20230928-1-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: 0.4663
• Loss: 2.7549

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
4.5242	0.46	200	0.3115	3.6451
4.0841	0.91	400	0.3325	3.4312
3.7774	1.37	600	0.3357	nan
3.7354	1.82	800	0.38	nan
3.586	2.28	1000	0.3763	3.3689
3.51	2.73	1200	0.4209	nan
3.4578	3.19	1400	0.4132	3.0195
3.2155	3.64	1600	0.4455	3.2697
3.1555	4.1	1800	0.4660	nan
3.0647	4.56	2000	0.4244	2.6126
3.1005	5.01	2200	0.4344	nan
3.0598	5.47	2400	0.4928	2.5990
3.032	5.92	2600	0.5036	2.6536
2.9123	6.38	2800	0.5025	2.7367
3.0423	6.83	3000	0.4803	2.8070
2.7696	7.29	3200	0.4973	nan
2.7321	7.74	3400	0.5105	2.8622
2.8714	8.2	3600	0.4936	2.5544
2.8736	8.66	3800	0.5274	2.5878
2.779	9.11	4000	0.5219	nan
2.6934	9.57	4200	0.4663	2.7549

Framework versions

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