Skip to main content

Syntheseus wrapper for retro* benchmark.

Project description

Syntheseus retro star benchmark

A wrapper for using the benchmark from retro* (Chen et al 2020) in syntheseus.

Usage:

from syntheseus_retro_star_benchmark import RetroStarReactionModel
model = RetroStarReactionModel()  # a syntheseus BackwardReactionModel object wrapping the pre-trained template classifier

from syntheseus_retro_star_benchmark import RetroStarInventory
inventory = RetroStarInventory()  # their inventory of ~23M purchasable molecules

from syntheseus_retro_star_benchmark import get_190_hard_test_smiles
test_smiles = get_190_hard_test_smiles()  # their recommended 190 test SMILES

from syntheseus_retro_star_benchmark import RetroStarValueMLP
value_fn = RetroStarValueMLP()  # their pre-trained search heuristic

Code was based on open source code from here. Some data was uploaded to figshare to ensure stable, consistent access.

Installation

Install either by cloning and using pip or running pip install syntheseus-retro-star-benchmark.

Development

Ensure to install all pre-commit hooks and run unit tests (provided by pytest).

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

syntheseus-retro-star-benchmark-0.1.0.tar.gz (16.0 kB view details)

Uploaded Source

Built Distribution

File details

Details for the file syntheseus-retro-star-benchmark-0.1.0.tar.gz.

File metadata

File hashes

Hashes for syntheseus-retro-star-benchmark-0.1.0.tar.gz
Algorithm Hash digest
SHA256 2c59608a2438801e716f2548cf3b5ffae311b4d4d7a14534f0fe0216671e1a4d
MD5 7804d47d64160c124a55c41701ed02bc
BLAKE2b-256 b0aeb5542840e8b389a56ab2fecd69291ceed14ab54b03d51d877653f6dfa221

See more details on using hashes here.

File details

Details for the file syntheseus_retro_star_benchmark-0.1.0-py3-none-any.whl.

File metadata

File hashes

Hashes for syntheseus_retro_star_benchmark-0.1.0-py3-none-any.whl
Algorithm Hash digest
SHA256 df3ee794c09967013c6cd6158aca374ab144b85bb619bb156b96eb546403203b
MD5 56798bf515f03eee74199ca13521b665
BLAKE2b-256 a4076f3fb235ad221f6773ca9da1942ee530c76fe7dc94ea5c774f1b1ba99b68

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page