Skip to content
This repository was archived by the owner on Oct 6, 2025. It is now read-only.

Conversation

@rmcpantoja
Copy link
Contributor

@rmcpantoja rmcpantoja commented Aug 15, 2023

Fixes #219, #234, #237, #307, #587.

XxNessuxX and others added 25 commits March 10, 2024 20:34
* Downgrade pip to avoid errors.

* Continue from last checkpoint.

* Disable streaming by default

* Update notebooks URLs

* Fix Typo

* Look for metdata in zip

---------

Co-authored-by: noname <[email protected]>
* update to latest pytorch/lightning/whatever

* add arguments for monitor and monitor mode

* fix parameter

* early stopping / proper progress bar / dynamic learning rate

* add options and adjust defaults

* improve logging and tensor support

* add optional plot with --show-plot

* allow saving plot to file as well as / instead of drawing it to screen

* remove requirement

* fix setting learning rate on existing models with new override parameter and add weight decay option

* graph learning rates

* hopefully fix graph

* add support for gradient clipping (probably)

* add option for reduceLRonplateau, set other params to defaults as before

* minor fix

* add random seed function

* HiFi-GAN parametrization according to newer torch.

---------

Co-authored-by: Matt Jeanes <[email protected]>
* Training notebook: software updates.

* Inference notebook: compatible with newest software.

* Exporter notebook: make compatible again.

* Training notebook: restore improvements of #2.
@synesthesiam
Copy link
Member

Development has moved: https://github.com/OHF-Voice/piper1-gpl
This is some incredible work! Would you be interested in having access to part of the new repo to maintain the notebook?

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Error while pre-processing

4 participants