branch: master
Commits on master
- 3e2430f use tqdm tqdm in mlperf training (#7929) 1 year ago
- 8647584 Don't take const in gcd and change the "nothing_changed" condition (#7926) 1 year ago
- 988d649 add TODO case to test_mod_congruence (#7925) 1 year ago
- 57262c8 update Tensor.scatter doc examples (#7924) 1 year ago
- cea5853 add Tensor.scatter (#7737) 1 year ago
- 38f34ca prepare mypy==1.13.0: legacy cast (#7866) 1 year ago
- 753f07e add circular pad mode to Tensor.pad (#7918) 1 year ago
- a58e289 Revert "prereqs for new block lin so PR works (#7919)" (#7921) 1 year ago
- c53261b prereqs for new block lin so PR works (#7919) 1 year ago
- a6171cb add stable diffusion v2 to mac benchmark (#7917) 1 year ago
- d318867 Factoring gcd out of mod (#7916) 1 year ago
- 84f96e4 hcq signal tiny refactor (#7913) 1 year ago
- 345457f webgpu cache packages (#7911) 1 year ago
- 6102e31 webgpu -n=auto (#7910) 1 year ago
- cab461c match lazy view in uop try 2 (#7905) 1 year ago
- ea57c52 base uop is always contiguous (#7907) 1 year ago
- ceda43c always swizzle load st in wmma [pr] (#7908) 1 year ago
- 4e5bf9d test assignment in jit (#7906) 1 year ago
- 0cd1cc2 PTX simplify: use a dict matcher for prefix [pr] (#7890) 1 year ago
- 10618ab Bring back WebGPU (#7063) 1 year ago
- ff3f2a9 Revert "move attention upcast (#7830)" (#7903) 1 year ago
- 04bee97 hotfix ctypes.c_ulong(size) for metal _alloc (#7902) 1 year ago
- 631dc98 validate llama quantize output (#7901) 1 year ago
- e8777cb assert view on uops without shape [pr] (#7898) 1 year ago
- a49ca0c clean up fully_flatten [pr] (#7885) 1 year ago
- e823de3 viz with bottom_up=True (#7894) 1 year ago
- 2ca41d6 ops metadata map try 2, early fuse [pr] (#7893) 1 year ago
- 9295c86 delete base op cast [pr] (#7891) 1 year ago
- 26784c4 delete cast arg 2 [pr] (#7881) 1 year ago
- 9d0038b small changes from block linearizer [pr] (#7888) 1 year ago