Dataset Curation

Learn how to get labeling recommendations using Tensorleap's Dataset Curation functionality.

After evaluation has completed, you can initiate 3 different curation process:

1. Unlabeled Data Recommendations - Tensorleap analyzes your model's performance and data distribution to identify high-impact unlabeled samples. By focusing your labeling effort on the most informative samples, you can significantly accelerate model improvement and reduce labeling costs.

2. Synthetic Data Optimization - Tensorleap aligns your synthetic data distribution with the target data source. By optimizing synthetic data generation, you can accelerate model improvement while significantly reducing labeling costs.

3. Pruning - Tensorleap analyzes your dataset distribution and rebalances it using your selected metadata tags. Apply filters to focus on a subset and optionally prioritize specific metadata dimensions to guide the pruning process.