bart-base-cnn-xsum-swe

This model is a fine-tuned version of Gabriel/bart-base-cnn-swe on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.1027
• Rouge1: 30.9467
• Rouge2: 12.2589
• Rougel: 25.4487
• Rougelsum: 25.4792
• Gen Len: 19.7379

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: 4e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 4
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
2.3076	1.0	6375	2.1986	29.7041	10.9883	24.2149	24.2406	19.7193
2.0733	2.0	12750	2.1246	30.4521	11.8107	24.9519	24.9745	19.6592
1.8933	3.0	19125	2.0989	30.9407	12.2682	25.4135	25.4378	19.7195
1.777	4.0	25500	2.1027	30.9467	12.2589	25.4487	25.4792	19.7379

Framework versions

• Transformers 4.22.2
• Pytorch 1.12.1+cu113
• Datasets 2.5.1
• Tokenizers 0.12.1