roberta-base-outputs

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

• Loss: 0.5836
• Accuracy: 0.6636
• F1: 0.6948
• Precision: 0.6409
• Recall: 0.7587

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: 1e-06
• train_batch_size: 8
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
0.6984	0.1778	1000	0.6931	0.5072	0.4296	0.5167	0.3677
0.6952	0.3556	2000	0.6932	0.4956	0.0032	0.6667	0.0016
0.6931	0.5333	3000	0.6922	0.5314	0.3417	0.5874	0.2409
0.6927	0.7111	4000	0.6901	0.5272	0.6625	0.5179	0.9192
0.6883	0.8889	5000	0.6792	0.5714	0.6346	0.5570	0.7373
0.6756	1.0667	6000	0.6521	0.6114	0.5702	0.6455	0.5107
0.6476	1.2444	7000	0.6317	0.627	0.6909	0.5939	0.8257
0.6278	1.4222	8000	0.6058	0.6474	0.6799	0.6276	0.7417
0.6134	1.6	9000	0.5959	0.6564	0.6909	0.6328	0.7607
0.6119	1.7778	10000	0.5870	0.6618	0.6933	0.6393	0.7571
0.6033	1.9556	11000	0.5836	0.6636	0.6948	0.6409	0.7587

Framework versions

• Transformers 4.41.1
• Pytorch 2.3.0+cu121
• Datasets 2.19.2
• Tokenizers 0.19.1