Analysis of the Application Efficiency of TensorFlow and PyTorch in Convolutional Neural Network
In this paper, we present an analysis of important aspects that arise during the development of neural network applications. Our aim is to determine if the choice of library can impact the system’s overall performance, either during training or design, and to extract a set of criteria that could be...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-11-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/22/22/8872 |
| _version_ | 1827643652580573184 |
|---|---|
| author | Ovidiu-Constantin Novac Mihai Cristian Chirodea Cornelia Mihaela Novac Nicu Bizon Mihai Oproescu Ovidiu Petru Stan Cornelia Emilia Gordan |
| author_facet | Ovidiu-Constantin Novac Mihai Cristian Chirodea Cornelia Mihaela Novac Nicu Bizon Mihai Oproescu Ovidiu Petru Stan Cornelia Emilia Gordan |
| author_sort | Ovidiu-Constantin Novac |
| collection | DOAJ |
| description | In this paper, we present an analysis of important aspects that arise during the development of neural network applications. Our aim is to determine if the choice of library can impact the system’s overall performance, either during training or design, and to extract a set of criteria that could be used to highlight the advantages and disadvantages of each library under consideration. To do so, we first extracted the previously mentioned aspects by comparing two of the most popular neural network libraries—PyTorch and TensorFlow—and then we performed an analysis on the obtained results, with the intent of determining if our initial hypothesis was correct. In the end, the results of the analysis are gathered, and an overall picture of what tasks are better suited for what library is presented. |
| first_indexed | 2024-03-09T18:00:07Z |
| format | Article |
| id | doaj.art-7814f1b1bb2b4ba6b9e0b3715acf1506 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-09T18:00:07Z |
| publishDate | 2022-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-7814f1b1bb2b4ba6b9e0b3715acf15062023-11-24T09:57:10ZengMDPI AGSensors1424-82202022-11-012222887210.3390/s22228872Analysis of the Application Efficiency of TensorFlow and PyTorch in Convolutional Neural NetworkOvidiu-Constantin Novac0Mihai Cristian Chirodea1Cornelia Mihaela Novac2Nicu Bizon3Mihai Oproescu4Ovidiu Petru Stan5Cornelia Emilia Gordan6Department of Computers and Information Technology, Electrical Engineering and Information Technology Faculty, University of Oradea, 410087 Oradea, RomaniaDepartment of Computers and Information Technology, Electrical Engineering and Information Technology Faculty, University of Oradea, 410087 Oradea, RomaniaDepartment of Electrical Engineering, Electrical Engineering and Information Technology Faculty, University of Oradea, 410087 Oradea, RomaniaDepartment of Electronics, Computers and Electrical Engineering, Faculty of Electronics, Telecommunication, and Computer Science, University of Pitesti, 110040 Pitesti, RomaniaDepartment of Electronics, Computers and Electrical Engineering, Faculty of Electronics, Telecommunication, and Computer Science, University of Pitesti, 110040 Pitesti, RomaniaDepartment of Automation, Faculty of Automation and Computer Science, Technical University of Cluj-Napoca, 400114 Cluj-Napoca, RomaniaDepartment of Electronics and Telecommunications, Electrical Engineering and Information Technology Faculty, University of Oradea, 410087 Oradea, RomaniaIn this paper, we present an analysis of important aspects that arise during the development of neural network applications. Our aim is to determine if the choice of library can impact the system’s overall performance, either during training or design, and to extract a set of criteria that could be used to highlight the advantages and disadvantages of each library under consideration. To do so, we first extracted the previously mentioned aspects by comparing two of the most popular neural network libraries—PyTorch and TensorFlow—and then we performed an analysis on the obtained results, with the intent of determining if our initial hypothesis was correct. In the end, the results of the analysis are gathered, and an overall picture of what tasks are better suited for what library is presented.https://www.mdpi.com/1424-8220/22/22/8872convolutional neural networkTensorFlowPyTorchnetwork trainingnetwork design |
| spellingShingle | Ovidiu-Constantin Novac Mihai Cristian Chirodea Cornelia Mihaela Novac Nicu Bizon Mihai Oproescu Ovidiu Petru Stan Cornelia Emilia Gordan Analysis of the Application Efficiency of TensorFlow and PyTorch in Convolutional Neural Network Sensors convolutional neural network TensorFlow PyTorch network training network design |
| title | Analysis of the Application Efficiency of TensorFlow and PyTorch in Convolutional Neural Network |
| title_full | Analysis of the Application Efficiency of TensorFlow and PyTorch in Convolutional Neural Network |
| title_fullStr | Analysis of the Application Efficiency of TensorFlow and PyTorch in Convolutional Neural Network |
| title_full_unstemmed | Analysis of the Application Efficiency of TensorFlow and PyTorch in Convolutional Neural Network |
| title_short | Analysis of the Application Efficiency of TensorFlow and PyTorch in Convolutional Neural Network |
| title_sort | analysis of the application efficiency of tensorflow and pytorch in convolutional neural network |
| topic | convolutional neural network TensorFlow PyTorch network training network design |
| url | https://www.mdpi.com/1424-8220/22/22/8872 |
| work_keys_str_mv | AT ovidiuconstantinnovac analysisoftheapplicationefficiencyoftensorflowandpytorchinconvolutionalneuralnetwork AT mihaicristianchirodea analysisoftheapplicationefficiencyoftensorflowandpytorchinconvolutionalneuralnetwork AT corneliamihaelanovac analysisoftheapplicationefficiencyoftensorflowandpytorchinconvolutionalneuralnetwork AT nicubizon analysisoftheapplicationefficiencyoftensorflowandpytorchinconvolutionalneuralnetwork AT mihaioproescu analysisoftheapplicationefficiencyoftensorflowandpytorchinconvolutionalneuralnetwork AT ovidiupetrustan analysisoftheapplicationefficiencyoftensorflowandpytorchinconvolutionalneuralnetwork AT corneliaemiliagordan analysisoftheapplicationefficiencyoftensorflowandpytorchinconvolutionalneuralnetwork |