Skip to content
Discussion options

You must be logged in to vote

Hi @kannaricci ,

Ideally you want to shuffle your data to ensure that the training batches are more representative of the dataset, and that it's not dependent on some order / index. If you set max_epochs>=0 during training, the training Corpus is shuffled automatically every epoch, so you don't need to worry about shuffling it by yourself :)

Replies: 1 comment 1 reply

Comment options

You must be logged in to vote
1 reply
@kanayer
Comment options

Answer selected by kanayer
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
usage General spaCy usage training Training and updating models
2 participants