Website Application for Creating Practical Machine Learning Models

About website

This website application designed for create simple machine learning models on datasets. You can:

• Upload your datasets
• Create different workspaces and work on different datasets
• Visulize your datasets
• Configure your datasets
• Create simple machine learning models for your datasets
• Inspect results of models
• Save your workspaces and continue after login
• Dowload your results

Requirements

• Required libraries are presented in the file requirements.txt. This important file can be used to create new virtual environment for Python - or for deploying the app.
• Used python version is : Python 3.7.3

How to run?

• This website can be run using flask commands.
  • Open any command prompt of your choice and set environment variables. Read the paragraph below for learning about environment variables.
  • Set the secret_key variable in the app.py carefully before deploying the app. In development process, this is not required.
  • Use the command flask init-db to initialize empty database and table for the user. Note that empty data table is already provided in the instance folder so this step is not required and it should only be used to initialize new empty table.
  • At last, to run the website, use the command flask run.

Important environment variables that should be set before:

• FLASK_APP -- This is important for determining where is the application. For more information, see.
• FLASK_ENV -- This variable is used for starting the application in debug mode .For more information, see.

For a tutorial, head to the img folder.

Resources

• Read flask documentation for more information about setting the environment variables, debugging and functions.