Python paddlepaddle基于图像分类网络VGG实现中草药识别并发布成应用

发表于 | 次阅读
字数统计: 7,854 字 | 阅读时长 ≈ 42 min

paddlepaddle基于图像分类网络VGG实现中草药识别并发布成应用

本文仅对代码进行记录，详细了解请看原文
原文链接：https://aistudio.baidu.com/aistudio/projectdetail/2310126?channelType=0&channel=0

原数据集地址： https://bj.bcebos.com/ai-studio-online/3c0c4d559b9941c38892855470a26bea17371e1938b34731acde5407a7c60410?authorization=bce-auth-v1%2F5cfe9a5e1454405eb2a975c43eace6ec%2F2022-09-04T15%3A26%3A48Z%2F-1%2F%2F0723b1538c1ae02efb5301af77b9a715e7702678611a09030dd53e0e602b48eb&responseContentDisposition=attachment%3B%20filename%3DChinese%20Medicine.zip

数据集集下载之后名称： Chinese Medicine.zip

模型训练的代码

若是进了解过程，不追求准确率，建议将每种类型的图像数据，删减到20张图片，训练速度不会那么慢
即，这里这个解压函数运行一次，之后去删除多余数据，然后将此处注掉，然后再运行后面的代码

1
2
# 调用解压函数解压数据集
#unzip_data(src_path,target_path)

完整训练代码如下

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
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
# 引入需要的模块
import os
import zipfile
import random
import json
import paddle
import sys
import numpy as np
from PIL import Image
import matplotlib.pyplot as plt
from paddle.io import Dataset
random.seed(200)

train_parameters = {
    "src_path":"./testworkfour/Chinese Medicine.zip",    #原始数据集路径
    "target_path":"./testworkfour/data/",                     #要解压的路径
    "train_list_path": "./testworkfour/data/train.txt",       #train.txt路径
    "eval_list_path": "./testworkfour/data/eval.txt",         #eval.txt路径
    "label_dict":{},                                          #标签字典
    "readme_path": "./testworkfour/data/readme.json",         #readme.json路径
    "class_dim": -1,                                          #分类数
}
src_path=train_parameters['src_path']
target_path=train_parameters['target_path']
train_list_path=train_parameters['train_list_path']
eval_list_path=train_parameters['eval_list_path']


def unzip_data(src_path,target_path):
    if(not os.path.isdir(target_path + "Chinese Medicine")):     
        z = zipfile.ZipFile(src_path, 'r')
        z.extractall(path=target_path)
        z.close()



def get_data_list(target_path,train_list_path,eval_list_path):
    '''
    生成数据列表
    '''
    #存放所有类别的信息
    class_detail = []
    #获取所有类别保存的文件夹名称
    data_list_path=target_path+"Chinese Medicine/"
    class_dirs = os.listdir(data_list_path)  
    #总的图像数量
    all_class_images = 0
    #存放类别标签
    class_label=0
    #存放类别数目
    class_dim = 0
    #存储要写进eval.txt和train.txt中的内容
    trainer_list=[]
    eval_list=[]
    #读取每个类别，['baihe', 'gouqi','jinyinhua','huaihua','dangshen']
    for class_dir in class_dirs:
        if class_dir != ".DS_Store":
            class_dim += 1
            #每个类别的信息
            class_detail_list = {}
            eval_sum = 0
            trainer_sum = 0
            #统计每个类别有多少张图片
            class_sum = 0
            #获取类别路径 
            path = data_list_path  + class_dir
            # 获取所有图片
            img_paths = os.listdir(path)
            for img_path in img_paths:                                  # 遍历文件夹下的每个图片
                name_path = path + '/' + img_path                       # 每张图片的路径
                if class_sum % 8 == 0:                                  # 每8张图片取一个做验证数据
                    eval_sum += 1                                       # test_sum为测试数据的数目
                    eval_list.append(name_path + "\t%d" % class_label + "\n")
                else:
                    trainer_sum += 1 
                    trainer_list.append(name_path + "\t%d" % class_label + "\n")#trainer_sum测试数据的数目
                class_sum += 1                                          #每类图片的数目
                all_class_images += 1                                   #所有类图片的数目
             
            # 说明的json文件的class_detail数据
            class_detail_list['class_name'] = class_dir             #类别名称
            class_detail_list['class_label'] = class_label          #类别标签
            class_detail_list['class_eval_images'] = eval_sum       #该类数据的测试集数目
            class_detail_list['class_trainer_images'] = trainer_sum #该类数据的训练集数目
            class_detail.append(class_detail_list)  
            #初始化标签列表
            train_parameters['label_dict'][str(class_label)] = class_dir
            class_label += 1 
            
    #初始化分类数
    train_parameters['class_dim'] = class_dim
  
    #乱序  
    random.shuffle(eval_list)
    with open(eval_list_path, 'a') as f:
        for eval_image in eval_list:
            f.write(eval_image) 
            
    random.shuffle(trainer_list)
    with open(train_list_path, 'a') as f2:
        for train_image in trainer_list:
            f2.write(train_image) 

    # 说明的json文件信息
    readjson = {}
    readjson['all_class_name'] = data_list_path                  #文件父目录
    readjson['all_class_images'] = all_class_images
    readjson['class_detail'] = class_detail
    jsons = json.dumps(readjson, sort_keys=True, indent=4, separators=(',', ': '))
    with open(train_parameters['readme_path'],'w') as f:
        f.write(jsons)
    print ('生成数据列表完成！')       





# 调用解压函数解压数据集
# unzip_data(src_path,target_path)


# 划分训练集与验证集，乱序，生成数据列表
#每次生成数据列表前，首先清空train.txt和eval.txt
with open(train_list_path, 'w') as f: 
    f.seek(0)
    f.truncate() 
with open(eval_list_path, 'w') as f: 
    f.seek(0)
    f.truncate() 
#生成数据列表   
get_data_list(target_path,train_list_path,eval_list_path)




# 定义数据读取器
class dataset(Dataset):
    def __init__(self, data_path, mode='train'):
        """
        数据读取器
        :param data_path: 数据集所在路径
        :param mode: train or eval
        """
        super().__init__()
        self.data_path = data_path
        self.img_paths = []
        self.labels = []

        if mode == 'train':
            with open(os.path.join(self.data_path, "train.txt"), "r", encoding="utf-8") as f:
                self.info = f.readlines()
            for img_info in self.info:
                img_path, label = img_info.strip().split('\t')
                self.img_paths.append(img_path)
                self.labels.append(int(label))

        else:
            with open(os.path.join(self.data_path, "eval.txt"), "r", encoding="utf-8") as f:
                self.info = f.readlines()
            for img_info in self.info:
                img_path, label = img_info.strip().split('\t')
                self.img_paths.append(img_path)
                self.labels.append(int(label))


    def __getitem__(self, index):
        """
        获取一组数据
        :param index: 文件索引号
        :return:
        """
        # 第一步打开图像文件并获取label值
        img_path = self.img_paths[index]
        img = Image.open(img_path)
        if img.mode != 'RGB':
            img = img.convert('RGB') 
        img = img.resize((224, 224), Image.BILINEAR)
        #img = rand_flip_image(img)
        img = np.array(img).astype('float32')
        img = img.transpose((2, 0, 1)) / 255
        label = self.labels[index]
        label = np.array([label], dtype="int64")
        return img, label

    def print_sample(self, index: int = 0):
        print("文件名", self.img_paths[index], "\t标签值", self.labels[index])

    def __len__(self):
        return len(self.img_paths)


 



#训练数据加载
train_dataset = dataset('./testworkfour/data',mode='train')
train_loader = paddle.io.DataLoader(train_dataset, batch_size=32, shuffle=True)
#评估数据加载
eval_dataset = dataset('./testworkfour/data',mode='eval')
eval_loader = paddle.io.DataLoader(eval_dataset, batch_size = 8, shuffle=False)


# 定义卷积池化网络
class ConvPool(paddle.nn.Layer):
    '''卷积+池化'''
    def __init__(self,
                 num_channels,
                 num_filters, 
                 filter_size,
                 pool_size,
                 pool_stride,
                 groups,
                 conv_stride=1, 
                 conv_padding=1,
                 ):
        super(ConvPool, self).__init__()  

        # groups代表卷积层的数量
        for i in range(groups):
            self.add_sublayer(   #添加子层实例
                'bb_%d' % i,
                paddle.nn.Conv2D(         # layer
                in_channels=num_channels, #通道数
                out_channels=num_filters,   #卷积核个数
                kernel_size=filter_size,   #卷积核大小
                stride=conv_stride,        #步长
                padding = conv_padding,    #padding
                )
            )
            self.add_sublayer(
                'relu%d' % i,
                paddle.nn.ReLU()
            )
            num_channels = num_filters
            

        self.add_sublayer(
            'Maxpool',
            paddle.nn.MaxPool2D(
            kernel_size=pool_size,           #池化核大小
            stride=pool_stride               #池化步长
            )
        )

    def forward(self, inputs):
        x = inputs
        for prefix, sub_layer in self.named_children():
            # print(prefix,sub_layer)
            x = sub_layer(x)
        return x


 
# VGG网络
class VGGNet(paddle.nn.Layer):
    def __init__(self):
        super(VGGNet, self).__init__()       
        # 5个卷积池化操作
        self.convpool01 = ConvPool(
            3, 64, 3, 2, 2, 2)  #3:通道数，64：卷积核个数，3:卷积核大小，2:池化核大小，2:池化步长，2:连续卷积个数
        self.convpool02 = ConvPool(
            64, 128, 3, 2, 2, 2)
        self.convpool03 = ConvPool(
            128, 256, 3, 2, 2, 3) 
        self.convpool04 = ConvPool(
            256, 512, 3, 2, 2, 3)
        self.convpool05 = ConvPool(
            512, 512, 3, 2, 2, 3)       
        self.pool_5_shape = 512 * 7* 7
        # 三个全连接层
        self.fc01 = paddle.nn.Linear(self.pool_5_shape, 4096)
        self.drop1 = paddle.nn.Dropout(p=0.5)
        self.fc02 = paddle.nn.Linear(4096, 4096)
        self.drop2 = paddle.nn.Dropout(p=0.5)
        self.fc03 = paddle.nn.Linear(4096, train_parameters['class_dim'])

    def forward(self, inputs, label=None):
        # print('input_shape:', inputs.shape) #[8, 3, 224, 224]
        """前向计算"""
        out = self.convpool01(inputs)
        # print('convpool01_shape:', out.shape)           #[8, 64, 112, 112]
        out = self.convpool02(out)
        # print('convpool02_shape:', out.shape)           #[8, 128, 56, 56]
        out = self.convpool03(out)
        # print('convpool03_shape:', out.shape)           #[8, 256, 28, 28]
        out = self.convpool04(out)
        # print('convpool04_shape:', out.shape)           #[8, 512, 14, 14]
        out = self.convpool05(out)
        # print('convpool05_shape:', out.shape)           #[8, 512, 7, 7]         

        out = paddle.reshape(out, shape=[-1, 512*7*7])
        out = self.fc01(out)
        out = self.drop1(out)
        out = self.fc02(out)
        out = self.drop2(out)
        out = self.fc03(out)
        
        if label is not None:
            acc = paddle.metric.accuracy(input=out, label=label)
            return out, acc
        else:
            return out
            
 

# 折线图，用于观察训练过程中loss和acc的走势
def draw_process(title,color,iters,data,label):
    plt.title(title, fontsize=24)
    plt.xlabel("iter", fontsize=20)
    plt.ylabel(label, fontsize=20)
    plt.plot(iters, data,color=color,label=label) 
    plt.legend()
    plt.grid()
    plt.show()



# 参数配置，要保留之前数据集准备阶段配置的参数，所以使用update更新字典
train_parameters.update({
    "input_size": [3, 224, 224],                              #输入图片的shape
    "num_epochs": 10,                                         #训练轮数
    "skip_steps": 10,                                         #训练时输出日志的间隔
    "save_steps": 100,                                         #训练时保存模型参数的间隔
    "learning_strategy": {                                    #优化函数相关的配置
        "lr": 0.0001                                          #超参数学习率
    },
    "checkpoints": "./testworkfour/checkpoints"          #保存的路径
})


 

model = VGGNet()
model.train()
# 配置loss函数
cross_entropy = paddle.nn.CrossEntropyLoss()
# 配置参数优化器
optimizer = paddle.optimizer.Adam(learning_rate=train_parameters['learning_strategy']['lr'],
                                  parameters=model.parameters()) 

steps = 0
Iters, total_loss, total_acc = [], [], []

for epo in range(train_parameters['num_epochs']):
    print('开始第{}次训练'.format(epo))
    for _, data in enumerate(train_loader()):
        steps += 1
        x_data = data[0]
        y_data = data[1]
        predicts, acc = model(x_data, y_data)
        loss = cross_entropy(predicts, y_data)
        loss.backward()
        optimizer.step()
        optimizer.clear_grad()
        if steps % train_parameters["skip_steps"] == 0:
            Iters.append(steps)
            total_loss.append(loss.numpy()[0])
            total_acc.append(acc.numpy()[0])
            #打印中间过程
            print('epo: {}, step: {}, loss is: {}, acc is: {}'\
                  .format(epo, steps, loss.numpy(), acc.numpy()))
        #保存模型参数
        if steps % train_parameters["save_steps"] == 0:
            save_path = train_parameters["checkpoints"]+"/"+"save_dir_" + str(steps) + '.pdparams'
            print('save model to: ' + save_path)
            paddle.save(model.state_dict(),save_path)
paddle.save(model.state_dict(),train_parameters["checkpoints"]+"/"+"save_dir_final.pdparams")
draw_process("trainning loss","red",Iters,total_loss,"trainning loss")
draw_process("trainning acc","green",Iters,total_acc,"trainning acc")

模型验证的代码

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
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
# 引入需要的模块
import os
import zipfile
import random
import json
import paddle
import sys
import numpy as np
from PIL import Image
import matplotlib.pyplot as plt
from paddle.io import Dataset

train_parameters = {
    "src_path":"./testworkfour/Chinese Medicine.zip",    #原始数据集路径
    "target_path":"./testworkfour/data/",                     #要解压的路径
    "train_list_path": "./testworkfour/data/train.txt",       #train.txt路径
    "eval_list_path": "./testworkfour/data/eval.txt",         #eval.txt路径
    "label_dict":{},                                          #标签字典
    "readme_path": "./testworkfour/data/readme.json",         #readme.json路径
    "class_dim": -1,                                          #分类数
}

src_path=train_parameters['src_path']
target_path=train_parameters['target_path']
train_list_path=train_parameters['train_list_path']
eval_list_path=train_parameters['eval_list_path']


def get_data_list(target_path,train_list_path,eval_list_path):
    '''
    生成数据列表
    '''
    #存放所有类别的信息
    class_detail = []
    #获取所有类别保存的文件夹名称
    data_list_path=target_path+"Chinese Medicine/"
    class_dirs = os.listdir(data_list_path)  
    #总的图像数量
    all_class_images = 0
    #存放类别标签
    class_label=0
    #存放类别数目
    class_dim = 0
    #存储要写进eval.txt和train.txt中的内容
    trainer_list=[]
    eval_list=[]
    #读取每个类别，['baihe', 'gouqi','jinyinhua','huaihua','dangshen']
    for class_dir in class_dirs:
        if class_dir != ".DS_Store":
            class_dim += 1
            #每个类别的信息
            class_detail_list = {}
            eval_sum = 0
            trainer_sum = 0
            #统计每个类别有多少张图片
            class_sum = 0
            #获取类别路径 
            path = data_list_path  + class_dir
            # 获取所有图片
            img_paths = os.listdir(path)
            for img_path in img_paths:                                  # 遍历文件夹下的每个图片
                name_path = path + '/' + img_path                       # 每张图片的路径
                if class_sum % 8 == 0:                                  # 每8张图片取一个做验证数据
                    eval_sum += 1                                       # test_sum为测试数据的数目
                    eval_list.append(name_path + "\t%d" % class_label + "\n")
                else:
                    trainer_sum += 1 
                    trainer_list.append(name_path + "\t%d" % class_label + "\n")#trainer_sum测试数据的数目
                class_sum += 1                                          #每类图片的数目
                all_class_images += 1                                   #所有类图片的数目
             
            # 说明的json文件的class_detail数据
            class_detail_list['class_name'] = class_dir             #类别名称
            class_detail_list['class_label'] = class_label          #类别标签
            class_detail_list['class_eval_images'] = eval_sum       #该类数据的测试集数目
            class_detail_list['class_trainer_images'] = trainer_sum #该类数据的训练集数目
            class_detail.append(class_detail_list)  
            #初始化标签列表
            train_parameters['label_dict'][str(class_label)] = class_dir
            class_label += 1 
            
    #初始化分类数
    train_parameters['class_dim'] = class_dim
  
    #乱序  
    random.shuffle(eval_list)
    with open(eval_list_path, 'a') as f:
        for eval_image in eval_list:
            f.write(eval_image) 
            
    random.shuffle(trainer_list)
    with open(train_list_path, 'a') as f2:
        for train_image in trainer_list:
            f2.write(train_image) 

    # 说明的json文件信息
    readjson = {}
    readjson['all_class_name'] = data_list_path                  #文件父目录
    readjson['all_class_images'] = all_class_images
    readjson['class_detail'] = class_detail
    jsons = json.dumps(readjson, sort_keys=True, indent=4, separators=(',', ': '))
    with open(train_parameters['readme_path'],'w') as f:
        f.write(jsons)
    print ('生成数据列表完成！')    


# 划分训练集与验证集，乱序，生成数据列表
#每次生成数据列表前，首先清空train.txt和eval.txt
with open(train_list_path, 'w') as f: 
    f.seek(0)
    f.truncate() 
with open(eval_list_path, 'w') as f: 
    f.seek(0)
    f.truncate() 
#生成数据列表   
get_data_list(target_path,train_list_path,eval_list_path)


def load_image(img_path):
    '''
    预测图片预处理
    '''
    img = Image.open(img_path) 
    if img.mode != 'RGB': 
        img = img.convert('RGB') 
    img = img.resize((224, 224), Image.BILINEAR)
    img = np.array(img).astype('float32') 
    img = img.transpose((2, 0, 1)) / 255 # HWC to CHW 及归一化
    return img


label_dic = train_parameters['label_dict']









# 定义卷积池化网络
class ConvPool(paddle.nn.Layer):
    '''卷积+池化'''
    def __init__(self,
                 num_channels,
                 num_filters, 
                 filter_size,
                 pool_size,
                 pool_stride,
                 groups,
                 conv_stride=1, 
                 conv_padding=1,
                 ):
        super(ConvPool, self).__init__()  

        # groups代表卷积层的数量
        for i in range(groups):
            self.add_sublayer(   #添加子层实例
                'bb_%d' % i,
                paddle.nn.Conv2D(         # layer
                in_channels=num_channels, #通道数
                out_channels=num_filters,   #卷积核个数
                kernel_size=filter_size,   #卷积核大小
                stride=conv_stride,        #步长
                padding = conv_padding,    #padding
                )
            )
            self.add_sublayer(
                'relu%d' % i,
                paddle.nn.ReLU()
            )
            num_channels = num_filters
            

        self.add_sublayer(
            'Maxpool',
            paddle.nn.MaxPool2D(
            kernel_size=pool_size,           #池化核大小
            stride=pool_stride               #池化步长
            )
        )

    def forward(self, inputs):
        x = inputs
        for prefix, sub_layer in self.named_children():
            # print(prefix,sub_layer)
            x = sub_layer(x)
        return x






# VGG网络
class VGGNet(paddle.nn.Layer):
    def __init__(self):
        super(VGGNet, self).__init__()       
        # 5个卷积池化操作
        self.convpool01 = ConvPool(
            3, 64, 3, 2, 2, 2)  #3:通道数，64：卷积核个数，3:卷积核大小，2:池化核大小，2:池化步长，2:连续卷积个数
        self.convpool02 = ConvPool(
            64, 128, 3, 2, 2, 2)
        self.convpool03 = ConvPool(
            128, 256, 3, 2, 2, 3) 
        self.convpool04 = ConvPool(
            256, 512, 3, 2, 2, 3)
        self.convpool05 = ConvPool(
            512, 512, 3, 2, 2, 3)       
        self.pool_5_shape = 512 * 7* 7
        # 三个全连接层
        self.fc01 = paddle.nn.Linear(self.pool_5_shape, 4096)
        self.drop1 = paddle.nn.Dropout(p=0.5)
        self.fc02 = paddle.nn.Linear(4096, 4096)
        self.drop2 = paddle.nn.Dropout(p=0.5)
        self.fc03 = paddle.nn.Linear(4096, train_parameters['class_dim'])

    def forward(self, inputs, label=None):
        # print('input_shape:', inputs.shape) #[8, 3, 224, 224]
        """前向计算"""
        out = self.convpool01(inputs)
        # print('convpool01_shape:', out.shape)           #[8, 64, 112, 112]
        out = self.convpool02(out)
        # print('convpool02_shape:', out.shape)           #[8, 128, 56, 56]
        out = self.convpool03(out)
        # print('convpool03_shape:', out.shape)           #[8, 256, 28, 28]
        out = self.convpool04(out)
        # print('convpool04_shape:', out.shape)           #[8, 512, 14, 14]
        out = self.convpool05(out)
        # print('convpool05_shape:', out.shape)           #[8, 512, 7, 7]         

        out = paddle.reshape(out, shape=[-1, 512*7*7])
        out = self.fc01(out)
        out = self.drop1(out)
        out = self.fc02(out)
        out = self.drop2(out)
        out = self.fc03(out)
        
        if label is not None:
            acc = paddle.metric.accuracy(input=out, label=label)
            return out, acc
        else:
            return out
            

# 定义数据读取器
class dataset(Dataset):
    def __init__(self, data_path, mode='train'):
        """
        数据读取器
        :param data_path: 数据集所在路径
        :param mode: train or eval
        """
        super().__init__()
        self.data_path = data_path
        self.img_paths = []
        self.labels = []

        if mode == 'train':
            with open(os.path.join(self.data_path, "train.txt"), "r", encoding="utf-8") as f:
                self.info = f.readlines()
            for img_info in self.info:
                img_path, label = img_info.strip().split('\t')
                self.img_paths.append(img_path)
                self.labels.append(int(label))

        else:
            with open(os.path.join(self.data_path, "eval.txt"), "r", encoding="utf-8") as f:
                self.info = f.readlines()
            for img_info in self.info:
                img_path, label = img_info.strip().split('\t')
                self.img_paths.append(img_path)
                self.labels.append(int(label))


    def __getitem__(self, index):
        """
        获取一组数据
        :param index: 文件索引号
        :return:
        """
        # 第一步打开图像文件并获取label值
        img_path = self.img_paths[index]
        img = Image.open(img_path)
        if img.mode != 'RGB':
            img = img.convert('RGB') 
        img = img.resize((224, 224), Image.BILINEAR)
        #img = rand_flip_image(img)
        img = np.array(img).astype('float32')
        img = img.transpose((2, 0, 1)) / 255
        label = self.labels[index]
        label = np.array([label], dtype="int64")
        return img, label

    def print_sample(self, index: int = 0):
        print("文件名", self.img_paths[index], "\t标签值", self.labels[index])

    def __len__(self):
        return len(self.img_paths)



#评估数据加载
eval_dataset = dataset('./testworkfour/data',mode='eval')
eval_loader = paddle.io.DataLoader(eval_dataset, batch_size = 8, shuffle=False)



# 模型评估
# 加载训练过程保存的最后一个模型
model__state_dict = paddle.load('./testworkfour/checkpoints/save_dir_final.pdparams')
model_eval = VGGNet()
model_eval.set_state_dict(model__state_dict) 
model_eval.eval()
accs = []
# 开始评估
for _, data in enumerate(eval_loader()):
    x_data = data[0]
    y_data = data[1]
    predicts = model_eval(x_data)
    acc = paddle.metric.accuracy(predicts, y_data)
    accs.append(acc.numpy()[0])
print('模型在验证集上的准确率为：',np.mean(accs))

模型使用的代码

批量的预测

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
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
# 引入需要的模块
import os
import zipfile
import random
import json
import paddle
import sys
import numpy as np
from PIL import Image
import matplotlib.pyplot as plt
from paddle.io import Dataset

train_parameters = {
    "src_path":"./testworkfour/Chinese Medicine.zip",    #原始数据集路径
    "target_path":"./testworkfour/data/",                     #要解压的路径
    "train_list_path": "./testworkfour/data/train.txt",       #train.txt路径
    "eval_list_path": "./testworkfour/data/eval.txt",         #eval.txt路径
    "label_dict":{},                                          #标签字典
    "readme_path": "./testworkfour/data/readme.json",         #readme.json路径
    "class_dim": -1,                                          #分类数
}

src_path=train_parameters['src_path']
target_path=train_parameters['target_path']
train_list_path=train_parameters['train_list_path']
eval_list_path=train_parameters['eval_list_path']


def get_data_list(target_path,train_list_path,eval_list_path):
    '''
    生成数据列表
    '''
    #存放所有类别的信息
    class_detail = []
    #获取所有类别保存的文件夹名称
    data_list_path=target_path+"Chinese Medicine/"
    class_dirs = os.listdir(data_list_path)  
    #总的图像数量
    all_class_images = 0
    #存放类别标签
    class_label=0
    #存放类别数目
    class_dim = 0
    #存储要写进eval.txt和train.txt中的内容
    trainer_list=[]
    eval_list=[]
    #读取每个类别，['baihe', 'gouqi','jinyinhua','huaihua','dangshen']
    for class_dir in class_dirs:
        if class_dir != ".DS_Store":
            class_dim += 1
            #每个类别的信息
            class_detail_list = {}
            eval_sum = 0
            trainer_sum = 0
            #统计每个类别有多少张图片
            class_sum = 0
            #获取类别路径 
            path = data_list_path  + class_dir
            # 获取所有图片
            img_paths = os.listdir(path)
            for img_path in img_paths:                                  # 遍历文件夹下的每个图片
                name_path = path + '/' + img_path                       # 每张图片的路径
                if class_sum % 8 == 0:                                  # 每8张图片取一个做验证数据
                    eval_sum += 1                                       # test_sum为测试数据的数目
                    eval_list.append(name_path + "\t%d" % class_label + "\n")
                else:
                    trainer_sum += 1 
                    trainer_list.append(name_path + "\t%d" % class_label + "\n")#trainer_sum测试数据的数目
                class_sum += 1                                          #每类图片的数目
                all_class_images += 1                                   #所有类图片的数目
             
            # 说明的json文件的class_detail数据
            class_detail_list['class_name'] = class_dir             #类别名称
            class_detail_list['class_label'] = class_label          #类别标签
            class_detail_list['class_eval_images'] = eval_sum       #该类数据的测试集数目
            class_detail_list['class_trainer_images'] = trainer_sum #该类数据的训练集数目
            class_detail.append(class_detail_list)  
            #初始化标签列表
            train_parameters['label_dict'][str(class_label)] = class_dir
            class_label += 1 
            
    #初始化分类数
    train_parameters['class_dim'] = class_dim
  
    #乱序  
    random.shuffle(eval_list)
    with open(eval_list_path, 'a') as f:
        for eval_image in eval_list:
            f.write(eval_image) 
            
    random.shuffle(trainer_list)
    with open(train_list_path, 'a') as f2:
        for train_image in trainer_list:
            f2.write(train_image) 

    # 说明的json文件信息
    readjson = {}
    readjson['all_class_name'] = data_list_path                  #文件父目录
    readjson['all_class_images'] = all_class_images
    readjson['class_detail'] = class_detail
    jsons = json.dumps(readjson, sort_keys=True, indent=4, separators=(',', ': '))
    with open(train_parameters['readme_path'],'w') as f:
        f.write(jsons)
    print ('生成数据列表完成！')    


# 划分训练集与验证集，乱序，生成数据列表
#每次生成数据列表前，首先清空train.txt和eval.txt
with open(train_list_path, 'w') as f: 
    f.seek(0)
    f.truncate() 
with open(eval_list_path, 'w') as f: 
    f.seek(0)
    f.truncate() 
#生成数据列表   
get_data_list(target_path,train_list_path,eval_list_path)


def load_image(img_path):
    '''
    预测图片预处理
    '''
    img = Image.open(img_path) 
    if img.mode != 'RGB': 
        img = img.convert('RGB') 
    img = img.resize((224, 224), Image.BILINEAR)
    img = np.array(img).astype('float32') 
    img = img.transpose((2, 0, 1)) / 255 # HWC to CHW 及归一化
    return img


label_dic = train_parameters['label_dict']
print(label_dic)







# 定义卷积池化网络
class ConvPool(paddle.nn.Layer):
    '''卷积+池化'''
    def __init__(self,
                 num_channels,
                 num_filters, 
                 filter_size,
                 pool_size,
                 pool_stride,
                 groups,
                 conv_stride=1, 
                 conv_padding=1,
                 ):
        super(ConvPool, self).__init__()  

        # groups代表卷积层的数量
        for i in range(groups):
            self.add_sublayer(   #添加子层实例
                'bb_%d' % i,
                paddle.nn.Conv2D(         # layer
                in_channels=num_channels, #通道数
                out_channels=num_filters,   #卷积核个数
                kernel_size=filter_size,   #卷积核大小
                stride=conv_stride,        #步长
                padding = conv_padding,    #padding
                )
            )
            self.add_sublayer(
                'relu%d' % i,
                paddle.nn.ReLU()
            )
            num_channels = num_filters
            

        self.add_sublayer(
            'Maxpool',
            paddle.nn.MaxPool2D(
            kernel_size=pool_size,           #池化核大小
            stride=pool_stride               #池化步长
            )
        )

    def forward(self, inputs):
        x = inputs
        for prefix, sub_layer in self.named_children():
            # print(prefix,sub_layer)
            x = sub_layer(x)
        return x






# VGG网络
class VGGNet(paddle.nn.Layer):
    def __init__(self):
        super(VGGNet, self).__init__()       
        # 5个卷积池化操作
        self.convpool01 = ConvPool(
            3, 64, 3, 2, 2, 2)  #3:通道数，64：卷积核个数，3:卷积核大小，2:池化核大小，2:池化步长，2:连续卷积个数
        self.convpool02 = ConvPool(
            64, 128, 3, 2, 2, 2)
        self.convpool03 = ConvPool(
            128, 256, 3, 2, 2, 3) 
        self.convpool04 = ConvPool(
            256, 512, 3, 2, 2, 3)
        self.convpool05 = ConvPool(
            512, 512, 3, 2, 2, 3)       
        self.pool_5_shape = 512 * 7* 7
        # 三个全连接层
        self.fc01 = paddle.nn.Linear(self.pool_5_shape, 4096)
        self.drop1 = paddle.nn.Dropout(p=0.5)
        self.fc02 = paddle.nn.Linear(4096, 4096)
        self.drop2 = paddle.nn.Dropout(p=0.5)
        self.fc03 = paddle.nn.Linear(4096, train_parameters['class_dim'])

    def forward(self, inputs, label=None):
        # print('input_shape:', inputs.shape) #[8, 3, 224, 224]
        """前向计算"""
        out = self.convpool01(inputs)
        # print('convpool01_shape:', out.shape)           #[8, 64, 112, 112]
        out = self.convpool02(out)
        # print('convpool02_shape:', out.shape)           #[8, 128, 56, 56]
        out = self.convpool03(out)
        # print('convpool03_shape:', out.shape)           #[8, 256, 28, 28]
        out = self.convpool04(out)
        # print('convpool04_shape:', out.shape)           #[8, 512, 14, 14]
        out = self.convpool05(out)
        # print('convpool05_shape:', out.shape)           #[8, 512, 7, 7]         

        out = paddle.reshape(out, shape=[-1, 512*7*7])
        out = self.fc01(out)
        out = self.drop1(out)
        out = self.fc02(out)
        out = self.drop2(out)
        out = self.fc03(out)
        
        if label is not None:
            acc = paddle.metric.accuracy(input=out, label=label)
            return out, acc
        else:
            return out
            





 

import time
# 加载训练过程保存的最后一个模型
model__state_dict = paddle.load('./testworkfour/checkpoints/save_dir_final.pdparams')
model_predict = VGGNet()
model_predict.set_state_dict(model__state_dict) 
model_predict.eval()
#infer_imgs_path = os.listdir("infer")
infer_imgs_path = os.listdir("./testworkfour/data/Chinese Medicine/dangshen")


# print(infer_imgs_path)

# 预测所有图片
for infer_img_path in infer_imgs_path:
    infer_img = load_image("./testworkfour/data/Chinese Medicine/dangshen/"+infer_img_path)
    infer_img = infer_img[np.newaxis,:, : ,:]  #reshape(-1,3,224,224)
    infer_img = paddle.to_tensor(infer_img)
    result = model_predict(infer_img)
    lab = np.argmax(result.numpy())
    print("样本: {},被预测为:{}".format(infer_img_path,label_dic[str(lab)]))
    # img = Image.open("./testworkfour/data/Chinese Medicine/baihe/"+infer_img_path)
    # plt.imshow(img)
    # plt.axis('off')
    # plt.show()
    # sys.stdout.flush()
    # time.sleep(0.5)

单个图片的预测

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

# 引入需要的模块
import os
import zipfile
import random
import json
import paddle
import sys
import numpy as np
from PIL import Image
import matplotlib.pyplot as plt
from paddle.io import Dataset

#所有的标签，以及数字对应的标签
label_dic = {'0': 'baihe', '1': 'dangshen', '2': 'gouqi', '3': 'huaihua', '4': 'jinyinhua'}
#所有的标签数量
class_dim = 0
for key in label_dic:      
    class_dim =class_dim+1
print(class_dim)



# 定义卷积池化网络
class ConvPool(paddle.nn.Layer):
    '''卷积+池化'''
    def __init__(self,
                 num_channels,
                 num_filters, 
                 filter_size,
                 pool_size,
                 pool_stride,
                 groups,
                 conv_stride=1, 
                 conv_padding=1,
                 ):
        super(ConvPool, self).__init__()  

        # groups代表卷积层的数量
        for i in range(groups):
            self.add_sublayer(   #添加子层实例
                'bb_%d' % i,
                paddle.nn.Conv2D(         # layer
                in_channels=num_channels, #通道数
                out_channels=num_filters,   #卷积核个数
                kernel_size=filter_size,   #卷积核大小
                stride=conv_stride,        #步长
                padding = conv_padding,    #padding
                )
            )
            self.add_sublayer(
                'relu%d' % i,
                paddle.nn.ReLU()
            )
            num_channels = num_filters
            

        self.add_sublayer(
            'Maxpool',
            paddle.nn.MaxPool2D(
            kernel_size=pool_size,           #池化核大小
            stride=pool_stride               #池化步长
            )
        )

    def forward(self, inputs):
        x = inputs
        for prefix, sub_layer in self.named_children():
            # print(prefix,sub_layer)
            x = sub_layer(x)
        return x

# VGG网络
class VGGNet(paddle.nn.Layer):
    def __init__(self):
        super(VGGNet, self).__init__()       
        # 5个卷积池化操作
        self.convpool01 = ConvPool(
            3, 64, 3, 2, 2, 2)  #3:通道数，64：卷积核个数，3:卷积核大小，2:池化核大小，2:池化步长，2:连续卷积个数
        self.convpool02 = ConvPool(
            64, 128, 3, 2, 2, 2)
        self.convpool03 = ConvPool(
            128, 256, 3, 2, 2, 3) 
        self.convpool04 = ConvPool(
            256, 512, 3, 2, 2, 3)
        self.convpool05 = ConvPool(
            512, 512, 3, 2, 2, 3)       
        self.pool_5_shape = 512 * 7* 7
        # 三个全连接层
        self.fc01 = paddle.nn.Linear(self.pool_5_shape, 4096)
        self.drop1 = paddle.nn.Dropout(p=0.5)
        self.fc02 = paddle.nn.Linear(4096, 4096)
        self.drop2 = paddle.nn.Dropout(p=0.5)
        self.fc03 = paddle.nn.Linear(4096, class_dim)

    def forward(self, inputs, label=None):
        # print('input_shape:', inputs.shape) #[8, 3, 224, 224]
        """前向计算"""
        out = self.convpool01(inputs)
        # print('convpool01_shape:', out.shape)           #[8, 64, 112, 112]
        out = self.convpool02(out)
        # print('convpool02_shape:', out.shape)           #[8, 128, 56, 56]
        out = self.convpool03(out)
        # print('convpool03_shape:', out.shape)           #[8, 256, 28, 28]
        out = self.convpool04(out)
        # print('convpool04_shape:', out.shape)           #[8, 512, 14, 14]
        out = self.convpool05(out)
        # print('convpool05_shape:', out.shape)           #[8, 512, 7, 7]         

        out = paddle.reshape(out, shape=[-1, 512*7*7])
        out = self.fc01(out)
        out = self.drop1(out)
        out = self.fc02(out)
        out = self.drop2(out)
        out = self.fc03(out)
        
        if label is not None:
            acc = paddle.metric.accuracy(input=out, label=label)
            return out, acc
        else:
            return out





def load_image(img_path):
    '''
    预测图片预处理
    '''
    img = Image.open(img_path)  
    if img.mode != 'RGB': 
        img = img.convert('RGB') 
    img = img.resize((224, 224), Image.BILINEAR)
    img = np.array(img).astype('float32') 
    print('numpy: ', img.shape)
    img = img.transpose((2, 0, 1)) / 255 # HWC to CHW 及归一化
    return img

# 加载训练过程保存的最后一个模型
model__state_dict = paddle.load('./testworkfour/checkpoints/save_dir_final.pdparams')
model_predict = VGGNet()
model_predict.set_state_dict(model__state_dict) 
model_predict.eval()


infer_img_path = "dangshen_6.jpg"
infer_img = load_image("./testworkfour/data/Chinese Medicine/dangshen/"+infer_img_path)
infer_img = infer_img[np.newaxis,:, : ,:]  #reshape(-1,3,224,224)
infer_img = paddle.to_tensor(infer_img)
result = model_predict(infer_img)
lab = np.argmax(result.numpy())
print("样本: {},被预测为:{}".format(infer_img_path,label_dic[str(lab)]))

发布成服务

发布成服务之后使用图片的base64 编码进行传输，返回预测结果

flask 安装

1
pip install flask -i http://pypi.douban.com/simple/ --trusted-host pypi.douban.com

启动模型 

1
python model_test_zcy.py

model_test_zcy.py的代码如下

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
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
import io
from flask import Flask, jsonify, request, abort
import json
import base64
import numpy as np
 
import os
from PIL import Image
import cv2
from io import BytesIO, StringIO

# 引入需要的模块
import os
import zipfile
import random
import json
import paddle
import sys
import numpy as np
from PIL import Image
import matplotlib.pyplot as plt
from paddle.io import Dataset

app = Flask(__name__)

#所有的标签，以及数字对应的标签
label_dic = {'0': 'baihe', '1': 'dangshen', '2': 'gouqi', '3': 'huaihua', '4': 'jinyinhua'}
#所有的标签数量
class_dim = 0
for key in label_dic:      
    class_dim =class_dim+1
print(class_dim)



@app.route('/zcy', methods=['GET', 'POST'])
def call_analysis():
    print("Hello, World!")
 
    data1 = request.data   #----获取的是字符串
    #print(data1)
    data2 = request.get_data()  #----获取的是字符串
    #print(data2)
    j_data =  json.loads(data2) #-----load将字符串解析成json
    # print(j_data)
    # print(j_data['base64']) 
    
    # decode_base64_matplot_img(j_data['base64'])
    # return ""
    return getAnalysis(j_data['base64'])


def getAnalysis(imgbase64):

   

    # 加载训练过程保存的最后一个模型
    model__state_dict = paddle.load('./testworkfour/checkpoints/save_dir_final.pdparams')
    model_predict = VGGNet()
    model_predict.set_state_dict(model__state_dict) 
    model_predict.eval()
 
    infer_img = load_image2(imgbase64)
    infer_img = infer_img[np.newaxis,:, : ,:]  #reshape(-1,3,224,224)
    infer_img = paddle.to_tensor(infer_img)
    result = model_predict(infer_img)
    print(result.numpy())
    lab = np.argmax(result.numpy())
    print("样本: {},被预测为:{}".format("1",label_dic[str(lab)]))
    return "样本被预测为:{}".format(label_dic[str(lab)])

# def getbase64(imgbase64):
 
    
#     # 传入为RGB格式下的base64，传出为RGB格式的numpy矩阵
#     byte_data = base64.b64decode(imgbase64)#将base64转换为二进制
#     encode_image = np.asarray(bytearray(byte_data), dtype="uint8")# 二进制转换为一维数组
#     img_array = cv2.imdecode(encode_image, cv2.IMREAD_COLOR)# 用cv2解码为三通道矩阵
#     img_array = cv2.cvtColor(img_array, cv2.COLOR_BGR2RGB)# BGR2RGB
 
#     img1 = img_array
#     imgs = [img1]
#     #img2 = cv2.imread('./data/images/zidane.jpg')[:,:,::-1]
#     #imgs = [img2]

#     results = model(imgs, size=640)
#     print('识别成功')
#     results.ims
#     results.render()
#     base64_image = ''
#     for img in results.ims:
#         buffered = BytesIO()
#         img_base64 = Image.fromarray(img)
#         img_base64.save(buffered, format='JPEG')
#         base64_image = base64.b64encode(buffered.getvalue()).decode('utf-8')
#         # with open('./result.txt', 'a+') as f:
#         #     f.write(base64_image)
#         #     f.close()
#     print(results)
#     results.save()
#     base64_image = 'data:image/jpeg;base64,%s' % base64_image
#     return base64_image




# 定义卷积池化网络
class ConvPool(paddle.nn.Layer):
    '''卷积+池化'''
    def __init__(self,
                 num_channels,
                 num_filters, 
                 filter_size,
                 pool_size,
                 pool_stride,
                 groups,
                 conv_stride=1, 
                 conv_padding=1,
                 ):
        super(ConvPool, self).__init__()  

        # groups代表卷积层的数量
        for i in range(groups):
            self.add_sublayer(   #添加子层实例
                'bb_%d' % i,
                paddle.nn.Conv2D(         # layer
                in_channels=num_channels, #通道数
                out_channels=num_filters,   #卷积核个数
                kernel_size=filter_size,   #卷积核大小
                stride=conv_stride,        #步长
                padding = conv_padding,    #padding
                )
            )
            self.add_sublayer(
                'relu%d' % i,
                paddle.nn.ReLU()
            )
            num_channels = num_filters
            

        self.add_sublayer(
            'Maxpool',
            paddle.nn.MaxPool2D(
            kernel_size=pool_size,           #池化核大小
            stride=pool_stride               #池化步长
            )
        )

    def forward(self, inputs):
        x = inputs
        for prefix, sub_layer in self.named_children():
            # print(prefix,sub_layer)
            x = sub_layer(x)
        return x

# VGG网络
class VGGNet(paddle.nn.Layer):
    def __init__(self):
        super(VGGNet, self).__init__()       
        # 5个卷积池化操作
        self.convpool01 = ConvPool(
            3, 64, 3, 2, 2, 2)  #3:通道数，64：卷积核个数，3:卷积核大小，2:池化核大小，2:池化步长，2:连续卷积个数
        self.convpool02 = ConvPool(
            64, 128, 3, 2, 2, 2)
        self.convpool03 = ConvPool(
            128, 256, 3, 2, 2, 3) 
        self.convpool04 = ConvPool(
            256, 512, 3, 2, 2, 3)
        self.convpool05 = ConvPool(
            512, 512, 3, 2, 2, 3)       
        self.pool_5_shape = 512 * 7* 7
        # 三个全连接层
        self.fc01 = paddle.nn.Linear(self.pool_5_shape, 4096)
        self.drop1 = paddle.nn.Dropout(p=0.5)
        self.fc02 = paddle.nn.Linear(4096, 4096)
        self.drop2 = paddle.nn.Dropout(p=0.5)
        self.fc03 = paddle.nn.Linear(4096, class_dim)

    def forward(self, inputs, label=None):
        # print('input_shape:', inputs.shape) #[8, 3, 224, 224]
        """前向计算"""
        out = self.convpool01(inputs)
        # print('convpool01_shape:', out.shape)           #[8, 64, 112, 112]
        out = self.convpool02(out)
        # print('convpool02_shape:', out.shape)           #[8, 128, 56, 56]
        out = self.convpool03(out)
        # print('convpool03_shape:', out.shape)           #[8, 256, 28, 28]
        out = self.convpool04(out)
        # print('convpool04_shape:', out.shape)           #[8, 512, 14, 14]
        out = self.convpool05(out)
        # print('convpool05_shape:', out.shape)           #[8, 512, 7, 7]         

        out = paddle.reshape(out, shape=[-1, 512*7*7])
        out = self.fc01(out)
        out = self.drop1(out)
        out = self.fc02(out)
        out = self.drop2(out)
        out = self.fc03(out)
        
        if label is not None:
            acc = paddle.metric.accuracy(input=out, label=label)
            return out, acc
        else:
            return out





def load_image(img_path):
    '''
    预测图片预处理
    '''
    img = Image.open(img_path) 
    if img.mode != 'RGB': 
        img = img.convert('RGB') 
    img = img.resize((224, 224), Image.BILINEAR)
    img = np.array(img).astype('float32') 
    img = img.transpose((2, 0, 1)) / 255 # HWC to CHW 及归一化
    return img

def load_image2(img_base64):

  
    '''
    预测图片预处理
    '''
    # byte_data = base64.b64decode(img_base64)#将base64转换为二进制
    # img = np.fromstring(byte_data, np.uint8)  # 转换np序列
 


    # img = np.asarray(bytearray(byte_data), dtype="uint8")# 二进制转换为一维数组
    # img = Image.fromarray(cv2.cvtColor(img, cv2.COLOR_BGR2RGB))

    # plt.imshow(img)
    # plt.axis('off')
    # plt.show()
    # sys.stdout.flush()

    img = base64.b64decode(img_base64)
    buffer = io.BytesIO() 
    img = Image.open(io.BytesIO(img))
    # img = Image.open(img_base64) 
    if img.mode != 'RGB': 
        img = img.convert('RGB') 
    img = img.resize((224, 224), Image.BILINEAR)
    img = np.array(img).astype('float32') 
    print('numpy: ', img.shape)
    img = img.transpose((2, 0, 1)) / 255 # HWC to CHW 及归一化
    return img

# 解码base64字符串为matplot图像
def decode_base64_matplot_img(base64_data):
    img = base64.b64decode(base64_data)
    img_array = np.fromstring(img, np.uint8)  # 转换np序列
    img_raw = cv2.imdecode(img_array, cv2.IMREAD_COLOR)  # 转换Opencv格式BGR
    img_matplot = cv2.cvtColor(img_raw, cv2.COLOR_BGR2RGB)  # BGR转RGB

    img_gray = cv2.imdecode(img_array, cv2.IMREAD_GRAYSCALE)  # 转换灰度图
    imggray_matplot = cv2.cvtColor(img_gray, cv2.COLOR_GRAY2RGB)  # 灰度图转RGB
    plt.figure()
    plt.title("Matplot RGB Origin Image")
    plt.axis("off")
    plt.imshow(img_matplot)

    plt.figure()
    plt.title("Matplot Gray Origin Image")
    plt.axis("off")
    plt.imshow(imggray_matplot)
    plt.show() 

if __name__ == '__main__':
    app.run(host='0.0.0.0', port=5000, debug = False)

#json模块编码： json.dumps()
#json模块解码：解码python json格式，用json.loads()

label_dic是模型的所有标签，将数据处理的那个地方直接拿取出来的，可直接使用
class_dim 是所有的标签数量

若没有报错则可正常启动

1
2
3
4
5
6
7
8
(base) [root@localhost paddle]# python model_test_zct.py 
 * Serving Flask app 'model_test_zcy'
 * Debug mode: off
WARNING: This is a development server. Do not use it in a production deployment. Use a production WSGI server instead.
 * Running on all addresses (0.0.0.0)
 * Running on http://127.0.0.1:5000
 * Running on http://192.168.23.134:5000
Press CTRL+C to quit

则可以使用代码外部调用

一辈子很短，努力的做好两件事就好；
第一件事是热爱生活，好好的去爱身边的人；
第二件事是努力学习，在工作中取得不一样的成绩，实现自己的价值，而不是仅仅为了赚钱；

继开 wechat
欢迎加我的微信，共同交流技术