scenario-TCR-XLMV_data-en-cardiff_eng_only_alpha2

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: 3.6119
• Accuracy: 0.5472
• F1: 0.5508

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: 24
• 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	1.0411	0.4652	0.4291
No log	2.07	120	1.0685	0.5035	0.4475
No log	3.1	180	1.1001	0.5485	0.5474
No log	4.14	240	1.0647	0.5516	0.5549
No log	5.17	300	1.2458	0.5489	0.5528
No log	6.21	360	1.2913	0.5719	0.5717
No log	7.24	420	1.5986	0.5437	0.5459
No log	8.28	480	1.6908	0.5498	0.5510
0.641	9.31	540	1.7310	0.5582	0.5587
0.641	10.34	600	1.9959	0.5388	0.5394
0.641	11.38	660	2.2660	0.5357	0.5401
0.641	12.41	720	2.3724	0.5507	0.5543
0.641	13.45	780	2.5843	0.5450	0.5464
0.641	14.48	840	2.7003	0.5534	0.5556
0.641	15.52	900	2.7255	0.5459	0.5491
0.641	16.55	960	2.9127	0.5481	0.5504
0.1116	17.59	1020	2.9543	0.5432	0.5462
0.1116	18.62	1080	3.0564	0.5560	0.5591
0.1116	19.66	1140	3.1501	0.5494	0.5530
0.1116	20.69	1200	3.2882	0.5467	0.5507
0.1116	21.72	1260	3.3562	0.5459	0.5496
0.1116	22.76	1320	3.4030	0.5538	0.5573
0.1116	23.79	1380	3.4897	0.5489	0.5523
0.1116	24.83	1440	3.5540	0.5476	0.5508
0.0147	25.86	1500	3.5772	0.5498	0.5530
0.0147	26.9	1560	3.6123	0.5481	0.5515
0.0147	27.93	1620	3.5954	0.5494	0.5529
0.0147	28.97	1680	3.6081	0.5489	0.5524
0.0147	30.0	1740	3.6119	0.5472	0.5508

Framework versions

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