yqzhishen 8784ed65d3 Support LYNXNet on main branch (#218) ...

* [DONE]New AUX_Decoder/Backbone Network : LYNXNet (#200)

* Update __init__.py

* Update LYNXNet

* add dropout

* Lynxnet outnorm (#206)

* post-norm

* fix

* add norm+mlp

* Update LYNXNet.py

* Update LYNXNetDecoder.py

* do not need mlp

* do not need mlp

* Add out norm for LYNXNET

* Add out norm for LYNXNETDecoder

* delete lynxnet aux_decoder (#212)

* refactor configuration options

* fix onnx exporter for lynxnet

* Add Pytorch version check when export onnx (#216)

* recommended lynxnet hyperparameters for variance models

* remove invalid items

* Refactor code

* Finish configuration schemas

---------

Co-authored-by: KakaruHayate <97896816+KakaruHayate@users.noreply.github.com>
Co-authored-by: yxlllc <llc1995@sina.com>

2024-11-16 00:09:35 +08:00

..

benchmarks

Finish benchmarks

2023-04-10 12:15:06 +08:00

exporters

Force vocoder to only accept mel_base = 'e'

2024-07-12 23:47:09 +08:00

modules

Support LYNXNet on main branch (#218)

2024-11-16 00:09:35 +08:00

__init__.py

Using VR model to estimate variance parameters (#196)

2024-06-19 23:59:52 +08:00

.gitignore

Ignore everything in assets/

2023-04-11 23:09:25 +08:00