Web-based GUI for Machine Learning Workflow/Pipeline
Professional software development project from the Fall 2023 semester at UAF.
Development Instructions
Flask Application
Local Debugging
[!NOTE]
This project utilizes Pipenv for managing virtual environments.
You can install it with pip install --user pipenv.
- Clone repository:
gh repo clone jbonda/ml-workflow-pipeline
- Navigate to the source directory:
cd ml-workflow-pipeline/ML_GUI
- Install dependencies:
pipenv shell
- Run program:
flask --app app.py --debug run
Production Environment
- Prerequisites: Miniconda → Linux
- Create virtual environment:
conda create --name {name} python={version}
- Activate virtual environment:
conda activate {name}
- Clone repository:
git clone https://github.com/jbonda/ml-workflow-pipeline.git
- Navigate to the source directory:
cd ml-workflow-pipeline/ML_GUI
- Install dependencies:
conda install -c anaconda {package}
- Run program:
gunicorn --config gunicorn_config.py app:app
System Benchmarking
###### Node.js
- Navigate to the source directory: `cd benchmark/src`
- Install dependencies: `npm i`
- Run the development script: `npm run devstart`
###### Hybrid
- Supplementary resources are available within the [`benchmark`](https://github.com/jbonda/ml-workflow-pipeline/tree/main/benchmark) directory.
###### .NET
- [ML.NET Tutorial - Get started in 10 minutes](https://dotnet.microsoft.com/en-us/learn/ml-dotnet/get-started-tutorial/intro)
## Project Diagrams
