tempo_eval.print_report

tempo_eval.print_report(output_dir='./', validate=True, corpus_names=None, estimates_dir=None, references_dir=None, include_relative_dir_in_item_id=True, format='kramdown', size=Size.L)[source]

Print reports for references and estimates and make them accessible via an index page.

By default the report covers all built-in references and estimates. When providing an estimates or reference directory, which must contain jams, the report covers only corpora that occur in the provided annotations or are specified in the corpus_names argument. By convention, tempo_eval uses the file’s os.path.relpath as item id (relative to the estimates or references directory). You can change this behavior to use os.path.basename by setting include_relative_dir_in_item_id to False. You can override either of these two mechanisms by adding a jam.file_metadata.identifiers['tempo_eval_id']-value to your jams.

Parameters

  • size (Size) – size of evaluation report—choose “L” for a comprehensive version, choose “S” for a shorter version with just the essentials

  • format (str) – kramdown, markdown or html

  • corpus_names (Optional[List[str]]) – list of corpora to generate reports for. See list_reference_corpus_names to get a list of valid names.

  • output_dir (str) – output directory

  • validate (bool) – validate jam while reading (has major performance impact)

  • references_dir (Optional[str]) – directory to search for reference jams

  • estimates_dir (Optional[str]) – directory to search for estimate jams

  • include_relative_dir_in_item_id (bool) – when using an extra reference or estimate directory, use the relative directory name as part of the item id or not.

Return type

str

Returns

output file name for the index page

Raises

ValueError – if the given format is not supported

Note

By default, validate is True in order to stay safe. But since it affects performance quite a bit, you might want to turn validation off when using this function to keep your sanity.