reinforcement-learning human-computer-interaction human-in-the-loop human-in-the-loop-machine-learning human-in-the-loop-rl

Simple-Human-in-the-Loop-ML-Interface

Description

A simple webpage-based human interface to facilitate human-in-the-loop machine learning research, which allows researchers to:

remotely interact with (provide feedback) learning agent on server/clusters
easily and efficiently annotate images
replay any trajectory with specific frame time
easily customize the interface to meet their needs (the interface is implemented with basic python Flask, HTML and JavaScript)

Getting Started

Run the app.py as normal Flask program

python app.py

Go to http://127.0.0.1:5000/
Control and provide feedback through keyboard or buttons
Create bounding boxes by simple click-hold-drag-release

Keyboard Input

You can define your own keyboard setting in function keyboardInputHandler in the file templates/index.html. Current setting is:

A: positive feedback (+1)
S: negative feedback (-1)
D: no feedback (0)
C: clear existing feedback
>: next frame
<: previous frame
P: auto play/stop auto play

Some Examples of Customizing the Interface

Object Tracking: object-tracking-example

Reference

Simple Image Annotator: https://github.com/sgp715/simple_image_annotator
Widening the Pipeline in Human-Guided Reinforcement Learning with Explanation and Context-Aware Data Augmentation: https://arxiv.org/abs/2006.14804

About

A simple human interface for human-in-the-loop machine learning research, which allows: 1. annote image on webpage, 2. collect human feedback through keyboard input, 3. direct interaction with agents on remote server

reinforcement-learning human-computer-interaction human-in-the-loop human-in-the-loop-machine-learning human-in-the-loop-rl

Languages

Language:HTML 54.5%Language:Jupyter Notebook 33.9%Language:Python 11.6%