scenario-KD-PR-MSV-D2_data-cl-cardiff_cl_only44

This model is a fine-tuned version of haryoaw/scenario-MDBT-TCR_data-cl-cardiff_cl_only on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.3590
• Accuracy: 0.4630
• F1: 0.4628

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: 32
• seed: 44
• 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	1.3195	0.4390	0.4186
1.1991	2.17	500	1.2866	0.4537	0.4542
1.1991	3.26	750	1.3622	0.4383	0.4293
1.0569	4.35	1000	1.3662	0.4252	0.4155
1.0569	5.43	1250	1.3920	0.4275	0.4229
0.9882	6.52	1500	1.3607	0.4560	0.4549
0.9882	7.61	1750	1.3742	0.4468	0.4429
0.9542	8.7	2000	1.4048	0.4329	0.4284
0.9542	9.78	2250	1.3944	0.4344	0.4227
0.9374	10.87	2500	1.3725	0.4414	0.4417
0.9374	11.96	2750	1.3790	0.4491	0.4464
0.9261	13.04	3000	1.3687	0.4552	0.4510
0.9261	14.13	3250	1.3778	0.4475	0.4438
0.9164	15.22	3500	1.3888	0.4583	0.4554
0.9164	16.3	3750	1.3785	0.4498	0.4415
0.9101	17.39	4000	1.3778	0.4568	0.4530
0.9101	18.48	4250	1.4075	0.4398	0.4261
0.9053	19.57	4500	1.3610	0.4576	0.4575
0.9053	20.65	4750	1.3679	0.4568	0.4570
0.9026	21.74	5000	1.3653	0.4614	0.4617
0.9026	22.83	5250	1.3701	0.4468	0.4465
0.8977	23.91	5500	1.3753	0.4498	0.4475
0.8977	25.0	5750	1.3781	0.4545	0.4532
0.8953	26.09	6000	1.3811	0.4468	0.4455
0.8953	27.17	6250	1.3609	0.4545	0.4527
0.893	28.26	6500	1.3696	0.4452	0.4433
0.893	29.35	6750	1.3590	0.4630	0.4628

Framework versions

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