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
8d6e1137
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
未验证
提交
8d6e1137
编写于
8月 31, 2021
作者:
Z
zhiboniu
提交者:
GitHub
8月 31, 2021
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
delay coco data load (#4095)
上级
f74aa666
变更
1
隐藏空白更改
内联
并排
Showing
1 changed file
with
23 addition
and
13 deletion
+23
-13
ppdet/data/source/keypoint_coco.py
ppdet/data/source/keypoint_coco.py
+23
-13
未找到文件。
ppdet/data/source/keypoint_coco.py
浏览文件 @
8d6e1137
...
...
@@ -63,6 +63,9 @@ class KeypointBottomUpBaseDataset(DetDataset):
self
.
ann_info
[
'num_joints'
]
=
num_joints
self
.
img_ids
=
[]
def
parse_dataset
(
self
):
pass
def
__len__
(
self
):
"""Get dataset length."""
return
len
(
self
.
img_ids
)
...
...
@@ -136,26 +139,30 @@ class KeypointBottomUpCocoDataset(KeypointBottomUpBaseDataset):
super
().
__init__
(
dataset_dir
,
image_dir
,
anno_path
,
num_joints
,
transform
,
shard
,
test_mode
)
ann_file
=
os
.
path
.
join
(
dataset_dir
,
anno_path
)
self
.
coco
=
COCO
(
ann_file
)
self
.
ann_file
=
os
.
path
.
join
(
dataset_dir
,
anno_path
)
self
.
shard
=
shard
self
.
test_mode
=
test_mode
def
parse_dataset
(
self
):
self
.
coco
=
COCO
(
self
.
ann_file
)
self
.
img_ids
=
self
.
coco
.
getImgIds
()
if
not
test_mode
:
if
not
self
.
test_mode
:
self
.
img_ids
=
[
img_id
for
img_id
in
self
.
img_ids
if
len
(
self
.
coco
.
getAnnIds
(
imgIds
=
img_id
,
iscrowd
=
None
))
>
0
]
blocknum
=
int
(
len
(
self
.
img_ids
)
/
shard
[
1
])
self
.
img_ids
=
self
.
img_ids
[(
blocknum
*
s
hard
[
0
]):(
blocknum
*
(
shard
[
0
]
+
1
))]
blocknum
=
int
(
len
(
self
.
img_ids
)
/
s
elf
.
s
hard
[
1
])
self
.
img_ids
=
self
.
img_ids
[(
blocknum
*
s
elf
.
shard
[
0
]):(
blocknum
*
(
self
.
shard
[
0
]
+
1
))]
self
.
num_images
=
len
(
self
.
img_ids
)
self
.
id2name
,
self
.
name2id
=
self
.
_get_mapping_id_name
(
self
.
coco
.
imgs
)
self
.
dataset_name
=
'coco'
cat_ids
=
self
.
coco
.
getCatIds
()
self
.
catid2clsid
=
dict
({
catid
:
i
for
i
,
catid
in
enumerate
(
cat_ids
)})
print
(
f
'=> num_images:
{
self
.
num_images
}
'
)
print
(
'=> num_images: {}'
.
format
(
self
.
num_images
)
)
@
staticmethod
def
_get_mapping_id_name
(
imgs
):
...
...
@@ -301,20 +308,23 @@ class KeypointBottomUpCrowdPoseDataset(KeypointBottomUpCocoDataset):
super
().
__init__
(
dataset_dir
,
image_dir
,
anno_path
,
num_joints
,
transform
,
shard
,
test_mode
)
ann_file
=
os
.
path
.
join
(
dataset_dir
,
anno_path
)
self
.
ann_file
=
os
.
path
.
join
(
dataset_dir
,
anno_path
)
self
.
shard
=
shard
self
.
test_mode
=
test_mode
self
.
coco
=
COCO
(
ann_file
)
def
parse_dataset
(
self
):
self
.
coco
=
COCO
(
self
.
ann_file
)
self
.
img_ids
=
self
.
coco
.
getImgIds
()
if
not
test_mode
:
if
not
self
.
test_mode
:
self
.
img_ids
=
[
img_id
for
img_id
in
self
.
img_ids
if
len
(
self
.
coco
.
getAnnIds
(
imgIds
=
img_id
,
iscrowd
=
None
))
>
0
]
blocknum
=
int
(
len
(
self
.
img_ids
)
/
shard
[
1
])
self
.
img_ids
=
self
.
img_ids
[(
blocknum
*
s
hard
[
0
]):(
blocknum
*
(
shard
[
0
]
+
1
))]
blocknum
=
int
(
len
(
self
.
img_ids
)
/
s
elf
.
s
hard
[
1
])
self
.
img_ids
=
self
.
img_ids
[(
blocknum
*
s
elf
.
shard
[
0
]):(
blocknum
*
(
self
.
shard
[
0
]
+
1
))]
self
.
num_images
=
len
(
self
.
img_ids
)
self
.
id2name
,
self
.
name2id
=
self
.
_get_mapping_id_name
(
self
.
coco
.
imgs
)
...
...
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