Yuzz1020
Yuzz1020
I think the author used gradient penalty instead of gradient clip during implementation .
Hi! I am facing the same problem. Is there any solution to this problem now? The environment I am using is python 2.7.15, tensorflow 1.6.
I am facing similar issues, have you solved this?
Hey, may I ask why I got this error when I try to run arch_search.py with setting as densenet-cell? Have you ever encountered this error? assert cell_architecture == Tree.build_tree_from_torch_module(net_config.building_block), 'Err'...
Hey, may I ask the accuracy you achieved? I have run the code with --setting=densenet-cell, but accuracy is only around 90.6% when max nets to train set as 300, which...
> the best net i got when searching phase is around 90%。after trainning for 300 epoch， i got 93 the densenet used for 300 epoch is deeper than the one...
> is you net for trainging the same with the net obtained by searching？ in the paper， the net for training is different with the net which is obtained by...
Hey, have you solved this error? I have got exact the same one as yours. Any idea on how to deal with it?
This is because when groups in conv layer is 1, Tree.build_tree_from_torch_module(net_config.building_block) will automatically consider it as Conv while cell_architecture built from net_config_for_sample will still consider it as GroupConv. I think...
Have you figured out the difference? I have the same question. I suppose the model architecture is not exactly the same between these two versions? And do you have any...