Network with parallel connections GoogLeNet / Inception V3 hands-on deep learning v2 pytorch

1. GoogLeNet with parallel connections

The first neural network can do more than 100 layers.
insert image description here

insert image description here
The name of the Inception block, taken from Inception, is to go deeper and deeper into the dream.
insert image description here

insert image description here
Inception reduces the number of parameters and reduces the amount of computation. It is smaller than convolution with 3x3 Conv and 5x5 Conv directly.
insert image description here
insert image description here
insert image description here
insert image description here
insert image description here
insert image description here
insert image description here

insert image description here
insert image description here
https://cv.gluon.ai/model_zoo/classification.html

Inception V3 accuracy is much higher than VGG, from 0.7 to close to 0.8, the memory will be much larger and more complicated.
Mushen's network ResNeSt: Split-Attention Networks is the most accurate on this graph.
insert image description here
insert image description here

2. Code implementation


insert image description here
insert image description here
insert image description here

3. Q&A

    1. 1 x 1 Conv convolution is to reduce the number of channels channel
    1. The deeper the convolutional layer, the more features are learned.

reference

https://www.bilibili.com/video/BV1b5411g7Xo?p=1

Guess you like

Origin blog.csdn.net/zgpeace/article/details/124162617