branch: master
Commits on master
- d4e4a08 fix: Tensor min function for unsigned ints (#7675) 1 year ago
- d1dfd59 assert specifying device to rand_like a multi tensor (#7678) 1 year ago
- 51432bf add rand_like test case with device specified (#7663) 1 year ago
- 23363de Add: failing tests for uint8 `min()` (#7669) 1 year ago
- 2950850 uop style prefer small dtype + cleanups [pr] (#7671) 1 year ago
- e84d089 delete ReduceOps, only use REDUCE_AXIS (#7667) 1 year ago
- 217c006 buffer access on UOp [pr] (#7665) 1 year ago
- 5da149d uop can have base [pr] (#7666) 1 year ago
- ca99c67 refactors from the delete lazy diff [pr] (#7664) 1 year ago
- e6cfaaa metal benchmark JIT=2 -> JIT=1 (#7661) 1 year ago
- 4c5f7dd flux set model path in args (#7660) 1 year ago
- 08706c2 more readable rand [pr] (#7659) 1 year ago
- 1884f02 add conv3x3 to speed_v_theoretical (#7658) 1 year ago
- 54c0abc cleaner code_for_op order [pr] (#7653) 1 year ago
- 962dafb use randn in speed_v_theoretical instead of rand (#7656) 1 year ago
- 397a2e6 no special case for int32 in truncate [pr] (#7657) 1 year ago
- 6159790 add gemv to speed_v_theoretical (#7654) 1 year ago
- e07d2d0 skip TestBeamSearch.test_large_ast (#7652) 1 year ago
- 0f02573 save lines in assign tracking [pr] (#7651) 1 year ago
- fbad490 move groups to uop [pr] (#7640) 1 year ago
- 4f1f823 add tiny test for randomness + remove ulong buffers (#7648) 1 year ago
- c06a5a9 Tensor.linspace raises for dtype.bool (#7649) 1 year ago
- 5eef59d add Tensor.linspace (#7609) 1 year ago
- 99f29e5 update speed_v_theoretical numbers (#7647) 1 year ago
- 035e39f remove copied is_dtype_supported from onnx [pr] (#7646) 1 year ago
- 9c63c3d These casts should only happen if these are supported (#7644) 1 year ago
- a88a15c setup perflevel in red CI (#7645) 1 year ago
- 773d5b6 beam benchmark tests (#7638) 1 year ago
- bfab032 fix HALF=1 in test_speed_v_torch (#7642) 1 year ago
- 4d81b79 qcom match texture/sampler descriptors to OpenCL (#7622) 1 year ago