Lower Accuracy Even with Batch norm.

[learner-codec]

Hi thanks for the amazing explanation of the implementation. However I failed to get the reported accuracy reported in the BLOG. With batch Norm I get an accuracy of around 36% which is also visible in the Notebook you have posted.

But on the second block. The accuracy becomes 70% somehow. Please Let me know if I am missing something.

[dpetrini]

I run the code in posted notebook and had same issue, Accuracy around 32% (instead of reported around 70%).

Even with MLP BN is set to True as pointed in blog post.

[dpetrini]

Hello,

Actually it seems that if you fix incorrect ordering of loss, as commented in issue " Incorrect Specification of loss function. #1 ", it will work like in the blog and will get correct accuracy.
It worked for me.