bart-abs-2409-1947-lr-3e-05-bs-8-maxep-6

This model is a fine-tuned version of sshleifer/distilbart-xsum-12-6 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 3.8980
• Rouge/rouge1: 0.462
• Rouge/rouge2: 0.2198
• Rouge/rougel: 0.4014
• Rouge/rougelsum: 0.4029
• Bertscore/bertscore-precision: 0.8957
• Bertscore/bertscore-recall: 0.8926
• Bertscore/bertscore-f1: 0.894
• Meteor: 0.4124
• Gen Len: 37.3727

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: 3e-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: 6
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge/rouge1	Rouge/rouge2	Rouge/rougel	Rouge/rougelsum	Bertscore/bertscore-precision	Bertscore/bertscore-recall	Bertscore/bertscore-f1	Meteor	Gen Len
0.1721	1.0	109	3.5761	0.4573	0.2036	0.3842	0.3851	0.8952	0.8917	0.8933	0.405	36.9727
0.1505	2.0	218	3.5724	0.45	0.1979	0.3854	0.3855	0.8956	0.8912	0.8933	0.3964	36.3273
0.1132	3.0	327	3.7251	0.4535	0.2052	0.385	0.3867	0.8955	0.8917	0.8934	0.4043	37.4273
0.0881	4.0	436	3.8461	0.4676	0.2097	0.397	0.3983	0.8967	0.8917	0.8941	0.4047	35.7091
0.0734	5.0	545	3.8385	0.458	0.21	0.3905	0.3919	0.8939	0.8927	0.8931	0.4105	38.0182
0.0612	6.0	654	3.8980	0.462	0.2198	0.4014	0.4029	0.8957	0.8926	0.894	0.4124	37.3727

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.0
• Datasets 3.0.0
• Tokenizers 0.19.1