my_awesome_billsum_model_34

This model is a fine-tuned version of google-t5/t5-small on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.6615
• Rouge1: 0.9649
• Rouge2: 0.8639
• Rougel: 0.9148
• Rougelsum: 0.916
• Gen Len: 4.7917

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	12	2.2768	0.4083	0.2813	0.3855	0.3853	17.3333
No log	2.0	24	1.7504	0.4318	0.2948	0.3978	0.3966	16.6042
No log	3.0	36	1.2490	0.4721	0.3506	0.4443	0.4447	15.3542
No log	4.0	48	0.9124	0.7673	0.6558	0.7251	0.7253	9.0833
No log	5.0	60	0.7653	0.9289	0.8292	0.8817	0.8823	5.7292
No log	6.0	72	0.7176	0.9649	0.8639	0.9148	0.916	4.7917
No log	7.0	84	0.6921	0.9649	0.8639	0.9148	0.916	4.7917
No log	8.0	96	0.6765	0.9649	0.8639	0.9148	0.916	4.7917
No log	9.0	108	0.6655	0.9649	0.8639	0.9148	0.916	4.7917
No log	10.0	120	0.6615	0.9649	0.8639	0.9148	0.916	4.7917

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1