Skip to Main Content
 

Global Search Box

 
 
 

ETD Abstract Container

Abstract Header

Measurement of machine learning performance with different condition and hyperparameter

Yin, Jiaqi
http://orcid.org/0000-0001-6413-3361
http://rave.ohiolink.edu/etdc/view?acc_num=osu1587693436870594

Abstract Details

2020, Master of Science, Ohio State University, Electrical and Computer Engineering.
In this article, we tested the performance of three major machine learning frameworks under deep learning: TensorFlow, Pytorch, and MXnet.The experimental environment of the whole test was measured by GPU GTX 1060 and CPU Inter i7-8650. During the whole experiment, we mainly measured the following indicators: memory utilization, CPU utilization, GPU utilization, GPU memory utilization, accuracy, and algorithm performance. In this paper, we compare the training performance of CPU/GPU; under different batch sizes, various indicators were measured respectively, and the accuracy under different batch size and learning rate were measured.
Xiaorui Wang (Advisor)
Irem Eryilmaz (Committee Member)
41 p.
Computer Science
Machine learning, TensorFlow, MXnet, PyTorch

Recommended Citations

Citations

  • Yin, J. (2020). Measurement of machine learning performance with different condition and hyperparameter [Master's thesis, Ohio State University]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=osu1587693436870594

    APA Style (7th edition)

  • Yin, Jiaqi. Measurement of machine learning performance with different condition and hyperparameter. 2020. Ohio State University, Master's thesis. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=osu1587693436870594.

    MLA Style (8th edition)

  • Yin, Jiaqi. "Measurement of machine learning performance with different condition and hyperparameter." Master's thesis, Ohio State University, 2020. http://rave.ohiolink.edu/etdc/view?acc_num=osu1587693436870594

    Chicago Manual of Style (17th edition)

osu1587693436870594
2,269
© 2020, all rights reserved.
This open access ETD is published by The Ohio State University and OhioLINK.