OncoPredict

Introduction

OncoPredict is a project aimed at detecting cancer from blood samples using machine learning and deep learning techniques.

Project Structure

• data/: Datasets used for training and testing.
• notebooks/: Jupyter notebooks for experiments and data exploration.
• src/: Source code for model training and evaluation.
• models/: Trained models and their configurations.
• results/: Results and performance metrics.
• tests/: Test scripts for validating data and models.

How to Run

1. Clone the repo: git clone https://github.com/your-username/OncoPredict.git
2. Install Python dependencies: pip install tensorflow pandas optuna scikit-learn matplotlib flask gunicorn
3. Install npm dependencies: npm install
4. Run the training script: npm run train

Hyperparameter Tuning

To run hyperparameter tuning using Optuna, follow these steps:

1. Install the required dependencies: pip install optuna
2. Run the training script with hyperparameter tuning: python src/train.py --tune

Generating and Viewing Evaluation Metrics and Visualizations

To generate and view detailed evaluation metrics and visualizations, follow these steps:

1. Ensure you have the required dependencies: pip install scikit-learn matplotlib
2. Run the training script: python src/train.py
3. After training, evaluation metrics will be saved in the results/metrics.txt file.
4. Confusion matrix and ROC curve visualizations will be saved in the results/ directory as confusion_matrix.png and roc_curve.png respectively.

Deploying the Model using Flask

To deploy the model using Flask, follow these steps:

1. Install the required dependencies: pip install flask gunicorn
2. Create a Procfile in the root directory with the following content:

   web: gunicorn src.app:app
   

3. Create a new file src/app.py with the Flask application code.
4. Deploy the application to Heroku or any other cloud platform that supports Flask.

Running the Flask App Locally

To run the Flask app locally, follow these steps:

1. Install the required dependencies: pip install flask
2. Run the Flask app: npm start
3. Send a POST request to the /predict endpoint with the input data to get predictions.