Permute problem

[TommyHuang821]

Hi David,

I have a question for Permute layer.
CAFFE blob : [# batch, # channel, height, width]
If Permute {dim:0, 2, 3, 1}, then the output should be [# batch, height, width, # channel]
But in this netscope CNN analysis, the output would be [# batch, width, # channel , height]
Which one is correct?

Thanks,
Tommy

[dgschwend]

Hi The permutation works as you explain. Netscope gives the correct result, please take another look at the dimensions. Input: N C H W Output: N H W C (as expected) Regards David

…

Am 05.12.2018 um 05:26 schrieb Chih-Sheng (Tommy) Huang ***@***.***>: Hi David, I have a question for Permute layer. CAFFE blob : [# batch, # channel, height, width] If Permute {dim:0, 2, 3, 1}, then the output should be [# batch, height, width, # channel] But in this netscope CNN analysis, the output would be [# batch, width, # channel , height] Which one is correct? Thanks, Tommy — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or mute the thread.

[TommyHuang821]

Hi David,

Many thanks for your reply.

BR,
Tommy