distilbart-cnn-12-6-eval-test-2

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: 4.7250
• Rouge1: 31.3552
• Rouge2: 4.2825
• Rougel: 15.1982
• Rougelsum: 27.9577
• Gen Len: 139.0

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: 1
• eval_batch_size: 1
• 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
4.4419	1.0	80	4.2847	30.8184	4.024	15.5589	27.647	133.6
3.5861	2.0	160	4.2721	30.7823	3.7736	14.992	28.0105	137.1
2.9885	3.0	240	4.4295	30.4747	3.8971	15.6055	27.9916	135.5
2.5254	4.0	320	4.5978	31.0505	4.1062	14.7292	27.9009	134.2
2.2404	5.0	400	4.7250	31.3552	4.2825	15.1982	27.9577	139.0

Framework versions

• Transformers 4.25.1
• Pytorch 1.11.0
• Datasets 2.2.1
• Tokenizers 0.12.1