Great Tit Hits Dataset #

(crowsetta format) #

The great-tit-hits-crowsetta.csv file contains song annotations in crowsetta's GenericSeq format. There are 1,161,033 rows, one per note, with the following columns:

  • label
  • onset_s
  • offset_s
  • onset_sample
  • offset_sample
  • notated_path
  • annot_path
  • sequence
  • annotation

See the crowsetta docs to learn how to use the library. To load the dataset in this format, simply run:

data root = [...]
from crowsetta.formats.seq import GenericSeq
dataset = GenericSeq.from_file(data_root / "great-tit-hits-crowsetta.csv")

The dataset was exported in this format using scrpts/auxiliary/to-crowsetta.py (see repo)/