Year 2020,
Volume: 1 Issue: 1, 10 - 15, 01.06.2020
Ayse Ayyuce Demirbas
,
Ahmet Çınar
References
- A. M. Turing (1950). Computing Machinery and Intelligence. Mind 49: 433-460
- D. E. Rumelhart, G. E. Hinton, R. J. Williams (1988): Learning Representations by Back-Propagating Errors, Cognitive modeling
Eric B. Olsen (2017) Proposal for a High Precision Tensor Processing Unit
- Frank Rosenblatt (1958), The Perceptron: A Probabilistic Model for Information Storage and Organization in the Brain, Psychological Review Vol. 65, No. 6
- Jia Deng, Wei Dong, Richard Socher, Li-Jia Li, Kai Li and Li Fei-Fei (2009) Dept. of Computer Science, Princeton University, USA, ImageNet: A Large-Scale Hierarchical Image Database
- Martín Abadi, Paul Barham, Jianmin Chen, Zhifeng Chen, Andy Davis, Jeffrey Dean, Matthieu Devin, Sanjay Ghemawat, Geoffrey Irving, Michael Isard, Manjunath Kudlur, Josh Levenberg, Rajat Monga, Sherry Moore, Derek G. Murray, Benoit Steiner, Paul Tucker,Vijay Vasudevan, Pete Warden, Martin Wicke, Yuan Yu, and Xiaoqiang Zheng, Google Brain, (2016) TensorFlow: A system for large-scale machine learning 12th USENIX Symposium on Operating Systems Design and Implementation (OSDI ’16) November 2–4, • Savannah, GA, USA
- Norman P. Jouppi, Cliff Young, Nishant Patil, David Patterson ve diğerleri (2017) In-Datacenter Performance Analysis of a Tensor Processing Unit Google, Inc. , Mountain View, CA USA
- Tao Sheng, Chen Feng, Shaojie Zhuo, Xiaopeng Zhang, Liang Shen, Mickey Aleksic (2018), A Quantization-Friendly Separable Convolution for MobileNets
- URL-1:
https://www.guru99.com/what-is-tensorflow.html [Erişim Tarihi: 10.03.2020]
- URL-2:
https://cloud.google.com/blog/products/ai-machine-learning/what-makes-tpus-fine-tuned-for-deep-learning [Erişim Tarihi: 10.02.2020]
- URL-3: https://medium.com/@antonpaquin/whats-inside-a-tpu-c013eb51973e [Erişim Tarihi: 10.03.2020]
- URL-4:
https://machinethink.net/blog/mobilenet-v2/ [Erişim Tarihi: 5.02.2020]
- URL-5:
https://towardsdatascience.com/review-mobilenetv1-depthwise-separable-convolutionlight-weight-model-a382df364b69 [Erişim Tarihi: 10.03.2020]
- URL-6: https://gist.github.com/ayyucedemirbas/6c2d6bd9324834432df02e8083be9031 [Erişim Tarihi: 23.03.2020]
- URL-7: https://gist.github.com/ayyucedemirbas/37ce6f12deb9db99715ac398a309285c [Erişim Tarihi: 5.03.2020]
- URL-8: https://gist.github.com/ayyucedemirbas/2901b48a1b33eec1fd4794a522c7e204 [Erişim Tarihi: 23.03.2020]
- URL-9: https://coral.withgoogle.com/docs/edgetpu/benchmarks/ [Erişim Tarihi: 15.03.2020]
- Warren McCulloch, Walter Pitts (1943): A Logical Calculus of the Ideas Immanent in Nervous Activity. Bulletin of Mathematical Biophysics, Vol. 5, p. 115, 133
- Wei Wang, Yutao Li, Ting Zou, Xin Wang, Jieyu You, Yanhong Luo (2020) A Novel Image Classification Approach via Dense-MobileNet Models
Nesne Sınıflandırma İşlemi İçin Tensor İşleme Birimi ve CPU Performans Karşılaştırması
Year 2020,
Volume: 1 Issue: 1, 10 - 15, 01.06.2020
Ayse Ayyuce Demirbas
,
Ahmet Çınar
Abstract
Tensor Processing Unit (TPU), Google tarafından derin öğrenme görevlerini hızlandırmak için özel olarak geliştirilmiş bir yongadır. Yakın zamana kadar TPU’lar sadece Google Cloud ve Google Colab Platformları üzerinden kullanılabilmekteydi. 2019 yılının başlarında Google firması bu ürünü Coral adı altında donanımsal olarak üretmiştir. Bu sayede, dizüstü bilgisayar ve Raspberry Pi 3 gibi düşük donanım özelliklerine sahip cihazlarda derin öğrenme uygulamaları daha hızlı bir şekilde
gerçekleştirilebilmektedir. Bu makalede MobileNet v1 modeli kullanılarak ilk olarak TPU bulunduran Google Coral USB Accelerator ile daha sonra ise CPU kullanılarak 5 kategori ve 4326 çiçek fotoğrafından oluşan bir Kaggle veriseti eğitilmiş ve eğitilen bu verisetine dayanarak görüntülerin sınıflandırılması sağlanmıştır. Bu yolla cihazın performansı incelenmiştir. CPU için başarı oranı %84,29252, TPU için başarı oranı ise %99,609 olarak elde edilmiştir.
References
- A. M. Turing (1950). Computing Machinery and Intelligence. Mind 49: 433-460
- D. E. Rumelhart, G. E. Hinton, R. J. Williams (1988): Learning Representations by Back-Propagating Errors, Cognitive modeling
Eric B. Olsen (2017) Proposal for a High Precision Tensor Processing Unit
- Frank Rosenblatt (1958), The Perceptron: A Probabilistic Model for Information Storage and Organization in the Brain, Psychological Review Vol. 65, No. 6
- Jia Deng, Wei Dong, Richard Socher, Li-Jia Li, Kai Li and Li Fei-Fei (2009) Dept. of Computer Science, Princeton University, USA, ImageNet: A Large-Scale Hierarchical Image Database
- Martín Abadi, Paul Barham, Jianmin Chen, Zhifeng Chen, Andy Davis, Jeffrey Dean, Matthieu Devin, Sanjay Ghemawat, Geoffrey Irving, Michael Isard, Manjunath Kudlur, Josh Levenberg, Rajat Monga, Sherry Moore, Derek G. Murray, Benoit Steiner, Paul Tucker,Vijay Vasudevan, Pete Warden, Martin Wicke, Yuan Yu, and Xiaoqiang Zheng, Google Brain, (2016) TensorFlow: A system for large-scale machine learning 12th USENIX Symposium on Operating Systems Design and Implementation (OSDI ’16) November 2–4, • Savannah, GA, USA
- Norman P. Jouppi, Cliff Young, Nishant Patil, David Patterson ve diğerleri (2017) In-Datacenter Performance Analysis of a Tensor Processing Unit Google, Inc. , Mountain View, CA USA
- Tao Sheng, Chen Feng, Shaojie Zhuo, Xiaopeng Zhang, Liang Shen, Mickey Aleksic (2018), A Quantization-Friendly Separable Convolution for MobileNets
- URL-1:
https://www.guru99.com/what-is-tensorflow.html [Erişim Tarihi: 10.03.2020]
- URL-2:
https://cloud.google.com/blog/products/ai-machine-learning/what-makes-tpus-fine-tuned-for-deep-learning [Erişim Tarihi: 10.02.2020]
- URL-3: https://medium.com/@antonpaquin/whats-inside-a-tpu-c013eb51973e [Erişim Tarihi: 10.03.2020]
- URL-4:
https://machinethink.net/blog/mobilenet-v2/ [Erişim Tarihi: 5.02.2020]
- URL-5:
https://towardsdatascience.com/review-mobilenetv1-depthwise-separable-convolutionlight-weight-model-a382df364b69 [Erişim Tarihi: 10.03.2020]
- URL-6: https://gist.github.com/ayyucedemirbas/6c2d6bd9324834432df02e8083be9031 [Erişim Tarihi: 23.03.2020]
- URL-7: https://gist.github.com/ayyucedemirbas/37ce6f12deb9db99715ac398a309285c [Erişim Tarihi: 5.03.2020]
- URL-8: https://gist.github.com/ayyucedemirbas/2901b48a1b33eec1fd4794a522c7e204 [Erişim Tarihi: 23.03.2020]
- URL-9: https://coral.withgoogle.com/docs/edgetpu/benchmarks/ [Erişim Tarihi: 15.03.2020]
- Warren McCulloch, Walter Pitts (1943): A Logical Calculus of the Ideas Immanent in Nervous Activity. Bulletin of Mathematical Biophysics, Vol. 5, p. 115, 133
- Wei Wang, Yutao Li, Ting Zou, Xin Wang, Jieyu You, Yanhong Luo (2020) A Novel Image Classification Approach via Dense-MobileNet Models