scenario-MDBT-TCR_data-cl-cardiff_cl_only

This model is a fine-tuned version of microsoft/mdeberta-v3-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 4.7992
• Accuracy: 0.5154
• F1: 0.5146

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: 5e-05
• train_batch_size: 32
• eval_batch_size: 64
• seed: 66
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
No log	1.09	250	1.2801	0.5069	0.5029
0.7278	2.17	500	1.6260	0.5077	0.5080
0.7278	3.26	750	1.6500	0.5193	0.5209
0.3512	4.35	1000	2.1813	0.5123	0.5144
0.3512	5.43	1250	2.5133	0.5154	0.5167
0.1838	6.52	1500	2.6502	0.5093	0.5093
0.1838	7.61	1750	3.0408	0.5015	0.5021
0.118	8.7	2000	3.3486	0.4877	0.4822
0.118	9.78	2250	3.5117	0.4923	0.4906
0.072	10.87	2500	3.5966	0.5046	0.5027
0.072	11.96	2750	3.3823	0.5100	0.5121
0.0545	13.04	3000	3.7627	0.5085	0.5053
0.0545	14.13	3250	3.9342	0.5108	0.5124
0.0336	15.22	3500	4.2215	0.5093	0.5061
0.0336	16.3	3750	4.2219	0.5046	0.5021
0.0272	17.39	4000	4.0061	0.5208	0.5227
0.0272	18.48	4250	4.3214	0.5116	0.5074
0.0198	19.57	4500	4.5333	0.5093	0.5075
0.0198	20.65	4750	4.3535	0.5247	0.5256
0.0161	21.74	5000	4.5169	0.5239	0.5238
0.0161	22.83	5250	4.4982	0.5285	0.5298
0.012	23.91	5500	4.5591	0.5170	0.5186
0.012	25.0	5750	4.7615	0.5085	0.5069
0.0066	26.09	6000	4.8457	0.5100	0.5079
0.0066	27.17	6250	4.7872	0.5131	0.5118
0.0069	28.26	6500	4.6257	0.5301	0.5303
0.0069	29.35	6750	4.7992	0.5154	0.5146

Framework versions

• Transformers 4.33.3
• Pytorch 2.1.1+cu121
• Datasets 2.14.5
• Tokenizers 0.13.3