v0.0.8

What's Changed

• feat: implement "text" featurizer by @kjappelbaum in #441
• feat: add RCSR code to text description and return numpy array by @kjappelbaum in #444

Full Changelog: v0.0.7...v0.0.8

v0.0.7

What's Changed

• make it easier for users to customize logging by @kjappelbaum in #415

Full Changelog: v0.0.6...v0.0.7

v0.0.6

What's Changed

• fix: print for the RDKitAdaptor Featurizers. by @kjappelbaum in #413
• make some additional imports optional

Full Changelog: v0.0.5...v0.0.6

v0.0.5

What's Changed

• add an additional example notebook by @kjappelbaum in #412

Full Changelog: v0.0.4...v0.0.5

v0.0.4

What's Changed

New features

This version comes with two new featurizers. First, a "meta"-featurizer that can apply any featurizer on both hosts and guests and then concatenates the results. Second, a flavor of atom-property labeled radial distribution functions that are centered on the guest atoms. That is, instead of considering the correlations between all atoms it considers only the correlations between the guest atoms and all framework atoms within a cutoff radius.

• implement host-guest featurizer by @kjappelbaum in #405
• implement guest-centered APRDF by @kjappelbaum in #405

Additionally, this release brings the first interface to an experimental dataset. We include the thermal stability dataset curated by the Kulik group.

• interface thermal stability dataset from Nandy et al. by @kjappelbaum in #411

Chores

• Use literal syntax instead of function calls to create data structure by @deepsource-autofix in #409
• Replace range(len(...)) with enumerate(...) by @deepsource-autofix in #410
• Use correct syntax for grouping parallel tests
• Start fixing typing issues

Full Changelog: v0.0.3...v0.0.4

mofdscribe v0.0.3

New features

• We implement interfaces for custom datasets based on a collection of files and/or a dataframe. As the built-in datasets, those datasets are compatible with mofdscribe features such as Splitters (#396)

Bugfixes

• pymatgen 2022-11-1 changed to order of args in Structure.to which was incompatible with our use (#391)

Chores

• Dropped numpy version pin (#399)
• Optimized CI using mamba and parallel tests (#393)

mofdscribe v0.0.2

This minor update of mofdscribe comes with a better interface for implementing custom StructureDataset and parallelize property computation in those datasets.

We also no longer drop nan for the QMOFDataset.

v0.0.1 - initial release

Bump version: 0.0.1-dev → 0.0.1