Deep Learning for Suicide and Depression Identification with Unsupervised Label Correction (ICANN 2021)

Frontiers in Neuroinformatics 2022: Local Label Point Correction for Edge Detection of Overlapping Cervical Cells

Code for Pseudo-Label Correction for Instance-Dependent Noise using Teacher-Student Framework

Deep Learning for Suicide and Depression Identification with Unsupervised Label Correction. 2021

Performed weakly supervised learning on CIFAR-10 images with noisy labels using convolutional neural networks (CNN).

label-utils is a Python utility for validating and correcting label files in datasets used for machine learning tasks. These scripts help ensure that the label files in your dataset are valid and correct, preventing issues during model training and evaluation.