DeepTracker: Visualizing the Training Process of Convolutional Neural Networks

TL;DR

DeepTracker is a visual analytics system designed to help experts understand and analyze the complex training processes of CNNs by visualizing training logs and data correlations, aiming to improve training efficiency.

Contribution

The paper introduces DeepTracker, a novel visualization tool that combines hierarchical indexing and cube-style visualization to explore CNN training dynamics in detail.

Findings

Enables exploration of CNN training logs at multiple levels of detail

Reveals complex correlations among training data, weights, and images

Demonstrates effectiveness through case studies on ResNet-50 and ImageNet

Abstract

Deep convolutional neural networks (CNNs) have achieved remarkable success in various fields. However, training an excellent CNN is practically a trial-and-error process that consumes a tremendous amount of time and computer resources. To accelerate the training process and reduce the number of trials, experts need to understand what has occurred in the training process and why the resulting CNN behaves as such. However, current popular training platforms, such as TensorFlow, only provide very little and general information, such as training/validation errors, which is far from enough to serve this purpose. To bridge this gap and help domain experts with their training tasks in a practical environment, we propose a visual analytics system, DeepTracker, to facilitate the exploration of the rich dynamics of CNN training processes and to identify the unusual patterns that are hidden behind the huge amount of training log. Specifically,we combine a hierarchical index mechanism and a set of hierarchical small multiples to help experts explore the entire training log from different levels of detail. We also introduce a novel cube-style visualization to reveal the complex correlations among multiple types of heterogeneous training data including neuron weights, validation images, and training iterations. Three case studies are conducted to demonstrate how DeepTracker provides its users with valuable knowledge in an industry-level CNN training process, namely in our case, training ResNet-50 on the ImageNet dataset. We show that our method can be easily applied to other state-of-the-art "very deep" CNN models.