Update transformers requirement from >=4.52.4 to >=4.57.6#8
Open
dependabot[bot] wants to merge 1 commit intomainfrom
Open
Update transformers requirement from >=4.52.4 to >=4.57.6#8dependabot[bot] wants to merge 1 commit intomainfrom
dependabot[bot] wants to merge 1 commit intomainfrom
Conversation
Updates the requirements on [transformers](https://github.com/huggingface/transformers) to permit the latest version. - [Release notes](https://github.com/huggingface/transformers/releases) - [Commits](huggingface/transformers@v4.52.4...v4.57.6) --- updated-dependencies: - dependency-name: transformers dependency-version: 4.57.6 dependency-type: direct:production ... Signed-off-by: dependabot[bot] <support@github.com>
FortPercent
pushed a commit
that referenced
this pull request
May 2, 2026
…ing null
`context_null_path` is required for working CFG (otherwise
cond + scale*(cond - 0) = (1+scale)*cond and reward variance collapses to
zero — silent grad_norm=0 across every previous smoke).
- Replace 3 legacy preprocessors + 2 launcher shells with one tool:
data_preprocess/prepare_wan_data.py
Accepts .txt (one prompt per line) or .json (list of {caption, ...}).
Loads the umT5 encoder lazily; per-row encode is skipped when
`context_path` already points at an existing .npy; the negative-prompt
encode is skipped when `context_null.npy` already exists. Re-running is
safe and a no-op if everything is already in place.
- Default negative prompt = Wan official Chinese template (overridable via
`--negative_prompt`).
- Dataset (`teleboost/utils/dataset/_dancegrpo_rl_dataset.py`): swap the
silent `null_context = torch.zeros_like(context)` fallback for a
KeyError that names the fix command. The fallback was a documented
training-killer pretending to be a "smoke convenience".
- README: new "Prepare data" section before "Run a smoke", updated table to
include all six 14B Wan22 smoke variants we ship; remove the
"*_wxe.sh removed" sentence (no longer relevant after cleanup).
- docs/install_from_scratch.md: rewrite the data-prep section accordingly,
rephrase gotcha #8 to describe the fail-fast behavior.
Removed:
- data_preprocess/preprocess_wan_data.py
- data_preprocess/preprocess_wan_embeddings.py
- data_preprocess/preprocess_wan_embeddings_fromlist.py
- data_preprocess/preprocess_wan_rl_embeddings{,_1p3B}.sh
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
FortPercent
pushed a commit
that referenced
this pull request
May 7, 2026
Smoke 全覆盖结果: #7 14B + HPSv2 PASS (33s/step cold, 26s warm) #8 1.3B + HPSv2 PASS (703s cold incl. inductor compile, 3.6s warm) #9 1.3B + Qwen-VL-7B PASS (31s cold, 6.7s warm; rewards=44→19.75 真实) #11 1.3B + Joint 4-reward PASS (80s cold, 12s warm; 4 worker init OK) 新增依赖: pip install decord (videophy reward 用)
FortPercent
pushed a commit
that referenced
this pull request
May 7, 2026
…ing null
`context_null_path` is required for working CFG (otherwise
cond + scale*(cond - 0) = (1+scale)*cond and reward variance collapses to
zero — silent grad_norm=0 across every previous smoke).
- Replace 3 legacy preprocessors + 2 launcher shells with one tool:
data_preprocess/prepare_wan_data.py
Accepts .txt (one prompt per line) or .json (list of {caption, ...}).
Loads the umT5 encoder lazily; per-row encode is skipped when
`context_path` already points at an existing .npy; the negative-prompt
encode is skipped when `context_null.npy` already exists. Re-running is
safe and a no-op if everything is already in place.
- Default negative prompt = Wan official Chinese template (overridable via
`--negative_prompt`).
- Dataset (`teleboost/utils/dataset/_dancegrpo_rl_dataset.py`): swap the
silent `null_context = torch.zeros_like(context)` fallback for a
KeyError that names the fix command. The fallback was a documented
training-killer pretending to be a "smoke convenience".
- README: new "Prepare data" section before "Run a smoke", updated table to
include all six 14B Wan22 smoke variants we ship; remove the
"*_wxe.sh removed" sentence (no longer relevant after cleanup).
- docs/install_from_scratch.md: rewrite the data-prep section accordingly,
rephrase gotcha #8 to describe the fail-fast behavior.
Removed:
- data_preprocess/preprocess_wan_data.py
- data_preprocess/preprocess_wan_embeddings.py
- data_preprocess/preprocess_wan_embeddings_fromlist.py
- data_preprocess/preprocess_wan_rl_embeddings{,_1p3B}.sh
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This file contains hidden or 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
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.
Updates the requirements on transformers to permit the latest version.
Release notes
Sourced from transformers's releases.
Commits
753d611Release 4.57.6ee9aed5Fixed incorrect model_type for qwen2vl and qwen2.5vl when config is saved and...542e65fRelease 4.57.5c3098a9Fix lr_scheduler_parsing (#41322)7d5c2b5QwenVL: add skipped keys insetattras well (#41808)bfd87b9Release 4.57.4018def0Grouped beam search from config params (#42472)a0699c9Handle decorator with optional arguments better (#42512)33ac6bdfix: make mistral base check conditional to fix offline loading (#42880)47b0e474.57.3Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting
@dependabot rebase.Dependabot commands and options
You can trigger Dependabot actions by commenting on this PR:
@dependabot rebasewill rebase this PR@dependabot recreatewill recreate this PR, overwriting any edits that have been made to it@dependabot show <dependency name> ignore conditionswill show all of the ignore conditions of the specified dependency@dependabot ignore this major versionwill close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)@dependabot ignore this minor versionwill close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)@dependabot ignore this dependencywill close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)