Open zheLim opened 5 years ago
BTW, CPU version of channel pruning only deals with regular 2D convolution. It cannot process dilation convolution. It would be better if an clarification is add to documents.
@zheLim Thanks for your suggestions.
Thanks for reply. Does solving a L2,1-norm regularized optimization problem get better result than lasso regression?
It runs faster under multi-GPU setting, and achieves higher accuracy on some models. We will provide more detailed results in the documentation.
Thanks a lot :)
Doc:
ChannelPrunedGpuLearner
;ChannelPrunedLearner
.Hey @jiaxiang-wu
How about the accuracy of ChannelPrunerGpuLearner on MobileNet ? Is this a structured pruning algorithm which leads to a regular pattern of sparsity?
ChannelPrunedGpuLearner
is a structured-pruning algorithm. The compressed model has regular sparsity patterns.@jiaxiang-wu Thanks much!
After having a glance on channel pruning gpu version, i found that gpu version may not strictly implement lasso regression. Neither Coordinate descent method nor LARS optimization algorithm are used. It will be great if you can add some description about gpu version on algorithm or something else.
BTW, The document of self-defined model is not clear enough.
In all, PocketFlow is a great job and i learn a lot and i am still learning from it :) .