Modelpublic

annahaz/xlm-roberta-base-misogyny-sexism-decay0.05-fr-indomain

sourceHugging Facemitupdated 4y agoView on Hugging Face

0likes13downloads

Model Card

xlm-roberta-base-misogyny-sexism-decay0.05-fr-indomain

This model is a fine-tuned version of xlm-roberta-base on the None dataset. It achieves the following results on the evaluation set:

—Loss: 1.2200
—Accuracy: 0.8708
—F1: 0.0040
—Precision: 0.1
—Recall: 0.0021
—Mae: 0.1292

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

—learning_rate: 2e-05
—trainbatchsize: 16
—evalbatchsize: 16
—seed: 42
—optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
—lrschedulertype: linear
—num_epochs: 7

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall	Mae
0.2085	1.0	2302	1.0000	0.8674	0.0212	0.1719	0.0113	0.1326
0.18	2.0	4604	1.0566	0.8527	0.0157	0.0520	0.0093	0.1473
0.1614	3.0	6906	1.1284	0.8673	0.0020	0.0222	0.0010	0.1327
0.1428	4.0	9208	1.1329	0.8714	0.0020	0.0714	0.0010	0.1286
0.1467	5.0	11510	1.1814	0.8708	0.0040	0.1	0.0021	0.1292
0.1375	6.0	13812	1.2020	0.8706	0.0020	0.05	0.0010	0.1294
0.1093	7.0	16114	1.2200	0.8708	0.0040	0.1	0.0021	0.1292

Framework versions

—Transformers 4.20.1
—Pytorch 1.12.0+cu102
—Datasets 2.3.2
—Tokenizers 0.12.1