distilbart-12-6-cnn-dm-abstractive-summarizer

This model is a fine-tuned version of sshleifer/distilbart-cnn-12-6 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.6988
• Rouge1: 0.42
• Rouge2: 0.2006
• Rougel: 0.3032
• Rougelsum: 0.3033
• Generated Length: 72.6006

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: 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: 5
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Generated Length
1.1498	1.0	1436	1.5556	0.4009	0.1815	0.2843	0.2843	72.1027
1.1051	2.0	2872	1.5456	0.4127	0.191	0.2941	0.2942	71.2483
0.892	3.0	4308	1.5954	0.4129	0.1924	0.2946	0.2947	72.453
0.7543	4.0	5744	1.6574	0.4177	0.1974	0.3018	0.3017	71.8179
0.651	5.0	7180	1.6988	0.42	0.2006	0.3032	0.3033	72.6006

Framework versions

• Transformers 4.39.3
• Pytorch 2.1.2
• Datasets 2.18.0
• Tokenizers 0.15.2