-
Notifications
You must be signed in to change notification settings - Fork 12.7k
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
[IDEA] Improvement the "Stateful RNN" section in chapter 16. #431
Comments
Hi @YRFT , Thanks for your great feedback. Indeed, I wasn't very rigorous with the I initially used Thanks again! |
In the "Stateful RNN" section in chapter 16, the code
steps_per_epoch = train_size // batch_size // n_steps
seems not appropriate.When we created the datasets, we used different values for
size
andshift
when callingwindow
method (withdrop_remainder=True
). As a result, the actual value of steps per ephoch is not always equal totrain_size // batch_size // n_steps
. For example:will output:
So, the resetting operations might not execute as expected.
I have created the datasets without calling the
repeat
method, and commentedsteps_per_epoch=steps_per_epoch
in thefit
method.The text was updated successfully, but these errors were encountered: