wiki_asp-software_7829_bart-base

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

• Loss: 3.0105
• Rouge1: 0.1449
• Rouge2: 0.0488
• Rougel: 0.1206
• Rougelsum: 0.1204
• Gen Len: 19.8626

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: 5e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 16
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	2.37	500	3.1540	0.1378	0.0439	0.1152	0.1152	19.8858
No log	4.74	1000	3.0456	0.141	0.0462	0.117	0.117	19.7587
No log	7.1	1500	3.0225	0.1428	0.0472	0.1183	0.1182	19.9145
3.2197	9.47	2000	3.0105	0.1449	0.0488	0.1206	0.1204	19.8626

Framework versions

• Transformers 4.38.2
• Pytorch 2.0.0+cu117
• Datasets 2.18.0
• Tokenizers 0.15.2