my_billsum_model

This model is a fine-tuned version of google-t5/t5-small on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.3699
• Rouge1: 0.1958
• Rouge2: 0.0949
• Rougel: 0.167
• Rougelsum: 0.167
• Gen Len: 19.0

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
• 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	62	2.7958	0.1228	0.0386	0.0997	0.1	19.0
No log	2.0	124	2.5846	0.1385	0.047	0.1139	0.114	19.0
No log	3.0	186	2.5034	0.1506	0.0563	0.1232	0.1234	19.0
No log	4.0	248	2.4548	0.1734	0.0756	0.1467	0.1468	19.0
No log	5.0	310	2.4231	0.1893	0.0877	0.1597	0.1597	19.0
No log	6.0	372	2.3991	0.1926	0.0913	0.1638	0.1638	19.0
No log	7.0	434	2.3862	0.1945	0.0944	0.166	0.166	19.0
No log	8.0	496	2.3764	0.195	0.094	0.1662	0.1663	19.0
2.7718	9.0	558	2.3714	0.1959	0.0952	0.1672	0.1672	19.0
2.7718	10.0	620	2.3699	0.1958	0.0949	0.167	0.167	19.0

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 2.21.0
• Tokenizers 0.19.1