longformer-epidemiology-2epoch

This model is a fine-tuned version of allenai/longformer-base-4096 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.6925
• Accuracy: 0.523
• F1: 0.3434

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: 20
• eval_batch_size: 20
• 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: 2

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
0.7018	0.1111	50	0.7033	0.477	0.3230
0.6963	0.2222	100	0.7050	0.477	0.3230
0.6963	0.3333	150	0.6921	0.523	0.3434
0.694	0.4444	200	0.6921	0.523	0.3434
0.7012	0.5556	250	0.6930	0.523	0.3434
0.697	0.6667	300	0.6969	0.477	0.3230
0.7008	0.7778	350	0.6923	0.523	0.3434
0.6944	0.8889	400	0.6924	0.523	0.3434
0.6945	1.0	450	0.7039	0.477	0.3230
0.6915	1.1111	500	0.6925	0.523	0.3434
0.6988	1.2222	550	0.6935	0.477	0.3230
0.6927	1.3333	600	0.6950	0.477	0.3230
0.6916	1.4444	650	0.6931	0.523	0.3434
0.6911	1.5556	700	0.6936	0.477	0.3230
0.6969	1.6667	750	0.6977	0.477	0.3230
0.6963	1.7778	800	0.6929	0.523	0.3434
0.6924	1.8889	850	0.6941	0.477	0.3230
0.6945	2.0	900	0.6935	0.477	0.3230

Framework versions

• Transformers 4.46.0
• Pytorch 2.4.0+cu121
• Datasets 2.21.0
• Tokenizers 0.20.1