menu
announcement

Spectrum is now read-only. Learn more about the decision in our official announcement.

snorkel

Building and managing training datasets for machine learning

Channels
Team

Data Slicing

May 9, 2021 at 4:37pm
Hi! I applied the labeling and transformation functions on my 3 million of sentences. Now, I’m trying to apply the SliceAwareClassifier but it needs more than 12GB of RAM for only 5000 sentences, is it normal? Are there other models to improve the slice performance?

June 11, 2021 at 12:38am
Hi (carola) - could you file an issue on Github with some code snippets/more details about your inputs? (e.g. how many slices, data sample)