The Beautiful Future

https://stackoverflow.com/questions/49536856/tensorflow-how-to-merge-batchnorm-into-convolution-for-faster-inference To the best of my knowledge, there is no built-in feature in TensorFlow for folding batch normalization. That being said, it's not that hard to do it manually. One note, there is no such thing as folding dropoutas dropout is simply deactivated at inference time. To fold batch norm..

https://github.com/tensorflow/tensorflow/tree/master/tensorflow/contrib/model_pruning기존 웨이트를 마스킹하는 방법을 사용, pruning.py가 제공함.conv = tf.nn.conv2d( images, pruning.apply_mask(weights), stride, padding)이렇게 감싸지 않고 이미 감싸진 funcion도 있음. layers.masked_conv2d 등 . @ 하이퍼파라미터- namestring타입, 디폴트 model_pruning썸마리, 오피들 이 네임공간에 있게된다?- begin_pruing_stepint 타입, 디폴트 0글로발 스텝에서 프르닝 시작 적용- end_pruning_stepint 타입, 디폴트 -..