bart-abs-1509-0313-lr-3e-05-bs-2-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: 3.3486
• Rouge/rouge1: 0.4705
• Rouge/rouge2: 0.2108
• Rouge/rougel: 0.3877
• Rouge/rougelsum: 0.3894
• Bertscore/bertscore-precision: 0.8936
• Bertscore/bertscore-recall: 0.8936
• Bertscore/bertscore-f1: 0.8934
• Meteor: 0.4277
• Gen Len: 38.4091

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: 2
• eval_batch_size: 2
• 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
1.1204	1.0	434	2.2199	0.4542	0.2173	0.3843	0.3855	0.8945	0.8893	0.8917	0.4072	37.2273
0.8222	2.0	868	2.3549	0.4613	0.2095	0.3935	0.3957	0.8994	0.8929	0.896	0.4089	36.8818
0.565	3.0	1302	2.6652	0.4686	0.2079	0.3905	0.3911	0.8943	0.8941	0.894	0.4207	39.6636
0.379	4.0	1736	2.9239	0.4614	0.2076	0.3937	0.3951	0.8962	0.8898	0.8928	0.401	34.8545
0.2543	5.0	2170	3.1849	0.4629	0.2086	0.3988	0.3998	0.8958	0.8914	0.8935	0.4076	36.1091
0.1761	6.0	2604	3.3486	0.4705	0.2108	0.3877	0.3894	0.8936	0.8936	0.8934	0.4277	38.4091

Framework versions

• Transformers 4.44.0
• Pytorch 2.4.0
• Datasets 2.21.0
• Tokenizers 0.19.1