MattChanTK

import random
import numpy as np
import math
from time import perf_counter
import os
import sys
from collections import deque

import gym
import cntk

MattChanTK

import random
import numpy as np
import math
from time import perf_counter
import os
import sys
from collections import deque

import gym
import cntk

MattChanTK

import random
import numpy as np
import math
from time import perf_counter
import os
import sys
from collections import deque

import gym
import cntk

MattChanTK

import random
import numpy as np
import math
from time import perf_counter
import os
import sys
from collections import deque

import gym
import cntk

MattChanTK

#!/bin/bash
#
# ==============================================================================
# Copyright (c) Microsoft. All rights reserved.
# Licensed under the MIT license. See LICENSE.md file in the project root
# for full license information.
# ==============================================================================

# Log steps, stop on error
# TODO cut down on logging

MattChanTK

def create_convolutional_neural_network(input_vars, out_dims, dropout_prob=0.0):

    convolutional_layer_1 = Convolution((5, 5), 32, strides=1, activation=cntk.ops.relu, pad=True, init=gaussian(), init_bias=0.1)(input_vars)
    pooling_layer_1 = MaxPooling((2, 2), strides=(2, 2), pad=True)(convolutional_layer_1)

    convolutional_layer_2 = Convolution((5, 5), 64, strides=1, activation=cntk.ops.relu, pad=True, init=gaussian(), init_bias=0.1)(pooling_layer_1)
    pooling_layer_2 = MaxPooling((2, 2), strides=(2, 2), pad=True)(convolutional_layer_2)

    fully_connected_layer = Dense(1024, activation=cntk.ops.relu, init=gaussian(), init_bias=0.1)(pooling_layer_2)
    dropout_layer = Dropout(dropout_prob)(fully_connected_layer)

MattChanTK

'''
-------------------
Classification Test
--------------------
'''
test_minibatch_size = 1000

sample_count = 0
test_results = []

MattChanTK

'''
-----------------------------------------
Training the Convolutional Neural Network
-----------------------------------------
'''
num_training_epoch = 1
training_progress_output_freq = 10

for epoch in range(num_training_epoch):

MattChanTK

'''
----------------------
Setting up the trainer
----------------------
'''
# Define the label as the other input parameter of the trainer
labels = cntk.ops.input_variable(output_dim, np.float32)

#Initialize the parameters for the trainer
train_minibatch_size = 50

MattChanTK

'''
---------------------------------------------
Constructing the Convolutional Neural Network
---------------------------------------------
'''
def create_convolutional_neural_network(input_vars, out_dims, dropout_prob=0.0):

    convolutional_layer_1 = Convolution((5, 5), 32, strides=1, activation=cntk.ops.relu, pad=True)(input_vars)
    pooling_layer_1 = MaxPooling((2, 2), strides=(2, 2), pad=True)(convolutional_layer_1)