xlm-roberta-text-cls-peft-prompt-tuning

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

• Loss: 0.7050
• Accuracy: 0.7294

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: 0.0001
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10.0

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.7278	0.9992	594	1.2574	0.5517
1.3236	2.0	1189	0.9894	0.6393
1.1827	2.9992	1783	0.8649	0.6897
1.1147	4.0	2378	0.8593	0.6897
1.0552	4.9992	2972	0.7745	0.6976
1.0143	6.0	3567	0.7181	0.7135
0.9872	6.9992	4161	0.7037	0.7507
1.002	8.0	4756	0.7111	0.7347
0.9816	8.9992	5350	0.6931	0.7241
0.9602	9.9916	5940	0.7050	0.7294

Framework versions

• PEFT 0.11.1
• Transformers 4.41.2
• Pytorch 2.1.0+cu121
• Datasets 2.19.2
• Tokenizers 0.19.1