scenario-TCR-4_data-en-cardiff_eng_only

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: 3.9676
• Accuracy: 0.5644
• F1: 0.5665

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: 34
• 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	1.72	100	1.1136	0.5428	0.5446
No log	3.45	200	1.2209	0.5556	0.5573
No log	5.17	300	1.5985	0.5692	0.5659
No log	6.9	400	2.1122	0.5503	0.5510
0.4646	8.62	500	2.3858	0.5613	0.5632
0.4646	10.34	600	2.4789	0.5679	0.5688
0.4646	12.07	700	3.0361	0.5335	0.5314
0.4646	13.79	800	3.1203	0.5697	0.5708
0.4646	15.52	900	3.3858	0.5520	0.5544
0.0601	17.24	1000	3.5188	0.5586	0.5615
0.0601	18.97	1100	3.7180	0.5600	0.5626
0.0601	20.69	1200	3.8266	0.5578	0.5607
0.0601	22.41	1300	4.0097	0.5489	0.5517
0.0601	24.14	1400	3.9363	0.5569	0.5599
0.006	25.86	1500	3.9628	0.5569	0.5597
0.006	27.59	1600	4.0212	0.5582	0.5613
0.006	29.31	1700	3.9676	0.5644	0.5665

Framework versions

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