sanchitintel
sanchitintel
## 🐛 Bug `torch.jit.frontend.NotSupportedError` while JIT tracing/scripting `FairseqLanguageModel` on CPU. ### To Reproduce Reopening #2741 #### Code sample Please refer to #2741 ### Expected behavior @myleott, this behavior is [expected...
## Problem history There seems to always have been a bug in `_vec_log_softmax_lastdim `. In particular, there were two issues with it - #### Bug 1 Before AVX512 support was...
### 🐛 Describe the bug ### Summary This issue had been discussed in https://github.com/pytorch/pytorch/pull/113004#discussion_r1383666731. The pattern-matching process seems to be sensitive to the presence or absence of `clone` nodes. A...
## Base PR for Inductor integration with oneDNN Graph int8 fusions (experimental) #### Platforms supported only builds on Linux with gcc version
### Feature request Port AdamW8bit support for CPU from `multi-backend-refactor` branch to the main branch ### Motivation Public cloud providers' machines with GPUs are usually expensive while datacenter-grade CPUs are...
Pattern has changed a lot due to changes in both torchao & PyTorch. Will have to rewrite. cc @ezyang @SherlockNoMad @EikanWang @jgong5 @wenzhe-nrv @voznesenskym @penguinwu @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @jiayisunx...
As suggested by @leslie-fang-intel in https://github.com/leslie-fang-intel/pytorch/commit/4c83e4e75138e8fa6e0d58438f75b7718dc8a0cc#diff-139642bd981df977f70f4c18c1c34bd1a85c1d6b9ffa06aaa98426ed83942a31R537 - all elements of `B` tiles (not referring to AMX tiles, but the tiles at the granularity of the micro-kernel) have contiguous elements since...