API

MODNet is built around two main classes, MODData and MODNetModel found in preprocessing and models modules.

• A MODData instance is used for representing a particular dataset. It contains a list of structures/compositions and the corresponding target properties, and can be used to perform feature selection.

• A MODNetModel instance is used for training and predicting of one or more properties or classes.

The complete hierarchical Python API of modnet can be found below.

• modnet package
  • Subpackages
    • modnet.featurizers package
      • Subpackages
      • Submodules
      • Module contents
    • modnet.hyper_opt package
      • Submodules
      • Module contents
    • modnet.matbench package
      • Submodules
      • Module contents
    • modnet.model_presets package
      • Submodules
      • Module contents
    • modnet.models package
      • Submodules
      • Module contents
  • Submodules
    • modnet.ext_data module
      • Usage
      • Dataset
      • load_ext_dataset()
    • modnet.preprocessing module
      • CompositionContainer
      • compute_mi()
      • map_mi()
      • nmi_target()
      • get_cross_nmi()
      • get_rr_p_parameter_default()
      • get_rr_c_parameter_default()
      • get_features_relevance_redundancy()
      • get_features_dyn()
      • merge_ranked()
      • MODData
    • modnet.sklearn module
      • MODNetFeaturizer
      • RR
      • MODNet
    • modnet.utils module
      • get_hash_of_file()
  • Module contents