feat: Add Intel XPU support for all models
- Add XPU accelerator, precision, and strategy classes
- Update DDP utilities to detect and use Intel XPU
- Add XPU trainer configs for RF3 and RFD3
- Update MPNN inference engine for XPU compatibility
- Update README with XPU documentation
Right now going to the GH pages site just shows the README, this is likely because GH is only building the README with Jekyll even though this workflow should override it without the added lines.
* Starting to put together the foundry and RFD3 external documentation
Set up the foundation for Sphinx to be able to build the external docs, first draft of a ppi_design_tutorial has been completed.
* Add RFdiffusion3 documentation and update toctree
Added initial documentation for RFdiffusion3 under models/rfd3/docs/index.rst and linked it in the main docs toctree. Updated toctree maxdepth for better navigation and added a symlink for the rfd3 model documentation.
* Update and expand RFdiffusion3 documentation
Added introductory and reference documentation files, improved the index structure with general information and examples, and enhanced the PPI design tutorial with additional notes, figures, and clarifications. Also fixed a typo in a PDB filename.
* Update RFD3 documentation and tutorial content
Expanded the Sphinx static path to include RFD3 assets and made minor formatting and clarity improvements in the main and RFD3-specific documentation. The PPI design tutorial was revised for clarity, improved step-by-step instructions, and better separation of setup and execution steps.
* Update RFD3 documentation and tutorials
Added source_suffix to Sphinx conf.py for Markdown support. Updated index.rst to include new documentation sections. Expanded intro_inference_calculations.md with detailed instructions on inference input formats, job configuration, and output files. Improved input.md formatting for appendices and FAQs. Revised ppi_design_tutorial.md for clarity, added details on settings, and expanded explanations for hotspots and batch inference.
* Update RFD3 docs: clarify input specs and file formats
Expanded and clarified the documentation for RFdiffusion3 input specifications, including more detailed explanations of the 'contig' string, input file types, and example YAML/JSON formats. Improved the intro to inference calculations to better explain the structure and usage of settings files, and updated descriptions for job configuration and output files. Added a placeholder for configuration options documentation.
* Update RFdiffusion3 input documentation and examples
Expanded and clarified the documentation for RFdiffusion3 input specification, including detailed explanations of CLI arguments, InputSpecification fields, the InputSelection mini-language, contig string formatting, and advanced options such as partial diffusion and CIF parser arguments. Added more examples, debugging recommendations, and an updated FAQ. Also updated the output file naming explanation for clarity. Removed the obsolete configuration_options.md file.
Note that images are still not being rendered correctly in many of the md files. Fix will be in future commit.
* PPI tutorial and RFD3 docs update
Created output files for PPI tutorial and listed their locations. Made edits to files to add labels to sections to remove sphinx warnings.
* Delete docs/source/conf.py~
This is an auto-save file from emacs - it does not need to be in the repo.
* Delete docs/source/index.rst~
This is an auto-saved copy of index.rst, it does not need to be in the repo
* Adding missing images and fixing docs symlink
* Fix grammatical error in RFdiffusion3 documentation
Fixed typo, clarified enzyme design language.
* Fix typos and enhance clarity in inference docs
Corrected typos and improved clarity in the documentation regarding inference settings and file formats.
* Improve 'unindex' field description in input_parsing.py
Updated the description for the 'unindex' field to clarify its purpose and usage.
* Make format
