scenario-TCR-XLMV_data-en-cardiff_eng_only_gamma2

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

• Loss: 3.4094
• Accuracy: 0.5516
• F1: 0.5553

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: 77
• 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	1.03	60	1.0398	0.4828	0.3902
No log	2.07	120	1.1798	0.4489	0.3679
No log	3.1	180	1.0463	0.4868	0.4351
No log	4.14	240	1.0244	0.5622	0.5553
No log	5.17	300	1.0819	0.5595	0.5478
No log	6.21	360	1.4170	0.5410	0.5407
No log	7.24	420	1.4249	0.5617	0.5653
No log	8.28	480	1.6285	0.5626	0.5627
0.6824	9.31	540	1.8719	0.5494	0.5516
0.6824	10.34	600	1.9037	0.5547	0.5574
0.6824	11.38	660	1.7645	0.5494	0.5516
0.6824	12.41	720	2.0301	0.5437	0.5459
0.6824	13.45	780	2.6619	0.5317	0.5330
0.6824	14.48	840	2.5606	0.5498	0.5520
0.6824	15.52	900	2.9065	0.5326	0.5347
0.6824	16.55	960	2.6860	0.5564	0.5597
0.132	17.59	1020	2.9277	0.5476	0.5495
0.132	18.62	1080	3.1905	0.5441	0.5472
0.132	19.66	1140	2.9974	0.5410	0.5446
0.132	20.69	1200	2.8902	0.5556	0.5575
0.132	21.72	1260	3.2156	0.5401	0.5432
0.132	22.76	1320	3.2772	0.5472	0.5501
0.132	23.79	1380	3.2211	0.5551	0.5569
0.132	24.83	1440	3.3844	0.5423	0.5450
0.0295	25.86	1500	3.3534	0.5494	0.5531
0.0295	26.9	1560	3.4030	0.5498	0.5534
0.0295	27.93	1620	3.4206	0.5511	0.5547
0.0295	28.97	1680	3.4273	0.5529	0.5565
0.0295	30.0	1740	3.4094	0.5516	0.5553

Framework versions

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