Restore MPTrj EDA to working order #130
Merged
Conversation
This file contains 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
CompRhys
added
reproducibility
We can find files inside the extxyz zipfile with magmoms stepping through with the debugger but they aren't being read? |
CompRhys
commented
Aug 28, 2024
| cbar_coords=(0.18, 0.85, 0.42, 0.02), | ||
| # annotate each element with its number of magmoms in MPtrj | ||
| anno_kwds=tile_count_anno, | ||
| # anno_kwds=tile_count_anno, |
There was a problem hiding this comment.
does this have a replacement?
CompRhys
commented
Aug 28, 2024
data/mp/eda_mp_trj.py
Outdated
| @@ -196,6 +216,7 @@ def tile_count_anno(hist_vals: list[Any]) -> dict[str, Any]: | |||
|
|
|||
|
|
|||
| # %% plot histogram of number of sites per element | |||
| # TODO fix the cbar position and weirdness with 6x10e0 | |||
CompRhys
temporarily deployed
to
github-pages
GitHub Actions
Inactive
…t the strange y axis labels on the Cl element tile
…zip (now with same number of frames as original = 1_580_395) assert len(mp_trj_atoms) == 145_923 assert len(df_mp_trj) == 1_580_395
janosh
reviewed
Sep 5, 2024
data/mp/eda_mp_trj.py
Outdated
| for atoms_list in tqdm(mp_trj_atoms.values(), total=len(mp_trj_atoms)) | ||
| for atoms in atoms_list | ||
| } | ||
| ).T.convert_dtypes() # convert object columns to float/int where possible | ||
| df_mp_trj.index.name = "frame_id" | ||
| assert len(df_mp_trj) == 1_580_312 # number of total frames | ||
| assert len(df_mp_trj) == 1_580_395 # number of total frames | ||
| if Key.formula not in df_mp_trj: | ||
| raise KeyError(f"{Key.formula!s} not in {df_mp_trj.columns=}") | ||
|
|
||
| # this is the unrelaxed (but MP2020 corrected) formation energy per atom of the actual | ||
| # relaxation step | ||
| df_mp_trj = df_mp_trj.rename(columns={"ef_per_atom": MbdKey.e_form_dft}) |
There was a problem hiding this comment.
When chunked the script runs fairly quickly, two options exist: i) rerun that script to keep the extra labels I didn't save, I think I didn't do this as some of the formation energies looked wrong to me as they were the same as the energy per atom. ii) just grab the reference energies for the compositions and compute it quickly with https://github.com/janosh/matbench-discovery/blob/4d3d33f63a963f0d54f5538af04dd74866456c8f/matbench_discovery/energy.py#L83C5-L83C24
happy to do either, the only lost information is the bandgap in the current file but stuff does need to be recomputed/referenced for e_form etc
There was a problem hiding this comment.
yeah, leaning towards ii) as it seems slightly less reliant on flawless data provenance
There was a problem hiding this comment.
option ii) implemented in a1f92db and confirmed to yield identical results as before
CompRhys
commented
Sep 5, 2024
…rmation energies on the fly using matbench_discovery.energy.get_e_form_per_atom
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Still missing Magmoms and there are some older pmv heatmap arguments that are no longer valid and consequently cause issues