vak.transforms.frame_labels.transforms

Class forms of transformations related to frame labels, i.e., vectors where each element represents a label for a frame, either a single sample in audio or a single time bin from a spectrogram.

These classes call functions from vak.transforms.frame_labels.functional. Not all functions in that module have a corresponding class, just key functions needed by dataloaders and models.

• FromSegments: transform to get frame labels from annotations

• ToLabels: transform to get back just string labels from frame labels, used to evaluate a model.

• ToSegments: transform to get segment onsets, offsets, and labels from frame labels.

  Used to convert model output to predictions. Inverse of from_segments.

• PostProcess: combines two post-processing transforms applied to frame labels, remove_short_segments and take_majority_vote, in one class.

Classes

FromSegments([unlabeled_label])	Transform that makes a vector of frame labels, given labeled segments in the form of onset times, offset times, and segment labels.
PostProcess(timebin_dur[, unlabeled_label, ...])	Apply post-processing transformations to a vector of frame labels.
ToLabels(labelmap)	Transforms that converts vector of frame labels to a string, one character for each continuous segment.
ToSegments(labelmap[, n_decimals_trunc])	Transform that converts a vector of frame labels into segments in the form of onset indices, offset indices, and labels.