Soumendu Kumar Ghosh
Soumendu Kumar Ghosh
I tried out your solution @YongWookHa, however, now I am getting an error to calculate loss function Error: `File "/home/min/a/ghosh37/distiller/distiller/apputils/image_classifier.py", line 588, in train loss = criterion(output, target) File "/home/min/a/ghosh37/distiller/env/lib64/python3.6/site-packages/torch/nn/modules/module.py",...
Yes, I did add values, but I was copying `model.values `only to single output instead of `output, aux_output`, and so when computing loss function on dict instead of a tensor,...
I am facing the same 'zip file is not a zip file' when trying to install pytorch on raspi pi 0. --------------------------------------------- EDIT: I tried pip3 install 'wheel links' and...
I am also having the same issue when I installed tensorflow 1.4
@rwightman Please let me know if you need any other information or log related to this?
Does the main branch support speculative decoding with chunking and batch size = 1 for long-form transcription?
Tested Whisper on NPU using nightly builds. Runs successfully. Noticed an error only max_new_tokens is 2048 and beyond.
Exporting whisper using the latest optimum-intel added a new dynamic shape in decoder model input, which was static in previous versions. This is causing the issue reported by @weberwcwei.
This issue can be closed now after 2024.5 release.
Even a clean environment didn't help me. I had to install torch=2.4.0 to get rid of the issue.