roberta-Reflections-goodareas-sweeps-current

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

• Loss: 0.1937
• Accuracy: 0.8562
• Precision: 0.3984
• Recall: 0.5632
• F1: 0.4667

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: 3.693911058164899e-06
• train_batch_size: 32
• eval_batch_size: 16
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
0.3925	1.0	52	0.1759	0.8883	0.0	0.0	0.0
0.3241	2.0	104	0.1606	0.8883	0.0	0.0	0.0
0.2914	3.0	156	0.1744	0.8883	0.0	0.0	0.0
0.2821	4.0	208	0.2609	0.8909	0.75	0.0345	0.0659
0.2739	5.0	260	0.1763	0.8935	0.75	0.0690	0.1263
0.2533	6.0	312	0.1390	0.8922	0.6154	0.0920	0.16
0.2482	7.0	364	0.2199	0.8755	0.4490	0.5057	0.4757
0.2362	8.0	416	0.2124	0.8652	0.4286	0.6207	0.5070
0.2375	9.0	468	0.1351	0.8973	0.5614	0.3678	0.4444
0.228	10.0	520	0.1650	0.8870	0.4945	0.5172	0.5056
0.2212	11.0	572	0.1771	0.8845	0.4851	0.5632	0.5213
0.2217	12.0	624	0.1756	0.8832	0.4792	0.5287	0.5027
0.2109	13.0	676	0.1942	0.8614	0.4118	0.5632	0.4757
0.2018	14.0	728	0.1795	0.8678	0.4298	0.5632	0.4876
0.2013	15.0	780	0.1817	0.8652	0.4211	0.5517	0.4776
0.1943	16.0	832	0.2071	0.8575	0.4077	0.6092	0.4885
0.2023	17.0	884	0.2143	0.8498	0.3897	0.6092	0.4753
0.1924	18.0	936	0.1966	0.8562	0.4031	0.5977	0.4815
0.183	19.0	988	0.1914	0.8614	0.4118	0.5632	0.4757
0.191	20.0	1040	0.1937	0.8562	0.3984	0.5632	0.4667

Framework versions

• Transformers 4.48.3
• Pytorch 2.5.1+cu124
• Datasets 2.21.0
• Tokenizers 0.21.0