Hot answers tagged

3 votes
Accepted

Why do we need 10 bits to represent the 1000 classes in AlexNet?

You need 10-bits ($2^{10} = 1024$) to represent 1000 classes.
2 votes
Accepted

Can the (sparse) categorical cross-entropy be greater than one?

Both the sparse categorical cross-entropy (SCE) and the categorical cross-entropy (CCE) can be greater than $1$. By the way, they are the same exact loss function: the only difference is really the ...
  • 34.9k
1 vote
Accepted

How is a neural network where the majority of inputs are 0 trained?

It's the loss function. I was using squared sum error, which I didn't think would have as a negative effect as it does, and I had to come to the explanation in my own time. Here's why: From the ...
  • 1,266
1 vote

Why do we need 10 bits to represent the 1000 classes in AlexNet?

It takes at least 10 bits to represent any number between $1-1000$ because $2^{10} = 1024$. This means that if one was trying to represent 1 of the 1000 classes, one would need at least 10 bits. ...

Only top scored, non community-wiki answers of a minimum length are eligible