Evaluation

This module evaluates music generation systems.

Music Generation

This module contains the implementation of the paper:

[1] Yang, L. C., & Lerch, A. (2020). On the evaluation of generative models in music. Neural Computing and Applications, 32(9), 4773-4784. https://doi.org/10.1007/s00521-018-3849-7

PitchMeasures(value)	An enumeration.
RhythmMeasures(value)	An enumeration.
get_all_dataset_measures(*args, **kwargs)
get_average_dataset_measures(*args, **kwargs)
get_eval_measures(notes[, measures])	Computes the features used to evaluate music.
euclidean_distance(measures1[, measures2])	Computes the pair-wise cross-validation with Euclidean distance of the features extracted from a dataset which are stored in a dict.
get_distribution(*args[, measure, show])	Plots the measure histogram of the input distances datasets (args).
compute_overlapped_area(eucl_dist1, ...)
compute_kld(eucl_dist1, eucl_dist2, measure)
compute_oa_kld(measures_dist1, measures_dist2)	Computes the KLD and OA of all the features in the input dicts and build a dict with the reults.
plot_measures(measures_dist1, measures_dist2)
model_features_violinplot(measures_dist1, ...)	Plots the distances distributions of the input features as a violin plot as the Fig.