Instructions to use mnaylor/bigbird-base-mimic-mortality with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use mnaylor/bigbird-base-mimic-mortality with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="mnaylor/bigbird-base-mimic-mortality")# Load model directly from transformers import AutoTokenizer, AutoModelForSequenceClassification tokenizer = AutoTokenizer.from_pretrained("mnaylor/bigbird-base-mimic-mortality") model = AutoModelForSequenceClassification.from_pretrained("mnaylor/bigbird-base-mimic-mortality") - Notebooks
- Google Colab
- Kaggle
BigBird for Mortality Prediction
Starting with Google's base BigBird model, we fine-tuned on binary mortality prediction in MIMIC admission notes. This model seeks to predict whether a certain patient will expire within a given ICU stay, based on the text available upon admission. Data prepared for this task as described in this project, using the simulated admission notes (taken from discharge summaries). This model will be used in an upcoming submission for IMLH at ICML 2021.
References
- Van Aken, et al., 2021: Clinical Outcome Prediction from Admission Notes using Self-Supervised Knowledge Integration
- Zaheer, et al., 2020: Big Bird: Transformers for Longer Sequences
- Downloads last month
- 10