From Tinkering to Engineering: Measurements in Tensorflow Playground

TL;DR

This paper introduces TFMeter, an extension of Tensorflow Playground that visualizes neural network architectures along with information-theoretic measurements to enhance understanding and reproducibility in neural network design.

Contribution

The paper presents TFMeter, a novel interactive tool that integrates measurements into neural network visualization, aiding better engineering intuition and experimental reproducibility.

Findings

Provides real-time measurements during network construction

Enhances understanding of neural network learning dynamics

Available online for public use

Abstract

In this article, we present an extension of the Tensorflow Playground, called Tensorflow Meter (short TFMeter). TFMeter is an interactive neural network architecting tool that allows the visual creation of different architectures of neural networks. In addition to its ancestor, the playground, our tool shows information-theoretic measurements while constructing, training, and testing the network. As a result, each change results in a change in at least one of the measurements, providing for a better engineering intuition of what different architectures are able to learn. The measurements are derived from various places in the literature. In this demo, we describe our web application that is available online at http://tfmeter.icsi.berkeley.edu/ and argue that in the same way that the original Playground is meant to build an intuition about neural networks, our extension educates users on available measurements, which we hope will ultimately improve experimental design and reproducibility in the field.