Learning Chemistry: Exploring the suitability of machine learning for the task of structure-based chemical ontology classification - Data

Each of the packed folders contains either data or the results from the experiments published in the respective paper titled "Learning Chemistry: Exploring the suitability of machine learning for the task of structure-based chemical ontology classification". Each dataset is formatted as a csv or python pickle file. 

• chemdata_classical: Data used as input for the classical approaches (LR, Random Forest, ...) 
• chemdata_lstm: Data used as input for the LSTM approaches. Intended to be used with cheleary
• classif_reports_classical: Classification metrics of all classical approaches
• pathlengths: Comparison between ClassyFire and classical approaches and LSTM
• predictions_classical: Predictions of all classical approaches
• predictions_lstm: Predictions and classification metrics of the LSTM