Skip to content

Latest commit

 

History

History
86 lines (58 loc) · 2.11 KB

readme.md

File metadata and controls

86 lines (58 loc) · 2.11 KB
Raw

Dataset Preprocesser Emission Reduction

Features

Changelog

  • Added more metrics as goal metrics
  • Added images datasets
  • Add method for reducing dataset (balanced vs uniform)
  • Added notebook containing analysis of the tool on actual data
  • Uploaded to github and dockerized

To do

  • Add more TS datasets
  • Clean UI
  • Add a section that shows user what are the environmental impacts when using the tool (carbon emission reduction, etc)

Run the tool

Docker and Docker-Compose (Recommended)

  • Install Docker
  • From the root of the project run the following

First install everything with

docker-compose --build

Now to run

docker-compose up -d

The first time the database needs to be set, run in order the following commands

curl -X 'GET' 'http://localhost:5000/api/load_experiments_on_db?reset_db=true' -H 'accept: application/json'

curl -X 'GET' 'http://localhost:5000/api/load_completeness_curves?load_only_missing=false' -H 'accept: application/json'

curl -X 'GET' 'http://localhost:5000/api/train_regressors' -H 'accept: application/json'

You can now visit the interface in https://localhost:8080

To finish the process after

docker-compose down

Installing everything manually

  • Install the following requirements
    • For the backend
      • python 3.10
      • From the backend/ directory, run pip install -r requirements.txt
    • For the frontend
      • install node 16
      • From the 'frontend/' directory, run npm install and npm run build

Now to run the backend

uvicorn main:app --port 3000

And the frontend

npm run serve

The first time the database needs to be set, run in order the following commands

curl -X 'GET' 'http://localhost:5000/api/load_experiments_on_db?reset_db=true' -H 'accept: application/json'

curl -X 'GET' 'http://localhost:5000/api/load_completeness_curves?load_only_missing=false' -H 'accept: application/json'

curl -X 'GET' 'http://localhost:5000/api/train_regressors' -H 'accept: application/json'

You can now visit the interface in https://localhost:8080