Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

I don't understand the use of Hard Attention in your code. #2

Open
bankanikhil29 opened this issue Jul 19, 2019 · 0 comments
Open

I don't understand the use of Hard Attention in your code. #2

bankanikhil29 opened this issue Jul 19, 2019 · 0 comments

Comments

@bankanikhil29
Copy link

bankanikhil29 commented Jul 19, 2019

Hard attention has such low accuracy on the validation data even after 10 epochs.
Is it just used for drawing the heat map?
Why is my accuracy affected in this case compared to my soft Attention?

From the papers I learnt that Soft Attention gives to multiple objects as compared to Hard attention. To cross check this I gave my model an image having 2 digits.

Soft Attention works well by giving attention to both the digits.

test_heat map

But in Hard attention I expected it to give attention to just one of the digits. Though the result is same compared to Soft Attention and prediction is affected.

download

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

No branches or pull requests

1 participant