Skip to content

Add maintainers to README #34

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
bananenpampe merged 2 commits into from
Mar 30, 2026
Merged

Add maintainers to README #34

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
75e9013
Add maintainers to README
Luthaf Mar 30, 2026
3b583ad
Apply suggestions from code review
Luthaf Mar 30, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
31 changes: 16 additions & 15 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -111,8 +111,8 @@ cs_iso_uncertainty = np.std(cs_committee_iso, axis=1, ddof=1)
cs_psa = np.linalg.eigvalsh(cs_tensor)
```

This snippet will estimate the predicted chemical shieldings of diamond to be highly uncertain,
as expected and desired, given that diamond as an inorganic material is not well
This snippet will estimate the predicted chemical shieldings of diamond to be highly uncertain,
as expected and desired, given that diamond as an inorganic material is not well
represented in the training data of the model.


Expand Down Expand Up @@ -163,11 +163,11 @@ pip install shiftml==<version>
<details>
<summary><strong>ShiftML3 predictions aren’t identical for magnetically equivalent atoms. Why?</strong></summary>

ShiftML3 is built on the **Point Edge Transformer (PET)** model, which is *not perfectly rotationally invariant*.
This can introduce tiny, random differences for atoms that are magnetically equivalent.
ShiftML3 is built on the **Point Edge Transformer (PET)** model, which is *not perfectly rotationally invariant*.
This can introduce tiny, random differences for atoms that are magnetically equivalent.
We have verified that these fluctuations are minor and do **not** harm overall accuracy.

> **Tip – get identical shielding predictions**
> **Tip – get identical shielding predictions**
> Average the predictions over all magnetically equivalent atoms.

</details>
Expand All @@ -177,10 +177,10 @@ We have verified that these fluctuations are minor and do **not** harm overall a
<details>
<summary><strong>ShiftML3 shows large errors versus my GIPAW-DFT shieldings. What’s going on?</strong></summary>

Chemical-shielding calculations are *very* sensitive to the **code and convergence parameters** used.
Chemical-shielding calculations are *very* sensitive to the **code and convergence parameters** used.
Only compare ShiftML3 to GIPAW-DFT data generated with *exactly* the same settings as the training set.

*Reference inputs* for Quantum Espresso with the correct parameters are available in this
*Reference inputs* for Quantum Espresso with the correct parameters are available in this
[Zenodo data repository](https://zenodo.org/records/7097427).

</details>
Expand All @@ -190,7 +190,7 @@ Only compare ShiftML3 to GIPAW-DFT data generated with *exactly* the same settin
<details>
<summary><strong>I used identical GIPAW-DFT parameters but still see big errors. What now?</strong></summary>

Check the model’s **uncertainty estimates** (committee variance; see “Advanced usage” above).
Check the model’s **uncertainty estimates** (committee variance; see “Advanced usage” above).
If the uncertainty is **several ×** the element’s test-set RMSE, the prediction is probably unreliable
for your structure.

Expand All @@ -201,11 +201,11 @@ for your structure.
<details>
<summary><strong>My calculated shieldings don’t correlate with experiment at all. Why?</strong></summary>

1. **Validate the baseline.**
Make sure reliable **GIPAW/PBE** results exist (or recompute them) and confirm they correlate with experiment.
1. **Validate the baseline.**
Make sure reliable **GIPAW/PBE** results exist (or recompute them) and confirm they correlate with experiment.
Inaccurate DFT—often the exchange–correlation functional—can be blamed.

2. **Check your structures.**
2. **Check your structures.**
If candidate geometries don’t reflect experimental conditions *or* the inter-atomic potential used to generate structures is poor,
both DFT and ML predictions will stray from reality.

Expand Down Expand Up @@ -250,11 +250,13 @@ pytest

## Contributors

Matthias Kellner\
Yuxuan Zhang\
Ruben Rodriguez Madrid\
Matthias Kellner
Yuxuan Zhang
Ruben Rodriguez Madrid
Guillaume Fraux

This project is [maintained](https://github.com/lab-cosmo/.github/blob/main/Maintainers.md) by [@bananenpampe](https://github.com/bananenpampe), who will reply to issues and pull requests opened on this repository as soon as possible. You can mention them directly if you did not receive an answer after a couple of days.

## References

This package is based on the following papers:
Expand All @@ -264,4 +266,3 @@ This package is based on the following papers:
- A Machine Learning Model of Chemical Shifts for Chemically and\
Structurally Diverse Molecular Solids - Cordova et al. [[3](https://doi.org/10.1021/acs.jpcc.2c03854)]
- A deep learning model for chemical shieldings in molecular organic solids including anisotropy - Kellner, Holmes, Rodriguez Madrid, Viscosi, Zhang, Emsley, Ceriotti [[4](https://doi.org/10.1021/acs.jpclett.5c01819)]

Loading