pegasus-cnn_dailymail-1000-lit-evalMA-ga

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

• Loss: 1.6852
• Rouge1: 25.789
• Rouge2: 11.0694
• Rougel: 20.7716
• Rougelsum: 22.4851
• Gen Len: 46.32

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: 1
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 4
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 3
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	250	1.7061	25.8286	10.8156	20.9502	22.6588	44.36
1.4533	2.0	500	1.6876	26.0862	11.5197	21.1282	23.0963	45.65
1.4533	3.0	750	1.6852	25.789	11.0694	20.7716	22.4851	46.32

Framework versions

• Transformers 4.16.2
• Pytorch 1.10.2
• Datasets 1.18.3
• Tokenizers 0.11.0