oktoberfest.pr.refine_intensity_predictor

oktoberfest.pr.refine_intensity_predictor(baseline_model_path, libraries, config, data_directory, result_directory, dataset_name, model_name, download_new_baseline_model=False)

Perform refinement/transfer learning on a baseline intensity predictor.

Parameters:

• baseline_model_path (Path) – Path of baseline model to refine

• libraries (list[Spectra]) – Spectral libraries to use as training data for refinement

• config (Config) – Config containing refinement learning options

• data_directory (Path) – Directory to save processed dataset and refined model to. The Parquet and ion_type and modification metadata files for the processed dataset will be stored in <output_directory>/<dataset_name>/, and the refined model as <output_directory>/<model_name>.keras

• result_directory (Path) – Directory to save CSV logs & report notebook to

• dataset_name (str) – Name of dataset

• model_name (str) – Name of refined model

• download_new_baseline_model (bool) – Whether to download a new baseline model from GitHub to the specified path

Return type:

None