Abstract
A novel framework is presented, which simplifies the integration of machine learning into systems for industrial inspection and testing. In contrast to most approaches utilizing a centralized setup, the proposed work follows an edge-computing paradigm. The scope is not limited to inspection tasks but includes all requirements connected to such tasks. The support for continual and distributed learning, as well as distributed accumulation of training data, is a crucial feature of the proposed system. An integrated user rights management allows for the collaboration of multiple people with different background of expertise and tasks on the same machine learning models. Through platform-independent design and the use of a progressive web app as a user-interface, this framework supports the deployment in heterogeneous systems. Separation of concerns and clean object-oriented design makes the framework highly extensible and adaptable to other domains.
Source code available: https://github.com/koi-learning.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Abadi, M., Agarwal, A., Barham, P., et al.: TensorFlow: large-scale machine learning on heterogeneous distributed systems. https://arxiv.org/pdf/1603.04467
Abadi, M., Barham, P., Chen, J., et al.: TensorFlow: a system for large-scale machine learning. In: Proceedings of OSDI ’16: 12th USENIX Symposium on Operating Systems Design and Implementation, pp. 265–283. USENIX Association, Berkeley (2016)
Caffe: Convolutional architecture for fast feature embedding (2014)
Paszke, A., Gross, S., Massa, F., et al.: PyTorch: an imperative style, high-performance deep learning library. In: Wallach, H., Larochelle, H., Beygelzimer, A., et al. (eds.) Advances in Neural Information Processing Systems, vol. 32, pp. 8026–8037. Curran Associates, Inc. (2019). https://proceedings.neurips.cc/paper/2019/file/bdbca288fee7f92f2bfa9f7012727740-Paper.pdf
Chollet, F., et al.: Keras (2015). https://keras.io. Accessed 22 Aug 2020
The Theano Development Team, Al-Rfou, R., Alain, G., et al.: Theano: A python framework for fast computation of mathematical expressions. arXiv p. arXiv:1605.02688 (2016)
Shultz, T.R., Fahlman, S.E., Craw, S., et al.: Continual learning. In: Sammut, C., Webb, G.I. (eds.) Encyclopedia of Machine Learning, p. 226. Springer, New York (2011). https://doi.org/10.1007/978-0-387-30164-8_171
Kakas, A.C., Cohn, D., Dasgupta, S., et al.: Active learning. In: Sammut, C., Webb, G.I. (eds.) Encyclopedia of Machine Learning, pp. 10–14. Springer, New York (2011). https://doi.org/10.1007/978-0-387-30164-8_6
GitHub: Nvidia/tensorrt. https://github.com/NVIDIA/TensorRT. Accessed 15 Aug 2020
GitHub: onnx/onnx. https://github.com/onnx/onnx. Accessed 15 Aug 2020
GitHub: cortexlabs/cortex. https://github.com/cortexlabs/cortex. Accessed 15 Aug 2020
TensorFlow: TensorFlow serving with docker TFX. https://www.tensorflow.org/tfx/serving/docker. Accessed 16 Aug 2020
Sergeev, A., Del Balso, M.: Horovod: fast and easy distributed deep learning in TensorFlow. https://arxiv.org/pdf/1802.05799
GitHub: allegroai/trains. https://github.com/allegroai/trains/. Accessed 15 Aug 2020
Simon Brown: The c4 model for visualising software architecture (2020). https://c4model.com/. Accessed 20 July 2020
Fielding, R.: Architectural styles and the design of network-based software architectures. https://www.ics.uci.edu/~fielding/pubs/dissertation/fielding_dissertation_2up.pdf. Accessed 21 July 2020
Russell, A.: Progressive web apps: escaping tabs without losing our soul - infrequently noted. https://infrequently.org/2015/06/progressive-apps-escaping-tabs-without-losing-our-soul/. Accessed 15 Aug 2020
Python Software Foundation: Python 3.8.5 documentation. https://docs.python.org/3/. Accessed 15 Aug 2020
Leach, P., Mealling, M., Salz, R.: RFC 4122 - a universally unique identifier UUID URN namespace. https://tools.ietf.org/pdf/rfc4122.pdf. Accessed 15 Aug 2020
The Pallets Projects: Flask. https://palletsprojects.com/p/flask/. Accessed 20 Aug 2020
The Polymer Project Authors: Litelement (2020). https://lit-element.polymer-project.org/. Accessed 22 Aug 2020
Abramov, D.: The Redux Development Team: Redux (2020). https://redux.js.org/. Accessed 22 Aug 2020
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Richter, J., Nau, J., Kirchhoff, M., Streitferdt, D. (2021). KOI: An Architecture and Framework for Industrial and Academic Machine Learning Applications. In: Simian, D., Stoica, L.F. (eds) Modelling and Development of Intelligent Systems. MDIS 2020. Communications in Computer and Information Science, vol 1341. Springer, Cham. https://doi.org/10.1007/978-3-030-68527-0_8
Download citation
DOI: https://doi.org/10.1007/978-3-030-68527-0_8
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-68526-3
Online ISBN: 978-3-030-68527-0
eBook Packages: Computer ScienceComputer Science (R0)