master
/ src / .ipynb_checkpoints / main-checkpoint.py

main-checkpoint.py @c038f7f — raw · history · blame 

  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
import os
import sys
import utils

from predictor import *

# sys.path.insert(0, os.path.abspath("../"))
sys.path.insert(0, os.path.dirname(__file__))


class sceneclassificationvgg19(object):

    def __init__(self, input={}):
        '''

        :param input:
        '''
        # self.checkpoint_dir = os.path.dirname(__file__) + "/checkpoint"
        pass

    def train(self, input={}):
        '''

        :param input:
        :return:
        '''

        # 必须参数
        path_train_orgin_images = input.get("path_train_orgin_images")  # 原始train图片数据集的目录
        path_val_orgin_images = input.get("path_val_orgin_images")  # 原始val图片数据集的目录
        path_orgin_json = input.get("path_orgin_json")  # 原始数据集中图片名和类别对应的json
        path_orgin_csv = input.get("path_orgin_csv")  # 原始数据集中类别csv表
        path_train_images = input.get("path_train_images")  # 此处的训练数据路径是分类后的classes路径
        path_val_images = input.get("path_val_images")  # 此处的训练数据路径是分类后的classes路径

        # 可选参数
        model_name = input.get("model_name", "small")
        learning_rate = input.get("learning_rate", 1e-3)
        batch_size = input.get("batch_size", 32)
        image_size = input.get("image_size", 224)
        optimizer = input.get("optimizer", None)  # 默认是Adam
        weights_mode = input.get("weights_mode", "loss")
        epoch = input.get("epoch", 100)

        # 数据预处理部分
        utils.split_train_and_val_to_classes(path_train_images, path_train_orgin_images, path_orgin_json,
                                             path_orgin_csv, False, True, False)
        utils.split_train_and_val_to_classes(path_val_images, path_val_orgin_images, path_orgin_json,
                                             path_orgin_csv, False, True, False)

        # 训练部分
        classes = len(os.listdir(path_train_images))  # 类型数len(os.listdir(path_train_images))
        classifier = BaseClassifier(name=model_name, image_size=image_size, learning_rate=learning_rate,
                                    batch_size=batch_size, classes=classes, weights_mode=weights_mode,
                                    optimizer=optimizer, epoch=epoch, path_train_images=path_train_images,
                                    path_val_images=path_val_images)

        classifier.train()

        pass

    def predict(self, input={}):
        '''

        :param input:
        :return:
        '''

        # 可选参数
        image_path = input.get("image_path", "./image/00f076d9b6ab784f69c0e43e77853d7c24d62342.jpg")

        path_train_images = None
        path_val_images = None
        model_name = "small"
        learning_rate = 1e-3
        batch_size = 32
        image_size = 224
        classes = 80  # 原始训练模型有80个场景类别
        optimizer = None  # 默认是Adam
        weights_mode = "loss"
        epoch = 100

        # single predictor
        predictor = KerasPredictor(BaseClassifier(name=model_name, image_size=image_size, learning_rate=learning_rate,
                                                  batch_size=batch_size, classes=classes, weights_mode=weights_mode,
                                                  optimizer=optimizer, epoch=epoch, path_train_images=path_train_images,
                                                  path_val_images=path_val_images), 'val')

        # integrated predictor
        # predictor = IntegratedPredictor([
        #     KerasPredictor(VGG16Classifier("vgg16_little"), 'test'),
        #     KerasPredictor(VGG16Classifier("vgg16_little"), 'val'),
        # ])

        prediction = predictor(image_path, return_with_prob=True)
        print(prediction)

        pass

    def load_model(self, file=os.path.dirname(__file__) + "/checkpoint" + "/sceneclassification10.pkl"):
        '''

        :param input:
        :return:
        '''
        pass


## Note: Uncomment this block before creating a crowdsourcing task

if __name__ == '__main__':
    func_name = sys.argv[0]
    # 预测选定图片的前三种类别及其准确率
    sceneclassificationvgg19(object).predict(
        input={"image_path": "./image/00f076d9b6ab784f69c0e43e77853d7c24d62342.jpg"})

    # sceneclassificationvgg19(object).train(
    #     input={
    #         "path_train_images": "./data/ai_challenger_scene_train_20170904/classes",
    #         "path_val_images": "./data/ai_challenger_scene_validation_20170908/classes"})

    # sceneclassificationvgg19(object).train(
    #     input={
    #         "path_train_images": "D:\DEEPLEARNING\project\sceneClassificationTest\data/ai_challenger_scene_train_20170904/classes",
    #         "path_val_images": "D:\DEEPLEARNING\project\sceneClassificationTest\data/ai_challenger_scene_validation_20170908/classes"})