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
竹叶青lvye
fcn.berkeleyvision.org
提交
0b07df96
F
fcn.berkeleyvision.org
项目概览
竹叶青lvye
/
fcn.berkeleyvision.org
与 Fork 源项目一致
从无法访问的项目Fork
通知
4
Star
0
Fork
0
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
Wiki
0
Wiki
分析
仓库
DevOps
项目成员
Pages
F
fcn.berkeleyvision.org
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
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
提交
Issue看板
体验新版 GitCode,发现更多精彩内容 >>
提交
0b07df96
编写于
9月 09, 2016
作者:
E
Evan Shelhamer
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
comment surgery helpers for clarity
上级
03c1daed
变更
1
显示空白变更内容
内联
并排
Showing
1 changed file
with
27 addition
and
5 deletion
+27
-5
surgery.py
surgery.py
+27
-5
未找到文件。
surgery.py
浏览文件 @
0b07df96
...
...
@@ -3,6 +3,18 @@ import caffe
import
numpy
as
np
def
transplant
(
new_net
,
net
,
suffix
=
''
):
"""
Transfer weights by copying matching parameters, coercing parameters of
incompatible shape, and dropping unmatched parameters.
The coercion is useful to convert fully connected layers to their
equivalent convolutional layers, since the weights are the same and only
the shapes are different. In particular, equivalent fully connected and
convolution layers have shapes O x I and O x I x H x W respectively for O
outputs channels, I input channels, H kernel height, and W kernel width.
Both `net` to `new_net` arguments must be instantiated `caffe.Net`s.
"""
for
p
in
net
.
params
:
p_new
=
p
+
suffix
if
p_new
not
in
new_net
.
params
:
...
...
@@ -18,12 +30,10 @@ def transplant(new_net, net, suffix=''):
print
'copying'
,
p
,
' -> '
,
p_new
,
i
new_net
.
params
[
p_new
][
i
].
data
.
flat
=
net
.
params
[
p
][
i
].
data
.
flat
def
expand_score
(
new_net
,
new_layer
,
net
,
layer
):
old_cl
=
net
.
params
[
layer
][
0
].
num
new_net
.
params
[
new_layer
][
0
].
data
[:
old_cl
][...]
=
net
.
params
[
layer
][
0
].
data
new_net
.
params
[
new_layer
][
1
].
data
[
0
,
0
,
0
,:
old_cl
][...]
=
net
.
params
[
layer
][
1
].
data
def
upsample_filt
(
size
):
"""
Make a 2D bilinear kernel suitable for upsampling of the given (h, w) size.
"""
factor
=
(
size
+
1
)
//
2
if
size
%
2
==
1
:
center
=
factor
-
1
...
...
@@ -34,6 +44,9 @@ def upsample_filt(size):
(
1
-
abs
(
og
[
1
]
-
center
)
/
factor
)
def
interp
(
net
,
layers
):
"""
Set weights of each layer in layers to bilinear kernels for interpolation.
"""
for
l
in
layers
:
m
,
k
,
h
,
w
=
net
.
params
[
l
][
0
].
data
.
shape
if
m
!=
k
and
k
!=
1
:
...
...
@@ -44,3 +57,12 @@ def interp(net, layers):
raise
filt
=
upsample_filt
(
h
)
net
.
params
[
l
][
0
].
data
[
range
(
m
),
range
(
k
),
:,
:]
=
filt
def
expand_score
(
new_net
,
new_layer
,
net
,
layer
):
"""
Transplant an old score layer's parameters, with k < k' classes, into a new
score layer with k classes s.t. the first k' are the old classes.
"""
old_cl
=
net
.
params
[
layer
][
0
].
num
new_net
.
params
[
new_layer
][
0
].
data
[:
old_cl
][...]
=
net
.
params
[
layer
][
0
].
data
new_net
.
params
[
new_layer
][
1
].
data
[
0
,
0
,
0
,:
old_cl
][...]
=
net
.
params
[
layer
][
1
].
data
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