summarizer-billsum_dataset

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

• Loss: 2.4835
• Rouge1: 0.1837
• Rouge2: 0.0818
• Rougel: 0.1536
• Rougelsum: 0.154
• 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: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	25	3.4284	0.1297	0.0383	0.109	0.1089	19.0
No log	2.0	50	3.0057	0.1222	0.0351	0.1031	0.1029	19.0
No log	3.0	75	2.8213	0.1242	0.0376	0.1042	0.1041	19.0
No log	4.0	100	2.7231	0.1283	0.0401	0.105	0.105	19.0
No log	5.0	125	2.6706	0.1371	0.049	0.1122	0.1122	19.0
No log	6.0	150	2.6307	0.1373	0.0473	0.1129	0.1128	19.0
No log	7.0	175	2.5988	0.1408	0.0496	0.1149	0.1148	19.0
No log	8.0	200	2.5731	0.1471	0.0509	0.1209	0.1212	19.0
No log	9.0	225	2.5557	0.156	0.0584	0.1293	0.1296	19.0
No log	10.0	250	2.5382	0.1642	0.0656	0.1357	0.1356	19.0
No log	11.0	275	2.5262	0.1695	0.0716	0.1402	0.1403	19.0
No log	12.0	300	2.5173	0.1773	0.0778	0.1475	0.1475	19.0
No log	13.0	325	2.5089	0.18	0.0801	0.1493	0.1496	19.0
No log	14.0	350	2.5013	0.1821	0.08	0.1515	0.1516	19.0
No log	15.0	375	2.4954	0.1823	0.0801	0.1527	0.1528	19.0
No log	16.0	400	2.4910	0.1832	0.0808	0.1532	0.1534	19.0
No log	17.0	425	2.4875	0.1842	0.082	0.154	0.1543	19.0
No log	18.0	450	2.4849	0.1841	0.0818	0.1539	0.1541	19.0
No log	19.0	475	2.4840	0.1837	0.0818	0.1536	0.154	19.0
2.7815	20.0	500	2.4835	0.1837	0.0818	0.1536	0.154	19.0

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Datasets 2.17.0
• Tokenizers 0.15.1