branch: master
Commits on master
- 5c56cac MI300 mfma support (#9417) 1 year ago
- 5500887 improve reproducibility of WebGPU CI puppeteer test (#9496) 1 year ago
- cde4fd3 do not view_left assign + elementwise sources always have a shape [pr] (#9491) 1 year ago
- 117b7a1 VALIDATE_WITH_CPU [pr] (#9488) 1 year ago
- 935cd01 simple failing test for graph_rewrite children [pr] (#9489) 1 year ago
- d20494e move buffer logic to Buffer [pr] (#9487) 1 year ago
- 3be2281 unbind Tensor variables last [pr] (#9486) 1 year ago
- b44f9c4 reorder do_realize [pr] (#9485) 1 year ago
- a82c933 am: rename soc21 to soc (#9482) 1 year ago
- b100fc0 split the rule that uses context in scheduler simplifier [pr] (#9484) 1 year ago
- 5e58f4b Tiny backend test_ops fix part 3 (#9483) 1 year ago
- 9fcef4d add masked_select to tensor.py (#9468) 1 year ago
- 4f8eac5 failed test case for threefry (#9469) 1 year ago
- 6dd8e5b refactor llvm compiler (#9403) 1 year ago
- 53d6f1e Add bitonic cat sort (#9422) 1 year ago
- f53be01 lower bert learning rate (#9481) 1 year ago
- e03c0aa more explicit DONT_PUSH_VIEWS [pr] (#9479) 1 year ago
- 3b00a77 fix view_left for unsafe pad ops [pr] (#9478) 1 year ago
- 813f713 merge_views for buffer ops + create valids last (#9472) 1 year ago
- bd1f71c simple failing test for extra ops in VALID [pr] (#9474) 1 year ago
- e26caf4 hotfix: skip test_mean_half_precision_underflow on amd ci (#9476) 1 year ago
- 824c5f4 dsp work try 3 (#9475) 1 year ago
- 242daa4 ptrcat (#9473) 1 year ago
- 52ae9af Fast DSP for MobileNetV2 (try 2) (#9467) 1 year ago
- 15ee742 add get_children_map to uop (#9470) 1 year ago
- d2cfbd8 bert lower learning rate and total steps (#9466) 1 year ago
- 09e7708 minimum change for rdna4 [pr] (#9455) 1 year ago
- be21616 reorder into swizzler + ast_fixup [pr] (#9456) 1 year ago
- cb7a7f6 quantization preprocessor from DSP, should be universal (#9437) 1 year ago
- ca5064a remove Kernel.float4_axis [pr] (#9448) 1 year ago