mnist_dnn Namespace Reference

Variables
int	batch_size = 128
int	num_classes = 10
int	epochs = 25
	x_test = x_test.reshape(10000, 784)
	y_test = keras.utils.to_categorical(y_test, num_classes)
	x_train = x_train.reshape(60000, 784)
	y_train = keras.utils.to_categorical(y_train, num_classes)
float	l2 = 0.0001
	model = Sequential()
	loss
	optimizer
	metrics
	history
	score = model.evaluate(x_test, y_test, verbose=0)
	include_optimizer
	json_string = model.to_json()

Detailed Description

Trains a simple deep NN on the MNIST dataset.

Gets to 98.40% test accuracy after 20 epochs
(there is *a lot* of margin for parameter tuning).
2 seconds per epoch on a K520 GPU.

Variable Documentation

int mnist_dnn.batch_size = 128

Definition at line 21 of file mnist_dnn.py.

int mnist_dnn.epochs = 25

Definition at line 23 of file mnist_dnn.py.

mnist_dnn.history

Initial value:

= model.fit(x_train, y_train,
                    batch_size=batch_size,
                    epochs=epochs,
                    verbose=1,
                    validation_data=(x_test, y_test))

Definition at line 57 of file mnist_dnn.py.

mnist_dnn.include_optimizer

Definition at line 67 of file mnist_dnn.py.

mnist_dnn.json_string = model.to_json()

Definition at line 69 of file mnist_dnn.py.

float mnist_dnn.l2 = 0.0001

Definition at line 40 of file mnist_dnn.py.

mnist_dnn.loss

Definition at line 51 of file mnist_dnn.py.

mnist_dnn.metrics

Definition at line 54 of file mnist_dnn.py.

mnist_dnn.model = Sequential()

Definition at line 41 of file mnist_dnn.py.

int mnist_dnn.num_classes = 10

Definition at line 22 of file mnist_dnn.py.

mnist_dnn.optimizer

Definition at line 53 of file mnist_dnn.py.

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

Definition at line 62 of file mnist_dnn.py.

mnist_dnn.x_test = x_test.reshape(10000, 784)

Definition at line 26 of file mnist_dnn.py.

mnist_dnn.x_train = x_train.reshape(60000, 784)

Definition at line 28 of file mnist_dnn.py.

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

Definition at line 26 of file mnist_dnn.py.

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

Definition at line 38 of file mnist_dnn.py.