pegasus-x-base-finetuned-multi-news

This model is a fine-tuned version of google/pegasus-x-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.3695
• Rouge1: 39.3711
• Rouge2: 13.3688
• Rougel: 22.2825
• Rougelsum: 33.911

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: 5.6e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 7

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum
2.7281	1.0	625	2.3958	37.1778	12.486	21.2186	31.8237
2.4474	2.0	1250	2.3748	37.9969	12.951	21.7277	32.5926
2.3396	3.0	1875	2.3739	38.8944	13.3746	22.1289	33.5442
2.2637	4.0	2500	2.3645	38.2964	13.048	21.9356	32.8345
2.2019	5.0	3125	2.3699	38.8395	13.1676	22.0546	33.4517
2.1603	6.0	3750	2.3641	39.3608	13.5131	22.3133	33.8639
2.1346	7.0	4375	2.3695	39.3711	13.3688	22.2825	33.911

Framework versions

• Transformers 4.28.1
• Pytorch 2.0.0+cu118
• Datasets 2.11.0
• Tokenizers 0.13.3