branch: master
Commits on master
- 6e86472 fix typing for test to run in py38 (#4930) 1 year ago
- 1326f29 fix Tensor.gather shape checking criteria (#4932) 1 year ago
- 898430c more typing in linearizer uoping utils (#4929) 1 year ago
- 828c98d add slides from code europe to docs 1 year ago
- 9a3c1e4 fix mul div failure (#4928) 1 year ago
- 11a03cb don't use uops.add while constructing (#4913) 1 year ago
- d894acb remove hardcoded -1s referencing late reduce (#4926) 1 year ago
- b833a11 allocate shared memory per block (#4924) 1 year ago
- ca4ccdd docsfix: nn.Tensor -> Tensor 1 year ago
- 3d13c23 llama3 `--download_model` (#4922) 1 year ago
- f902af4 increase metal ci test timeout to 20 minutes (#4920) 1 year ago
- fdbb430 skip unsupported dtype in fuzz_linearizer (#4917) 1 year ago
- 7f3d9e6 revert hsa autogen removal (#4914) 1 year ago
- 58cf6ea add missing dir level for amd mockgpu (#4911) 1 year ago
- b886d25 improve test_dropout_on_shard (#4912) 1 year ago
- 7f03420 only install comgr in AMD CI (#4909) 1 year ago
- 35e53c0 add sharded arange test (#4908) 1 year ago
- 798ea61 widen test_ops [low, high] and more strict atol (#4906) 1 year ago
- 97b05f5 revert the .detach() in layernorm (#4904) 1 year ago
- 8b5bcf3 process replay in all of CI (#4884) 1 year ago
- 9715a71 replace set with dedup (#4901) 1 year ago
- c8cd637 test case for Tensor.var reducing over size = 1 axis (#4902) 1 year ago
- c0fb7ee cleanup lazy const fold for binary (#4900) 1 year ago
- 5bf1f7d nv better error messages for ioctls (#4899) 1 year ago
- b9f26ee hotfix: import datasets in nn init 1 year ago
- b56ae56 cosmetic changes to uop _match (#4897) 1 year ago
- b1db2d0 tqdm replacement (#4846) 1 year ago
- 05d7ab7 set tensor core opt options in Renderer (#4896) 1 year ago
- f42183b hotfix: relax cifar to 93.2 1 year ago
- 1dde829 UOps.IF* to graph spec (#4894) 1 year ago