From e36b5eeb8d95ed9b1cb6c9bd13488e336122153d Mon Sep 17 00:00:00 2001 From: Tulio <tulio.tutui@gmail.com> Date: Wed, 22 Mar 2023 21:39:09 +0100 Subject: [PATCH] Restructure README and fix question 1 --- README.md | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/README.md b/README.md index fc32938..2bad3c0 100644 --- a/README.md +++ b/README.md @@ -12,12 +12,14 @@ The plot of the losses has a unexpected behavior: the generator loss increases i # Part 2 -### Question 1 + +## Questions +### Question 1: Knowing the input and output images will be 256x256, what will be the dimension of the encoded vector x8 ? -Since this parameter is hard coded, the size of x8 will always be 512x512 +Since this parameter is hard coded, the size of x8 will always be 1x1. This is due to the fact that each encoder layer decreases the size by a factor of 2, and since we have 8 encoder layer (including `inconv`), we have $256/2^8 = 1$. -### Question 2 +### Question 2: As you can see, U-net has an encoder-decoder architecture with skip connections. Explain why it works better than a traditional encoder-decoder. The standard encoder-decoder architecture has a limitation that the decoder may not be able to reconstruct details from the compressed image, leading to information loss. Thus, the U-net introduces skip connections that allow the decoder to better recover details and spacial information from the source image. \ No newline at end of file -- GitLab