This Repository Contains MLZoomcamp Midterm Project Files.
Dataset is taken from https://www.kaggle.com/alakaaay/diabetes-uci-dataset
Given dataset has been collected using direct questionnaires from the patients of Sylhet Diabetes Hospital in Sylhet, Bangladesh, and approved by a doctor. As we train a model for the project, We are predicting a person going to have diabetes or not. With attributes that the dataset have, we can predict how much a person is going to have diabetes or not with a number between 0 to 1, Actually with multiplying it with 100, It would give the percentage of a person having diabetes in future.
There are some modification to it to explore more options in machine learning
- Binary category was added to gender in order to use One-hot encoding (dummy variable) option in training.
Content Attribute information:
- Age: 20-65
- Sex: Male/Female
- Polyuria: Yes/No
- Polydipsia: Yes/No
- sudden weight loss: Yes/No
- weakness: Yes/No
- Polyphagia: Yes/No
- Genital thrush: Yes/No
- visual blurring: Yes/No
- Itching: Yes/No
- Irritability: Yes/No
- delayed healing: Yes/No
- partial paresis: Yes/No
- muscle stiffness: Yes/No
- Alopecia: Yes/No
- Obesity: Yes/No
- Class: Positive/Negative
The project is now deployed on heroku cloud servers and to test it just run the file test.ipnyb. Or you can just send json data into the link: https://diabetes-prediction-server.herokuapp.com/predict and get the results.
To deploy this project on your local computer or your server just follow the instructions below:
- First Install Docker on your system
- Install pipenv on your system.
pipenv install requirments.txt
docker built -t diabetes-prediction .
docker run -it -p 9696:9696 diabetes-prediction:latest