prox-doc-xlm-roberta-base

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: 0.3884
• Accuracy: 0.8212

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: 64
• eval_batch_size: 128
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.05
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	0	0	0.6939	0.4947
0.4562	0.1002	251	0.4377	0.7956
0.369	0.2003	502	0.4421	0.8068
0.4306	0.3005	753	0.4042	0.8123
0.384	0.4006	1004	0.4016	0.8180
0.4115	0.5008	1255	0.3996	0.8177
0.4205	0.6010	1506	0.4070	0.8079
0.39	0.7011	1757	0.3883	0.8227
0.3902	0.8013	2008	0.3951	0.8179
0.369	0.9014	2259	0.3884	0.8212

Framework versions

• Transformers 4.51.3
• Pytorch 2.4.1+cu124
• Datasets 3.5.0
• Tokenizers 0.21.1