green_as_train_context_roberta-large_20e

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

• Loss: 1.4371
• Val Accuracy: 0.8913
• Val Precision: 0.7554
• Val Recall: 0.5910
• Val F1: 0.6632

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-06
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Val Accuracy	Val Precision	Val Recall	Val F1
0.1908	1.0	1012	0.4035	0.8904	0.7844	0.5448	0.6430
0.152	2.0	2024	0.4631	0.8930	0.7440	0.6235	0.6784
0.12	3.0	3036	0.5046	0.8879	0.7028	0.6605	0.6810
0.0757	4.0	4048	0.7762	0.8902	0.7438	0.6003	0.6644
0.0557	5.0	5060	0.8961	0.8846	0.7273	0.5802	0.6455
0.0319	6.0	6072	0.8864	0.8916	0.7338	0.6296	0.6777
0.0235	7.0	7084	0.8025	0.8902	0.7348	0.6157	0.6700
0.0125	8.0	8096	1.1034	0.8916	0.7559	0.5926	0.6644
0.0114	9.0	9108	1.1414	0.8882	0.7422	0.5864	0.6552
0.0147	10.0	10120	1.2555	0.8902	0.7401	0.6065	0.6667
0.0068	11.0	11132	1.2923	0.8879	0.7526	0.5679	0.6473
0.0112	12.0	12144	1.3150	0.8890	0.8024	0.5139	0.6265
0.0059	13.0	13156	1.1883	0.8899	0.7396	0.6049	0.6655
0.0056	14.0	14168	1.3822	0.8871	0.7824	0.5216	0.6259
0.0029	15.0	15180	1.4309	0.8888	0.7741	0.5448	0.6395
0.0021	16.0	16192	1.3541	0.8916	0.7529	0.5972	0.6661
0.004	17.0	17204	1.3666	0.8907	0.7384	0.6142	0.6706
0.0022	18.0	18216	1.4396	0.8896	0.7525	0.5818	0.6562
0.0028	19.0	19228	1.4340	0.8910	0.7539	0.5910	0.6626
0.0001	20.0	20240	1.4371	0.8913	0.7554	0.5910	0.6632

Framework versions

• Transformers 4.38.2
• Pytorch 2.1.2
• Datasets 2.18.0
• Tokenizers 0.15.2