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: 5.9844
• Accuracy: 0.4483
• F1: 0.4469

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	1.1475	0.4552	0.4538
0.861	2.1739	500	1.5478	0.4699	0.4682
0.861	3.2609	750	1.9490	0.4599	0.4584
0.4638	4.3478	1000	2.3832	0.4537	0.4532
0.4638	5.4348	1250	2.4965	0.4568	0.4562
0.2113	6.5217	1500	3.4612	0.4506	0.4511
0.2113	7.6087	1750	3.5718	0.4660	0.4645
0.1221	8.6957	2000	3.8516	0.4367	0.4309
0.1221	9.7826	2250	3.9459	0.4421	0.4349
0.0796	10.8696	2500	4.2407	0.4591	0.4590
0.0796	11.9565	2750	4.5609	0.4498	0.4446
0.06	13.0435	3000	4.4842	0.4537	0.4500
0.06	14.1304	3250	4.6808	0.4529	0.4497
0.0356	15.2174	3500	5.2708	0.4290	0.4221
0.0356	16.3043	3750	4.7855	0.4383	0.4380
0.0282	17.3913	4000	5.2598	0.4645	0.4644
0.0282	18.4783	4250	5.4851	0.4691	0.4695
0.0231	19.5652	4500	5.8382	0.4406	0.4359
0.0231	20.6522	4750	5.4555	0.4514	0.4501
0.0179	21.7391	5000	5.5153	0.4452	0.4435
0.0179	22.8261	5250	6.0191	0.4421	0.4389
0.0108	23.9130	5500	5.7614	0.4514	0.4490
0.0108	25.0	5750	5.9259	0.4522	0.4496
0.0087	26.0870	6000	5.9545	0.4452	0.4438
0.0087	27.1739	6250	5.9292	0.4491	0.4469
0.0082	28.2609	6500	5.9245	0.4630	0.4614
0.0082	29.3478	6750	5.9844	0.4483	0.4469

Framework versions

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