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
BaiXuePrincess
Paddle
提交
c44005f0
P
Paddle
项目概览
BaiXuePrincess
/
Paddle
与 Fork 源项目一致
Fork自
PaddlePaddle / Paddle
通知
1
Star
1
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
P
Paddle
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
未验证
提交
c44005f0
编写于
2月 09, 2023
作者:
W
wanghuancoder
提交者:
GitHub
2月 09, 2023
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
fixoptminizer _set_auxiliary_var bug (#50335)
上级
0036316e
变更
4
隐藏空白更改
内联
并排
Showing
4 changed file
with
16 addition
and
0 deletion
+16
-0
python/paddle/distributed/fleet/meta_optimizers/ascend/ascend_optimizer.py
...tributed/fleet/meta_optimizers/ascend/ascend_optimizer.py
+4
-0
python/paddle/distributed/fleet/meta_optimizers/meta_optimizer_base.py
.../distributed/fleet/meta_optimizers/meta_optimizer_base.py
+4
-0
python/paddle/distributed/fleet/meta_parallel/sharding/group_sharded_optimizer_stage2.py
.../meta_parallel/sharding/group_sharded_optimizer_stage2.py
+4
-0
python/paddle/incubate/optimizer/lookahead.py
python/paddle/incubate/optimizer/lookahead.py
+4
-0
未找到文件。
python/paddle/distributed/fleet/meta_optimizers/ascend/ascend_optimizer.py
浏览文件 @
c44005f0
...
@@ -236,6 +236,10 @@ class AscendOptimizer(Optimizer):
...
@@ -236,6 +236,10 @@ class AscendOptimizer(Optimizer):
ret_list
.
append
(
var
)
ret_list
.
append
(
var
)
return
ret_list
return
ret_list
def
_set_auxiliary_var
(
self
,
key
,
val
):
super
().
_set_auxiliary_var
(
key
,
val
)
self
.
inner_opt
.
_set_auxiliary_var
(
key
,
val
)
def
minimize
(
def
minimize
(
self
,
self
,
loss
,
loss
,
...
...
python/paddle/distributed/fleet/meta_optimizers/meta_optimizer_base.py
浏览文件 @
c44005f0
...
@@ -25,6 +25,10 @@ class MetaOptimizerBase(Optimizer):
...
@@ -25,6 +25,10 @@ class MetaOptimizerBase(Optimizer):
self
.
meta_optimizers_white_list
=
[]
self
.
meta_optimizers_white_list
=
[]
self
.
meta_optimizers_black_list
=
[]
self
.
meta_optimizers_black_list
=
[]
def
_set_auxiliary_var
(
self
,
key
,
val
):
super
().
_set_auxiliary_var
(
key
,
val
)
self
.
inner_opt
.
_set_auxiliary_var
(
key
,
val
)
def
_set_basic_info
(
def
_set_basic_info
(
self
,
loss
,
role_maker
,
user_defined_optimizer
,
user_defined_strategy
self
,
loss
,
role_maker
,
user_defined_optimizer
,
user_defined_strategy
):
):
...
...
python/paddle/distributed/fleet/meta_parallel/sharding/group_sharded_optimizer_stage2.py
浏览文件 @
c44005f0
...
@@ -203,6 +203,10 @@ class GroupShardedOptimizerStage2(Optimizer):
...
@@ -203,6 +203,10 @@ class GroupShardedOptimizerStage2(Optimizer):
# Update optimizer parameters and adjust parameter storage and use according to rank.
# Update optimizer parameters and adjust parameter storage and use according to rank.
self
.
_update_opt_status
()
self
.
_update_opt_status
()
def
_set_auxiliary_var
(
self
,
key
,
val
):
super
().
_set_auxiliary_var
(
key
,
val
)
self
.
_optim
.
_set_auxiliary_var
(
key
,
val
)
@
paddle
.
autograd
.
no_grad
()
@
paddle
.
autograd
.
no_grad
()
def
_sync_params_and_buffers
(
self
):
def
_sync_params_and_buffers
(
self
):
"""
"""
...
...
python/paddle/incubate/optimizer/lookahead.py
浏览文件 @
c44005f0
...
@@ -144,6 +144,10 @@ class LookAhead(Optimizer):
...
@@ -144,6 +144,10 @@ class LookAhead(Optimizer):
self
.
_global_step_var
=
None
self
.
_global_step_var
=
None
self
.
_k_var
=
None
self
.
_k_var
=
None
def
_set_auxiliary_var
(
self
,
key
,
val
):
super
().
_set_auxiliary_var
(
key
,
val
)
self
.
inner_optimizer
.
_set_auxiliary_var
(
key
,
val
)
@
framework
.
dygraph_only
@
framework
.
dygraph_only
@
imperative_base
.
no_grad
@
imperative_base
.
no_grad
def
step
(
self
):
def
step
(
self
):
...
...
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