-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathrun_FH.py
191 lines (160 loc) · 6.45 KB
/
run_FH.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
import os, shutil, random, glob
import copy
#添加tensorflow的库
import tensorflow as tf
from tensorflow import keras
from tensorflow.keras import layers
#添加matplotlib的库进行界面可视化
import matplotlib.pyplot as plt
from tensorflow.keras.applications.resnet import ResNet50
from tensorflow.keras.applications.resnet import ResNet101
from tensorflow.keras.applications.resnet import ResNet152
from keras.applications.inception_v3 import InceptionV3
from keras.applications.efficientnet import EfficientNetB0
from tensorflow.keras.callbacks import ModelCheckpoint
from tensorflow.keras.callbacks import ReduceLROnPlateau
from tensorflow.keras.callbacks import EarlyStopping
#
print("Num GPUs Available: ", len(tf.config.experimental.list_physical_devices('GPU')))
#变量
resize = 224 #输入图片尺寸参数
#total_num = 2190 #训练集总计和
#train_num = 1971 #训练集实际训练数字，20的验证集
epochs = 50 #迭代次数
batch_size = 4 #每次训练多少张
# -----------------------------------------------------------------------------
dir_data = 'C:\\flower_category' #训练集路径
dir_category = os.path.join(dir_data, 'category2')
category_list = os.listdir(dir_category)
category_list.sort(key=lambda x:int(x.split('.')[0]))
# #健壮性的断言
# #assert os.path.exists(dir_mask), 'Could not find ' + dir_mask
# #assert os.path.exists(dir_nomask), 'Could not find ' + dir_nomask
# #定义了文件指针对整个文件夹遍历一遍，将图像读出来
i = 0
for c in category_list:
category_list[i] = [os.path.abspath(fp) for fp in glob.glob(os.path.join(os.path.join(dir_category, c), '*.jpg'))]
i = i+1
# fpath_nomask = [os.path.abspath(fp) for fp in glob.glob(os.path.join(dir_nomask, '*.jpg'))]
# #文件数
num_data = len(category_list[0])
# num_nomask = len(fpath_nomask)
# #设置标签
label_data = copy.copy(category_list)
for i in range(219):
label_data[i] = [i] * num_data
#label_mask = [0] * num_data
#label_nomask = [1] * num_data
#
# print('#mask: ', num_mask)
# print('#nomask: ', num_nomask)
# #划分多少为验证集
RATIO_TEST = 0.1
#
num_data_test = int(num_data * RATIO_TEST)
# num_nomask_test = int(num_nomask * RATIO_TEST)
#
# # train
fpath_train = []
for i in range(219):
fpath_train = fpath_train + category_list[i][num_data_test:]
label_train = []
for i in range(219):
label_train = label_train + label_data[i][num_data_test:]
# validation
fpath_vali = []
for i in range(219):
fpath_vali = fpath_vali + category_list[i][:num_data_test]
#fpath_vali = fpath_mask[:num_data_test] + fpath_nomask[:num_nomask_test]
label_vali = []
for i in range(219):
label_vali = label_vali + label_data[i][:num_data_test]
#label_vali = label_mask[:num_data_test] + label_nomask[:num_nomask_test]
#
num_train = len(fpath_train)
num_vali = len(fpath_vali)
#
print(num_train)
print(num_vali)
# =============================================================================
#预处理
def preproc(fpath, label):
image_byte = tf.io.read_file(fpath)
image = tf.io.decode_image(image_byte)
image_resize = tf.image.resize_with_pad(image, resize, resize) #缩放到224*224
image_norm = tf.cast(image_resize, tf.float32) / 255. #归一化
image = image_norm
image = tf.image.random_brightness(image, 0.10)
image = tf.image.random_flip_left_right(image)
image = tf.image.random_contrast(image,0.3,0.7)
image = tf.image.random_hue(image,0.1)
label_onehot = tf.one_hot(label, 219)
return image, label_onehot
dataset_train = tf.data.Dataset.from_tensor_slices((fpath_train, label_train)) #将数据进行预处理
dataset_train = dataset_train.shuffle(num_train).repeat() #打乱顺序
dataset_train = dataset_train.map(preproc, num_parallel_calls=tf.data.experimental.AUTOTUNE)
dataset_train = dataset_train.batch(batch_size).prefetch(tf.data.experimental.AUTOTUNE) #一批次处理多少份
dataset_vali = tf.data.Dataset.from_tensor_slices((fpath_vali, label_vali))
dataset_vali = dataset_vali.shuffle(num_vali).repeat()
dataset_vali = dataset_vali.map(preproc, num_parallel_calls=tf.data.experimental.AUTOTUNE)
dataset_vali = dataset_vali.batch(batch_size).prefetch(tf.data.experimental.AUTOTUNE)
# -----------------------------------------------------------------------------
model = keras.Sequential()
model.add(EfficientNetB0(include_top=False, pooling='avg', weights='imagenet'))
model.add(layers.Flatten())
#model.add(layers.Dense(1024, activation='relu'))
#model.add(layers.Dropout(0.5))
model.add(layers.Dense(512, activation='relu'))
model.add(layers.Dropout(0.5))
model.add(layers.Dense(219, activation='softmax'))
model.summary
filepath='C:/flower_category/model'
checkpoint = ModelCheckpoint(filepath, monitor='val_acc', verbose=1,
save_best_only=True,mode='max',period=2)
reduce_lr = ReduceLROnPlateau(monitor='val_loss', factor=0.95,
patience=2,verbose=1, min_lr=0.001)
escb = EarlyStopping(monitor="val_loss",patience=3,verbose=1,
restore_best_weights=True)
callbacks_list = [checkpoint,reduce_lr,escb]
model.compile(loss='categorical_crossentropy',
optimizer='sgd',
metrics=['acc'])
history = model.fit(dataset_train,
steps_per_epoch = num_train//batch_size,
epochs = epochs, #迭代次数
validation_data = dataset_vali,
validation_steps = num_vali//batch_size,
verbose = 1,
callbacks=callbacks_list)
#评分标准
# scores = model.evaluate(train_data, train_label, verbose=1)
scores = model.evaluate(dataset_train, steps=num_train//batch_size, verbose=1)
print(scores)
# scores = model.evaluate(test_data, test_label, verbose=1)
scores = model.evaluate(dataset_vali, steps=num_vali//batch_size, verbose=1)
print(scores)
#保存模型
#model.save('C:/flower_category/model/ResNet50.h5')
# Record loss and acc
history_dict = history.history
train_loss = history_dict['loss']
train_accuracy = history_dict['acc']
val_loss = history_dict['val_loss']
val_accuracy = history_dict['val_acc']
# Draw loss
plt.figure()
plt.plot(range(epochs), train_loss, label='train_loss')
plt.plot(range(epochs), val_loss, label='val_loss')
plt.legend()
plt.xlabel('epochs')
plt.ylabel('loss')
# Draw acc
plt.figure()
plt.plot(range(epochs), train_accuracy, label='train_accuracy')
plt.plot(range(epochs), val_accuracy, label='val_acc')
plt.legend()
plt.xlabel('epochs')
plt.ylabel('accuracy')
# Display
plt.show()
print('Task finished')