Modelpublic

AnonymousCS/populism_classifier_326

sourceHugging Faceapache-2.0updated 1y agoView on Hugging Face

0likes14downloads

Model Card

populismclassifier326

This model is a fine-tuned version of AnonymousCS/populism_english_bert_large_cased on the None dataset. It achieves the following results on the evaluation set:

—Loss: 0.2419
—Accuracy: 0.9929
—1-f1: 0.8727
—1-recall: 0.8421
—1-precision: 0.9057
—Balanced Acc: 0.9197

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: 1e-05
—trainbatchsize: 64
—evalbatchsize: 64
—seed: 42
—optimizer: Use adamwtorch with betas=(0.9,0.999) and epsilon=1e-08 and optimizerargs=No additional optimizer arguments
—lrschedulertype: linear
—num_epochs: 20
—mixedprecisiontraining: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	1-f1	1-recall	1-precision	Balanced Acc
0.1833	1.0	124	0.1742	0.9590	0.5091	0.7368	0.3889	0.8512
0.1031	2.0	248	0.1743	0.9833	0.7130	0.7193	0.7069	0.8552
0.0298	3.0	372	0.1666	0.9848	0.7458	0.7719	0.7213	0.8815
0.0065	4.0	496	0.2741	0.9889	0.78	0.6842	0.9070	0.8411
0.0008	5.0	620	0.1625	0.9899	0.8246	0.8246	0.8246	0.9097
0.0109	6.0	744	0.2292	0.9924	0.8649	0.8421	0.8889	0.9195
0.0002	7.0	868	0.2173	0.9904	0.8348	0.8421	0.8276	0.9184
0.0007	8.0	992	0.2419	0.9929	0.8727	0.8421	0.9057	0.9197

Framework versions

—Transformers 4.46.3
—Pytorch 2.4.1+cu121
—Datasets 3.1.0
—Tokenizers 0.20.3