dmagee

I'm also trying to get this working. The thing I'm stuck on is GPUTensor.cu. It gives a heap error. If you comment out some of the template instantiations (I tried...

> I'm also trying to get this working. The thing I'm stuck on is GPUTensor.cu. It gives a heap error. If you comment out some of the template instantiations (I...

Ok, more advice from my experiments. It turned out I was using an older version of cudnn (cudnn-10.0-v7.3.1) which isn't really designed to work with cuda11.X, and I do suspect...

Sorry for spaming everyone, but now with cudnn-11.1-v8.0.5.39 I'm getting an exception thrown on the cudnnConvolutionForward call in CuDnnConvolutionEngine.cu. The output is: ``` ... Starting minibatch loop. About to throw...

No idea if I'm talking to myself, but the exception I reported above is due to the fact that the workspace size calculation in CNTK seems broken (too small) in...

> Thanks for sharing the info here @dmagee! Sounds like trying to setup cntk with the latest Cuda is a non trivial task. No worries. You're absolutely right. the Nvidia...

Another issue I found was in cudnnCommon.cpp on the line: auto err = cudnnDestroy(*src); This causes an crash somewhere in the nvidia cudnn library. Essentially a single instance of cudnnHandle_t...

[Copying some discussion on this matter from Openslide mailing list for completeness] It looks as though you are indeed seeing #132. You may be able to convince yourself one way...

Ok, so to add to my algorithm a second stage I think will fix the issues pus a modification to the first: Step 1 (almost as before): Set tile 0,0...

Just had a chance to look at this (early days yet, but got it all compiled and linked to our viewer anyway). Some questions/observations: I Managed to print out the...