SocialSciencePegasusLargeModel

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

• Loss: 5.7391
• Rouge1: 43.2515
• Rouge2: 13.5819
• Rougel: 29.2476
• Rougelsum: 39.2268
• Bertscore Precision: 76.5154
• Bertscore Recall: 81.3593
• Bertscore F1: 78.854
• Bleu: 0.1036
• Gen Len: 191.3589

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: 1
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 16
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Bertscore Precision	Bertscore Recall	Bertscore F1	Bleu	Gen Len
6.1745	0.3943	300	5.9613	40.1903	12.4753	28.1708	36.7059	75.8626	80.8932	78.2884	0.0959	191.3589
5.8826	0.7885	600	5.7391	43.2515	13.5819	29.2476	39.2268	76.5154	81.3593	78.854	0.1036	191.3589

Framework versions

• Transformers 4.41.2
• Pytorch 2.1.2
• Datasets 2.2.1
• Tokenizers 0.19.1