longformer_summarise

This model is a fine-tuned version of allenai/led-base-16384 on the scientific_papers dataset. It achieves the following results on the evaluation set:

• Loss: 2.3003
• Rouge2 Precision: 0.1654
• Rouge2 Recall: 0.0966
• Rouge2 Fmeasure: 0.1118

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: 2
• eval_batch_size: 2
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 1
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge2 Precision	Rouge2 Recall	Rouge2 Fmeasure
2.909	0.08	10	2.8969	0.09	0.1439	0.0953
2.615	0.16	20	2.6182	0.1232	0.0865	0.0924
2.581	0.24	30	2.4687	0.1357	0.0733	0.09
2.1294	0.32	40	2.5215	0.1495	0.0932	0.1044
2.8083	0.4	50	2.3870	0.1794	0.1054	0.1224
3.0704	0.48	60	2.3676	0.1572	0.0989	0.1108
2.4716	0.56	70	2.3554	0.1707	0.1039	0.1198
2.454	0.64	80	2.3411	0.1619	0.0943	0.1115
2.3046	0.72	90	2.3105	0.1547	0.0965	0.1116
1.7467	0.8	100	2.3417	0.1551	0.0877	0.1046
2.7696	0.88	110	2.3226	0.1543	0.0954	0.1085
2.4999	0.96	120	2.3003	0.1654	0.0966	0.1118

Framework versions

• Transformers 4.21.3
• Pytorch 1.12.1+cu113
• Datasets 1.2.1
• Tokenizers 0.12.1