tinygrad/extra/torch_backend
chenyu 986e865830
fix TINY_BACKEND=1 cumsum (#14138)
* fix TINY_BACKEND=1 cumsum

old hack was wrong, need to apply contiguous on the input

* test time

* test_linalg_svd is slow
2026-01-14 09:54:49 -05:00
..
.gitignore torch backend works for ResNet-18 (#9200) 2025-02-22 22:16:23 +08:00
backend.py fix TINY_BACKEND=1 cumsum (#14138) 2026-01-14 09:54:49 -05:00
backend2.py torch and numpy dtype interop [pr] (#9224) 2025-02-24 18:26:49 +08:00
example.py fix: revive torch backend (#13280) 2025-11-19 15:26:50 -08:00
test.py fix TINY_BACKEND=1 cumsum (#14138) 2026-01-14 09:54:49 -05:00
test_compile.py start torch.compile support (#9279) 2025-02-27 10:29:51 +08:00
test_inplace.py move frontend dir to nn [pr] (#12470) 2025-10-07 10:42:22 +08:00
test_kernel_fusion.py fix: revive torch backend (#13280) 2025-11-19 15:26:50 -08:00
test_multigpu.py move frontend dir to nn [pr] (#12470) 2025-10-07 10:42:22 +08:00
torch_tests.py rig up torch's testing framework [pr] (#9254) 2025-02-26 18:46:22 +08:00
wrapped_tensor.cpp fix: revive torch backend (#13280) 2025-11-19 15:26:50 -08:00