for ex: I have four classes class 1: 200 samples class 2: 100 samples class 3: 20 samples class 4: 10 samples so during predictions, most of my samples are predicting class 1 or 2 because samples are high even I have unlabeled samples of class3 and class4. How can I overcome this in HF? is it…

How to give equal importance of all labels while dealing with unbalanced samples

marshmellow77 January 28, 2022, 6:37am 5

Topic		Replies	Views
How to finetune the facebook/bart-large-mnli model using HF Trainer? Beginners	1	212	May 22, 2023
Weights & Biases supporting Whisper Fine-tuning :partying_face: Community Calls	4	554	December 9, 2022
NLP Training data Intermediate	0	70	March 14, 2024
10_000 samples & 10_000 labels Research	0	448	July 31, 2021
Finetuning a Tensorflow model for Multilabel classification Beginners	2	500	August 25, 2023