Flamingo 0.3.0

What's Changed

• restart from last batch by @felipeZ in #34
• Dev by @felipeZ in #36
• allow the user to change the batch size by @felipeZ in #37
• run molecular filters in parallel #38 by @felipeZ in #39
• Solve issue reading billions of molecules by @felipeZ in #40
• removed unused print statement by @felipeZ in #42
• implement a filter for maximum number of functional groups by @felipeZ in #44
• add more functional groups to the single functional group filter by @felipeZ in #45
• Create single functional group filter by @felipeZ in #46
• replace set by frozenset by @felipeZ in #47
• add support for python 3.9 by @felipeZ in #48
• update documentation by @felipeZ in #49
• minor syntax fixes by @felipeZ in #50
• Drug-likeness new feature by @felipeZ in #52
• Run drug likeness only if requested by @felipeZ in #53
• Skip bond calculation as a feature by @felipeZ in #55
• Catch errors calling RDKit to compute drug-likeness properties by @felipeZ in #56
• skip smile if it doesn't have an inverse mol2smile representation by @felipeZ in #57
• reintroduce distance to the molecule featurizer by @felipeZ in #58
• Reset the molecular index when computing the druglikeness by @felipeZ in #60
• Create interface to cosmo-rs by @felipeZ in #61
• run fast sigma using a single Core by @felipeZ in #62
• remove atomic and bond features by @felipeZ in #64
• REL: Prepare for a flamingo release on pypi by @BvB93 in #65

Full Changelog: v0.2.1...0.3.0

Version 0.2.1

Changes

• Use all the available CPU to compute bulkiness with CAT by calling the imap_unordered Pool's method.
• Remove the batch_size input parameter and fix it to 1000.

Version 0.2.0

0.2.0 [03/11/2020]

• Workflow to compute properties using CAT

Version 0.1.0

0.1.0 (14/10/2020)

Added

• Move swan functionality to compute and filter properties flamingo