Heart Attack Predictions

Model deployed on this site using Deploifai

---

Running locally

1. Install dependencies with python 3.8 and run train.py script to train the model. This will create an artifacts directory that contains the model, scaler, and encoder

2. Run main.py and visit http://localhost:7860[http://localhost:7860] to view use the model using Gradio

Running as a Docker container

1. Ensure that you have created the model by running the training script

2. Build the image

docker build -t heartattack .

or use BuildKit to specify the platform for deployment

docker buildx build --platform linux/amd64 -t heartattack .

3. Run a container using the image

docker run -it -p 7860:7860 --name HeartAttack heartattack

4. Visit http://localhost:7860

---

Tools used

• Python 3.8
• Pandas
• Scikit-learn
• Gradio
• Deploifai

Dataset obtained from https://www.kaggle.com/datasets/rashikrahmanpritom/heart-attack-analysis-prediction-dataset