Jigsaw-Solver/model.py at master · shivaverma/Jigsaw-Solver · 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
import keras
from keras.layers import TimeDistributed as td
from keras.layers import Conv2D, Flatten, Dense, ZeroPadding2D, Activation
from keras.layers import MaxPooling2D, Dropout, BatchNormalization, Reshape


def get_model():

    model = keras.models.Sequential()

    model.add(td(ZeroPadding2D(2), input_shape=(4, 100, 100, 3)))

    model.add(td(Conv2D(50, kernel_size=(5,5), padding='same', activation='relu', strides=2)))
    model.add(td(BatchNormalization()))
    model.add(td(MaxPooling2D()))

    model.add(td(Conv2D(100, kernel_size=(5,5), padding='same', activation='relu', strides=2)))
    model.add(td(BatchNormalization()))
    model.add(td(Dropout(0.3)))

    model.add(td(Conv2D(100, kernel_size=(3,3), padding='same', activation='relu', strides=2)))
    model.add(td(BatchNormalization()))
    model.add(td(Dropout(0.3)))

    model.add(td(Conv2D(200, kernel_size=(3,3), padding='same', activation='relu', strides=1)))
    model.add(td(BatchNormalization()))
    model.add(td(Dropout(0.3)))

    model.add(Flatten())

    model.add(Dense(600, activation='relu'))
    model.add(BatchNormalization())
    model.add(Dense(400, activation='relu'))
    model.add(BatchNormalization())
    model.add(Dropout(0.3))
    model.add(Dense(16))
    model.add(Reshape((4, 4)))
    model.add(Activation('softmax'))

    return model