scenario-KD-PR-CDF-EN-FROM-CL-D2_data-en-cardiff_eng_only_gamma-jason

This model is a fine-tuned version of FacebookAI/xlm-roberta-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 16.3831
• Accuracy: 0.3955
• F1: 0.3889

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: 88458
• 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	12.6768	0.3280	0.2299
No log	3.45	200	12.6987	0.3836	0.3671
No log	5.17	300	12.8320	0.3858	0.3602
No log	6.9	400	13.1201	0.3620	0.3230
13.3889	8.62	500	13.0591	0.3920	0.3845
13.3889	10.34	600	14.1639	0.4004	0.3832
13.3889	12.07	700	14.7159	0.3810	0.3536
13.3889	13.79	800	14.3341	0.3907	0.3815
13.3889	15.52	900	14.9958	0.3792	0.3564
9.5405	17.24	1000	14.4299	0.3902	0.3887
9.5405	18.97	1100	15.3658	0.3999	0.3973
9.5405	20.69	1200	15.1707	0.3990	0.3940
9.5405	22.41	1300	15.8556	0.3955	0.3816
9.5405	24.14	1400	15.7963	0.3915	0.3825
6.9987	25.86	1500	15.8559	0.4056	0.4012
6.9987	27.59	1600	16.1802	0.3854	0.3757
6.9987	29.31	1700	16.3831	0.3955	0.3889

Framework versions

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