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
MegEngine 天元
MegEngine
提交
48822557
MegEngine
项目概览
MegEngine 天元
/
MegEngine
9 个月 前同步成功
通知
392
Star
4702
Fork
582
代码
文件
提交
分支
Tags
贡献者
分支图
Diff
Issue
0
列表
看板
标记
里程碑
合并请求
0
DevOps
流水线
流水线任务
计划
Wiki
0
Wiki
分析
仓库
DevOps
项目成员
Pages
MegEngine
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
Issue
0
Issue
0
列表
看板
标记
里程碑
合并请求
0
合并请求
0
Pages
DevOps
DevOps
流水线
流水线任务
计划
分析
分析
仓库分析
DevOps
Wiki
0
Wiki
成员
成员
收起侧边栏
关闭侧边栏
动态
分支图
创建新Issue
流水线任务
提交
Issue看板
前往新版Gitcode,体验更适合开发者的 AI 搜索 >>
提交
48822557
编写于
4月 03, 2020
作者:
M
Megvii Engine Team
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
docs(mge/bn): fix docs and tests of batchnorm
GitOrigin-RevId: 8a96aa5fc221df1dbcf185b24d82cf7cf2a3cc24
上级
2c4d1afe
变更
1
隐藏空白更改
内联
并排
Showing
1 changed file
with
9 addition
and
3 deletion
+9
-3
python_module/megengine/module/batchnorm.py
python_module/megengine/module/batchnorm.py
+9
-3
未找到文件。
python_module/megengine/module/batchnorm.py
浏览文件 @
48822557
...
@@ -126,7 +126,7 @@ class BatchNorm2d(_BatchNorm):
...
@@ -126,7 +126,7 @@ class BatchNorm2d(_BatchNorm):
By default, during training this layer keeps running estimates of its
By default, during training this layer keeps running estimates of its
computed mean and variance, which are then used for normalization during
computed mean and variance, which are then used for normalization during
evaluation. The running estimates are kept with a default :attr:`momentum`
evaluation. The running estimates are kept with a default :attr:`momentum`
of 0.
1
.
of 0.
9
.
If :attr:`track_running_stats` is set to ``False``, this layer will not
If :attr:`track_running_stats` is set to ``False``, this layer will not
keep running estimates, and batch statistics are instead used during
keep running estimates, and batch statistics are instead used during
...
@@ -154,7 +154,7 @@ class BatchNorm2d(_BatchNorm):
...
@@ -154,7 +154,7 @@ class BatchNorm2d(_BatchNorm):
:type momentum: float
:type momentum: float
:param momentum: the value used for the `running_mean` and `running_var`
:param momentum: the value used for the `running_mean` and `running_var`
computation.
computation.
Default: 0.
1
Default: 0.
9
:type affine: bool
:type affine: bool
:param affine: a boolean value that when set to ``True``, this module has
:param affine: a boolean value that when set to ``True``, this module has
learnable affine parameters. Default: ``True``
learnable affine parameters. Default: ``True``
...
@@ -174,12 +174,18 @@ class BatchNorm2d(_BatchNorm):
...
@@ -174,12 +174,18 @@ class BatchNorm2d(_BatchNorm):
# With Learnable Parameters
# With Learnable Parameters
m = M.BatchNorm2d(4)
m = M.BatchNorm2d(4)
inp = mge.tensor(np.random.rand(
64, 4, 32, 32
))
inp = mge.tensor(np.random.rand(
1, 4, 3, 3).astype("float32"
))
oup = m(inp)
oup = m(inp)
print(m.weight, m.bias)
# Without Learnable Parameters
# Without Learnable Parameters
m = M.BatchNorm2d(4, affine=False)
m = M.BatchNorm2d(4, affine=False)
oup = m(inp)
oup = m(inp)
print(m.weight, m.bias)
.. testoutput::
Tensor([1. 1. 1. 1.]) Tensor([0. 0. 0. 0.])
None None
"""
"""
def
_check_input_ndim
(
self
,
inp
):
def
_check_input_ndim
(
self
,
inp
):
...
...
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