A cleaned version of XceptionNet in Keras, Xception is one of the best models used for Image Classification.
Xception is a Convolutional Neural Network, which is capable of producing state-of-the-art results. Xception is comparatively a more efficient model and the weight files produced are also smaller compared to other models.
You can use this cleaned version of the Xception model to produce state-of-the-art results, you can also train it on more datasets, the current weight files are the weights produced by training Xception on the ImageNet database.
The current model downloads and caches the weights file, you can also download the weights files seperately and load the weights from the correct path of the downloaded weights file. You can download it from: https://github.com/fchollet/deep-learning-models/releases/download/v0.4/xception_weights_tf_dim_ordering_tf_kernels.h5
Made with ♥ by Yan Chummar, at 14 Years of Age.