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
s920243400
PaddleDetection
提交
6fdd5de2
P
PaddleDetection
项目概览
s920243400
/
PaddleDetection
与 Fork 源项目一致
Fork自
PaddlePaddle / PaddleDetection
通知
2
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
P
PaddleDetection
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
提交
6fdd5de2
编写于
6月 06, 2018
作者:
Y
yi.wu
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
update
上级
8893cf12
变更
1
显示空白变更内容
内联
并排
Showing
1 changed file
with
8 addition
and
5 deletion
+8
-5
benchmark/fluid/fluid_benchmark.py
benchmark/fluid/fluid_benchmark.py
+8
-5
未找到文件。
benchmark/fluid/fluid_benchmark.py
浏览文件 @
6fdd5de2
...
@@ -266,7 +266,10 @@ def train(avg_loss, infer_prog, optimizer, train_reader, test_reader, batch_acc,
...
@@ -266,7 +266,10 @@ def train(avg_loss, infer_prog, optimizer, train_reader, test_reader, batch_acc,
# FIXME(wuyi): For use_reader_op, if the current
# FIXME(wuyi): For use_reader_op, if the current
# pass is not the last, the last batch of this pass
# pass is not the last, the last batch of this pass
# is also equal to args.batch_size.
# is also equal to args.batch_size.
num_samples
+=
len
(
args
.
batch_size
)
if
args
.
use_reader_op
:
num_samples
+=
args
.
batch_size
else
:
num_samples
+=
len
(
data
)
train_losses
.
append
(
loss
)
train_losses
.
append
(
loss
)
print
(
"Pass: %d, Iter: %d, Loss: %f
\n
"
%
print
(
"Pass: %d, Iter: %d, Loss: %f
\n
"
%
(
pass_id
,
iters
,
np
.
mean
(
train_losses
)))
(
pass_id
,
iters
,
np
.
mean
(
train_losses
)))
...
@@ -350,9 +353,6 @@ def train_parallel(avg_loss, infer_prog, optimizer, train_reader, test_reader,
...
@@ -350,9 +353,6 @@ def train_parallel(avg_loss, infer_prog, optimizer, train_reader, test_reader,
if
iters
==
args
.
skip_batch_num
:
if
iters
==
args
.
skip_batch_num
:
start_time
=
time
.
time
()
start_time
=
time
.
time
()
num_samples
=
0
num_samples
=
0
# NOTE: if use reader ops, the input data is not splited to multiple cards
if
args
.
use_reader_op
and
iters
>=
args
.
iterations
/
args
.
gpus
:
break
if
args
.
use_fake_data
or
args
.
use_reader_op
:
if
args
.
use_fake_data
or
args
.
use_reader_op
:
try
:
try
:
loss
,
=
exe
.
run
([
avg_loss
.
name
])
loss
,
=
exe
.
run
([
avg_loss
.
name
])
...
@@ -362,6 +362,9 @@ def train_parallel(avg_loss, infer_prog, optimizer, train_reader, test_reader,
...
@@ -362,6 +362,9 @@ def train_parallel(avg_loss, infer_prog, optimizer, train_reader, test_reader,
loss
,
=
exe
.
run
([
avg_loss
.
name
],
feed
=
feeder
.
feed
(
data
))
loss
,
=
exe
.
run
([
avg_loss
.
name
],
feed
=
feeder
.
feed
(
data
))
if
args
.
update_method
==
"pserver"
:
if
args
.
update_method
==
"pserver"
:
exe
.
bcast_params
()
exe
.
bcast_params
()
if
args
.
use_reader_op
:
num_samples
+=
args
.
batch_size
else
:
num_samples
+=
len
(
data
)
num_samples
+=
len
(
data
)
iters
+=
1
iters
+=
1
if
batch_id
%
1
==
0
:
if
batch_id
%
1
==
0
:
...
...
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