search

Chemellia / ChemistryFeaturization.jl

Interface package for featurizing atomic structures
https://chemistryfeaturization.chemellia.org/dev/
MIT License
41 stars 14 forks source link

Orbital Feature Descriptors and more #100

Closed thazhemadam closed 3 years ago

thazhemadam commented 3 years ago

Initial work up for introducing support for OrbitalFeatureDescriptors, that can encode Atoms into OFMs (orbital field matrices).

Still a work in progress, so open to any and all ideas/criticisms.

codecov-commenter commented 3 years ago

Codecov Report

Merging #100 (9c581c1) into main (7c3ba5b) will increase coverage by 6.15%. The diff coverage is 95.32%.

Impacted file tree graph

@@            Coverage Diff             @@
##             main     #100      +/-   ##
==========================================
+ Coverage   80.81%   86.97%   +6.15%     
==========================================
  Files          12       17       +5     
  Lines         318      476     +158     
==========================================
+ Hits          257      414     +157     
- Misses         61       62       +1
Impacted Files Coverage Δ
src/atoms/atoms.jl 100.00% <ø> (ø)
src/data.jl 0.00% <0.00%> (ø)
src/featurizations/graphnodefeaturization.jl 75.47% <0.00%> (ø)
src/features/orbitalfeature.jl 80.64% <80.64%> (ø)
src/ChemistryFeaturization.jl 100.00% <100.00%> (+33.33%) :arrow_up:
src/atoms/atomgraph.jl 76.19% <100.00%> (+1.99%) :arrow_up:
src/codecs/onehotonecold.jl 100.00% <100.00%> (ø)
src/codecs/simplecodec.jl 100.00% <100.00%> (ø)
src/features/elementfeature.jl 88.23% <100.00%> (+5.88%) :arrow_up:
src/features/features.jl 100.00% <100.00%> (+33.33%) :arrow_up:
... and 10 more

Continue to review full report at Codecov.

Legend - Click here to learn more Δ = absolute <relative> (impact), ø = not affected, ? = missing data Powered by Codecov. Last update 7c3ba5b...9c581c1. Read the comment docs.

thazhemadam commented 3 years ago

Currently, this is a little lacking in documentation, but this PR is pretty behind already and I didn't want to unnecessarily introduce merge conflicts with #116.

rkurchin commented 3 years ago

Sounds good to me! Go ahead and merge in whatever order you want with #116 (I'll let you do it in case there's an order that will make it less messy haha)