Add 9-species model weights link to FAQ (#303)

* Add model weights link * Generate new screengrabs with rich-codex * Clarify that these weights should only be used for benchmarking --------- Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com> Co-authored-by: Wout Bittremieux <[email protected]>
Noble-Lab · Feb 24, 2024 · 6e705e5 · 6e705e5
1 parent b7f8ff9
commit 6e705e5
Show file tree

Hide file tree

Showing 2 changed files with 160 additions and 11 deletions.
diff --git a/docs/faq.md b/docs/faq.md
@@ -145,6 +145,13 @@ You can use any of the scheduler classes available in [`torch.optim.lr_scheduler
 
 ## Miscellaneous
 
+**Where can I find Casanovo model weights trained on the nine-species benchmark?**
+
+You can find the Casanovo weights corresponding to the nine-species benchmark [on Zenodo](https://doi.org/10.5281/zenodo.10694984), compatible with Casanovo v4.x.x.
+These weights correspond to training and validation on eight species using the default configurations, with the remaining species held out for testing, as indicated by the file names.
+Note that these weights are only intended for evaluation purposes on this specific benchmark dataset.
+For general-purpose usage of Casanovo, use its [default weights](https://casanovo.readthedocs.io/en/latest/getting_started.html#download-model-weights) instead, as these will give significantly improved performance.
+
 **How can I generate a precision–coverage curve?**
 
 You can evaluate a trained Casanovo model compared to ground-truth peptide labels using a precision–coverage curve.