Analysis of Layer Efficiency and Layer Reduction on Pre-trained CNN Models

碩士 === 國立臺灣科技大學 === 電機工程系 === 106 === Deep learning still encounters several issues like overfitting and oversize due to the use of a large number of layers. The huge size greatly constrains performance and portability of deep learning models in limited environments like embedded and IoT devices. In...

Full description

Bibliographic Details
Main Author:	Brilian Tafjira Nugraha
Other Authors:	Shun-Feng Su
Format:	Others
Language:	en_US
Published:	2018
Online Access:	http://ndltd.ncl.edu.tw/handle/2cvk37

id	ndltd-TW-106NTUS5442128
record_format	oai_dc
spelling	ndltd-TW-106NTUS54421282019-05-16T00:59:40Z http://ndltd.ncl.edu.tw/handle/2cvk37 Analysis of Layer Efficiency and Layer Reduction on Pre-trained CNN Models Analysis of Layer Efficiency and Layer Reduction on Pre-trained CNN Models Brilian Tafjira Nugraha Brilian Tafjira Nugraha 碩士國立臺灣科技大學電機工程系 106 Deep learning still encounters several issues like overfitting and oversize due to the use of a large number of layers. The huge size greatly constrains performance and portability of deep learning models in limited environments like embedded and IoT devices. In this study, we reported our analysis of activation and gradient output and weight in each layer of the pre-trained models of VGG-16 and custom AlexNet to measure the efficiency of its layers. The efficiencies are estimated by using our measurements and compared with the manual layer reduction to validate the most relevant method. The method for multiple layer reductions is used for validation. With this found approach, the time of one-layer reduction and re-training processes on both models can save up to 9 folds and 5 folds respectively without significant accuracy reduction. Shun-Feng Su 蘇順豐 2018 學位論文 ; thesis 48 en_US
collection	NDLTD
language	en_US
format	Others
sources	NDLTD
description	碩士 === 國立臺灣科技大學 === 電機工程系 === 106 === Deep learning still encounters several issues like overfitting and oversize due to the use of a large number of layers. The huge size greatly constrains performance and portability of deep learning models in limited environments like embedded and IoT devices. In this study, we reported our analysis of activation and gradient output and weight in each layer of the pre-trained models of VGG-16 and custom AlexNet to measure the efficiency of its layers. The efficiencies are estimated by using our measurements and compared with the manual layer reduction to validate the most relevant method. The method for multiple layer reductions is used for validation. With this found approach, the time of one-layer reduction and re-training processes on both models can save up to 9 folds and 5 folds respectively without significant accuracy reduction.
author2	Shun-Feng Su
author_facet	Shun-Feng Su Brilian Tafjira Nugraha Brilian Tafjira Nugraha
author	Brilian Tafjira Nugraha Brilian Tafjira Nugraha
spellingShingle	Brilian Tafjira Nugraha Brilian Tafjira Nugraha Analysis of Layer Efficiency and Layer Reduction on Pre-trained CNN Models
author_sort	Brilian Tafjira Nugraha
title	Analysis of Layer Efficiency and Layer Reduction on Pre-trained CNN Models
title_short	Analysis of Layer Efficiency and Layer Reduction on Pre-trained CNN Models
title_full	Analysis of Layer Efficiency and Layer Reduction on Pre-trained CNN Models
title_fullStr	Analysis of Layer Efficiency and Layer Reduction on Pre-trained CNN Models
title_full_unstemmed	Analysis of Layer Efficiency and Layer Reduction on Pre-trained CNN Models
title_sort	analysis of layer efficiency and layer reduction on pre-trained cnn models
publishDate	2018
url	http://ndltd.ncl.edu.tw/handle/2cvk37
work_keys_str_mv	AT briliantafjiranugraha analysisoflayerefficiencyandlayerreductiononpretrainedcnnmodels AT briliantafjiranugraha analysisoflayerefficiencyandlayerreductiononpretrainedcnnmodels
_version_	1719172437773385728

Analysis of Layer Efficiency and Layer Reduction on Pre-trained CNN Models

Similar Items