absa-roberta-model

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

• Loss: 0.1735
• Accuracy: 0.9264
• Precision: 0.9650
• Recall: 0.8497
• F1: 0.8963

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: 5e-05
• train_batch_size: 256
• eval_batch_size: 256
• 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: linear
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
0.7609	0.1667	20	0.2774	0.8618	0.5589	0.6059	0.5814
0.1981	0.3333	40	0.1765	0.9260	0.9647	0.8484	0.8954
0.1699	0.5	60	0.1754	0.9264	0.9650	0.8497	0.8963
0.1715	0.6667	80	0.1773	0.9264	0.9650	0.8497	0.8963
0.1727	0.8333	100	0.1744	0.9264	0.9650	0.8497	0.8963
0.1717	1.0	120	0.1759	0.9264	0.9650	0.8497	0.8963
0.1717	1.1667	140	0.1741	0.9264	0.9650	0.8497	0.8963
0.1649	1.3333	160	0.1732	0.9264	0.9650	0.8497	0.8963
0.174	1.5	180	0.1769	0.9264	0.9650	0.8497	0.8963
0.1472	1.6667	200	0.1740	0.9264	0.9650	0.8497	0.8963
0.1725	1.8333	220	0.1733	0.9264	0.9650	0.8497	0.8963
0.1693	2.0	240	0.1738	0.9264	0.9650	0.8497	0.8963
0.1705	2.1667	260	0.1733	0.9264	0.9650	0.8497	0.8963
0.1761	2.3333	280	0.1739	0.9264	0.9650	0.8497	0.8963
0.1608	2.5	300	0.1738	0.9264	0.9650	0.8497	0.8963
0.1593	2.6667	320	0.1739	0.9264	0.9650	0.8497	0.8963
0.1568	2.8333	340	0.1737	0.9264	0.9650	0.8497	0.8963
0.1675	3.0	360	0.1735	0.9264	0.9650	0.8497	0.8963

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.1