scenario-NON-KD-PO-COPY-CDF-CL-D2_data-cl-cardiff_cl_only66

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: 5.9979
• Accuracy: 0.4498
• F1: 0.4497

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: 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	1.0870	250	1.2493	0.4568	0.4501
0.812	2.1739	500	1.5854	0.4637	0.4628
0.812	3.2609	750	1.8772	0.4614	0.4605
0.4271	4.3478	1000	2.3694	0.4414	0.4339
0.4271	5.4348	1250	2.6689	0.4537	0.4473
0.1992	6.5217	1500	3.0050	0.4537	0.4527
0.1992	7.6087	1750	3.1201	0.4468	0.4406
0.1147	8.6957	2000	3.9025	0.4360	0.4298
0.1147	9.7826	2250	4.0949	0.4390	0.4331
0.0816	10.8696	2500	4.3006	0.4306	0.4218
0.0816	11.9565	2750	4.5881	0.4606	0.4569
0.0558	13.0435	3000	4.4255	0.4576	0.4577
0.0558	14.1304	3250	5.1150	0.4606	0.4600
0.0388	15.2174	3500	4.6378	0.4568	0.4571
0.0388	16.3043	3750	5.2331	0.4498	0.4458
0.0269	17.3913	4000	5.3200	0.4491	0.4481
0.0269	18.4783	4250	5.2543	0.4599	0.4583
0.0175	19.5652	4500	5.3747	0.4552	0.4548
0.0175	20.6522	4750	5.4521	0.4460	0.4448
0.0181	21.7391	5000	5.3489	0.4606	0.4604
0.0181	22.8261	5250	5.8017	0.4552	0.4543
0.0093	23.9130	5500	5.6669	0.4560	0.4560
0.0093	25.0	5750	5.5959	0.4529	0.4517
0.0076	26.0870	6000	5.8141	0.4576	0.4554
0.0076	27.1739	6250	5.8656	0.4560	0.4556
0.006	28.2609	6500	5.9365	0.4583	0.4577
0.006	29.3478	6750	5.9979	0.4498	0.4497

Framework versions

• Transformers 4.44.2
• Pytorch 2.1.1+cu121
• Datasets 2.14.5
• Tokenizers 0.19.1