mirror of
https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI.git
synced 2024-12-29 19:15:04 +08:00
Update Changelog_CN.md
This commit is contained in:
parent
dce1baefb3
commit
64fbc89047
@ -1,6 +1,6 @@
|
||||
20230409
|
||||
|
||||
修正训练参数,提升显卡平均利用率,A100最高从25%提升至90%左右,V100:50%->90%左右,2060S:60%->90%左右,P40:25%->90%左右,训练速度显著提升
|
||||
修正训练参数,提升显卡平均利用率,A100最高从25%提升至90%左右,V100:50%->90%左右,2060S:60%->85%左右,P40:25%->95%左右,训练速度显著提升
|
||||
|
||||
修正参数:总batch_size改为每张卡的batch_size
|
||||
|
||||
@ -11,3 +11,5 @@
|
||||
修复分布式训练每个rank都保存一次ckpt的问题
|
||||
|
||||
特征提取进行nan特征过滤
|
||||
|
||||
修复静音输入输出随机辅音or噪声的问题(老版模型需要重做训练集重训)
|
||||
|
Loading…
Reference in New Issue
Block a user