How to choose the window size of CNN in deep learning

0 votes

In Convolutional Neural Network (CNN), a filter is select for weights sharing. For example, in the following pictures, a 3x3 window with the stride (distance between adjacent neurons) 1 is chosen.

So my question is: How to choose the window size? If I use 4x4 with the stride being 2, how much difference will it cause? Thanks a lot in advance

Mar 26, 2022 in Machine Learning by Nandini
• 5,480 points
944 views

1 answer to this question.

0 votes

There is no definitive solution to this question: filter size is one of the hyperparameters that must be tuned in most cases. There are, however, some important remarks that may be of assistance to you. Smaller filters with a greater number of them are frequently desired.
Four 5x5 filters, for example, have 100 parameters (ignoring bias), while ten 3x3 filters have 90. You may still capture the variety of features in the image with the larger of filters, but with fewer parameters.
Modern CNNs take this concept a step further by using 3x1 and 1x3 convolutional layers in succession. This further reduces the number of parameters, but has no effect on performance.
The stride chosen is similarly significant, but it has an impact on the tensor shape after convolution, and hence the entire network. The basic norm is to use stride=1 in regular convolutions and padding to preserve the spatial size, and stride=2 when downsampling the image.

To learn more, visit our deep learning course

answered Mar 30, 2022 by Dev
• 6,000 points

Related Questions In Machine Learning

+1 vote
1 answer

How to load CNN model in keras?

Hi@akhtar, You can load your pre-created model in ...READ MORE

answered May 18, 2020 in Machine Learning by MD
• 95,440 points
1,658 views
0 votes
1 answer

How to save CNN model in keras?

Hi@akhtar, You can save your CNN model in keras. For ...READ MORE

answered May 19, 2020 in Machine Learning by MD
• 95,440 points
1,684 views
0 votes
1 answer

How to import the BatchNormalization function in Keras?

Hi@akhtar, The general use case is to use ...READ MORE

answered Jul 29, 2020 in Machine Learning by MD
• 95,440 points
3,069 views
0 votes
1 answer

What is Depth of a convolutional neural network?

The depth of a Deep Neural Network ...READ MORE

answered Mar 25, 2022 in Machine Learning by Nandini
• 5,480 points
1,041 views
+1 vote
2 answers

What is the difference between correlation and covariance?

Correlation and Co-variance both are used as ...READ MORE

answered Jul 24, 2018 in Data Analytics by Abhi
• 3,720 points
3,715 views
0 votes
1 answer

How do I become a data scientist step by step?

I am assuming that you are a ...READ MORE

answered Jul 26, 2018 in Data Analytics by Abhi
• 3,720 points
542 views
0 votes
1 answer

Which machine learning classifier to choose, in general?

Choice of Machine Learning Classifier depends upon ...READ MORE

answered Feb 21, 2022 in Machine Learning by Dev
• 6,000 points
320 views
0 votes
1 answer
webinar REGISTER FOR FREE WEBINAR X
REGISTER NOW
webinar_success Thank you for registering Join Edureka Meetup community for 100+ Free Webinars each month JOIN MEETUP GROUP