pegasus-large-finetuned-cnn_dailymail

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

• Loss: 1.0469
• Rouge1: 45.2373
• Rouge2: 22.4813
• Rougel: 31.8329
• Rougelsum: 41.6862
• Bleu 1: 34.8304
• Bleu 2: 23.4162
• Bleu 3: 17.4357
• Meteor: 35.0815
• Lungime rezumat: 56.5898
• Lungime original: 48.7656

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 4

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Bleu 1	Bleu 2	Bleu 3	Meteor	Lungime rezumat	Lungime original
1.2058	1.0	7165	1.0640	44.5245	22.1439	31.4275	40.9808	33.9802	22.8183	16.9712	34.1035	55.245	48.7656
1.0602	2.0	14330	1.0534	44.7088	22.1286	31.3398	41.0804	34.1571	22.9231	17.0479	35.1782	59.6166	48.7656
1.0144	3.0	21495	1.0479	45.0257	22.3325	31.7313	41.4189	34.6084	23.227	17.2859	34.7757	56.1443	48.7656
0.9875	4.0	28660	1.0469	45.2373	22.4813	31.8329	41.6862	34.8304	23.4162	17.4357	35.0815	56.5898	48.7656

Framework versions

• Transformers 4.40.0
• Pytorch 2.2.2+cu118
• Datasets 2.19.0
• Tokenizers 0.19.1