mirror of
https://github.com/zebrajr/pytorch.git
synced 2025-12-06 12:20:52 +01:00
Modified the flag PYTORCH_TUNABLEOP_NUMERICAL_CHECK, so that it accepts the numerical tolerances in the format atol_rtol as compared to the previous 0 and 1. Retains previous functionality with default values as well. Pull Request resolved: https://github.com/pytorch/pytorch/pull/163079 Approved by: https://github.com/naromero77amd, https://github.com/jeffdaily |
||
|---|---|---|
| .. | ||
| amp | ||
| __init__.py | ||
| _device_limits.py | ||
| _gpu_trace.py | ||
| _memory_viz.py | ||
| _pin_memory_utils.py | ||
| _sanitizer.py | ||
| _utils.py | ||
| comm.py | ||
| gds.py | ||
| graphs.py | ||
| jiterator.py | ||
| memory.py | ||
| nccl.py | ||
| nvtx.py | ||
| profiler.py | ||
| random.py | ||
| sparse.py | ||
| streams.py | ||
| tunable.py | ||