scenario-TCR_data-en-cardiff_eng_only1

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: 3.8346
• Accuracy: 0.5551
• F1: 0.5565

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: 777
• 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	1.1084	0.3638	0.2340
No log	2.07	120	1.0633	0.5181	0.4969
No log	3.1	180	1.0162	0.5106	0.5135
No log	4.14	240	1.0900	0.5573	0.5521
No log	5.17	300	1.2575	0.5644	0.5581
No log	6.21	360	1.4901	0.5489	0.5486
No log	7.24	420	1.7893	0.5560	0.5570
No log	8.28	480	1.8097	0.5767	0.5786
0.6774	9.31	540	1.9894	0.5679	0.5669
0.6774	10.34	600	2.1280	0.5445	0.5443
0.6774	11.38	660	2.4589	0.5653	0.5661
0.6774	12.41	720	2.8356	0.5370	0.5374
0.6774	13.45	780	3.0624	0.5414	0.5370
0.6774	14.48	840	3.1735	0.5410	0.5427
0.6774	15.52	900	3.1008	0.5481	0.5491
0.6774	16.55	960	3.2542	0.5388	0.5411
0.0909	17.59	1020	3.4400	0.5401	0.5415
0.0909	18.62	1080	3.5071	0.5467	0.5491
0.0909	19.66	1140	3.3959	0.5392	0.5375
0.0909	20.69	1200	3.5250	0.5507	0.5531
0.0909	21.72	1260	3.5264	0.5569	0.5588
0.0909	22.76	1320	3.6723	0.5516	0.5529
0.0909	23.79	1380	3.6306	0.5595	0.5614
0.0909	24.83	1440	3.7656	0.5476	0.5480
0.0131	25.86	1500	3.7895	0.5578	0.5594
0.0131	26.9	1560	3.7447	0.5591	0.5601
0.0131	27.93	1620	3.8071	0.5551	0.5572
0.0131	28.97	1680	3.8322	0.5556	0.5572
0.0131	30.0	1740	3.8346	0.5551	0.5565

Framework versions

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