Garuda is a Web-UI based automated machine learning (Auto-ML) platform. The main goal of the tool is to provide an easy-to-use and less puzzling way to build, optimize, and deploy machine learning models for any problem domain. This tool was originally developed for synthetic biology applications. It gives an access for non machine learning experts; with no programming background, to work with a variety built-in machine learning models to solve a problem of their own, and/or machine learning experts to automate the repetitive works in building and experimenting the modeling process, so that they can focus on what matters most (i.e. feature engineering, EDA. etc.)
At this point, machine learning algorithms supported by Garuda are primarily based on some of the most popular Python machine learning libraries, such as: scikit-learn, xgboost, light-gbm, and keras/tensor-flow. The back-end server is built on top of Flask, and the front-end UI is built with Bootstrap/jQuery.
This is an on-going work that I update regularly. I am committed to release the software as an open-source, and I can always use some help from contributors who are passionate in developing tools that will provide access machine learning for everyone.
Click the following video links to see Garuda in action.
Garuda Demo #1 - Microfluidics Design Automation
Garuda Demo #2 - Genetic Parts Prediction (Cello Data)
