20230928-5-xlm-roberta-base-new

This model is a fine-tuned version of xlm-roberta-base on an unknown dataset. It achieves the following results on the evaluation set:

• Accuracy: 0.5236
• Loss: nan

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: 10

Training results

Training Loss	Epoch	Step	Accuracy	Validation Loss
4.5001	0.46	200	0.2965	nan
4.0546	0.91	400	0.3261	3.5917
3.7694	1.37	600	0.3403	nan
3.6655	1.82	800	0.2936	nan
3.5615	2.28	1000	0.3601	3.3849
3.3854	2.73	1200	0.4212	3.2257
3.321	3.19	1400	0.4053	nan
3.3052	3.64	1600	0.4615	nan
3.1626	4.1	1800	0.4169	2.8506
3.1598	4.56	2000	0.4441	2.8069
3.0887	5.01	2200	0.4215	3.0388
3.0471	5.47	2400	0.4562	2.8999
3.0477	5.92	2600	0.4587	2.8482
2.6846	6.38	2800	0.4778	2.6451
2.9033	6.83	3000	0.4688	2.5056
2.9835	7.29	3200	0.4794	2.7805
2.8992	7.74	3400	0.5	2.7021
2.8387	8.2	3600	0.5176	nan
2.8935	8.66	3800	0.4929	nan
2.7382	9.11	4000	0.4913	nan
2.8238	9.57	4200	0.5236	nan

Framework versions

• Transformers 4.33.3
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.13.3