Kaihui-AMD b42cc81be1 Preserve platform-specific PyTorch builds during update (#14902) ... When running `pip install -r requirements.txt` during an update, the bare `torch`/`torchvision`/`torchaudio` entries can cause pip to overwrite ROCm, DirectML, XPU, or other platform-specific torch builds with the default PyPI (CUDA) packages. This adds pip constraints to pin torch packages to their current versions when a non-standard build variant (e.g. +rocm, +xpu) is detected, preventing the update from breaking the user's GPU setup. Fixes #14902 Co-authored-by: Cursor <cursoragent@cursor.com>		2026-07-13 10:39:59 +08:00
..
update_windows	Preserve platform-specific PyTorch builds during update (#14902)	2026-07-13 10:39:59 +08:00
windows_amd_base_files	Add script in AMD portable to launch with dynamic vram. (#13667)	2026-05-01 20:19:46 -04:00
windows_intel_base_files	Basic intel standalone package .bat (#13333)	2026-04-08 21:39:29 -04:00
windows_nightly_base_files	fp16 accumulation can now be enabled on the stable package. (#7750)	2025-04-23 05:28:24 -04:00
windows_nvidia_base_files	Add helpful message to portable. (#11671)	2026-01-06 14:43:24 -05:00