DeepGraph: A PyCharm Tool for Visualizing and Understanding Deep Learning Models

Qiang Hu, Lei Ma, Jianjun Zhao

研究成果: 著書/レポートタイプへの貢献会議での発言

1 引用 (Scopus)

抄録

As more and more domain specific big data become available, there comes a strong need on the fast development and deployment of deep learning (DL) systems with high quality for domain specific applications, including many safety-critical scenarios. In traditional software engineering, software visualization plays an important role to enhance developers' performance with many tools available. However, there are limited visualization supports existing for DL systems, especially in integrated development environments (IDEs) that allow a developer to visualize the source code of a deep neural network (DNN) and its graph architecture. In this paper, we propose DeepGraph, a visualization tool for visualizing and understanding a deep neural network. DeepGraph analyzes the training program to construct the graph representation of a DNN, and establishes and maintains the linkage (mapping) between the source code of the training program and its corresponding neural network architecture. We implemented DeepGraph as a PyCharm plugin and performed preliminary empirical study to demonstrate its usefulness for understanding deep nueral networks.

元の言語英語
ホスト出版物のタイトルProceedings - 25th Asia-Pacific Software Engineering Conference, APSEC 2018
出版者IEEE Computer Society
ページ628-632
ページ数5
ISBN(電子版)9781728119700
DOI
出版物ステータス出版済み - 7 2 2018
イベント25th Asia-Pacific Software Engineering Conference, APSEC 2018 - Nara, 日本
継続期間: 12 4 201812 7 2018

出版物シリーズ

名前Proceedings - Asia-Pacific Software Engineering Conference, APSEC
2018-December
ISSN(印刷物)1530-1362

会議

会議25th Asia-Pacific Software Engineering Conference, APSEC 2018
日本
Nara
期間12/4/1812/7/18

Fingerprint

Visualization
Learning systems
Network architecture
Software engineering
Neural networks
Deep neural networks
Deep learning
Big data

All Science Journal Classification (ASJC) codes

  • Software

これを引用

Hu, Q., Ma, L., & Zhao, J. (2018). DeepGraph: A PyCharm Tool for Visualizing and Understanding Deep Learning Models. : Proceedings - 25th Asia-Pacific Software Engineering Conference, APSEC 2018 (pp. 628-632). [8719435] (Proceedings - Asia-Pacific Software Engineering Conference, APSEC; 巻数 2018-December). IEEE Computer Society. https://doi.org/10.1109/APSEC.2018.00079

DeepGraph : A PyCharm Tool for Visualizing and Understanding Deep Learning Models. / Hu, Qiang; Ma, Lei; Zhao, Jianjun.

Proceedings - 25th Asia-Pacific Software Engineering Conference, APSEC 2018. IEEE Computer Society, 2018. p. 628-632 8719435 (Proceedings - Asia-Pacific Software Engineering Conference, APSEC; 巻 2018-December).

研究成果: 著書/レポートタイプへの貢献会議での発言

Hu, Q, Ma, L & Zhao, J 2018, DeepGraph: A PyCharm Tool for Visualizing and Understanding Deep Learning Models. : Proceedings - 25th Asia-Pacific Software Engineering Conference, APSEC 2018., 8719435, Proceedings - Asia-Pacific Software Engineering Conference, APSEC, 巻. 2018-December, IEEE Computer Society, pp. 628-632, 25th Asia-Pacific Software Engineering Conference, APSEC 2018, Nara, 日本, 12/4/18. https://doi.org/10.1109/APSEC.2018.00079
Hu Q, Ma L, Zhao J. DeepGraph: A PyCharm Tool for Visualizing and Understanding Deep Learning Models. : Proceedings - 25th Asia-Pacific Software Engineering Conference, APSEC 2018. IEEE Computer Society. 2018. p. 628-632. 8719435. (Proceedings - Asia-Pacific Software Engineering Conference, APSEC). https://doi.org/10.1109/APSEC.2018.00079
Hu, Qiang ; Ma, Lei ; Zhao, Jianjun. / DeepGraph : A PyCharm Tool for Visualizing and Understanding Deep Learning Models. Proceedings - 25th Asia-Pacific Software Engineering Conference, APSEC 2018. IEEE Computer Society, 2018. pp. 628-632 (Proceedings - Asia-Pacific Software Engineering Conference, APSEC).
@inproceedings{66e35fdea59f4bc58408858fae316db4,
title = "DeepGraph: A PyCharm Tool for Visualizing and Understanding Deep Learning Models",
abstract = "As more and more domain specific big data become available, there comes a strong need on the fast development and deployment of deep learning (DL) systems with high quality for domain specific applications, including many safety-critical scenarios. In traditional software engineering, software visualization plays an important role to enhance developers' performance with many tools available. However, there are limited visualization supports existing for DL systems, especially in integrated development environments (IDEs) that allow a developer to visualize the source code of a deep neural network (DNN) and its graph architecture. In this paper, we propose DeepGraph, a visualization tool for visualizing and understanding a deep neural network. DeepGraph analyzes the training program to construct the graph representation of a DNN, and establishes and maintains the linkage (mapping) between the source code of the training program and its corresponding neural network architecture. We implemented DeepGraph as a PyCharm plugin and performed preliminary empirical study to demonstrate its usefulness for understanding deep nueral networks.",
author = "Qiang Hu and Lei Ma and Jianjun Zhao",
year = "2018",
month = "7",
day = "2",
doi = "10.1109/APSEC.2018.00079",
language = "English",
series = "Proceedings - Asia-Pacific Software Engineering Conference, APSEC",
publisher = "IEEE Computer Society",
pages = "628--632",
booktitle = "Proceedings - 25th Asia-Pacific Software Engineering Conference, APSEC 2018",
address = "United States",

}

TY - GEN

T1 - DeepGraph

T2 - A PyCharm Tool for Visualizing and Understanding Deep Learning Models

AU - Hu, Qiang

AU - Ma, Lei

AU - Zhao, Jianjun

PY - 2018/7/2

Y1 - 2018/7/2

N2 - As more and more domain specific big data become available, there comes a strong need on the fast development and deployment of deep learning (DL) systems with high quality for domain specific applications, including many safety-critical scenarios. In traditional software engineering, software visualization plays an important role to enhance developers' performance with many tools available. However, there are limited visualization supports existing for DL systems, especially in integrated development environments (IDEs) that allow a developer to visualize the source code of a deep neural network (DNN) and its graph architecture. In this paper, we propose DeepGraph, a visualization tool for visualizing and understanding a deep neural network. DeepGraph analyzes the training program to construct the graph representation of a DNN, and establishes and maintains the linkage (mapping) between the source code of the training program and its corresponding neural network architecture. We implemented DeepGraph as a PyCharm plugin and performed preliminary empirical study to demonstrate its usefulness for understanding deep nueral networks.

AB - As more and more domain specific big data become available, there comes a strong need on the fast development and deployment of deep learning (DL) systems with high quality for domain specific applications, including many safety-critical scenarios. In traditional software engineering, software visualization plays an important role to enhance developers' performance with many tools available. However, there are limited visualization supports existing for DL systems, especially in integrated development environments (IDEs) that allow a developer to visualize the source code of a deep neural network (DNN) and its graph architecture. In this paper, we propose DeepGraph, a visualization tool for visualizing and understanding a deep neural network. DeepGraph analyzes the training program to construct the graph representation of a DNN, and establishes and maintains the linkage (mapping) between the source code of the training program and its corresponding neural network architecture. We implemented DeepGraph as a PyCharm plugin and performed preliminary empirical study to demonstrate its usefulness for understanding deep nueral networks.

UR - http://www.scopus.com/inward/record.url?scp=85066814056&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85066814056&partnerID=8YFLogxK

U2 - 10.1109/APSEC.2018.00079

DO - 10.1109/APSEC.2018.00079

M3 - Conference contribution

AN - SCOPUS:85066814056

T3 - Proceedings - Asia-Pacific Software Engineering Conference, APSEC

SP - 628

EP - 632

BT - Proceedings - 25th Asia-Pacific Software Engineering Conference, APSEC 2018

PB - IEEE Computer Society

ER -