Skip to content
体验新版
项目
组织
正在加载...
登录
切换导航
打开侧边栏
Overbill1683
Stable Diffusion Webui
提交
8284ebd9
S
Stable Diffusion Webui
项目概览
Overbill1683
/
Stable Diffusion Webui
12 个月 前同步成功
通知
1779
Star
81
Fork
0
代码
文件
提交
分支
Tags
贡献者
分支图
Diff
分析
仓库
DevOps
项目成员
Pages
S
Stable Diffusion Webui
项目概览
项目概览
详情
发布
仓库
仓库
文件
提交
分支
标签
贡献者
分支图
比较
Pages
分析
分析
仓库分析
DevOps
成员
成员
收起侧边栏
关闭侧边栏
动态
分支图
提交
体验新版 GitCode,发现更多精彩内容 >>
提交
8284ebd9
编写于
7月 26, 2023
作者:
A
AUTOMATIC1111
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
fix autograd which i broke for no good reason when implementing SDXL
上级
deed8439
变更
1
隐藏空白更改
内联
并排
Showing
1 changed file
with
2 addition
and
2 deletion
+2
-2
modules/sd_hijack_clip.py
modules/sd_hijack_clip.py
+2
-2
未找到文件。
modules/sd_hijack_clip.py
浏览文件 @
8284ebd9
...
@@ -273,9 +273,9 @@ class FrozenCLIPEmbedderWithCustomWordsBase(torch.nn.Module):
...
@@ -273,9 +273,9 @@ class FrozenCLIPEmbedderWithCustomWordsBase(torch.nn.Module):
# restoring original mean is likely not correct, but it seems to work well to prevent artifacts that happen otherwise
# restoring original mean is likely not correct, but it seems to work well to prevent artifacts that happen otherwise
batch_multipliers
=
torch
.
asarray
(
batch_multipliers
).
to
(
devices
.
device
)
batch_multipliers
=
torch
.
asarray
(
batch_multipliers
).
to
(
devices
.
device
)
original_mean
=
z
.
mean
()
original_mean
=
z
.
mean
()
z
*=
batch_multipliers
.
reshape
(
batch_multipliers
.
shape
+
(
1
,)).
expand
(
z
.
shape
)
z
=
z
*
batch_multipliers
.
reshape
(
batch_multipliers
.
shape
+
(
1
,)).
expand
(
z
.
shape
)
new_mean
=
z
.
mean
()
new_mean
=
z
.
mean
()
z
*=
(
original_mean
/
new_mean
)
z
=
z
*
(
original_mean
/
new_mean
)
return
z
return
z
...
...
编辑
预览
Markdown
is supported
0%
请重试
或
添加新附件
.
添加附件
取消
You are about to add
0
people
to the discussion. Proceed with caution.
先完成此消息的编辑!
取消
想要评论请
注册
或
登录