Hugh Perkins

Interesting observation. Thanks! :) On 11 June 2017 02:41:07 BST, cwon7609 wrote: >DeepCL would be more popular if it was a backend to Keras > >-- >You are receiving this...

Seems like something that could make a nice standalone project for someone perhaps?

The second. Cpu sends commands to the gpu. The data stays on the gpu in between layers, but the command to run the next layer comes from the cpu. Nuance:...

:)

I believe that stride 1 is the default? On Wed, Jul 22, 2020, 05:32 Willem Chen wrote: > I must use conv kernel 3x3 with stride 1 > So how...

Understood. Involves a fair amount of work. Will ponder. Anyone else who arrives in this thread, and wants non-square kernels, please add a `+1` into this thread, to motivate me...

Oh, hmmm, kernels.... is this for Inception?

Ah, hmmm, right. Presumably you'd want the input 'images' to also be non-squared too? Presumably, quite very not-square, like 1-dimensional?

Hmmm... does sound potentially quite useful generally. By the way, just out of curiosity do you know any papers that use this technique?

Thanks!