Skip to content

Commit 54aee35

Browse files
committed
fix
1 parent 9438684 commit 54aee35

File tree

1 file changed

+3
-2
lines changed

1 file changed

+3
-2
lines changed

README.md

Lines changed: 3 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -10,7 +10,7 @@
1010
## Setup for train
1111
#### 1. Download model and files.
1212
There is a **download_model.bat** in [CNTK/FasterRCNN/](https://github.com/springkim/FasterRCNN_SpringEdition/tree/master/CNTK/FasterRCNN).
13-
This file can download **VGG16_ImageNet_Caffe.model** . because our **FasterRCNN** use **VGG16** instead of **Alexnet**.
13+
This file can download **VGG16_ImageNet_Caffe.model** .
1414

1515
And move to [CNTK/FasterRCNN/bin](https://github.com/springkim/FasterRCNN_SpringEdition/tree/master/CNTK/FasterRCNN/bin) and run **download_runfile.bat**.
1616
This file download **FasterRCNN_Train_SE.exe** and dependency dlls.
@@ -48,6 +48,7 @@ Finally, It looks like below.
4848
│ └ <images that wrote in train_img_file.txt>
4949
├ train_img_file.txt
5050
├ train_roi_file.txt
51+
├ VGG16_ImageNet_Caffe.model
5152
└ train.bat
5253
```
5354
Now you can run **train.bat** for trainning.
@@ -65,7 +66,7 @@ And open **FasterRCNN_SE_Detection_Example.sln** as [Visual Studio 2015](https:/
6566
We needs only 1 header file (**FasterRCNN_SE.h**) for detect. Of course your exe file needs **FasterRCNN_Detect_SE.exe** and requirement dlls(22 files).
6667

6768
Model loading time is about 20~30s.
68-
Detectio time is about 0.15s.
69+
Detection time is about 0.15s.
6970
###### Result
7071
```
7172
Selected GPU[0] GeForce GTX 1080 as the process wide default device.

0 commit comments

Comments
 (0)