clinical-t5

This is a finetuned T5-small model from Google, a checkpoint with 60 million parameters, for clinical note summarization. It was finetuned with the augmented-clinical-notes dataset, available in the Hugging Face.

Intended uses & limitations

The model was created for learning purposes. Hence, although being briefly evaluated in this notebook, it should be further refined.

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: 4

Framework versions

  • Transformers 4.30.0
  • Pytorch 2.2.1+cu121
  • Datasets 2.19.1
  • Tokenizers 0.13.3
Downloads last month
79
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Dataset used to train hossboll/clinical-t5