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
2301_76417823
VisualDL
提交
362dae55
V
VisualDL
项目概览
2301_76417823
/
VisualDL
与 Fork 源项目一致
Fork自
PaddlePaddle / VisualDL
通知
1
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
V
VisualDL
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 >>
未验证
提交
362dae55
编写于
1月 16, 2018
作者:
Y
Yan Chunwei
提交者:
GitHub
1月 16, 2018
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
fix image one sample bug (#156)
上级
9f0872be
变更
3
隐藏空白更改
内联
并排
Showing
3 changed file
with
24 addition
and
2 deletion
+24
-2
demo/mxnet/mxnet_demo.py
demo/mxnet/mxnet_demo.py
+19
-1
visualdl/python/storage.py
visualdl/python/storage.py
+1
-1
visualdl/server/lib.py
visualdl/server/lib.py
+4
-0
未找到文件。
demo/mxnet/mxnet_demo.py
浏览文件 @
362dae55
import
numpy
as
np
import
mxnet
as
mx
import
logging
import
mxnet
as
mx
...
...
@@ -22,6 +24,8 @@ with logger.mode("train"):
# scalar0 is used to record scalar metrics while MXNet is training. We will record accuracy.
# In the visualization, we can see the accuracy is increasing as more training steps happen.
scalar0
=
logger
.
scalar
(
"scalars/scalar0"
)
image0
=
logger
.
image
(
"images/image0"
,
1
)
histogram0
=
logger
.
histogram
(
"histogram/histogram0"
,
num_buckets
=
100
)
# Record training steps
cnt_step
=
0
...
...
@@ -42,6 +46,19 @@ def add_scalar():
cnt_step
+=
1
return
_callback
def
add_image_histogram
():
def
_callback
(
iter_no
,
sym
,
arg
,
aux
):
image0
.
start_sampling
()
weight
=
arg
[
'fullyconnected1_weight'
].
asnumpy
()
shape
=
[
100
,
50
]
data
=
weight
.
flatten
()
image0
.
add_sample
(
shape
,
list
(
data
))
histogram0
.
add_record
(
iter_no
,
list
(
data
))
image0
.
finish_sampling
()
return
_callback
# Start to build CNN in MXNet, train MNIST dataset. For more info, check MXNet's official website:
# https://mxnet.incubator.apache.org/tutorials/python/mnist.html
...
...
@@ -81,7 +98,8 @@ lenet_model.fit(train_iter,
eval_metric
=
'acc'
,
# integrate our customized callback method
batch_end_callback
=
[
add_scalar
()],
num_epoch
=
2
)
epoch_end_callback
=
[
add_image_histogram
()],
num_epoch
=
5
)
test_iter
=
mx
.
io
.
NDArrayIter
(
mnist
[
'test_data'
],
None
,
batch_size
)
prob
=
lenet_model
.
predict
(
test_iter
)
...
...
visualdl/python/storage.py
浏览文件 @
362dae55
...
...
@@ -140,7 +140,7 @@ class LogWriter(object):
}
return
type2scalar
[
type
](
tag
)
def
image
(
self
,
tag
,
num_samples
,
step_cycle
):
def
image
(
self
,
tag
,
num_samples
,
step_cycle
=
1
):
"""
Create an image writer that used to write image data.
"""
...
...
visualdl/server/lib.py
浏览文件 @
362dae55
import
pprint
import
re
import
sys
import
time
import
urllib
from
tempfile
import
NamedTemporaryFile
...
...
@@ -131,6 +132,7 @@ def get_invididual_image(storage, mode, tag, step_index, max_size=80):
with
storage
.
mode
(
mode
)
as
reader
:
res
=
re
.
search
(
r
".*/([0-9]+$)"
,
tag
)
# remove suffix '/x'
offset
=
0
if
res
:
offset
=
int
(
res
.
groups
()[
0
])
tag
=
tag
[:
tag
.
rfind
(
'/'
)]
...
...
@@ -206,4 +208,6 @@ def retry(ntimes, function, time2sleep, *args, **kwargs):
try
:
return
function
(
*
args
,
**
kwargs
)
except
:
error_info
=
'
\n
'
.
join
(
map
(
str
,
sys
.
exc_info
()))
logger
.
error
(
"Unexpected error: %s"
%
error_info
)
time
.
sleep
(
time2sleep
)
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