Skip to content
Dockerized machine learning web application that predicts the final amount of wins per nba team.
Jupyter Notebook Other
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
media
src
stored_data
.dockerignore
.gitattributes
.gitignore
Dockerfile
LICENSE
README.md
model_building.ipynb

README.md

NBA Winning

Objective

Predict each NBA playoff team's final W count.

About

21 years of regular season NBA team stats data was scraped, starting at 1997 till 2018. The scraped data was used in fitting the ML model. The SVM model is able to predict the final season's Win score per team with a mean STD of 3.5 wins. The model metrics information is stored at this location ./src/static/model/support_vector_machines.txt. To see the entire data pre-processing and model selection that was used for this project you can clone this project and start this jupyter notebook: ./model_building.ipynb

Starting a Docker Container

Installation of DockerCE (Community Edition)

Create the Docker image for this project

In your shell CLI (I am using bash) navigate to the root (where the Dockerfile is located).

$ cd into/this/projects/root/directory/
$ ls
... ... Dockerfile ...

Build the Docker image

$ docker build -t image-name .
Sending build context to Docker daemon [size]
Step 1/* : FROM [base]
...
Successfully built ************
Successfully tagged ***:latest

Run the Docker container from the image

$ docker run -dp 5000:5000 --rm image-name
**************************************

Check to see of the daemon process is working in the background

$ docker ps -a
ContainerID	Image	     Command	 Created   Status		    Ports				           Names
***********	image-name "/bin/sh" 3 seconds Up 3 seconds 0.0.0.0:5000->5000/tcp Rdm-name

Sample

After successfully deploying the container to the port 5000 open up an Internet browser and type http://localhost:5000 into the URL bar and the following webpage will magically appear.

Sample GIF

to terminate the docker daemon process and then remove the image from your system:

$ docker stop (ContainerID or Name)
*************
$ docker rmi image-name
Deleted: sha256:*************************************
Deleted: sha256:*************************************
Deleted: sha256:*************************************
...

Development Process

Step 1 - Data Gathering and Cleaning

Step 2 - ML Model Build The code and process I used for developing the various model's used in this application is located >> ./model_building.ipynb

Step 3 - Flask Python Flask is the framework used to establish the various routes, APIs, as well as the templating system to tie the backend with what you see in the browser application.

Step 4 - Dockerizing This application has been configured to easily build a docker image from which you're able to containerize and launch. Everything needed to build the Docker ubuntu image is included inside of the Dockerfile.

Credit

tbd

You can’t perform that action at this time.