scenario-KD-PO-CDF-CL-D2_data-cl-cardiff_cl_only_beta-jason

This model is a fine-tuned version of haryoaw/scenario-TCR_data-cl-cardiff_cl_only2 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 16.2119
• Accuracy: 0.3974
• F1: 0.3962

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: 6666
• 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	11.7960	0.3650	0.3427
13.5624	2.17	500	11.1810	0.3927	0.3922
13.5624	3.26	750	10.7353	0.3889	0.3725
11.0045	4.35	1000	11.0340	0.4113	0.4092
11.0045	5.43	1250	11.1511	0.3958	0.3764
9.4397	6.52	1500	11.3893	0.4128	0.4098
9.4397	7.61	1750	11.7867	0.4174	0.4136
8.0563	8.7	2000	12.7215	0.4020	0.3934
8.0563	9.78	2250	13.1991	0.4159	0.4158
7.071	10.87	2500	13.4791	0.3966	0.3938
7.071	11.96	2750	12.9321	0.4005	0.3939
5.9842	13.04	3000	13.5185	0.3873	0.3775
5.9842	14.13	3250	14.5623	0.4028	0.3984
5.0429	15.22	3500	14.7614	0.4012	0.3954
5.0429	16.3	3750	13.8500	0.4151	0.4149
4.4814	17.39	4000	14.0842	0.4051	0.4022
4.4814	18.48	4250	14.4055	0.3904	0.3804
4.0867	19.57	4500	15.5442	0.3858	0.3809
4.0867	20.65	4750	14.6236	0.3966	0.3951
3.5928	21.74	5000	14.9268	0.4005	0.3920
3.5928	22.83	5250	15.2065	0.3897	0.3884
3.2444	23.91	5500	16.5178	0.3889	0.3860
3.2444	25.0	5750	15.1592	0.3920	0.3879
3.0552	26.09	6000	15.4594	0.3974	0.3952
3.0552	27.17	6250	15.8492	0.3843	0.3842
2.8654	28.26	6500	15.8468	0.3827	0.3806
2.8654	29.35	6750	16.2119	0.3974	0.3962

Framework versions

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