Visualizing Dataflow Graphs of Deep Learning Models in TensorFlow

By Kanit Wongsuphasawat, Daniel Smilkov, James Wexler, Jimbo Wilson, Dandelion Mané, Doug Fritz, Dilip Krishnan, Fernanda B. Viégas, Martin Wattenberg. IEEE Transactions on Visualization and Computer Graphics, 2018

348 citations recorded by the source corpus.

We present a design study of the TensorFlow Graph Visualizer, part of the TensorFlow machine intelligence platform. This tool helps users understand complex machine learning architectures by visualizing their underlying dataflow graphs. The tool works by applying a series of graph transformations that enable standard layout techniques to produce a legible interactive diagram. To declutter the graph, we decouple non-critical nodes from the layout. To provide an overview, we build a clustered graph using the hierarchical structure annotated in the source code. To support exploration of nested structure on demand, we perform edge bundling to enable stable and responsive cluster expansion. Finally, we detect and highlight repeated structures to emphasize a model's modular composition. To demonstrate the utility of the visualizer, we describe example usage scenarios and report user feedback. Overall, users find the visualizer useful for understanding, debugging, and sharing the structures of their models.

Subfields: Computer Vision and Pattern Recognition, Information Systems, Information Systems and Management

View Paper: https://www.semanticscholar.org/paper/b71fe09f4077acdeae25681e8f0cc4483d56e4ec

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Information Systems

In Information Systems and Management

Citations in the corpus

Visual Interaction with Deep Learning Models through Collaborative Semantic Inference (2019, methodology): “Passive Observation [Understanding] Activation Atlas [12], Classification Visualization [15], DeepEyes [65], Feature classifiers [8], Tensorboard [84], Weight Visualization [76] Deconvolution [90], LIME [66], Neuron Analysis [20], Rationals [54], Saliency [55], Structured Interpretation [22], Predic”
A Multidisciplinary Survey and Framework for Design and Evaluation of Explainable AI Systems (2018, background): “eir desired attribute values. In the case of tuning DNN networks, visual analytics tools enhance designers’ ability to modify networks [165], improve training [129], and to compare different networks [211]. 6.3 AI Experts In our categorization, AI experts are machine learning scientists and engineer”
Visual Analytics in Deep Learning: An Interrogative Survey for the Next Frontiers (2018, positive): “It is encouraging to see many of the visual analytics systems recognize this importance and report on design studies conducted with AI experts before building a tool to understand the users and their needs [14], [15], [39], [43], [52].”
RuleMatrix: Visualizing and Understanding Classifiers with Rules (2018, methodology): “In the meantime, there are concrete demands in the industry to apply visualization to assist the development of machine learning systems [16, 46].”
DeepVID: Deep Visual Interpretation and Diagnosis for Image Classifiers via Knowledge Distillation (2019, background): “The model-based solutions focus on revealing the internal architecture of a DNN and the interaction between different components [19], [20], [21].”
Evaluating the Quality of Machine Learning Explanations: A Survey on Methods and Metrics (2021, methodology): “For example, TensorFlow Graph Visualizer [59] is used to visualise data flow graphs of deep learning models in TensorFlow to help users understand, debug, and share the structure of their deep learning models.”
Explainable Artificial Intelligence: a Systematic Review (2020, methodology): “Similarly, a graphical representation can be employed to represent the inner structure of a model, such as the graphs proposed in [136] where each node is a layer of the network and the edges the connections between layers.”
Manifold: A Model-Agnostic Framework for Interpretation and Diagnosis of Machine Learning Models (2018, background): “Typical solutions include visualizing the internal structure or intermediate states of the model to enhance the understanding and interpretation [12,25,41], evaluating and analyzing the performance of models or algorithms [5, 22, 33], and interactively improving the models at different development s”