Skip to content

Выпускная квалификационная работа

Notifications You must be signed in to change notification settings

korotaS/grad_project_2021

Repository files navigation

Graduation project 2021 at SE FCS HSE

"Cross-platform Desktop Application for Automated Machine Learning Pipeline Building"

Prerequisites

In order to run the application correctly, you need to have:

  • node.js 14+ and npm 6+
  • python 3.6+

Installation

macOS

sh init-macos.sh

Windows

init-win.bat

Run app

With computations on local machine

Don't forget to enter the python environment!

npm start

With computations on remote machine

To start the app:

npm start

Then start flask server on remote host by running:

python src/python/main.py --expose --port 1234

Then in the app click "Change local to remote" button, enter host/port of the remote machine and click "Connect". Now you can train models on the remove server!

Tips

  • In order to create correct dataset structure, check datasets.md
  • You can run inference of the model with appropriate preprocessing class, see example in this notebook

About

Выпускная квалификационная работа

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages