ceos/unsloth
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
unsloth/studio/backend
History
Datta Nimmaturi 04359be333
 [Studio] Try installing causal-conv1d from prebuilt wheels if avialable (#4547) 
* Try installing causal-conv1d from prebuilt wheels if avialable

* Prefer installing mamba-ssm from wheel to speed up things

* undo python stack install changes

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Revert "undo python stack install changes"

This reverts commit d943551092.

* add comments

* Fix wheel installer: model detection, platform tags, torch pin, error handling

- Add nemotron-h (hyphen) and granite-4.0-h / granitemoehybrid to model
  detection for both causal-conv1d and mamba-ssm. These hybrid Mamba models
  were silently skipped since nemotron_h (underscore) never matches real
  HF model IDs like nvidia/Nemotron-H-8B-Base, and granite was missing
  entirely despite being a supported model in model_config.py and loader.py.
- Fix _causal_conv1d_platform_tag to detect linux_aarch64 via
  platform.machine() instead of hardcoding linux_x86_64. Both upstream
  releases publish aarch64 wheels. Drop win_amd64 since neither repo
  publishes Windows wheels (avoids a wasted HTTP probe on every run).
- Pin torch to >=2.6.0,<2.11.0 instead of <=2.10.0 to add a version floor
  and document the wheel coverage range with upstream release links.
- Strip non-numeric suffixes from torch minor version so nightly builds
  like 2.7a0 correctly resolve to wheel tag torch2.7 instead of torch2.7a0.
- Use stderr=_sp.PIPE instead of stderr=_sp.STDOUT in the env probe so
  torch import warnings do not corrupt the JSON output.
- Add timeout=30 to the env probe subprocess to prevent indefinite hangs.
- Catch Exception (not just ImportError) on the existing-install check so
  ABI-broken installs with OSError/RuntimeError are retried rather than
  silently accepted.
- Guard uv invocation with shutil.which("uv") to prevent FileNotFoundError
  crash when uv is not on PATH. Wrap the top-level ensure calls in
  try/except so failures do not kill the training worker.
- Hoist _SSM_MODEL_SUBSTRINGS to module level.
- Remove redundant --torch-backend=auto flag from direct wheel URL install.

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Add LFM2 to causal-conv1d detection; stop training on install failure

- Add "lfm2" to _model_wants_causal_conv1d so Studio picks up the
  fast kernel path for Liquid Foundation Model 2.
- Replace silent logger.warning on SSM dependency install failure
  with an error event that tells the user to choose another model
  and stops the training job immediately.

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Catch subprocess timeout in torch probe; narrow import guard to ImportError

- _probe_causal_conv1d_env: wrap subprocess.run in try/except for
  TimeoutExpired so a slow torch import returns None (falls back to
  PyPI) instead of killing the training job.
- _install_package_wheel_first: narrow except Exception to except
  ImportError on the __import__ check so unexpected errors from a
  broken module still propagate.

* Remove unconditional torch pin from install_python_stack

The torch>=2.6.0,<2.11.0 pin was added to ensure prebuilt
causal-conv1d / mamba-ssm wheels exist, but it runs at install
time for all users regardless of model choice. This can downgrade
or unnecessarily upgrade torch. The worker already handles wheel
compatibility at training time by probing the environment and
falling back to PyPI, so the install-time pin is not needed.

---------

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Daniel Han <danielhanchen@gmail.com>
2026-03-25 02:22:26 -07:00
..
assets studio: improve onboarding UX, tooltips, and training defaults (#4355) 2026-03-17 07:46:07 -07:00
auth fix: remove old comments (#4292) 2026-03-14 16:50:13 +04:00
core [Studio] Try installing causal-conv1d from prebuilt wheels if avialable (#4547) 2026-03-25 02:22:26 -07:00
loggers Final cleanup 2026-03-12 18:28:04 +00:00
models feat(studio): training history persistence and past runs viewer (#4501) 2026-03-25 00:58:55 -07:00
plugins Bump Data Designer to 0.5.4 (removes litellm dependency) (#4569) 2026-03-25 02:01:43 -07:00
requirements Bump Data Designer to 0.5.4 (removes litellm dependency) (#4569) 2026-03-25 02:01:43 -07:00
routes feat(studio): training history persistence and past runs viewer (#4501) 2026-03-25 00:58:55 -07:00
state Final cleanup 2026-03-12 18:28:04 +00:00
storage feat(studio): training history persistence and past runs viewer (#4501) 2026-03-25 00:58:55 -07:00
tests feat: support GGUF export for non-PEFT models + fix venv_t5 switching for local checkpoints (#4455) 2026-03-20 12:13:18 +04:00
utils feat(studio): training history persistence and past runs viewer (#4501) 2026-03-25 00:58:55 -07:00
__init__.py Final cleanup 2026-03-12 18:28:04 +00:00
_platform_compat.py Fix Studio crash on Anaconda/conda-forge Python (#4484) 2026-03-22 05:36:55 -07:00
colab.py Fix Studio crash on Anaconda/conda-forge Python (#4484) 2026-03-22 05:36:55 -07:00
main.py feat(studio): training history persistence and past runs viewer (#4501) 2026-03-25 00:58:55 -07:00
run.py Fix Studio port conflict detection for loopback addresses (#4532) 2026-03-23 22:34:47 -07:00