Raaj comments

Results 46 comments of


Raaj

Tesla GV100 results?

Caffe doesnt support Tensorcores right? So its useless anyway

Error When Compiling with CudNN 5 and CUDA 8

It doesnt compile either

ReadProtoFromBinaryFile(param_file, param) Failed to parse NetParameter file

Getting the same error

cuDNN5 currently not supported

Can you tell me the files I will need to modify in my caffe? I have my own version of caffe based on the new cudnn stuff

cuDNN5 currently not supported

Looks to be these two commits. Pardon my poor knowledge of git but, assuming I have my version of caffe, and I merely want to merge changes from commit 7fec25bad4405fb97b9a48d7307d8a8bb8bacae9...

cuDNN5 currently not supported

Ok nevermind, just added a .patch to those commit links and applied them. seems to work

cuDNN5 currently not supported

It appears my merge may not have worked. I get the following error: when attempting to run the toolbox ``` F1019 14:33:19.133157 17310 upgrade_proto.cpp:95] Check failed: ReadProtoFromBinaryFile(param_file, param) Failed to...

404 not found on get_balances

Still doesnt work, now it says API keys required

Recurrent neural layers support (RNN, LSTM) via Caffe backend (Direct inputs to LSTM)

Hi all, How does one actually use the LSTM layer. I keep getting errors, saying certain parameters are invalid ``` layer { name: "lstm1" type: "Lstm" bottom: "data" bottom: "clip"...

ImportError: cannot import name 'pyopenpose' from partially initialized module 'openpose' (most likely due to a circular import)

Hi, yes if you have a different version of python (not system python), then you may need to manually set those two parameters above. If you do `sudo make install`,...