In CNN, the convolution and pooling layers may change the size of the feature map. A simple CNN is given below with input image size of 32x32x3. In each layer, several filters are applied with different filter size (f), stride (s), and padding (p). Compute the missing values (a, b, c, d, e, f). a= b= POOL1 Max Pool d filters OTÖTÖTT? f=2 s=2 14x14xc s=1 axaxb p=0 C= d= e= 32 x 32 x 3 6 filters f=5 s=1 CONV1 p=0 f=5 CONV2 p=0 Max Pool f=2 POOL2 0= 10x10x16 5=2 exexf p=0
In CNN, the convolution and pooling layers may change the size of the feature map. A simple CNN is given below with input image size of 32x32x3. In each layer, several filters are applied with different filter size (f), stride (s), and padding (p). Compute the missing values (a, b, c, d, e, f). a= b= POOL1 Max Pool d filters OTÖTÖTT? f=2 s=2 14x14xc s=1 axaxb p=0 C= d= e= 32 x 32 x 3 6 filters f=5 s=1 CONV1 p=0 f=5 CONV2 p=0 Max Pool f=2 POOL2 0= 10x10x16 5=2 exexf p=0
Related questions
Question
THIS IS NOT A GRADED ASSIGNMENT. THIS IS PRACTICE QUESTIONS.
Transcribed Image Text:In CNN, the convolution and pooling layers may change the size of the feature
map. A simple CNN is given below with input image size of 32x32x3. In each layer,
several filters are applied with different filter size (f), stride (s), and padding (p).
Compute the missing values (a, b, c, d, e, f).
a=
b=
POOL1
Max Pool
d filters
OrÖTÖTT?
f=2
s=2 14x14xc s=1
axaxb
p=0
C=
d=
e=
f=
32 x 32 x 3
6 filters
f=5
s=1
CONV1
p=0
f=5
CONV2
p=0
Max Pool
f=2
POOL2
10 x 10 x 16 = 2 exexf
p=0
Expert Solution
This question has been solved!
Explore an expertly crafted, step-by-step solution for a thorough understanding of key concepts.
This is a popular solution!
Trending now
This is a popular solution!
Step by step
Solved in 3 steps
