GoogLeNet - 편집 역사

2017년 7월 3일 (월) 02:29에 Admin님의 편집

2017-07-03T02:29:17Z

Admin: /* graphical view */

2017-07-03T02:28:37Z

graphical view

2017년 6월 28일 (수) 03:59에 Admin님의 편집

2017-06-28T03:59:10Z

차이 보기

2017년 5월 23일 (화) 02:23에 Admin님의 편집

2017-05-23T02:23:21Z

2017년 5월 23일 (화) 02:22에 Admin님의 편집

2017-05-23T02:22:54Z

Admin: Admin님이 GoogleNet 문서를 GoogLeNet 문서로 이동했습니다

2017-05-23T02:20:10Z

Admin님이 GoogleNet 문서를 GoogLeNet 문서로 이동했습니다

Admin: 새 문서: * 17 Sep 2014 * [https://arxiv.org/pdf/1409.4842.pdf pdf] * abstract
We propose a deep convolutional neural network architecture codenamed "Inception", which was responsibl...

2017-05-23T02:18:40Z

새 문서: * 17 Sep 2014 * [https://arxiv.org/pdf/1409.4842.pdf pdf] * abstract<blockquote>We propose a deep convolutional neural network architecture codenamed "Inception", which was responsibl...

새 문서

* 17 Sep 2014
* [https://arxiv.org/pdf/1409.4842.pdf pdf]
* abstract<blockquote>We propose a deep convolutional neural network architecture codenamed "Inception", which was responsible for setting the new state of the art for classification and detection in the ImageNet Large-Scale Visual Recognition Challenge 2014 (ILSVRC 2014). The main hallmark of this architecture is the improved utilization of the computing resources inside the network. This was achieved by a carefully crafted design that allows for increasing the depth and width of the network while keeping the computational budget constant. To optimize quality, the architectural decisions were based on the Hebbian principle and the intuition of multi-scale processing. One particular incarnation used in our submission for ILSVRC 2014 is called GoogLeNet, a 22 layers deep network, the quality of which is assessed in the context of classification and detection.</blockquote>

← 이전 판		2017년 7월 3일 (월) 02:29 판
1번째 줄:		1번째 줄:
		+	==paper review==
	* 17 Sep 2014		* 17 Sep 2014
	* [https://arxiv.org/pdf/1409.4842.pdf pdf]		* [https://arxiv.org/pdf/1409.4842.pdf pdf]

← 이전 판		2017년 7월 3일 (월) 02:28 판
585번째 줄:		585번째 줄:
	</g>		</g>
	</svg></html>		</svg></html>
		+
		+	위 net의 코드를 [https://github.com/dmlc/mxnet/blob/master/example/image-classification/symbols/inception-bn.py github]에서 볼 수 있다. <del>아무리 봐도 tensorflow보다 간결하고 직관적이란 말이야...</del>
		+
		+	==a==

← 이전 판		2017년 5월 23일 (화) 02:23 판
2번째 줄:		2번째 줄:
	* [https://arxiv.org/pdf/1409.4842.pdf pdf]		* [https://arxiv.org/pdf/1409.4842.pdf pdf]
	* abstract<blockquote>We propose a deep convolutional neural network architecture codenamed "Inception", which was responsible for setting the new state of the art for classification and detection in the ImageNet Large-Scale Visual Recognition Challenge 2014 (ILSVRC 2014). The main hallmark of this architecture is the improved utilization of the computing resources inside the network. This was achieved by a carefully crafted design that allows for increasing the depth and width of the network while keeping the computational budget constant. To optimize quality, the architectural decisions were based on the [https://en.wikipedia.org/wiki/Hebbian_theory Hebbian principle] and the intuition of multi-scale processing. One particular incarnation used in our submission for ILSVRC 2014 is called GoogLeNet, a 22 layers deep network, the quality of which is assessed in the context of classification and detection.</blockquote>		* abstract<blockquote>We propose a deep convolutional neural network architecture codenamed "Inception", which was responsible for setting the new state of the art for classification and detection in the ImageNet Large-Scale Visual Recognition Challenge 2014 (ILSVRC 2014). The main hallmark of this architecture is the improved utilization of the computing resources inside the network. This was achieved by a carefully crafted design that allows for increasing the depth and width of the network while keeping the computational budget constant. To optimize quality, the architectural decisions were based on the [https://en.wikipedia.org/wiki/Hebbian_theory Hebbian principle] and the intuition of multi-scale processing. One particular incarnation used in our submission for ILSVRC 2014 is called GoogLeNet, a 22 layers deep network, the quality of which is assessed in the context of classification and detection.</blockquote>
		+	* Hebbian principle?

@@ 1번째 줄: / 1번째 줄: @@
 * 17 Sep 2014
 * [https://arxiv.org/pdf/1409.4842.pdf pdf]
-* abstract<blockquote>We propose a deep convolutional neural network architecture codenamed "Inception", which was responsible for setting the new state of the art for classification and detection in the ImageNet Large-Scale Visual Recognition Challenge 2014 (ILSVRC 2014). The main hallmark of this architecture is the improved utilization of the computing resources inside the network. This was achieved by a carefully crafted design that allows for increasing the depth and width of the network while keeping the computational budget constant. To optimize quality, the architectural decisions were based on the Hebbian principle and the intuition of multi-scale processing. One particular incarnation used in our submission for ILSVRC 2014 is called GoogLeNet, a 22 layers deep network, the quality of which is assessed in the context of classification and detection.</blockquote>
+* abstract<blockquote>We propose a deep convolutional neural network architecture codenamed "Inception", which was responsible for setting the new state of the art for classification and detection in the ImageNet Large-Scale Visual Recognition Challenge 2014 (ILSVRC 2014). The main hallmark of this architecture is the improved utilization of the computing resources inside the network. This was achieved by a carefully crafted design that allows for increasing the depth and width of the network while keeping the computational budget constant. To optimize quality, the architectural decisions were based on the [https://en.wikipedia.org/wiki/Hebbian_theory Hebbian principle] and the intuition of multi-scale processing. One particular incarnation used in our submission for ILSVRC 2014 is called GoogLeNet, a 22 layers deep network, the quality of which is assessed in the context of classification and detection.</blockquote>

← 이전 판	2017년 5월 23일 (화) 02:20 판
(차이 없음)