test-dialogue-summarization

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

• Loss: 2.0710
• Rouge: {'rouge1': 46.5916, 'rouge2': 21.9208, 'rougeL': 22.0124, 'rougeLsum': 22.0124}
• Bert Score: 0.8784
• Bleurt 20: -0.7903
• Gen Len: 15.58

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: 7
• eval_batch_size: 7
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge	Bert Score	Bleurt 20	Gen Len
2.5288	1.0	186	2.1809	{'rouge1': 48.087, 'rouge2': 21.7173, 'rougeL': 21.5447, 'rougeLsum': 21.5447}	0.877	-0.8143	15.63
2.3277	2.0	372	2.1230	{'rouge1': 47.3856, 'rouge2': 21.3069, 'rougeL': 21.6399, 'rougeLsum': 21.6399}	0.8788	-0.786	15.54
2.2381	3.0	558	2.0912	{'rouge1': 45.9843, 'rouge2': 21.1854, 'rougeL': 21.4006, 'rougeLsum': 21.4006}	0.8776	-0.817	15.235
2.2123	4.0	744	2.0761	{'rouge1': 46.5269, 'rouge2': 21.7291, 'rougeL': 21.8936, 'rougeLsum': 21.8936}	0.8785	-0.7809	15.515
2.2443	5.0	930	2.0710	{'rouge1': 46.5916, 'rouge2': 21.9208, 'rougeL': 22.0124, 'rougeLsum': 22.0124}	0.8784	-0.7903	15.58

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.0