Pinned

1. cleanlab

   The standard package for machine learning with noisy labels and finding mislabeled data. Works with most datasets and models.

   Python 2.2k 211

2. benchmarking-keras-pytorch

   🔥 Reproducibly benchmarking Keras and PyTorch models

   Python 358 40

3. hypopt

   ⏸ Parallelized hyper-param optimization with validation set, not crossval

   Python 71 16

4. confidentlearning-reproduce

   Official data release to reproduce Confident Learning paper results

   Jupyter Notebook 24 10

5. rankpruning

   🧹 Formerly for binary classification with noisy labels. Replaced by cleanlab.

   Jupyter Notebook 78 12

6. facebookresearch/EgoCom-Dataset

   EgoCom: A Multi-person Multi-modal Egocentric Communications Dataset

   Jupyter Notebook 31 3