Abmiguity-factor

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

• Loss: 0.8537
• Rouge1: 0.5239
• Rouge2: 0.2727
• Rougel: 0.3876
• Rougelsum: 0.3876
• Gen Len: 90.5

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: 8
• eval_batch_size: 8
• 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	1	1.2998	0.3903	0.1429	0.2699	0.2699	69.0
No log	2.0	2	1.1258	0.4737	0.202	0.3449	0.3449	77.0
No log	3.0	3	1.0220	0.4627	0.2003	0.3372	0.3372	87.5
No log	4.0	4	0.9522	0.472	0.2042	0.3429	0.3429	85.5
No log	5.0	5	0.9162	0.4951	0.2238	0.3814	0.3814	95.0
No log	6.0	6	0.8882	0.4951	0.2238	0.3814	0.3814	95.0
No log	7.0	7	0.8659	0.5171	0.2652	0.4122	0.4122	97.5
No log	8.0	8	0.8537	0.5239	0.2727	0.3876	0.3876	90.5

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 3.0.1
• Tokenizers 0.19.1