Skip to content
This repository has been archived by the owner on Mar 23, 2024. It is now read-only.

CAMs show abnormal peak at lower left corner for all input samples #48

Open
dipayandas97 opened this issue Mar 20, 2020 · 4 comments
Open

Comments

@dipayandas97
Copy link

I constructed a chexnet model using DenseNet121 and loaded our pre-trained weights. On generating CAMs of the final conv layer, an unwanted region of high activations are found on all output CAMs. This is regardless of input samples.

a1
a2
n1
n2

First two are outputs from abnormal class and last two from normal class

@armiro
Copy link

armiro commented May 2, 2020

Same here.
@dipayandas97 could you find any probable reason for that or how to correct it?

@mwestt
Copy link

mwestt commented May 8, 2020

I would also be interested to hear some rationale behind what's causing this, as I've just encountered it.

@naomity
Copy link

naomity commented Jan 12, 2021

same issue even for my own implementation/training, any possible reasons?

@Aliktk
Copy link

Aliktk commented Jun 2, 2021

@armiro @naomity @dipayandas97 can you please tell me about all the dependencies including Cuda and NumPy etc version I will be very thankful. I am trying from almost one month. Please help me in this regard

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

No branches or pull requests

5 participants