Modelpublic

sarasarasara/parsBERT_text_emotion_classification-finetuned-BMD-20241201_103841

sourceHugging Facemitupdated 2y agoView on Hugging Face

0likes5downloads

Model Card

parsBERTtextemotionclassification-finetuned-BMD-20241201103841

This model is a fine-tuned version of Nhoseinnia/parsBERT_text_emotion_classification on an unknown dataset. It achieves the following results on the evaluation set:

—Loss: 3.1468
—Accuracy: 0.4706
—F1: 0.4671

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: 32
—evalbatchsize: 32
—seed: 42
—optimizer: Use adamwtorch with betas=(0.9,0.999) and epsilon=1e-08 and optimizerargs=No additional optimizer arguments
—lrschedulertype: linear
—lrschedulerwarmup_ratio: 0.1
—num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
0.1509	1.0	4	1.2847	0.5	0.5038
0.1209	2.0	8	1.2551	0.5882	0.5903
0.1129	3.0	12	1.4353	0.5294	0.5297
0.0882	4.0	16	1.9818	0.4412	0.4457
0.0856	5.0	20	1.7055	0.5294	0.5190
0.0337	6.0	24	1.9646	0.4706	0.4799
0.0292	7.0	28	2.3365	0.4412	0.4442
0.0254	8.0	32	2.1677	0.3824	0.3819
0.0127	9.0	36	2.5429	0.4118	0.4147
0.011	10.0	40	2.6179	0.4412	0.4466
0.0084	11.0	44	2.5994	0.4412	0.4455
0.0058	12.0	48	2.6673	0.5	0.5029
0.0089	13.0	52	2.6855	0.5	0.5040
0.0046	14.0	56	2.9154	0.4412	0.4416
0.0037	15.0	60	3.0083	0.4412	0.4416
0.0035	16.0	64	2.9644	0.4412	0.4416
0.0037	17.0	68	2.8735	0.4706	0.4671
0.0031	18.0	72	2.7300	0.5	0.4976
0.0028	19.0	76	2.8202	0.5	0.4976
0.0058	20.0	80	3.1716	0.4412	0.4471
0.0023	21.0	84	3.3853	0.4412	0.4471
0.0024	22.0	88	3.3990	0.4412	0.4416
0.0023	23.0	92	3.4125	0.4412	0.4416
0.0025	24.0	96	3.3417	0.4412	0.4471
0.0021	25.0	100	3.2096	0.4706	0.4671
0.002	26.0	104	3.1596	0.4706	0.4671
0.002	27.0	108	3.1351	0.4706	0.4671
0.002	28.0	112	3.1332	0.4706	0.4671
0.0022	29.0	116	3.1421	0.4706	0.4671
0.0021	30.0	120	3.1468	0.4706	0.4671

Framework versions

—Transformers 4.46.3
—Pytorch 2.5.1+cu121
—Datasets 3.1.0
—Tokenizers 0.20.3