What's Changed
- Update site to show ALIGNN results [`38`](https://github.com/janosh/matbench-discovery/pull/38)
- ALIGNN test on WBM data [`37`](https://github.com/janosh/matbench-discovery/pull/37)
- Add PDF versions of all SI figures [`36`](https://github.com/janosh/matbench-discovery/pull/36)
- Display `largest-error-scatter-select.svelte` in `/si` [`35`](https://github.com/janosh/matbench-discovery/pull/35)
- Fix KeyError('wyckoff_spglib') in `fetch_process_wbm_dataset.py` [`34`](https://github.com/janosh/matbench-discovery/pull/34)
- Fix `compute_struct_fingerprints.py`: cannot insert `material_id`, already exists [`31`](https://github.com/janosh/matbench-discovery/pull/31)
- Fix `fetch_process_wbm_dataset.py` unnecessarily loading WBM initial structures from disk [`29`](https://github.com/janosh/matbench-discovery/pull/29)
- Fix loading `2023-02-07-ppd-mp.pkl.gz` [`26`](https://github.com/janosh/matbench-discovery/pull/26)
- Fix downloading compressed Figshare data [`21`](https://github.com/janosh/matbench-discovery/pull/21)
- Run scripts in CI [`19`](https://github.com/janosh/matbench-discovery/pull/19)
- Prompt to download missing data files [`15`](https://github.com/janosh/matbench-discovery/pull/15)
- Add `upload_to_figshare.py` and replace local data-loading with cached Figshare downloads [`13`](https://github.com/janosh/matbench-discovery/pull/13)
- DRY workflows [`8`](https://github.com/janosh/matbench-discovery/pull/8)
- Add `Files`, `DataFiles` and `PredFiles` classes in `matbench_discovery/data.py` [`7`](https://github.com/janosh/matbench-discovery/pull/7)
- fix path issues and using JSON pandas reader for CSV files in train_(cgcnn|wrenformer).py (closes 43) [`43`](https://github.com/janosh/matbench-discovery/issues/43)
- Fix KeyError('wyckoff_spglib') in `fetch_process_wbm_dataset.py` (34) [`32`](https://github.com/janosh/matbench-discovery/issues/32) [`#33`](https://github.com/janosh/matbench-discovery/issues/33)
- Fix `compute_struct_fingerprints.py`: cannot insert `material_id`, already exists (31) [`30`](https://github.com/janosh/matbench-discovery/issues/30)
- Fix `fetch_process_wbm_dataset.py` unnecessarily loading WBM initial structures from disk (29) [`28`](https://github.com/janosh/matbench-discovery/issues/28)
- fix AttributeError: 'float' object has no attribute 'split' in fetch_process_wbm_dataset.py (closes 27) [`27`](https://github.com/janosh/matbench-discovery/issues/27)
- Fix loading `2023-02-07-ppd-mp.pkl.gz` (26) [`23`](https://github.com/janosh/matbench-discovery/issues/23)
- Fix downloading compressed Figshare data (21) [`20`](https://github.com/janosh/matbench-discovery/issues/20) [`#17`](https://github.com/janosh/matbench-discovery/issues/17)
- Run scripts in CI (19) [`16`](https://github.com/janosh/matbench-discovery/issues/16)
- Prompt to download missing data files (15) [`14`](https://github.com/janosh/matbench-discovery/issues/14) [`#14`](https://github.com/janosh/matbench-discovery/issues/14)
- Add `upload_to_figshare.py` and replace local data-loading with cached Figshare downloads (13) [`10`](https://github.com/janosh/matbench-discovery/issues/10) [`#12`](https://github.com/janosh/matbench-discovery/issues/12) [`#11`](https://github.com/janosh/matbench-discovery/issues/11)
- support model_name as array in metadata.yml files to share metadata between multiple types of a model [`c8eaebd`](https://github.com/janosh/matbench-discovery/commit/c8eaebd673973fccf7db74cec0f8a8a6f0883932)
- include legacy MP energy corrections in data/wbm/2022-10-19-wbm-summary.csv, use them to remove old and apply new corrections in test_megnet.py [`8166801`](https://github.com/janosh/matbench-discovery/commit/81668016b610a6567058a4066b739586e54cf43b)
- fix join_chgnet_results.py by removing code to apply MP2020 corrections [`01658ad`](https://github.com/janosh/matbench-discovery/commit/01658ad73e4c6cb6a320d4f906f19fb443d08c9f)
- add columns wyckoff_spglib and spacegroup_symbol to data/mp/2023-01-10-mp-energies.csv [`01e88f0`](https://github.com/janosh/matbench-discovery/commit/01e88f007faec207ecf7a6f59663f62aecbaf394)
- add model metrics table when only validating the 10k most stable predictions for each model [`c4ca186`](https://github.com/janosh/matbench-discovery/commit/c4ca1860b970bfc469278b915b80fdbf4c2d9626)
- run model combos (m3gnet|chgnet)+megnet through test_megnet.py [`fbe847b`](https://github.com/janosh/matbench-discovery/commit/fbe847bbcb4500277f4e291ce965611667df87d2)
- split analyze_model_failure_cases.py into two scripts, new one is analyze_elements.py [`0fad3bd`](https://github.com/janosh/matbench-discovery/commit/0fad3bd478f2290a26b90351d8b11c5b6cc1491a)
- add scatter plot of largest errors averaged over models vs DFT hull distance [`457abf0`](https://github.com/janosh/matbench-discovery/commit/457abf05476f8835a29e29a1be40e6acb085523b)
- add scripts/compute_struct_fingerprints.py to generate matminer SiteStats fingerprints for all MP+WBM structures [`7946b5e`](https://github.com/janosh/matbench-discovery/commit/7946b5e433713ef8f46b839562f09d889331e63b)
- split model run times into train and test contribs [`b8a18d8`](https://github.com/janosh/matbench-discovery/commit/b8a18d8fd357ea4ec77dae837fdb8317fe978b28)
- add models/chgnet/{test_chgnet,join_m3gnet_results}.py [`da39074`](https://github.com/janosh/matbench-discovery/commit/da390741b0d0f03fbcca5bcbac725c2211a80427)
- git surgery purging all large CSV files from repo history [`4a90dee`](https://github.com/janosh/matbench-discovery/commit/4a90dee85965aebbcfcd547353c548eeb19651ce)
- add scripts/compute_projections.py [`eec1cb4`](https://github.com/janosh/matbench-discovery/commit/eec1cb4073c2db65cd238157f656af963582ffc5)
- make horizontal versions of roc-models-2x4.pdf and model-run-times-bar.pdf [`551050e`](https://github.com/janosh/matbench-discovery/commit/551050e17cc98cf786707448fd304bb06177a914)
- add changelog.md [`0621282`](https://github.com/janosh/matbench-discovery/commit/0621282b0ed6ca397b6293e8d6f2bcbaac736468)
- add models/chgnet/ctk_struct_traj.py and models/wrenformer/analyze_wrenformer.py [`c9fed5a`](https://github.com/janosh/matbench-discovery/commit/c9fed5ae7bddc02c3fbb2160a00b4e8822b1a179)
- add data/mp/get_mp_traj.py to save a snapshot of all MP ionic steps on 2023-03-15 to be released as MBD canonical training set [`76647e3`](https://github.com/janosh/matbench-discovery/commit/76647e37b9fdfbbc77fb2317ad2a0cf3369638a9)
- move load_df_wbm_preds+PRED_FILES from matbench_discovery/{data->preds}.py [`387184c`](https://github.com/janosh/matbench-discovery/commit/387184caa50c27e2574a86a4bef65a1628434378)
- add site/src/figs/(largest-fp-diff-each-error-models|largest-each-errors-fp-diff-models).svelte [`4b6e83a`](https://github.com/janosh/matbench-discovery/commit/4b6e83a13816f9b8dcc911af5cbe1d79f05bf876)
- move /paper/preprint to /paper [`8a0c3bb`](https://github.com/janosh/matbench-discovery/commit/8a0c3bb47609cf784d95e7499394785487383d82)
- add global STABILITY_THRESHOLD to consistently parametrize across the codebase when materials count as thermodynamically stable [`70b2b1d`](https://github.com/janosh/matbench-discovery/commit/70b2b1d81476e5913802d8286f281af9385a8754)
- plot easy vs hard structures (for all models) norm of SiteStats fingerprint difference before/after relaxation [`a6bfa74`](https://github.com/janosh/matbench-discovery/commit/a6bfa749aba0076a20af9e8fa83422744a4d45ea)
- add missing doc strings [`1ae72c0`](https://github.com/janosh/matbench-discovery/commit/1ae72c0c2f395e90d0a2ba6d1a87f3e7332af805)
- fix direct data file download URLs on /contribute page [`83736ad`](https://github.com/janosh/matbench-discovery/commit/83736ad5a94b823f69d7c61314daf9fbaeac331d)
- add git submodule paper to prepare journal submission in separate repo [`3ac72e6`](https://github.com/janosh/matbench-discovery/commit/3ac72e66df1b2e163c08840fabe41e078d7045f6)
- add figs/cumulative-mae-rmse.svelte displayed in /si [`8798786`](https://github.com/janosh/matbench-discovery/commit/879878603af492888448d59c209a41f74f3fe6f5)
- rename ElemCountInset to PtableInset and add props precision, show_percent, unit [`abfc0ac`](https://github.com/janosh/matbench-discovery/commit/abfc0ac1e6bd956975466426f8760ef9a489be99)
- add scripts/ctk_structure_viewer.py [`fa1a439`](https://github.com/janosh/matbench-discovery/commit/fa1a439b0c73222e583c14b9167c143074d57315)
- +layout.svelte add <svelte:head> with <title> and <meta description> [`5d38918`](https://github.com/janosh/matbench-discovery/commit/5d3891858e88c8efce694389c55dcf597b5464ce)
- refactor data loading in model test scripts [`d564ade`](https://github.com/janosh/matbench-discovery/commit/d564ade5359daeca286ac41c59b02716572c5a40)
- add CmdPalette invoked with cmd+k for keyboard-only site navigation [`59010fa`](https://github.com/janosh/matbench-discovery/commit/59010fa272c29c9d598ca89d81ecc7a116114672)
- improve preprint results + discussion wording [`d27f1c4`](https://github.com/janosh/matbench-discovery/commit/d27f1c466a3c6a8677b51b908f13d60b040c8dfa)
- fix outdated numbers in data/wbm/readme.md [`79c8042`](https://github.com/janosh/matbench-discovery/commit/79c804281c5aa49b223fa7944f8dee8633dc7f44)
- change SI plot of largest model errors: Predicted vs. DFT hull distance colored by model disagreement [`a5b3211`](https://github.com/janosh/matbench-discovery/commit/a5b3211f571f49e5c72d948d971bbd3db19066f0)
- show figs/element-prevalence-vs-error.svelte on /si page [`6771462`](https://github.com/janosh/matbench-discovery/commit/6771462f997fc8cf87e19a7327d349670d2ad44d)
- rm frontmatter.yml after consolidating with citation.cff [`228be00`](https://github.com/janosh/matbench-discovery/commit/228be00c7f4408328c28919cd053d84bf616b5e0)
- rename data/wbm/(analysis->eda).py [`0f2410d`](https://github.com/janosh/matbench-discovery/commit/0f2410de1d619c8318d9a06ce3d693d19d76b223)
- make wide tables horizontally scrollable on mobile screens [`fc20ccf`](https://github.com/janosh/matbench-discovery/commit/fc20ccf5f33dc50ebfa429d4b9ceee95715d7e03)
- toggle site tips on cmd+j, update deps, tweak copy (abstract, metrics-table-first-10k, readme) [`137c414`](https://github.com/janosh/matbench-discovery/commit/137c414badc8a4bd69748396077a46292a09c647)
- add date_published field to model metadata [`cddda31`](https://github.com/janosh/matbench-discovery/commit/cddda31d9395574e63ac3f8a39a9e7df7570fa31)
- fix /contribute [`dfa6c24`](https://github.com/janosh/matbench-discovery/commit/dfa6c24b12273816d3fa30442003e884c64d3b58)
- allow toggling MP+WBM element prevalence ptable heat maps between count modes 'occurrence' and 'composition' [`1fe5bcb`](https://github.com/janosh/matbench-discovery/commit/1fe5bcbf0752e6eb0d04b0e3365bc570237fbf60)
- add material_id to figs/mp-elemental-ref-energies.svelte hover tooltip [`c87f0b2`](https://github.com/janosh/matbench-discovery/commit/c87f0b2087b4d0ed7fb43331cf82e57fbca75ef8)
- bump rollup/plugin-yaml to v4.1 and refactor citation.cff import using new extensions option [`cc08c78`](https://github.com/janosh/matbench-discovery/commit/cc08c78bcc2f1433c6cc52cb97b16b05a41d0014)
- tweak contributing guide [`b3ea8d3`](https://github.com/janosh/matbench-discovery/commit/b3ea8d3980d62032ed33ae75fb0c7539fafcebe3)
- upload pbenner's ALIGNN checkpoint to Figshare and link from models/alignn/readme.md [`da56377`](https://github.com/janosh/matbench-discovery/commit/da5637725e3f785cdf8bd82ff6a3be95858b2eae)
- start preprint [`fd8355d`](https://github.com/janosh/matbench-discovery/commit/fd8355dca9fe87ef9196980c06907ef377fdcd97)
- rename pages /about-test-test-set to /about-the-data and /how-to-contribute to /contribute [`e41ca32`](https://github.com/janosh/matbench-discovery/commit/e41ca32e768ea06f404b216353521f89e6bee2e2)
**Full Changelog**: https://github.com/janosh/matbench-discovery/compare/v0.1.4...v0.1.5