t5-medical-text-simplification

This model is a fine-tuned version of mrm8488/t5-small-finetuned-text-simplification on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.4158
• Bleu: {'bleu': 0.24913061085239344, 'precisions': [0.6300697552884507, 0.46170603353322726, 0.3783389479827051, 0.3190805662507599], 'brevity_penalty': 0.5754971743889961, 'length_ratio': 0.6441136869219061, 'translation_length': 44011, 'reference_length': 68328}
• Sari: {'sari': 21.772869578730884}
• Fkgl: 10.2474

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: 5
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Sari	Fkgl
1.5524	1.0	1578	1.4317	{'bleu': 0.24854970426705067, 'precisions': [0.626776178839714, 0.45794346978557504, 0.37443247809101465, 0.3154227136604469], 'brevity_penalty': 0.5792493345645447, 'length_ratio': 0.646821215314366, 'translation_length': 44196, 'reference_length': 68328}	{'sari': 21.542679628603977}	10.2949
1.5282	2.0	3156	1.4249	{'bleu': 0.24886563197246125, 'precisions': [0.6285792076961474, 0.4604086221222934, 0.3770192256766061, 0.3176616771658094], 'brevity_penalty': 0.5767757332645675, 'length_ratio': 0.6450357101042032, 'translation_length': 44074, 'reference_length': 68328}	{'sari': 21.665573517166536}	10.2937
1.4997	3.0	4734	1.4176	{'bleu': 0.24852094682922746, 'precisions': [0.629403208945048, 0.4605591734808794, 0.377421066595914, 0.3182660566398332], 'brevity_penalty': 0.5753144561890373, 'length_ratio': 0.6439819693244351, 'translation_length': 44002, 'reference_length': 68328}	{'sari': 21.700716936778782}	10.2544
1.5028	4.0	6312	1.4176	{'bleu': 0.24876653336273433, 'precisions': [0.6299538437052363, 0.4615309246785058, 0.37816241471767237, 0.3188943296728769], 'brevity_penalty': 0.5748880487421792, 'length_ratio': 0.6436746282636694, 'translation_length': 43981, 'reference_length': 68328}	{'sari': 21.750120178010484}	10.2531
1.4976	5.0	7890	1.4158	{'bleu': 0.24913061085239344, 'precisions': [0.6300697552884507, 0.46170603353322726, 0.3783389479827051, 0.3190805662507599], 'brevity_penalty': 0.5754971743889961, 'length_ratio': 0.6441136869219061, 'translation_length': 44011, 'reference_length': 68328}	{'sari': 21.772869578730884}	10.2474

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.1