Home - DSGT-DLP/Deep-Learning-Playground GitHub Wiki
Welcome to Deep Learning Playground
About
Deep Learning Playground is a project aimed at democratizing access to Machine Learning and Deep Learning in a low-code/no-code manner. Our team strongly believes that Machine Learning and Deep Learning can and will be done without the necessity to write code. In order to accelerate to the future, we have built interfaces where users can load in their datasets, select the configurations, drag/drop the building blocks. With one click of a button, the user will get helpful analytics in order to help them prototype their model and we give code that they can directly copy/paste into their individual project.
TLDR: Make ML and DL as easy as drag/drop and clicking buttons just like Scratch.
How to Run on localhost
See README.
We use NPM/NodeJS for package management on our frontend and Anaconda for the backend
prod
mode
How to Run Docker container in See How to Run Docker container
Functionalities
Functionality is best shown through the website. You can access the GIF on our About
page.
You have the ability to upload or link your data file, drag/drop the building blocks (ie: layers), specify your loss function and other important metadata and your model will train and give you the results + model architecture files
Contact Information for our Dev Team
See https://docs.google.com/spreadsheets/d/1fTgHjGjfxqjAjwfxhxVbPynJS8r5ahXU6enNK9tpG5w/edit?usp=sharing