CDH1 Cancer Research

ADD LINK TO PAPER.

Reproducing Inference

Requirements

• Python 3.8 or later
• CPU

Setup

After cloning the repository to your local machine, install the code in editable mode from the repositories root directory cdh1-cancer-res/ by executing pip install -e .

Test the setup

To test that the installation works, script a single aggregator model. A successful test should return Created torchscripted checkpoint at <PATH>. To run the setup test, execute the following command: python -m cdh1_cancer_res.model

Run the demo prediction

• 1. Script the ensemble model:
  • python -m cdh1_cancer_res.script_ensemble
  • This will create the torchscripted ensemble model we will require to run our demo predictions.
• 2. Predict the CDH1 on the provided demo embeddings using the torchscripted ensemble model:
  • python -m cdh1_cancer_res.predict
  • The prediction step should print our the ground truth values, the model's continuous and binarized prediction.

---

Paige.AI CDH1 Cancer Research (c) by Paige.AI

Paige.AI CDH1 Cancer Research is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

You should have received a copy of the license along with this work. If not, see http://creativecommons.org/licenses/by-nc-nd/4.0/.