Pr8 integration - #20
Merged
Merged
Conversation
Updated dataset information and performance metrics for hand kinematic regression.
…s avoids namespace issues with third party libraries (e.g., datasets from HuggingFace)
…lity with old pytorch versions (<2.5)
…ory optimization for DDP - finetuning dataset preload in RAM option and documentation - regression task support for EMG finetuning - additional dataset configurations (log dir, individual h5 file)
…ers, EMG finetuning dataset handling, and improved training script logging
…iency and logging
…, and sequence length
…ssues with huggingface datasets)
- Added label smoothing and classification type options to the fine-tuning YAML. - Updated input normalization settings for pretraining and fine-tuning. - Revised documentation to clarify preprocessing and fine-tuning details. - Improved error handling for classification type in the finetune task.
The PR read cfg.wandb unconditionally. Only the two TinyMyo experiment
configs define a wandb key, and there is no top-level default in
config/defaults.yaml, so under OmegaConf struct mode every non-TinyMyo
experiment (LUNA_*, FEMBA_*, LuMamba_*, PanLUNA_*) raised
ConfigAttributeError at startup before any training began.
Read the key with cfg.get("wandb", None) and gate WandbLogger creation on
the result. TinyMyo runs keep the W&B logger; other models fall back to
the TensorBoard logger alone. No config files are changed.
Co-Authored-By: Claude Opus 5 (1M context) <[email protected]>
pyproject.toml holds repo-wide configuration that is not TinyMyo-specific: the project metadata and dependency list, the uv source pin for warmup-scheduler, and the black (line-length 120) and isort (black profile) settings the whole repo is formatted against. Deleting it would have dropped that tooling config for every model. Restored verbatim from main. It already declares wandb as a dependency, so the PR's new import in run_train.py resolves. Co-Authored-By: Claude Opus 5 (1M context) <[email protected]>
The PR changed --safetensor_path from default="model.safetensors" to required=True, which breaks existing callers that rely on the default. The argument is shared tooling, not TinyMyo-specific. This was the PR's only change to the file, so restoring it from main discards nothing else the PR intended. Co-Authored-By: Claude Opus 5 (1M context) <[email protected]>
Member
Author
|
Co-authored-by: Matteo Fasulo |
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
2 participants
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.
Supersedes #8