Scribd Logo
Upload

Welcome to Scribd!

  • 26 views

    CNN - Continued Convolution Neural Networks (Continued)

    Uploaded by

    Ijaz Fazil
    CNN

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    CNN - Continued Convolution Neural Networks (Continued)

    Uploaded by

    Ijaz Fazil
    26 views5 pages
    CNN

    Original Title

    3._CNN_-_Continued

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    CNN

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    26 views5 pages

    CNN - Continued Convolution Neural Networks (Continued)

    Uploaded by

    Ijaz Fazil
    CNN

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 5

    3. CNN - Continued | Lecture 12. Convolutional ... https://courses.edx.org/courses/course-v1:MITx+...

    Lecture 12. Convolutional Neural


    Course  Unit 3 Neural networks (2.5 weeks)  Networks 
    3. CNN - Continued

    Audit Access Expires May 11, 2020


    You lose all access to this course, including your progress, on May 11, 2020.
    Upgrade by Apr 1, 2020 to get unlimited access to the course as long as it
    exists on the site. Upgrade now

    3. CNN - Continued
    Convolution Neural Networks (Continued)

    1 of 5 2020-03-17, 12:42 p.m.


    3. CNN - Continued | Lecture 12. Convolutional ... https://courses.edx.org/courses/course-v1:MITx+...

     0:00 / 0:00  1.50x    

    Video Transcripts
    Download video �le Download SubRip (.srt) �le
    Download Text (.txt) �le

    CNN - Numerical Example


    1/1 point (graded)
    In this problem, we are going to work out the outputs of a tiny toy example of CNN
    that is made up of just one conv layer consisting of just one �lter F of shape 2 × 2
    followed by a max-pooling layer of shape 2 × 2. The input image is of shape 3 × 3

    The output of the CNN is calculated as Pool (ReLU (Conv (I))) where ReLU is
    the recti�ed linear activation function given by:

    ReLU (x) = max (0, x)

    Also assume that the stride for the convolution and pool layers is 1

    For the following values of the image I and �lter weights F enter below the value of
    the output of the CNN (hint - it will be a single integer):

    2 of 5 2020-03-17, 12:42 p.m.


    3. CNN - Continued | Lecture 12. Convolutional ... https://courses.edx.org/courses/course-v1:MITx+...

    ⎡1 0 2⎤
    I = ⎢3 1 0⎥
    ⎣0 0 4⎦

    1 0
    F =[ ]
    0 1

    5  Answer: 5

    Solution:

    First let's calculate the output of the convolutional layer

    ⎡1 0 2⎤
    I = ⎢3 1 0⎥
    ⎣0 0 4⎦

    1 0
    F =[ ]
    0 1

    ⎡1 0 2⎤ 1 0
    Conv (I) = ⎢ 3 1 0 ⎥ . [ ]
    ⎣0 0 4⎦ 0 1

    2 0
    Conv (I) = [ ]
    3 5

    3 of 5 2020-03-17, 12:42 p.m.


    3. CNN - Continued | Lecture 12. Convolutional ... https://courses.edx.org/courses/course-v1:MITx+...

    2 0
    ReLU (Conv (I)) = ReLU ([ ])
    3 5

    2 0
    ReLU (Conv (I)) = [ ]
    3 5

    2 0
    Pool (ReLU (Conv (I))) = Pool ([ ])
    3 5

    Pool (ReLU (Conv (I))) = 5

    You have used 3 of 3 attempts

     Answers are displayed within the problem

    CNN Meaning
    1/1 point (graded)
    If you are trying to recognize a large number of features, you should have a small
    number of �lters.

    true

    false

    Solution:

    4 of 5 2020-03-17, 12:42 p.m.


    3. CNN - Continued | Lecture 12. Convolutional ... https://courses.edx.org/courses/course-v1:MITx+...

    Each �lter represents a distinct set of weights, which corresponds to searching for a
    particular feature in the image. If you have a large number of features, you want
    many �lters.

    You have used 2 of 2 attempts

     Answers are displayed within the problem

    Discussion Hide Discussion


    Topic: Unit 3 Neural networks (2.5 weeks):Lecture 12.
    Convolutional Neural Networks / 3. CNN - Continued

    Add a Post

    Show all posts by recent activity

    There are no posts in this topic yet.

    Learn About Veri�ed Certi�cates


    © All Rights Reserved

    5 of 5 2020-03-17, 12:42 p.m.

    You might also like