mnist_double_cnn Namespace Reference

Variables
int	batch_size = 128
int	num_classes = 10
int	epochs = 100
	img_rows
	img_cols
	x_test = x_test.reshape(x_test.shape[0], 1, img_rows, img_cols)
	y_test = keras.utils.to_categorical(y_test, num_classes)
	x_train = x_train.reshape(x_train.shape[0], 1, img_rows, img_cols)
tuple	input_shape = (1, img_rows, img_cols)
	y_train = keras.utils.to_categorical(y_train, num_classes)
string	kernel_init = 'he_uniform'
	model = Sequential()
	loss
	optimizer
	metrics
	verbose
	validation_data
	score = model.evaluate(x_test, y_test, verbose=0)
	json_string = model.to_json()

Detailed Description

Trains a simple convnet on the MNIST dataset.

Gets to 99.25% test accuracy after 12 epochs
(there is still a lot of margin for parameter tuning).
16 seconds per epoch on a GRID K520 GPU.

Variable Documentation

mnist_double_cnn.batch_size = 128

Definition at line 18 of file mnist_double_cnn.py.

mnist_double_cnn.epochs = 100

Definition at line 20 of file mnist_double_cnn.py.

mnist_double_cnn.img_cols

Definition at line 23 of file mnist_double_cnn.py.

mnist_double_cnn.img_rows

Definition at line 23 of file mnist_double_cnn.py.

tuple mnist_double_cnn.input_shape = (1, img_rows, img_cols)

Definition at line 31 of file mnist_double_cnn.py.

mnist_double_cnn.json_string = model.to_json()

Definition at line 84 of file mnist_double_cnn.py.

string mnist_double_cnn.kernel_init = 'he_uniform'

Definition at line 50 of file mnist_double_cnn.py.

mnist_double_cnn.loss

Definition at line 66 of file mnist_double_cnn.py.

mnist_double_cnn.metrics

Definition at line 68 of file mnist_double_cnn.py.

mnist_double_cnn.model = Sequential()

Definition at line 51 of file mnist_double_cnn.py.

int mnist_double_cnn.num_classes = 10

Definition at line 19 of file mnist_double_cnn.py.

mnist_double_cnn.optimizer

Definition at line 67 of file mnist_double_cnn.py.

mnist_double_cnn.score = model.evaluate(x_test, y_test, verbose=0)

Definition at line 78 of file mnist_double_cnn.py.

mnist_double_cnn.validation_data

Definition at line 77 of file mnist_double_cnn.py.

mnist_double_cnn.verbose

Definition at line 76 of file mnist_double_cnn.py.

mnist_double_cnn.x_test = x_test.reshape(x_test.shape[0], 1, img_rows, img_cols)

Definition at line 26 of file mnist_double_cnn.py.

mnist_double_cnn.x_train = x_train.reshape(x_train.shape[0], 1, img_rows, img_cols)

Definition at line 29 of file mnist_double_cnn.py.

mnist_double_cnn.y_test = keras.utils.to_categorical(y_test, num_classes)

Definition at line 26 of file mnist_double_cnn.py.

mnist_double_cnn.y_train = keras.utils.to_categorical(y_train, num_classes)

Definition at line 46 of file mnist_double_cnn.py.