Adapted from analysis competed by: Emerson, Crick, Allie, Janowcz, Ziva Subelji and Sam Thorne
Permission to adapt DSCI100 analysis for the purposes of DSCI310 can be see here.
We built a
This model could potentially aid in medical diagnosies for patients with heart conditions based on factors that are easy to measure. The information this model provides can be used as a precaution to spread public awareness of heart disease risk based on your age, blood pressure and heart rate. This model should not be used to diagnose individuals directly but used as a tool to assess the possibility that the patient is, or will experience heart disease.
The data we used to build our model came from the processed versions of data for Cleveland, Hungary, Switzerland and VA Long Beach patient information from 1988. All these data sets can be found here.
The full report can be found here
The full HTML report can be found here
There are three ways to run the report on your local computer. You can use a dockerhub image, build the image locally or manually download the dependencies onto your local machine.
To follow instructions using the dockerhub image or to build the image locally, you first need to download Docker. Once you have done this you can follow the instructions below:
- Clone this GitHub repository onto your local computer using:
git clone <paste-HTTPS/SSH-here>in your terminal. - Navigate to the root of this project directory using
cd dsci-310-group-18. - Run the following: (you might be prompted to login to docker to move on to the next step)
docker pull leuneri/dsci-310-group-18:latest
- This step should automatically open jupyter lab on your computer inside the project directory (or prompt you to copy a link to open jupyter lab). Our report can be found as an .ipynb inside
notebooks/along with any code within theR/directory. To open jupyter run the following:
docker run -p 8888:8888 leuneri/dsci-310-group-18:latest
- Navigate to the terminal within jupyter lab.
- Run the following within the jupyter terminal:
make notebooks/_build/html/analysis.html
- Copy the link generated inside your web browser and you will be able to view the full report as a .html file.
If you wish to re-run the above steps run make clean in the jupyter terminal first to clear any files generated from the above steps.
- Clone this GitHub repository onto your local computer using:
git clone <paste-HTTPS/SSH-here>in your terminal. - Navigate to the root of this project directory using
cd dsci-310-group-18. - Run the following: (THIS STEP TAKES ~25 MINUTES TO RUN SO KEEP THIS IN MIND BEFORE COPYING INTO YOUR TERMINAL)
docker build --tag dsci-310-group-18:latest .
- Once that has loaded you can run (This will open jupyter lab):
docker run -p 8888:8888 dsci-310-group-18:latest
From there you can follow steps 5-7 of the docker image instructions.
If you wish to re-run the above steps run make clean in the jupyter terminal first to clear any files generated from the above steps.
- Clone this GitHub repository onto your local computer using :
git clone <past-HTTPS/SSH-here>into your terminal. - Navigate to the root of this project directory using
cd dsci-310-group-18. - Install all the dependencies and softwards listed below
- Run the following in your terminal:
make notebooks/_build/html/analysis.html
- Copy the link generated into your web browser to view the complete report.
To remove the generated files run make clean to restart with a 'clean' repository.
Software includes:
R version 4.2.3, Jupyterlab 3.6.3, GNU Make 4.3 & JupyterBook 0.13.2
R language packages include:
| Package | Version |
|---|---|
| cowplot | 1.1.1 |
| tidymodels | 1.0.0 |
| tidyverse | 1.3.2 |
| repr | 1.1.6 |
| ggplot2 | 3.3.6 |
| caret | 6.0.94 |
| e1071 | 1.7.13 |
| testthat | 3.1.7 |
| kknn | 1.3.1 |
| group18package | 0.0.0.9000 |
group18package can be found here
This project and it's assets are licensed under a CC-BY license.
All code of this project is licensed under the MIT License.
Complete license file can be found here