Skip to content
This repository has been archived by the owner on Sep 27, 2021. It is now read-only.

labeling for Resnet #7

Open
Varchita-Beena opened this issue Sep 8, 2019 · 1 comment
Open

labeling for Resnet #7

Varchita-Beena opened this issue Sep 8, 2019 · 1 comment

Comments

@Varchita-Beena
Copy link

While applying ResNet are you passing labels?
If yes then for every patch the label has been included? As there are of 49 patches per image.
So I would like to know patch wise labeling(i.e 49 times same label is repeated) is there or image wise?

@Magauiya
Copy link

Magauiya commented Oct 29, 2019

While applying ResNet are you passing labels?
If yes then for every patch the label has been included? As there are of 49 patches per image.
So I would like to know patch wise labeling(i.e 49 times same label is repeated) is there or image wise?

When you train ResNet, no need to pass labels since CPC is an unsupervised learning method. In code, you also can notice that the "train" mode does not use labels from the Fashion_MNIST.
However, when you train the last linear layer (validation mode), you need those labels and they are not repeated 49 times. It is clearly seen if you print out labels_onehot variable

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

2 participants