Josh Katofsky

Same issue

While training, it should save your model weights periodically - files which can then be used to construct new textgenrnn objects. Is that not working for you?

Made a PR to fix, since it may not always be easy to tweak the local installation of textgenrnn, for example when using this module on a cloud service and...

@ShadiestGoat now both the master branch and the TF 2.0 branch are not using the outdated import. Not sure about the status of the PyPi package, but you can always...

Set the keyword variable `gen_epochs=0` on the train method call. This will not run `generate()` on every epoch. EDIT: if you are referring to the progress bar, you can set...

Can set the `max_length` argument on the generate functions. This defines the longest possible output, but the network will still output anything smaller than it.

Thanks for this solution!