A Channel Pruning Algorithm Based on Depth-Wise Separable Convolution Unit
Deep learning has made significant progress in many fields such as image identification, speech recognition and natural language processing, especially in the field of computer vision. The better performance of the neural network often built on deeper, wider network structure, more network parameter...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2019-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/8918317/ |
| _version_ | 1797360168153907200 |
|---|---|
| author | Ke Zhang Ken Cheng Jingjing Li Yuanyuan Peng |
| author_facet | Ke Zhang Ken Cheng Jingjing Li Yuanyuan Peng |
| author_sort | Ke Zhang |
| collection | DOAJ |
| description | Deep learning has made significant progress in many fields such as image identification, speech recognition and natural language processing, especially in the field of computer vision. The better performance of the neural network often built on deeper, wider network structure, more network parameters and more storage and often computational expensive. As a result, it is hard to deploy neural network to mobile and embedded devices. Therefore, compressing of convolutional neural networks is very necessary and practical. In this paper, we propose a channel pruning algorithm for depth-wise separable convolution units and introduce a new channel selection algorithm based on information gain and a method for quickly recovering network performance after pruning. The proposed method is implemented on MobileNet and validated on several popular datasets. The experimental results show that our method can achieve better experimental results on several image classification datasets, and also achieve good detection results on the PASCAL VOC image detection dataset. |
| first_indexed | 2024-03-08T15:35:22Z |
| format | Article |
| id | doaj.art-3504dca2d6d34ef9955300068aadadf6 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-03-08T15:35:22Z |
| publishDate | 2019-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-3504dca2d6d34ef9955300068aadadf62024-01-10T00:04:18ZengIEEEIEEE Access2169-35362019-01-01717329417330910.1109/ACCESS.2019.29569768918317A Channel Pruning Algorithm Based on Depth-Wise Separable Convolution UnitKe Zhang0https://orcid.org/0000-0001-9696-4944Ken Cheng1Jingjing Li2https://orcid.org/0000-0002-5504-2529Yuanyuan Peng3School of Computer Science and Engineering, University of Electronic Science and Technology of China, Chengdu, ChinaSchool of Information and Communication Engineering, University of Electronic Science and Technology of China, Chengdu, ChinaSchool of Computer Science and Engineering, University of Electronic Science and Technology of China, Chengdu, ChinaSchool of Computer Science and Engineering, University of Electronic Science and Technology of China, Chengdu, ChinaDeep learning has made significant progress in many fields such as image identification, speech recognition and natural language processing, especially in the field of computer vision. The better performance of the neural network often built on deeper, wider network structure, more network parameters and more storage and often computational expensive. As a result, it is hard to deploy neural network to mobile and embedded devices. Therefore, compressing of convolutional neural networks is very necessary and practical. In this paper, we propose a channel pruning algorithm for depth-wise separable convolution units and introduce a new channel selection algorithm based on information gain and a method for quickly recovering network performance after pruning. The proposed method is implemented on MobileNet and validated on several popular datasets. The experimental results show that our method can achieve better experimental results on several image classification datasets, and also achieve good detection results on the PASCAL VOC image detection dataset.https://ieeexplore.ieee.org/document/8918317/Deep learningchannel pruningconvolutional neural networksdepth-wise separable convolution unit |
| spellingShingle | Ke Zhang Ken Cheng Jingjing Li Yuanyuan Peng A Channel Pruning Algorithm Based on Depth-Wise Separable Convolution Unit IEEE Access Deep learning channel pruning convolutional neural networks depth-wise separable convolution unit |
| title | A Channel Pruning Algorithm Based on Depth-Wise Separable Convolution Unit |
| title_full | A Channel Pruning Algorithm Based on Depth-Wise Separable Convolution Unit |
| title_fullStr | A Channel Pruning Algorithm Based on Depth-Wise Separable Convolution Unit |
| title_full_unstemmed | A Channel Pruning Algorithm Based on Depth-Wise Separable Convolution Unit |
| title_short | A Channel Pruning Algorithm Based on Depth-Wise Separable Convolution Unit |
| title_sort | channel pruning algorithm based on depth wise separable convolution unit |
| topic | Deep learning channel pruning convolutional neural networks depth-wise separable convolution unit |
| url | https://ieeexplore.ieee.org/document/8918317/ |
| work_keys_str_mv | AT kezhang achannelpruningalgorithmbasedondepthwiseseparableconvolutionunit AT kencheng achannelpruningalgorithmbasedondepthwiseseparableconvolutionunit AT jingjingli achannelpruningalgorithmbasedondepthwiseseparableconvolutionunit AT yuanyuanpeng achannelpruningalgorithmbasedondepthwiseseparableconvolutionunit AT kezhang channelpruningalgorithmbasedondepthwiseseparableconvolutionunit AT kencheng channelpruningalgorithmbasedondepthwiseseparableconvolutionunit AT jingjingli channelpruningalgorithmbasedondepthwiseseparableconvolutionunit AT yuanyuanpeng channelpruningalgorithmbasedondepthwiseseparableconvolutionunit |