Try to search your question here, if you can't find : Ask Any Question Now ?

What effect does 1×1 convolution has on quality of feature being learnt?

HomeCategory: stackoverflowWhat effect does 1×1 convolution has on quality of feature being learnt?
pragati asked 3 weeks ago

I was learning about inception module from deeplearning.ai by Andrew Ng, wherein we use 1×1 convolution to reduce computational cost. for example, if we directly apply 5×5 convolution we need to multiply for about 120M times whereas by introducing 1×1 convolution we reduce it by the factor of 10.

My question what effect does this have on the feature maps that are being learned or will the quality of feature maps differ if we had not used 1×1 convolution.

below is the screenshot of how we reduce multiplications from 120M to about 12M:

enter image description here

1 Answers
Best Answer
Mannu answered 3 weeks ago
Your Answer

19 + 4 =

Popular Tags

WP Facebook Auto Publish Powered By : XYZScripts.com