summarize

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: 2.6935
• Evaluation: {'evaluation_runtime': 28.518348455429077, 'samples_per_second': 33.3118869588378, 'steps_per_second': 33.3118869588378}
• Rounded Rouge: {'rouge1': 0.1705, 'rouge2': 0.0588, 'rougeL': 0.1354, 'rougeLsum': 0.1355}

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: 4
• eval_batch_size: 4
• 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	Evaluation	Rounded Rouge
3.1701	1.0	500	2.8229	{'evaluation_runtime': 30.270989179611206, 'samples_per_second': 31.383183230756966, 'steps_per_second': 31.383183230756966}	{'rouge1': 0.1615, 'rouge2': 0.0525, 'rougeL': 0.128, 'rougeLsum': 0.1281}
2.9661	2.0	1000	2.7672	{'evaluation_runtime': 28.879830598831177, 'samples_per_second': 32.894929793613414, 'steps_per_second': 32.894929793613414}	{'rouge1': 0.1676, 'rouge2': 0.0567, 'rougeL': 0.1326, 'rougeLsum': 0.1327}
2.9128	3.0	1500	2.7414	{'evaluation_runtime': 28.787310361862183, 'samples_per_second': 33.00065160858421, 'steps_per_second': 33.00065160858421}	{'rouge1': 0.1693, 'rouge2': 0.0575, 'rougeL': 0.1342, 'rougeLsum': 0.1343}
2.8783	4.0	2000	2.7240	{'evaluation_runtime': 28.755173683166504, 'samples_per_second': 33.03753301814126, 'steps_per_second': 33.03753301814126}	{'rouge1': 0.1694, 'rouge2': 0.0581, 'rougeL': 0.1343, 'rougeLsum': 0.1344}
2.8548	5.0	2500	2.7137	{'evaluation_runtime': 30.050004959106445, 'samples_per_second': 31.613971488284534, 'steps_per_second': 31.613971488284534}	{'rouge1': 0.171, 'rouge2': 0.0591, 'rougeL': 0.1354, 'rougeLsum': 0.1354}
2.8353	6.0	3000	2.7047	{'evaluation_runtime': 29.376569986343384, 'samples_per_second': 32.33869714679546, 'steps_per_second': 32.33869714679546}	{'rouge1': 0.1703, 'rouge2': 0.0587, 'rougeL': 0.135, 'rougeLsum': 0.135}
2.8229	7.0	3500	2.6996	{'evaluation_runtime': 27.381307363510132, 'samples_per_second': 34.69520236517353, 'steps_per_second': 34.69520236517353}	{'rouge1': 0.1714, 'rouge2': 0.0592, 'rougeL': 0.1357, 'rougeLsum': 0.1357}
2.8154	8.0	4000	2.6958	{'evaluation_runtime': 27.409220457077026, 'samples_per_second': 34.65986934899169, 'steps_per_second': 34.65986934899169}	{'rouge1': 0.17, 'rouge2': 0.0587, 'rougeL': 0.1351, 'rougeLsum': 0.1352}
2.8068	9.0	4500	2.6943	{'evaluation_runtime': 27.376741409301758, 'samples_per_second': 34.7009889086807, 'steps_per_second': 34.7009889086807}	{'rouge1': 0.1702, 'rouge2': 0.0588, 'rougeL': 0.1352, 'rougeLsum': 0.1353}
2.8	10.0	5000	2.6935	{'evaluation_runtime': 28.518348455429077, 'samples_per_second': 33.3118869588378, 'steps_per_second': 33.3118869588378}	{'rouge1': 0.1705, 'rouge2': 0.0588, 'rougeL': 0.1354, 'rougeLsum': 0.1355}

Framework versions

• Transformers 4.38.2
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2