William

slower training speed

10

Hi, I have compared the training speed between version 0.15 and 0.16, and found that 0.16 trains at a significantly slower speed with exactly the same settings. Why is that...

Quantization output varies (and other relevant issues)

3

Hi, Is there any random number involved in the quantization process except the random value in stochastic rounding scheme? I ask because I noticed that if I turn on quantization...

Sparsification of Depthwise Separable Convolution

2

Hi, I see in [https://github.com/tidsp/caffe-jacinto/blob/caffe-0.16/src/caffe/net.cpp#L2078](url) that the sparsification process excludes thin layers and depthwise separable layers. I understand why thin layers are excluded as accuracy after sparsification may drop drastically....

Weighted Softmax Loss

1

Hi, May I know what is the right way to use the 'label_weights' parameter in softmax loss layer to assign loss weights to different labels? To be more specific, how...