In [1]:

    

import numpy as np
import json
from keras.models import Model
from keras.layers import Input
from keras.layers.convolutional import Conv2D
from keras.layers.pooling import MaxPooling2D, AveragePooling2D
from keras.layers.normalization import BatchNormalization
from keras import backend as K


    

    




Using TensorFlow backend.

In [2]:

    

def format_decimal(arr, places=8):
    return [round(x * 10**places) / 10**places for x in arr]


    

In [3]:

    

data_in_shape = (8, 8, 2)

conv_0 = Conv2D(4, 3, 3, activation='relu', border_mode='valid', subsample=(1, 1), dim_ordering='tf', bias=True)
pool_0 = MaxPooling2D(pool_size=(2, 2), strides=None, border_mode='valid', dim_ordering='tf')

input_layer = Input(shape=data_in_shape)
x = conv_0(input_layer)
output_layer = pool_0(x)
model = Model(input=input_layer, output=output_layer)

np.random.seed(6000)
data_in = 2 * np.random.random(data_in_shape) - 1

# set weights to random (use seed for reproducibility)
weights = []
for i, w in enumerate(model.get_weights()):
    np.random.seed(6000 + i)
    weights.append(2 * np.random.random(w.shape) - 1)
model.set_weights(weights)

result = model.predict(np.array([data_in]))

print({
    'input': {'data': format_decimal(data_in.ravel().tolist()), 'shape': list(data_in_shape)},
    'weights': [{'data': format_decimal(weights[i].ravel().tolist()), 'shape': list(weights[i].shape)} for i in range(len(weights))],
    'expected': {'data': format_decimal(result[0].ravel().tolist()), 'shape': list(result[0].shape)}
})


    

    




{'weights': [{'shape': [3, 3, 2, 4], 'data': [-0.33704585, -0.76599005, 0.43704017, -0.5908736, 0.52324417, 0.08437049, -0.4699497, -0.17460028, -0.7879552, 0.47957678, 0.35859039, -0.1946386, -0.04222754, 0.43957667, 0.33056406, 0.17665698, 0.03271313, 0.96557731, -0.14530269, -0.42844754, -0.9892456, -0.6573214, 0.9645642, 0.53937169, 0.03144698, 0.17482522, -0.01958232, 0.35636633, 0.03950257, 0.29152626, -0.86949394, 0.47902437, -0.11383247, -0.67173347, -0.27645687, -0.56208137, -0.38046996, -0.89436374, 0.76776252, 0.607511, 0.0836086, 0.20298698, -0.65466554, 0.05431541, 0.2909275, 0.7421486, 0.55794809, -0.0110642, 0.05647404, -0.4401333, 0.53312743, 0.75781162, 0.91153872, 0.94005194, -0.17789221, -0.017994, -0.28008426, 0.52780133, -0.04725771, 0.60518146, 0.60948463, 0.07878468, -0.76871217, -0.06945041, 0.57079619, -0.51862638, 0.85541681, 0.3306379, 0.61299955, 0.61458121, -0.94657194, -0.10072256]}, {'shape': [4], 'data': [0.17620778, -0.91407573, -0.25756853, -0.54019677]}], 'expected': {'shape': [3, 3, 4], 'data': [2.35519338, 1.53862083, 0.87997359, 0.0, 1.48876822, 0.81337726, 1.00681305, 0.81853366, 1.51867306, 0.51967287, 0.52789873, 0.26385701, 1.21009409, 0.45989263, 1.84282112, 0.90611005, 0.55074763, 1.10996759, 1.62130594, 0.69155359, 0.19971609, 0.0, 0.41694176, 0.0, 0.0, 0.92262471, 1.62677503, 0.03982544, 2.06196785, 0.1178565, 2.04240298, 1.91396677, 2.27832818, 0.73671174, 1.12158275, 0.99239504]}, 'input': {'shape': [8, 8, 2], 'data': [-0.33704585, -0.76599005, 0.43704017, -0.5908736, 0.52324417, 0.08437049, -0.4699497, -0.17460028, -0.7879552, 0.47957678, 0.35859039, -0.1946386, -0.04222754, 0.43957667, 0.33056406, 0.17665698, 0.03271313, 0.96557731, -0.14530269, -0.42844754, -0.9892456, -0.6573214, 0.9645642, 0.53937169, 0.03144698, 0.17482522, -0.01958232, 0.35636633, 0.03950257, 0.29152626, -0.86949394, 0.47902437, -0.11383247, -0.67173347, -0.27645687, -0.56208137, -0.38046996, -0.89436374, 0.76776252, 0.607511, 0.0836086, 0.20298698, -0.65466554, 0.05431541, 0.2909275, 0.7421486, 0.55794809, -0.0110642, 0.05647404, -0.4401333, 0.53312743, 0.75781162, 0.91153872, 0.94005194, -0.17789221, -0.017994, -0.28008426, 0.52780133, -0.04725771, 0.60518146, 0.60948463, 0.07878468, -0.76871217, -0.06945041, 0.57079619, -0.51862638, 0.85541681, 0.3306379, 0.61299955, 0.61458121, -0.94657194, -0.10072256, -0.20867325, 0.67206388, -0.7091287, -0.37040119, -0.19479755, -0.27238751, 0.40815762, 0.46682499, 0.14357759, 0.70909437, 0.72072034, -0.34537329, 0.94900334, 0.7811956, -0.05788226, 0.55793823, -0.58275979, -0.9408907, -0.99961866, -0.61331503, 0.31771212, -0.97225187, 0.5547143, 0.29870985, 0.09646246, 0.18945358, -0.90285618, -0.13699646, 0.91833066, 0.39441019, 0.33963873, -0.78466329, 0.60209752, -0.94062026, 0.18010792, 0.92858172, -0.52799242, 0.41509646, 0.54362401, 0.58928085, 0.30124276, -0.25104719, 0.25849856, -0.40750885, 0.03137015, 0.47267263, 0.96452281, 0.78649294, 0.41429247, -0.85004428, 0.1775019, -0.475431, 0.04413765, 0.0808915, -0.25057979, 0.17138563]}}

In [ ]: