 'path'
images_list = os.listdir(path)
print('# of classes:', len(images_list))
n_classes = len(images_list)
print('importing classes')
for n in range(n_classes):
pic_list = os.listdir(path + '/' + str(n))
for pic in pic_list:
cur_img = cv2.imread(path + '/' + str(n) + '/' + pic)
cur_img = cv2.resize(cur_img, (32, 32))
images.append(cur_img)
classes.append(n)
print(n, end=' ')
print()
images = np.asarray(images)
classes = np.asarray(classes)
X_train, X_test, y_train, y_test = train_test_split(images, classes, test_size=0.2, random_state=1)
X_train, X_validation, y_train, y_validation = train_test_split(X_train, y_train, test_size=0.2, random_state=1)
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,
)
test_datagen = ImageDataGenerator(rescale=1./255)
y_train = to_categorical(y_train, n_classes)
y_test = to_categorical(y_test, n_classes)
y_validation = to_categorical(y_validation, n_classes)
model = models.Sequential()
model.add(layers.Conv2D(60, (5,5), input_shape=(32, 32, 3), activation='relu'))
model.add(layers.Conv2D(60, (5,5), activation='relu'))
model.add(layers.MaxPooling2D((2, 2)))
model.add(layers.Conv2D(30, (3, 3), activation='relu'))
model.add(layers.Conv2D(30, (3, 3), activation='relu'))
model.add(layers.MaxPooling2D((2, 2)))
model.add(layers.Dropout(0.5))
model.add(layers.Flatten())
model.add(layers.Dense(units=500, activation='relu'))
model.add(layers.Dropout(0.5))
model.add(layers.Dense(units=n_classes, activation='softmax'))
model.compile(optimizer='adam',
loss='categorical_crossentropy',
metrics=['acc'])
print(model.summary())
callbacks = [
keras.callbacks.TensorBoard(
log_dir = 'my_log_dir',
histogram_freq = 1,
embeddings_freq = 1
),
keras.callbacks.EarlyStopping(
monitor='val_acc',
patience=2
),
keras.callbacks.ModelCheckpoint(
filepath='my_model.h5',
monitor='val_loss',
save_best_only=True
)
]
train_generator = train_datagen.flow(
X_train,
y_train,
batch_size=64
)
validation_generator = test_datagen.flow(
X_validation,
y_validation
)
model.fit_generator(
train_generator,
epochs=20,
validation_data=validation_generator,
shuffle=1,
callbacks=callbacks
)