Category of model metadata which should be null for defined metadata.For custom metadata is should be one of the following values "Performance,Training Profile,Training and Validation Datasets,Training Environment,other".
Description of model metadata
Key of the model Metadata. The key can either be user defined or OCI defined. List of OCI defined keys: * useCaseType * libraryName * libraryVersion * estimatorClass * hyperParameters * testartifactresults
Allowed values for useCaseType: binary_classification, regression, multinomial_classification, clustering, recommender, dimensionality_reduction/representation, time_series_forecasting, anomaly_detection, topic_modeling, ner, sentiment_analysis, image_classification, object_localization, other
Allowed values for libraryName: scikit-learn, xgboost, tensorflow, pytorch, mxnet, keras, lightGBM, pymc3, pyOD, spacy, prophet, sktime, statsmodels, cuml, oracle_automl, h2o, transformers, nltk, emcee, pystan, bert, gensim, flair, word2vec, ensemble, other
Defines properties of each model metadata.