Skip to content

v2.0.0

Choose a tag to compare

View all tags
@jvalegre jvalegre released this 17 Feb 11:42
· 88 commits to master since this release
2.0.0
e5d92ea
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

Adaptation of the code to avoid overfitting and to use with low-data problems

  • Fixed a bug in one-hot encoding in the one-hot test
  • Adding the possibility to disable the automatic standarization of descriptors (--std False)
  • Changing CV_test (now it standardizes the full database with sklearn functions)
  • Fixing a bug with the sklearn-intelex accelerator
  • Fixing a threading bug with matplotlib in SHAP
  • train:validation split was replaced by a repeated k-fold CV
  • The program always holds out a test set
  • The average results of the repeated k-fold CV are used to measure predictive ability and to predict new results
  • The BayesianOptimization() is used to find the bets model, using a combined metric that depends on interpolation and extrapolation of diferent types of CVs
  • This version does not work with classification problems and the AQME and EVALUATE modules were disabled until v2.0.1.
  • Updated ROBERT score, which is more robust towards small data problems
Assets 2
Loading