Using the TCGA-PAAD dataset, The model predicts relative risk (hazard ratios) for patients based on clinical and genomic features.
Install Requirements:
pip3 install -r requirements.txt
If using nix:
nix develop
Demo Inference: Picks 5 random samples from the dataset to run inference. (pre-trained model is already present in bin/)
python3 inference.py
To see the training process or re-train the model, run training.ipynb