xlm-roberta-base-finetuned-partypredictor

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: 1.6783
• Accuracy: 0.2495

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: 2e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 16

Training results

Training Loss	Epoch	Step	Accuracy	Validation Loss
1.7766	0.76	5000	0.1331	1.8909
1.7572	1.52	10000	0.1331	1.7809
1.7543	2.28	15000	0.1031	1.8126
1.7273	3.05	20000	0.1331	1.8048
1.7435	3.81	25000	0.2675	1.7892
1.7606	4.99	30000	0.3121	1.7848
1.7546	5.82	35000	0.3121	1.7737
1.7417	6.65	40000	0.3121	1.7699
1.7007	7.48	45000	0.1529	1.7088
1.7542	7.87	50000	0.1331	1.8058
1.75	8.66	55000	0.1331	1.8347
1.7505	10.05	60000	1.8079	0.1231
1.7545	10.88	65000	1.7756	0.3121
1.7322	11.72	70000	1.7371	0.2707
1.7082	12.56	75000	1.6886	0.2419
1.7035	13.4	80000	1.6844	0.2638
1.6889	14.23	85000	1.6728	0.2525
1.6779	15.07	90000	1.6737	0.2490
1.6821	15.91	95000	1.6783	0.2495

Framework versions

• Transformers 4.26.0
• Pytorch 1.13.1
• Datasets 2.9.0
• Tokenizers 0.13.2