TL;DR - when training, a config.json is created and I would like to know how to define useful label ids, instead of the default “LABEL_0”. I’ve defined a dataset loader, which includes a label feature. I’m calling functions such as: model = ResNetForImageClassification.from_pretrained(...) train_d…

# define mappings as dictionaries id2label = {"0": "not toxic", "1": "toxic"} label2id = {"not toxic": "0", "toxic": "1"} # define config config = AutoConfig.from_pretrained(model_name, label2id=label2id, id2label=id2label) model = AutoModelForSequenceClassification.from_pretrained(model_name, conf…

How to set useful id2label and label2id in config.json using Trainer

Beginners

xshubhamx April 7, 2024, 4:54pm 2

I am facing the same issue. Pleasse let me know if you find a solution

How to separate predictions map to create id2label.json

Topic		Replies	Views
Problems in deployment when I configure my own labels Amazon SageMaker	6	2634	July 12, 2023
Get label to id / id to label mapping Beginners	2	3342	November 8, 2021
Column names of custom dataset for use with trainer Beginners	3	5229	March 31, 2024
Creating Trainer object is deleting my 'labels' feature Beginners	3	1430	January 21, 2021
Popping `inputs[labels]` when self.label_smoother is not None (in trainer.py) Beginners	2	1237	November 11, 2021

How to set useful id2label and label2id in config.json using Trainer

Related topics