preksha12

Actually I am trying to calculate FLOPS for NTU-RGBD dataset with skeleton data.

Is your prepare_data.py working ?In my case it is working but it is not generating image_mean_binaryproto file .So what should be the solution?

Convn function is also used in subsampling operation in cnnff function but in subsampling layer(s) we only perform pooling so convolution operation is not executed on this layer but still...