Bugfix:

- onnx export Documentation - notice for AMD user to use DirectML Version
hpx502766238 · Jun 29, 2023 · 8090196 · 8090196
1 parent 03f5697
commit 8090196
Show file tree

Hide file tree

Showing 3 changed files with 5 additions and 1 deletion.
diff --git a/README.md b/README.md
@@ -92,6 +92,8 @@ Windows 版と Mac 版を提供しています。
 
 - RVC で使用する場合の GUI の各項目説明は[こちら](tutorials/tutorial_rvc_ja_latest.md)をご覧ください
 
+- AMDのGPUはDirectML版ONNXでのみ使用できる場合があります。（実験的実装です。）
+
 - ダウンロードはこちらから。
 
 | Version    | OS  | フレームワーク                            | link                                                                                                                                                            | サポート VC                                                                   | サイズ |

diff --git a/README_en.md b/README_en.md
@@ -87,6 +87,8 @@ We offer Windows and Mac versions.
 
 - Please refer to [here](tutorials/tutorial_rvc_en_latest.md) for the description of each item of GUI to be used in RVC.
 
+- AMD GPUs may only be used with DirectML-version ONNX in some cases (as it is an experimental implementation).
+
 - Download (When you cannot download from google drive, try [hugging_face](https://huggingface.co/wok000/vcclient000/tree/main))
 
 | Version    | OS  | Framework                                 | link                                                                                                                                                             | support VC                                                                    | size   |

diff --git a/server/voice_changer/RVC/inferencer/rvc_models/infer_pack/commons.py b/server/voice_changer/RVC/inferencer/rvc_models/infer_pack/commons.py
@@ -95,7 +95,7 @@ def subsequent_mask(length):
 
 
 # @torch.jit.script
-@torch.jit._script_if_tracing
+# @torch.jit._script_if_tracing
 def fused_add_tanh_sigmoid_multiply(input_a, input_b, n_channels):
     n_channels_int = n_channels[0]
     in_act = input_a + input_b