BART_corrector_15

This model is a fine-tuned version of ainize/bart-base-cnn on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0214
• Rouge1: 80.3263
• Rouge2: 78.1274
• Rougel: 80.3215
• Rougelsum: 80.3039
• Gen Len: 19.3993

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	Gen Len
0.0597	1.0	2365	0.0367	79.3503	76.3308	79.32	79.3005	19.3992
0.0322	2.0	4730	0.0276	79.9515	77.4211	79.9331	79.9164	19.3983
0.0212	3.0	7095	0.0241	80.1413	77.8084	80.129	80.1098	19.3992
0.0148	4.0	9460	0.0219	80.2625	78.035	80.2579	80.2372	19.4
0.0111	5.0	11825	0.0214	80.3263	78.1274	80.3215	80.3039	19.3993

Framework versions

• Transformers 4.21.1
• Pytorch 1.12.1+cu113
• Datasets 2.4.0
• Tokenizers 0.12.1