scenario-TCR-XLMV_data-en-cardiff_eng_only_beta

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: 3.4054
• Accuracy: 0.5467
• F1: 0.5510

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: 112233
• 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	1.0887	0.4449	0.3540
No log	2.07	120	1.0211	0.4700	0.3777
No log	3.1	180	1.0598	0.5141	0.4790
No log	4.14	240	1.0131	0.5644	0.5652
No log	5.17	300	1.1073	0.5586	0.5595
No log	6.21	360	1.3697	0.5635	0.5542
No log	7.24	420	1.4910	0.5379	0.5385
No log	8.28	480	1.7325	0.5507	0.5542
0.6649	9.31	540	1.8878	0.5489	0.5505
0.6649	10.34	600	2.2758	0.5309	0.5320
0.6649	11.38	660	2.3053	0.5357	0.5357
0.6649	12.41	720	2.3674	0.5542	0.5574
0.6649	13.45	780	2.7705	0.5309	0.5332
0.6649	14.48	840	2.7515	0.5520	0.5522
0.6649	15.52	900	2.9868	0.5423	0.5447
0.6649	16.55	960	2.7489	0.5582	0.5597
0.1079	17.59	1020	2.8748	0.5525	0.5560
0.1079	18.62	1080	3.0165	0.5467	0.5511
0.1079	19.66	1140	3.2954	0.5340	0.5356
0.1079	20.69	1200	3.1051	0.5441	0.5488
0.1079	21.72	1260	3.2199	0.5441	0.5467
0.1079	22.76	1320	3.1660	0.5454	0.5500
0.1079	23.79	1380	3.2637	0.5445	0.5474
0.1079	24.83	1440	3.2934	0.5538	0.5576
0.0279	25.86	1500	3.2834	0.5476	0.5506
0.0279	26.9	1560	3.3734	0.5467	0.5507
0.0279	27.93	1620	3.4145	0.5437	0.5476
0.0279	28.97	1680	3.4043	0.5454	0.5496
0.0279	30.0	1740	3.4054	0.5467	0.5510

Framework versions

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