cat-dogs_tensorflow/DataProvider.py at master · pawelpeksa/cat-dogs_tensorflow · GitHub

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
from Utils import Utils
import tensorflow as tf
import skimage.io
import numpy
from keras.preprocessing.image import ImageDataGenerator, array_to_img, img_to_array, load_img
from Configuration import Configuration as Config
import matplotlib.pyplot as plt
import scipy.misc

class DataProvider:
    def __init__(self, train_folder, submission_folder, verification_folder):

        self.verification_images_count = 0
        self.train_images_count = 0

        self._train_folder = train_folder
        self._submission_folder = submission_folder
        self._verification_folder = verification_folder

        self._submission_file_paths = list()

        self._train_image_files = list()
        self._train_labels = list()

        self._test_image_files = list()
        self._test_labels = list()

        self.load_train_data_info()
        self.load_submission_data_info()

    def extract_image_file_names_with_labels(self, folder):
        files = Utils.files_in_path(folder)

        image_files = []
        image_labels = []

        for imageFilePath in files:
            label_name = imageFilePath[0:3]

            label = [0, 0]

            if label_name == "cat":
                label[0] = 1
            elif label_name == "dog":
                label[1] = 1

            image_files.append(folder + "/" + imageFilePath)
            image_labels.append(label)

        return image_files, image_labels

    def test_data_batch(self):
        test_image_files = self._test_image_files[0:Config.batch_size]
        test_labels = self._test_labels[0:Config.batch_size]

        self._test_image_files = self._test_image_files[Config.batch_size:]
        self._test_labels = self._test_labels[Config.batch_size:]

        images = list()

        for file_path in test_image_files:
            image = self.load_test_image(file_path)
            images.append(image)

        return numpy.array(images), numpy.array(test_labels)


    def next_data_batch(self):
        if len(self._train_image_files) < Config.batch_size:
            self.load_train_data_info()

        train_image_files = self._train_image_files[0:Config.batch_size]
        train_labels = self._train_labels[0:Config.batch_size]

        self._train_image_files = self._train_image_files[Config.batch_size:]
        self._train_labels = self._train_labels[Config.batch_size:]

        images = list()

        for file_path in train_image_files:
            image = self.load_train_image(file_path)
            images.append(image)

        return numpy.array(images), numpy.array(train_labels)


    def load_verification_data_info(self):
        self._test_image_files, self._test_labels = self.extract_image_file_names_with_labels(self._verification_folder)
        self.verification_images_count = len(self._test_image_files)

    def load_train_data_info(self):
        self._train_image_files, self._train_labels = self.extract_image_file_names_with_labels(self._train_folder)
        self._test_image_files, self._test_labels = self.extract_image_file_names_with_labels(self._verification_folder)

        self.verification_images_count = len(self._test_image_files)
        self.train_images_count = len(self._train_image_files)

    def load_submission_data_info(self):
        for i in range(1, 125001):
            file_path = self._submission_folder + "/" + str(i) + ".jpg"
            self._submission_file_paths.append(file_path)

    def submission_data_batch(self, batch_size):
        sliced_list = self._submission_file_paths[:batch_size]
        self._submission_file_paths = self._submission_file_paths[batch_size:]

        images = list()

        for file_path in sliced_list:
            image = self.load_image(file_path)
            images.append(image)

        return numpy.array(images), sliced_list

    @staticmethod
    def flatten(matrix):
        return [item for vector in matrix for item in vector]

    @staticmethod
    def add_noise(image):

        datagen = ImageDataGenerator(
            rotation_range=40,
            width_shift_range=0.2,
            height_shift_range=0.2,
            rescale=1./255,
            shear_range=0.2,
            zoom_range=0.2,
            horizontal_flip=True,
            fill_mode='nearest'
        )

        x = image.reshape((1,) + image.shape)

        for batch in datagen.flow(x, batch_size=1):
            return batch[0]

    @staticmethod
    def scale(image):
        datagen = ImageDataGenerator(
            rescale=1./255
        )

        x = image.reshape((1,) + image.shape)

        for batch in datagen.flow(x, batch_size=1):
            return batch[0]

    @staticmethod
    def load_train_image(image_file):
        image = skimage.io.imread(image_file).astype(numpy.float32)
        image = DataProvider.add_noise(image)
        image = DataProvider.flatten(image)
        return image

    @staticmethod
    def load_test_image(image_file):
        image = skimage.io.imread(image_file).astype(numpy.float32)
        image = DataProvider.scale(image)
        image = DataProvider.flatten(image)
        return image