Wide Residual Networks in Keras

A package of Wide Residual Networks for image recognition in Keras.

keras-wrn is the Keras package for Wide Residual Networks. It's fast and flexible.

Wide ResNets are faster to train and more accurate than traditional ResNets, even when pre-activation structure is used.

Quick example

import keras

import keras_wrn

shape, classes = (32, 32, 3), 10

model = keras_wrn.build_model(shape, classes, 16, 4)

model.compile("adam", "categorical_crossentropy", ["accuracy"])

(x_train, y_train), (_, _) = keras.datasets.cifar10.load_data()

y_train = keras.utils.np_utils.to_categorical(y_train)

model.fit(x_train, y_train, epochs=10)

Contribute

Hey there! New ideas are welcome: open/close issues, fork the repo and share your code with a Pull Request.

Clone this project to your computer:

git clone https://github.com/EricAlcaide/keras-wrn

By participating in this project, you agree to abide by the thoughtbot code of conduct

Meta

• Author's GitHub Profile: Eric Alcaide
• Twitter: @eric_alcaide
• Email: [email protected]