Predict should accept batched inputs

SonyCSLParis / pesto

Self-supervised learning for fast pitch estimation

GNU Lesser General Public License v3.0

168 stars 15 forks source link

Predict should accept batched inputs #7

Closed aRI0U closed 9 months ago

aRI0U commented 10 months ago

Currently predict is limited to input tensors of dimension 2 (channels, num_samples). In a preprocessing context, it could be useful to make it accept audio batches