Refactor diffusionsat source code to be more clean and update with latest huggingface/diffusers - #23
Open
Bili-Sakura wants to merge 13 commits into
Open
Refactor diffusionsat source code to be more clean and update with latest huggingface/diffusers#23Bili-Sakura wants to merge 13 commits into
huggingface/diffusers#23Bili-Sakura wants to merge 13 commits into
Conversation
Replace src/diffusers references with libs/diffusers across Makefile, setup.py, pyproject.toml, and README.md. Update installation to use pip install diffusers instead of editable install from src.
…ort and simplifying initialization. Remove legacy code and streamline forward method for enhanced clarity and performance.
…etadata handling and conditioning embedding. Simplify initialization and forward methods, while ensuring compatibility with varying downsample strides.
…ion handling. Remove unused ControlNetOutput class and streamline Conv3d initialization for better readability. Update logging imports for consistency.
…cing compatibility and simplifying the forward method. Remove unused methods and streamline imports for improved readability.
…ffusersStableDiffusionPipeline and DiffusersControlNetPipeline, enhancing functionality and reducing code duplication. Streamline method definitions for improved readability and maintainability.
…bleDiffusionPipeline, enhancing functionality and reducing code duplication. Streamline method definitions for improved readability and maintainability.
…nd pipeline_controlnet.py to align with updated module structure, enhancing code organization and maintainability.
…text_to_image.py by removing unused image_grid and compute_snr definitions. Update compute_snr call to use noise_scheduler for improved functionality and maintainability.
…eamline import statements and remove unused code in controlnet_texas_samples.ipynb and single-image.ipynb. Update metadata handling and enhance code readability.
Author
|
This is still on going. I will test model loading, inference and training in days. Furthermore, I will integrate the diffusionsat code as a custom pipeline so that only |
…neration - Introduced `pipeline_diffusionsat.py` for standard text-to-image generation with metadata support. - Added `pipeline_diffusionsat_controlnet.py` for ControlNet generation, incorporating conditional metadata handling. - Created README.md to document usage and setup instructions for the new pipelines.
Author
|
Also, I wonder can I re-distribute the model checkpoint from zenodo to huggingface/models for community convinient usage. I made a quick test repo on https://huggingface.co/BiliSakura . Tell me if any copyright concerns. |
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
1 participant
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.
For details of changes, see git commit history.