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
openvinotoolkit
mmaction2
提交
49232481
M
mmaction2
项目概览
openvinotoolkit
/
mmaction2
大约 1 年 前同步成功
通知
2
Star
5
Fork
3
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
DevOps
流水线
流水线任务
计划
Wiki
0
Wiki
分析
仓库
DevOps
项目成员
Pages
M
mmaction2
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
DevOps
DevOps
流水线
流水线任务
计划
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
流水线任务
提交
Issue看板
体验新版 GitCode,发现更多精彩内容 >>
提交
49232481
编写于
7月 06, 2020
作者:
J
JoannaLXY
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
fix
上级
0753eb82
变更
3
隐藏空白更改
内联
并排
Showing
3 changed file
with
3 addition
and
59 deletion
+3
-59
configs/localization/bmn/README.md
configs/localization/bmn/README.md
+1
-28
configs/localization/bsn/README.md
configs/localization/bsn/README.md
+1
-29
docs/merge_docs.sh
docs/merge_docs.sh
+1
-2
未找到文件。
configs/localization/bmn/README.md
浏览文件 @
49232481
...
...
@@ -8,35 +8,8 @@
|-|-|-|-|-|-|-|-|
|
[
bmn_400x100_9e_2x8_activitynet_feature
](
/configs/localization/bmn/bmn_400x100_2x8_9e_activitynet_feature.py
)
| None |75.28|67.22|5420|3.27|
[
ckpt
](
)|
[
log
]
()|
## Data
1.
Put the rescaled feature data folder
`csv_mean_100`
under
`$MMACTION/data/activitynet_feature_cuhk/`
.
The raw feature data could be found at [here](https://github.com/wzmsltw/BSN-boundary-sensitive-network).
2.
Put the annotaion files under
`$MMACTION/data/ActivityNet`
.
The annotation files could be found at [here]().
3.
Finally, make sure your folder structure same with the tree structure below.
If your folder structure is different, you can also change the corresponding paths in config files.
```
mmaction
├── mmaction
├── tools
├── config
├── data
│ ├── activitynet_feature_cuhk
│ │ ├── csv_mean_100
│ ├── ActivityNet
│ │ ├── anet_anno_train.json
│ │ ├── anet_anno_val.json
│ │ ├── anet_anno_test.json
...
```
## Checkpoint
Put the
`tem_best.pth.tar`
and
`pem_best.pth.tar`
under
`checkpoints/`
.
The ckpts could be found at
[
here
](
).
For more details on data preparation, you can refer to
[
Prepaing Activitynet
](
../../../tools/data/activitynet/preparing_activitynet.md
)
.
## Train
You can use the following command to train a model.
...
...
configs/localization/bsn/README.md
浏览文件 @
49232481
...
...
@@ -8,36 +8,8 @@
|-|-|-|-|-|-|-|-|
|bsn_400x100_1x16_20e_activitynet_feature | None |74.65|66.45|41(TEM)+25(PEM)|0.074(TEM)+0.036(PEM)|
[
ckpt_tem
](
)
[
ckpt_pem
]
|
[
log_tem
](
)
[
log_pem
]
()|
## Data
1.
Put the rescaled feature data folder
`csv_mean_100`
under
`$MMACTION/data/activitynet_feature_cuhk/`
.
The raw feature data could be found at [here](https://github.com/wzmsltw/BSN-boundary-sensitive-network).
2.
Put the annotaion files under
`$MMACTION/data/ActivityNet`
.
The annotation files could be found at [here]().
3.
Finally, make sure your folder structure same with the tree structure below.
If your folder structure is different, you can also change the corresponding paths in config files.
```
mmaction
├── mmaction
├── tools
├── config
├── data
│ ├── activitynet_feature_cuhk
│ │ ├── csv_mean_100
│ ├── ActivityNet
│ │ ├── anet_anno_train.json
│ │ ├── anet_anno_val.json
│ │ ├── anet_anno_test.json
...
```
## Checkpoint
1.
Put the
`tem_best.pth.tar`
and
`pem_best.pth.tar`
under
`checkpoints/`
.
The ckpts could be found at [here]() (TODO).
For more details on data preparation, you can refer to
[
Prepaing Activitynet
](
../../../tools/data/activitynet/preparing_activitynet.md
)
.
## Train
You can use the following commands to train a model.
...
...
docs/merge_docs.sh
浏览文件 @
49232481
...
...
@@ -5,7 +5,6 @@ cat ../configs/recognition/*/*.md > recognition_models.md
cat
./tutorials/finetune.md ./tutorials/new_dataset.md ./tutorials/data_pipeline.md ./tutorials/new_modules.md
>
tutorials.md
cat
../tools/data/
*
/
*
.md
>
prepare_data.md
cat
../tools/data/
*
/
*
.md
>
prepare_data.md
sed
-i
's/#/##&/'
localization_models.md
...
...
@@ -13,7 +12,7 @@ sed -i 's/#/##&/' recognition_models.md
sed
-i
's/#/#&/'
tutorials.md
sed
's/# Preparing/# /
'
prepare_data.md
sed
-i
's/# Preparing/# /g
'
prepare_data.md
sed
-i
's/#/##&/'
prepare_data.md
sed
-i
'1i\# Tutorials'
tutorials.md
...
...
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