my_billsum_model

This model is a fine-tuned version of Shubhu07/my_billsum_model on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.2772
• Rouge1: 0.2051
• Rouge2: 0.0968
• Rougel: 0.1722
• Rougelsum: 0.1722
• Gen Len: 20.0

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	50	2.4554	0.1484	0.0499	0.1211	0.1213	20.0
No log	2.0	100	2.3995	0.1764	0.0717	0.146	0.1461	20.0
No log	3.0	150	2.3573	0.1978	0.0877	0.1655	0.1656	20.0
No log	4.0	200	2.3325	0.2036	0.0937	0.17	0.17	20.0
No log	5.0	250	2.3136	0.2061	0.0953	0.1728	0.1727	20.0
No log	6.0	300	2.2987	0.2056	0.096	0.1725	0.1724	20.0
No log	7.0	350	2.2898	0.204	0.0957	0.1714	0.1715	20.0
No log	8.0	400	2.2824	0.2047	0.0962	0.1721	0.1721	20.0
No log	9.0	450	2.2788	0.2053	0.0968	0.1723	0.1723	20.0
2.5896	10.0	500	2.2772	0.2051	0.0968	0.1722	0.1722	20.0

Framework versions

• Transformers 4.51.1
• Pytorch 2.6.0+cu124
• Datasets 3.5.0
• Tokenizers 0.21.1