scenario-KD-PO-CDF-CL-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: 33.2615
• Accuracy: 0.4306
• F1: 0.4291

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	21.9865	0.4159	0.4018
23.1773	2.17	500	20.3976	0.4468	0.4402
23.1773	3.26	750	24.2630	0.4398	0.4284
13.2399	4.35	1000	26.2345	0.4375	0.4366
13.2399	5.43	1250	28.2139	0.4174	0.4046
7.5118	6.52	1500	27.0005	0.4406	0.4383
7.5118	7.61	1750	33.6835	0.4329	0.4221
4.7676	8.7	2000	31.6054	0.4213	0.4195
4.7676	9.78	2250	33.1980	0.4290	0.4281
3.4176	10.87	2500	33.2081	0.4174	0.4078
3.4176	11.96	2750	33.2910	0.4267	0.4216
2.4248	13.04	3000	32.5098	0.4329	0.4298
2.4248	14.13	3250	33.3356	0.4290	0.4284
2.0388	15.22	3500	34.5635	0.4290	0.4234
2.0388	16.3	3750	35.3404	0.4298	0.4285
1.6264	17.39	4000	34.5365	0.4182	0.4091
1.6264	18.48	4250	33.6755	0.4306	0.4269
1.3659	19.57	4500	33.4139	0.4236	0.4143
1.3659	20.65	4750	31.2188	0.4375	0.4377
1.0643	21.74	5000	32.8825	0.4352	0.4324
1.0643	22.83	5250	33.1957	0.4174	0.4154
0.8897	23.91	5500	32.9339	0.4367	0.4362
0.8897	25.0	5750	32.8236	0.4267	0.4257
0.7473	26.09	6000	33.7690	0.4244	0.4194
0.7473	27.17	6250	34.4919	0.4167	0.4166
0.6047	28.26	6500	32.1175	0.4313	0.4295
0.6047	29.35	6750	33.2615	0.4306	0.4291

Framework versions

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