#987

Build # #987

Build Type

push

travis-ci

Committed by GitHub

Commit Message

build(deps): bump decode-uri-component in /waves/javascript_demo

Run Details

910 of 910 relevant lines covered (100.0%)

1.0 hits per line

Source File
Press 'n' to go to next uncovered line, 'b' for previous

100.0

/neural_network/mnist_network.py

from neural_network import InputNode
from neural_network import L2ErrorNode
from neural_network import LinearNode
from neural_network import NeuralNetwork
from neural_network import ReluNode
from neural_network import SigmoidNode
from random import shuffle
import os


def load_1s_and_7s(filename):
    print('Loading data {}...'.format(filename))
    examples = []
    with open(filename, 'r') as infile:
        for line in infile:
            if line[0] in ['1', '7']:
                tokens = [int(x) for x in line.split(',')]
                label = tokens[0]
                example = [x / 255 for x in tokens[1:]]  # scale to [0,1]
                if label == 1:
                    examples.append([example, 0])
                elif label == 7:
                    examples.append([example, 1])
    print('Data loaded.')
    return examples


def print_example(example):
    for i, pixel in enumerate(example):
        if i % 28 == 0:
            print()
        print('%4d' % int(pixel * 255), end='')


def show_random_examples(network, test, n=5):
    test = test[:]
    shuffle(test)
    for i in range(n):
        example, label = test[i]
        print_example(example)
        print("\nExample with label {} is predicted to have label {}".format(
            label, network.evaluate(example)))


def build_network():
    input_nodes = InputNode.make_input_nodes(28*28)

    first_layer = [LinearNode(input_nodes) for i in range(10)]
    first_layer_relu = [ReluNode(L) for L in first_layer]

    second_layer = [LinearNode(first_layer_relu) for i in range(10)]
    second_layer_relu = [ReluNode(L) for L in second_layer]

    linear_output = LinearNode(second_layer_relu)
    output = SigmoidNode(linear_output)
    error_node = L2ErrorNode(output)
    network = NeuralNetwork(
        output, input_nodes, error_node=error_node, step_size=0.05)

    return network


cant_find_files = '''
Was unable to find the files {}, {}.

You may have to extract them from the gzipped tarball mnist/mnist.tar.gz.
'''


def train_mnist(data_dirname, num_epochs=5):
    train_file = os.path.join(data_dirname, 'mnist_train.csv')
    test_file = os.path.join(data_dirname, 'mnist_test.csv')
    try:
        train = load_1s_and_7s(train_file)
        test = load_1s_and_7s(test_file)
    except Exception:  # pragma: no cover
        print(cant_find_files.format(train_file, test_file))
        raise

    network = build_network()
    n = len(train)
    epoch_size = int(n/10)

    for i in range(num_epochs):
        shuffle(train)
        validation = train[:epoch_size]
        real_train = train[epoch_size: 2*epoch_size]

        print("Starting epoch of {} examples with {} validation".format(
            len(real_train), len(validation)))

        network.train(real_train, max_steps=len(real_train))

        print("Finished epoch. Validation error={:.3f}".format(
            network.error_on_dataset(validation)))

    print("Test error={:.3f}".format(network.error_on_dataset(test)))
    show_random_examples(network, test)
    return network


if __name__ == "__main__":
    data_dirname = os.path.join(os.path.dirname(__file__), 'mnist')
    network = train_mnist(data_dirname)

1	from neural_network import InputNode	1✔
2	from neural_network import L2ErrorNode	1✔
3	from neural_network import LinearNode	1✔
4	from neural_network import NeuralNetwork	1✔
5	from neural_network import ReluNode	1✔
6	from neural_network import SigmoidNode	1✔
7	from random import shuffle	1✔
8	import os	1✔
9
10
11	def load_1s_and_7s(filename):	1✔
12	print('Loading data {}...'.format(filename))	1✔
13	examples = []	1✔
14	with open(filename, 'r') as infile:	1✔
15	for line in infile:	1✔
16	if line[0] in ['1', '7']:	1✔
17	tokens = [int(x) for x in line.split(',')]	1✔
18	label = tokens[0]	1✔
19	example = [x / 255 for x in tokens[1:]] # scale to [0,1]	1✔
20	if label == 1:	1✔
21	examples.append([example, 0])	1✔
22	elif label == 7:	1✔
23	examples.append([example, 1])	1✔
24	print('Data loaded.')	1✔
25	return examples	1✔
26
27
28	def print_example(example):	1✔
29	for i, pixel in enumerate(example):	1✔
30	if i % 28 == 0:	1✔
31	print()	1✔
32	print('%4d' % int(pixel * 255), end='')	1✔
33
34
35	def show_random_examples(network, test, n=5):	1✔
36	test = test[:]	1✔
37	shuffle(test)	1✔
38	for i in range(n):	1✔
39	example, label = test[i]	1✔
40	print_example(example)	1✔
41	print("\nExample with label {} is predicted to have label {}".format(	1✔
42	label, network.evaluate(example)))
43
44
45	def build_network():	1✔
46	input_nodes = InputNode.make_input_nodes(28*28)	1✔
47
48	first_layer = [LinearNode(input_nodes) for i in range(10)]	1✔
49	first_layer_relu = [ReluNode(L) for L in first_layer]	1✔
50
51	second_layer = [LinearNode(first_layer_relu) for i in range(10)]	1✔
52	second_layer_relu = [ReluNode(L) for L in second_layer]	1✔
53
54	linear_output = LinearNode(second_layer_relu)	1✔
55	output = SigmoidNode(linear_output)	1✔
56	error_node = L2ErrorNode(output)	1✔
57	network = NeuralNetwork(	1✔
58	output, input_nodes, error_node=error_node, step_size=0.05)
59
60	return network	1✔
61
62
63	cant_find_files = '''	1✔
64	Was unable to find the files {}, {}.
65
66	You may have to extract them from the gzipped tarball mnist/mnist.tar.gz.
67	'''
68
69
70	def train_mnist(data_dirname, num_epochs=5):	1✔
71	train_file = os.path.join(data_dirname, 'mnist_train.csv')	1✔
72	test_file = os.path.join(data_dirname, 'mnist_test.csv')	1✔
73	try:	1✔
74	train = load_1s_and_7s(train_file)	1✔
75	test = load_1s_and_7s(test_file)	1✔
76	except Exception: # pragma: no cover
77	print(cant_find_files.format(train_file, test_file))
78	raise
79
80	network = build_network()	1✔
81	n = len(train)	1✔
82	epoch_size = int(n/10)	1✔
83
84	for i in range(num_epochs):	1✔
85	shuffle(train)	1✔
86	validation = train[:epoch_size]	1✔
87	real_train = train[epoch_size: 2*epoch_size]	1✔
88
89	print("Starting epoch of {} examples with {} validation".format(	1✔
90	len(real_train), len(validation)))
91
92	network.train(real_train, max_steps=len(real_train))	1✔
93
94	print("Finished epoch. Validation error={:.3f}".format(	1✔
95	network.error_on_dataset(validation)))
96
97	print("Test error={:.3f}".format(network.error_on_dataset(test)))	1✔
98	show_random_examples(network, test)	1✔
99	return network	1✔
100
101
102	if __name__ == "__main__":
103	data_dirname = os.path.join(os.path.dirname(__file__), 'mnist')
104	network = train_mnist(data_dirname)

pim-book / programmers-introduction-to-mathematics / #987

Source File Press 'n' to go to next uncovered line, 'b' for previous

Source File
Press 'n' to go to next uncovered line, 'b' for previous