bart-large-cnn-finetuned-roundup-8

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: 1.4519
• Rouge1: 49.5671
• Rouge2: 27.0118
• Rougel: 30.8538
• Rougelsum: 45.5503
• Gen Len: 141.75

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: 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: 8
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	132	1.3159	48.5275	28.0817	30.6646	45.5024	142.0
No log	2.0	264	1.2377	47.0791	27.4386	28.9458	44.1536	142.0
No log	3.0	396	1.2474	49.3567	29.5904	30.8029	46.6083	142.0
0.9623	4.0	528	1.2914	47.8795	27.0611	29.8538	44.4494	142.0
0.9623	5.0	660	1.2982	49.9921	28.4839	31.5688	46.9734	142.0
0.9623	6.0	792	1.3521	46.7269	25.8672	29.7325	43.8279	142.0
0.9623	7.0	924	1.4102	47.4995	26.0066	29.4342	44.1102	141.8
0.3734	8.0	1056	1.4519	49.5671	27.0118	30.8538	45.5503	141.75

Framework versions

• Transformers 4.18.0
• Pytorch 1.11.0+cu113
• Datasets 2.1.0
• Tokenizers 0.12.1