bart-large-cnn-finetuned-pubmed-finetuned-roundup-e8

This model is a fine-tuned version of theojolliffe/bart-large-cnn-finetuned-pubmed on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.1034
• Rouge1: 48.4605
• Rouge2: 28.5961
• Rougel: 32.5389
• Rougelsum: 45.7358
• Gen Len: 142.0

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: 2e-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: 8
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	25	1.4278	47.952	29.4059	34.273	45.7244	142.0
No log	2.0	50	1.4351	48.7561	29.4049	30.631	46.4074	142.0
No log	3.0	75	1.5375	50.0069	31.4237	32.0834	47.679	142.0
No log	4.0	100	1.6647	49.6919	28.8821	31.9357	47.0396	142.0
No log	5.0	125	1.8070	47.8472	26.6979	30.7049	44.5848	142.0
No log	6.0	150	1.9981	47.8352	27.0966	31.4529	46.5251	142.0
No log	7.0	175	2.0904	48.6272	30.5493	32.7827	46.8462	142.0
No log	8.0	200	2.1034	48.4605	28.5961	32.5389	45.7358	142.0

Framework versions

• Transformers 4.18.0
• Pytorch 1.11.0+cu113
• Datasets 2.1.0
• Tokenizers 0.12.1