scenario-NON-KD-PR-COPY-CDF-EN-D2_data-en-cardiff_eng_only_gamma

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: 4.3453
• Accuracy: 0.4877
• F1: 0.4869

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: 11423
• 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	1.1420	0.4696	0.4490
No log	3.45	200	1.4019	0.4894	0.4836
No log	5.17	300	1.7536	0.4674	0.4550
No log	6.9	400	2.3793	0.4757	0.4684
0.5065	8.62	500	2.6051	0.4877	0.4878
0.5065	10.34	600	3.0214	0.4775	0.4751
0.5065	12.07	700	3.1854	0.5066	0.5058
0.5065	13.79	800	3.6398	0.4824	0.4765
0.5065	15.52	900	3.7498	0.4890	0.4838
0.0693	17.24	1000	4.0213	0.4683	0.4595
0.0693	18.97	1100	3.8916	0.4943	0.4931
0.0693	20.69	1200	4.0797	0.4788	0.4755
0.0693	22.41	1300	4.0526	0.4974	0.4978
0.0693	24.14	1400	4.3302	0.4850	0.4797
0.015	25.86	1500	4.2868	0.4938	0.4939
0.015	27.59	1600	4.3197	0.4907	0.4907
0.015	29.31	1700	4.3453	0.4877	0.4869

Framework versions

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