Skip to content

Commit

Permalink
Merge pull request #26 from JatinKumar9/patch-8
Browse files Browse the repository at this point in the history
Update Oceanverse.md
  • Loading branch information
Ashutosh-iitrpr authored Jun 21, 2024
2 parents 2319ee4 + 9d100d1 commit e48c003
Showing 1 changed file with 5 additions and 5 deletions.
10 changes: 5 additions & 5 deletions Oceanverse.md
Original file line number Diff line number Diff line change
Expand Up @@ -2366,8 +2366,8 @@ c) What kind of values you need in a filter used for blurring an image?

143.What is purpose of applying a pooling layer on an image. How is it different from convolution layer?

144.a)Apply 2*2 Max pooling on the following image.
b) Apply 2*2 avg pooling on the following image.
144.a)Apply $$ 2 \* 2 $$ Max pooling on the following image.
b) Apply $$ 2 \* 2 $$ avg pooling on the following image.
c) What are the dimensions of the image after pooling? Does pooling change the depth of the image?
d)Why don't we use Min pooling?

Expand All @@ -2379,9 +2379,9 @@ d)Why don't we use Min pooling?

147.For a weight W with a gradient ∂L/∂W=0.01, a learning rate α=0.1, and an initial weight W0=0.5, compute the updated weight using gradient descent.

148.a) Differentiate $$ \frac{1}{{1 + e^{-x}}} $$ with respect to \(x\). \\
b) Differentiate $$ -\log(2x^2) $$ with respect to \(x\). \\
c) Differentiate $$ \frac{e^i}{{\sum e^k}} $$ with respect to \(i\) and \(j\).
148.a) Differentiate \( \frac{1}{{1 + e^{-x}}} \) with respect to \(x\).
b) Differentiate \( -\log(2x^2) \) with respect to \(x\).
c) Differentiate \( \frac{e^i}{{\sum e^k}} \) with respect to \(i\) and \(j\).

149.Given is a single neuron, let the loss L be (z_cap - z)^2, find ∂L/∂w_11 and ∂L/∂p2.
![fig4](/aicamp/images/n0.jpg)
Expand Down

0 comments on commit e48c003

Please sign in to comment.