bart-large-cnn-weaksup-100-NOpad-early1

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

• Loss: 2.0768
• Rouge1: 28.7953
• Rouge2: 10.9535
• Rougel: 20.6447
• Rougelsum: 24.3516
• Gen Len: 68.5

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: 3
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	100	1.8905	31.2906	13.5675	21.5533	27.2536	64.2
No log	2.0	200	2.0768	28.7953	10.9535	20.6447	24.3516	68.5

Framework versions

• Transformers 4.16.2
• Pytorch 1.10.2
• Datasets 1.18.3
• Tokenizers 0.11.0