Fréchet ChEMNet Distance
Project description
Fréchet ChemNet Distance
Code for the paper "Fréchet ChemNet Distance: A Metric for Generative Models for Molecules in Drug Discovery" JCIM / ArXiv
Installation
You can install the FCD using
pip install fcd
rdkit is best install via conda
conda install rdkit -c rdkit
Requirements
python>3<3.8
numpy
tensorflow>1.8
keras>2.1
scipy
rdkit is causing troubles with python3.8 for me. You might wanna opt for 3.7.
For the effect of versions on results see tests/test_results.csv
.
Using the current versions of tensorflow (2.1.0) and keras (2.3.1) results differ from previous versions but
are probably negligible.
Version 1.1 changes
- Got rid of unneeded imports
load_ref_model
doesn't need an argument any more to load a model.canonical
andcanonical_smiles
now returnNone
for invalid smiles.- Added
get_fcd
as a quick way to get a the fcd score from two lists of smiles.
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
FCD-1.1.tar.gz
(53.1 MB
view hashes)
Built Distribution
FCD-1.1-py3-none-any.whl
(53.1 MB
view hashes)