https://hwiyong.tistory.com/45
1x1 convolution이란,
GoogLeNet 즉, Inception network에서는 1x1 convolution을 통해 연산량을 줄였습니다. 이만큼 더 깊은 신경망을 만들어 우수한 결과를 만들어냈죠. 1x1 conv에는 크게 3가지의 장점이 있는 것 같습니다. #channel..
hwiyong.tistory.com
https://dataplay.tistory.com/28
8. CNN - 1x1, 3x3, 5x5, 예제 코드
앞으로의 포스팅에서는 CNN에 관련 되어 있는 몇가지 구조에 대해서 설명을 하고, 실험도 해본 결과를 포스팅 해보겠습니다. 1. Factorizing Convolutions. VGGNet, GoogleNet 등을 보시면 알게되는 개념 입니다...
dataplay.tistory.com
http://blog.naver.com/laonple/220692793375
[Part Ⅴ. Best CNN Architecture] 5. GoogLeNet [2] - 라온피플 머신러닝 아카데미 -
Part I. Machine Learning Part V. Best CNN Architecture Part VII. Semantic ...
blog.naver.com
A Gentle Introduction to 1x1 Convolutions to Manage Model Complexity
Pooling can be used to down sample the content of feature maps, reducing their width and height whilst maintaining their salient features. A problem with deep convolutional neural networks is that the number of feature maps often increases with the depth o
machinelearningmastery.com
https://iamaaditya.github.io/2016/03/one-by-one-convolution/
One by One [ 1 x 1 ] Convolution - counter-intuitively useful
Whenever I discuss or show GoogleNet architecture, one question always comes up -
iamaaditya.github.io
https://stats.stackexchange.com/questions/194142/what-does-1x1-convolution-mean-in-a-neural-network
What does 1x1 convolution mean in a neural network?
I am currently doing the Udacity Deep Learning Tutorial. In Lesson 3, they talk about a 1x1 convolution. This 1x1 convolution is used in Google Inception Module. I'm having trouble understanding wh...
stats.stackexchange.com