Upgrading Hordelib #1
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Co-Authored-By: Divided by Zer0 <mail@dbzer0.com>
This deflects a problem where custom models passing a full path in windows (including a drive letter) would end up calling code in ComfyUI where a relative path was expected (`folder_paths.get_full_path`).
…th 6 updates Updates the requirements on [pytest](https://github.com/pytest-dev/pytest), [mypy](https://github.com/python/mypy), [black](https://github.com/psf/black), [ruff](https://github.com/astral-sh/ruff), [tox](https://github.com/tox-dev/tox) and [pre-commit](https://github.com/pre-commit/pre-commit) to permit the latest version. Updates `pytest` from 8.1.1 to 8.2.0 - [Release notes](https://github.com/pytest-dev/pytest/releases) - [Changelog](https://github.com/pytest-dev/pytest/blob/main/CHANGELOG.rst) - [Commits](pytest-dev/pytest@8.1.1...8.2.0) Updates `mypy` from 1.9.0 to 1.10.0 - [Changelog](https://github.com/python/mypy/blob/master/CHANGELOG.md) - [Commits](python/mypy@1.9.0...v1.10.0) Updates `black` from 24.3.0 to 24.4.2 - [Release notes](https://github.com/psf/black/releases) - [Changelog](https://github.com/psf/black/blob/main/CHANGES.md) - [Commits](psf/black@24.3.0...24.4.2) Updates `ruff` from 0.3.3 to 0.4.2 - [Release notes](https://github.com/astral-sh/ruff/releases) - [Changelog](https://github.com/astral-sh/ruff/blob/main/CHANGELOG.md) - [Commits](astral-sh/ruff@v0.3.3...v0.4.2) Updates `tox` to 4.15.0 - [Release notes](https://github.com/tox-dev/tox/releases) - [Changelog](https://github.com/tox-dev/tox/blob/main/docs/changelog.rst) - [Commits](tox-dev/tox@4.14.1...4.15.0) Updates `pre-commit` to 3.7.0 - [Release notes](https://github.com/pre-commit/pre-commit/releases) - [Changelog](https://github.com/pre-commit/pre-commit/blob/main/CHANGELOG.md) - [Commits](pre-commit/pre-commit@v3.6.2...v3.7.0) --- updated-dependencies: - dependency-name: pytest dependency-type: direct:development update-type: version-update:semver-minor dependency-group: python-packages - dependency-name: mypy dependency-type: direct:development update-type: version-update:semver-minor dependency-group: python-packages - dependency-name: black dependency-type: direct:development update-type: version-update:semver-minor dependency-group: python-packages - dependency-name: ruff dependency-type: direct:development update-type: version-update:semver-minor dependency-group: python-packages - dependency-name: tox dependency-type: direct:development dependency-group: python-packages - dependency-name: pre-commit dependency-type: direct:development dependency-group: python-packages ... Signed-off-by: dependabot[bot] <support@github.com>
* updated pipelines * feat: remove the requirement of a design pipeline * ci: run pre-commit with `--all-files` * chore: fix type issues exposed by mypy `v1.10.0` --------- Co-authored-by: tazlin <tazlin.on.github@gmail.com>
* feat :adds stable_cascade_2pass * ci: improved 2pass tests * tests: support raising exception on similarity fail * fix: clean up hires_fix resize logic; tests: 2pass tests --------- Co-authored-by: tazlin <tazlin.on.github@gmail.com>
- ComfyUI uses stdlib `logging` to emit messages now (instead of print). - The original purpose of `OutputCollector` was to handle print specifically (and later, the TQDM messages for the purpose of emitting progress on inference so processes didn't appear stuck). - However, now that ComfyUI (and other libraries always have) use `logging` we can intercept those messages and emit them through loguru. See `comfy_horde.py` for more info.
Co-Authored-By: Divided by Zer0 <mail@dbzer0.com>
ComfyUI's default mode of running (`main.py`) includes calls to `cleanup_models(...)` and `soft_empty_cache(...)` on a timer. I suspect issues that have arisen lately are rooted in the fact that horde-engine does not currently do something similar. I am adding this (default on) option to the `HordeLib` class to call these after every pipeline run.
This is a patent "temporary" fix to prevent extremely large models, such as cascade models, from being forced to fully load. I will track the resolution of this in an issue: Haidra-Org#312
This is an attempt to get the tests to run on high vram cards
* fix: ignore models/ dir * feat: Flux support + img2img + batches * fix: standardized baseline * comment out dev tests * fix/refactor: download/validate schnell during CI runs Also adjusts the model name fixture logic to support the dynamic downloading/validating of future models with surrounding comments/warnings to extend the list downloading/validating relies on. * fix: use latest compat. `horde_model_reference` * fix: support for flux loras * missing image --------- Co-authored-by: tazlin <tazlin.on.github@gmail.com>
The numba library emits many megabytes of log messages at an unacceptable rate. This has performance implications due to the IO intensive path every step of the numba pre-compilation must take in order to emit these messages.
This is due to the CI consistently failing here.
The first in a batch should match the generation with a same payload but with `n: 1`.
This fixes the problem originating in `lora.py` of comfyui with the line `do_pad_weight = len(v) > 1 and v[1]["pad_weight"]` (no `is None` check).
Previously, CPU was the default, but it now requires explicitly specifying CPU/GPU. Historically, I have been unable to get GPU support to work easily. For now, I am going to keep the status quo to avoid causing other problems.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.