ML-Redefining-Cancer-Treatment 🩺

• Personalized Medicine: Redefining Cancer Treatment.

• Predict the effect of Genetic Variants to enable Personalized Medicine.

• Medical Domain AI Solution 💊

• Credit & thanks to teclov.

• Kaggle link: https://www.kaggle.com/c/msk-redefining-cancer-treatment

• Most exiting part is, real world problem statement and dataset availability.

Tech stack 👇

Python | Machine Learning | NLP | VS Code 💻

Dataset 👇

1. In the project "training_text" file is missing under "training" directory (due to upload file size limit). Please find it in the kaggle link provided above.

How to run ? 👇

1. Clone/Download the project from github.
2. Open project in Pycharm/VS Code.
3. Run "run.py" file or Run by highlighting snippets of code.

Application directory structure 👇

Lessons learnt 📖

• Important business metrics to consider before modelling such as Interpretability, Latency, Due to medical domain very less error etc.
• Merge dataframes based on IDs.
• EDA & Viz.
• Log loss evaluation metrics.
• Laplace smoothing.
• Callibirated classifier.
• Stochastic gradient descent.
• Response encoding technique.
• Feature importance.
• Finally, python 😄

License 👇

Connect with me 👇