scenario-KD-PR-MSV-D2_data-cl-cardiff_cl_only55

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: 1.3677
• Accuracy: 0.4468
• F1: 0.4461

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: 55
• 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	1.2834	0.4591	0.4503
1.2011	2.17	500	1.3504	0.4483	0.4403
1.2011	3.26	750	1.3205	0.4468	0.4415
1.0583	4.35	1000	1.3622	0.4406	0.4355
1.0583	5.43	1250	1.3754	0.4375	0.4333
0.9917	6.52	1500	1.3816	0.4398	0.4380
0.9917	7.61	1750	1.3783	0.4475	0.4434
0.9569	8.7	2000	1.3621	0.4529	0.4497
0.9569	9.78	2250	1.3658	0.4460	0.4450
0.939	10.87	2500	1.3567	0.4630	0.4616
0.939	11.96	2750	1.3631	0.4676	0.4675
0.9262	13.04	3000	1.3922	0.4421	0.4304
0.9262	14.13	3250	1.3950	0.4275	0.4135
0.9176	15.22	3500	1.3653	0.4660	0.4658
0.9176	16.3	3750	1.3690	0.4498	0.4449
0.9114	17.39	4000	1.3964	0.4398	0.4336
0.9114	18.48	4250	1.3738	0.4614	0.4600
0.9058	19.57	4500	1.3574	0.4730	0.4728
0.9058	20.65	4750	1.3716	0.4606	0.4577
0.9018	21.74	5000	1.3684	0.4583	0.4578
0.9018	22.83	5250	1.3791	0.4568	0.4534
0.8988	23.91	5500	1.3708	0.4560	0.4549
0.8988	25.0	5750	1.3438	0.4730	0.4728
0.8963	26.09	6000	1.3544	0.4622	0.4615
0.8963	27.17	6250	1.3508	0.4622	0.4590
0.8937	28.26	6500	1.3531	0.4637	0.4633
0.8937	29.35	6750	1.3677	0.4468	0.4461

Framework versions

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