Open MatSci ML Toolkit is a framework for prototyping and scaling out deep learning models for materials discovery supporting widely used materials science datasets, and built on top of PyTorch Lightning, the Deep Graph Library, and PyTorch Geometric.
A number pages are still missing that describe pretty basic functionality for matsciml. This issue is being used to track those pages so we don't forget.
[ ] The "How do I" page is a stub - there's only model loading right now, but can be expanded to multitask stuff.
[ ] Training pipeline page is mostly blank. This should describe the task interface.
[ ] Datasets page is missing an OQMD entry
[ ] Missing unified references page - would be good to consolidate all the references we have for every model, dataset, technique in one page.
[x] #285
[x] Models (description of architecture, expected transforms, etc.)
[x] Transforms (expected arguments)
[x] Tasks
[x] Documentation for how to write documentation :) closed by #281
[ ] Inference tasks
[ ] Embedding generation
[ ] ASE interface
[ ] General testing workflow without averaging (i.e. pair plot of actual vs. predicted)
[x] #282
Request attributes
[ ] Would this be a refactor of existing code?
[ ] Does this proposal require new package dependencies?
[ ] Would this change break backwards compatibility?
[ ] Does this proposal include a new model?
[ ] Does this proposal include a new dataset?
[ ] Does this proposal include a new task/workflow?
Feature/behavior summary
A number pages are still missing that describe pretty basic functionality for
matsciml. This issue is being used to track those pages so we don't forget.Request attributes
Related issues
No response
Solution description
Write the docs :P
Additional notes
No response