pretrain_model

This model is a fine-tuned version of roberta-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.6196
• Precision: 0.6607
• Recall: 0.6589
• F1: 0.6598
• Accuracy: 0.6575

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: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.6965	0.1377	500	0.6910	0.526	1.0	0.6894	0.526
0.6963	0.2755	1000	0.6921	0.526	1.0	0.6894	0.526
0.6957	0.4132	1500	0.6666	0.6154	0.7300	0.6678	0.618
0.6914	0.5510	2000	0.6834	0.7069	0.4677	0.5629	0.618
0.6768	0.6887	2500	0.6838	0.6412	0.6388	0.64	0.622
0.6786	0.8264	3000	0.6539	0.7273	0.4259	0.5372	0.614
0.663	0.9642	3500	0.6743	0.6560	0.5437	0.5946	0.61
0.6564	1.1019	4000	0.6381	0.6763	0.6198	0.6468	0.644
0.6468	1.2397	4500	0.6010	0.6613	0.7871	0.7188	0.676
0.6275	1.3774	5000	0.6103	0.7246	0.5703	0.6383	0.66
0.6275	1.5152	5500	0.6018	0.7311	0.5894	0.6526	0.67
0.6141	1.6529	6000	0.5947	0.7269	0.6578	0.6906	0.69
0.617	1.7906	6500	0.5872	0.7165	0.6920	0.7041	0.694
0.6059	1.9284	7000	0.5816	0.7227	0.7034	0.7129	0.702

Framework versions

• Transformers 4.46.3
• Pytorch 2.5.1+cu121
• Datasets 3.1.0
• Tokenizers 0.20.3