scenario-NON-KD-PR-COPY-CDF-CL-D2_data-cl-cardiff_cl_only_delta

This model is a fine-tuned version of xlm-roberta-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 5.4388
• Accuracy: 0.4390
• F1: 0.4380

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: 11213
• 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	1.1982	0.4576	0.4521
0.9123	2.17	500	1.4363	0.4722	0.4650
0.9123	3.26	750	1.5173	0.4715	0.4672
0.5863	4.35	1000	1.8168	0.4545	0.4539
0.5863	5.43	1250	2.0339	0.4645	0.4643
0.3146	6.52	1500	2.0526	0.4753	0.4739
0.3146	7.61	1750	2.5574	0.4560	0.4543
0.1799	8.7	2000	2.7053	0.4537	0.4542
0.1799	9.78	2250	3.2816	0.4468	0.4462
0.1166	10.87	2500	3.5971	0.4414	0.4418
0.1166	11.96	2750	3.5830	0.4491	0.4477
0.0894	13.04	3000	3.7770	0.4537	0.4537
0.0894	14.13	3250	4.0171	0.4475	0.4464
0.0625	15.22	3500	4.3230	0.4383	0.4350
0.0625	16.3	3750	4.4061	0.4421	0.4402
0.0419	17.39	4000	4.5390	0.4468	0.4460
0.0419	18.48	4250	4.7343	0.4452	0.4445
0.0328	19.57	4500	4.5586	0.4514	0.4527
0.0328	20.65	4750	4.9107	0.4437	0.4424
0.0225	21.74	5000	5.1509	0.4313	0.4276
0.0225	22.83	5250	4.8634	0.4444	0.4436
0.0209	23.91	5500	5.1513	0.4352	0.4318
0.0209	25.0	5750	5.0801	0.4552	0.4555
0.0117	26.09	6000	5.2642	0.4468	0.4444
0.0117	27.17	6250	5.3801	0.4367	0.4342
0.0092	28.26	6500	5.4445	0.4367	0.4343
0.0092	29.35	6750	5.4388	0.4390	0.4380

Framework versions

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