KLH1472

Results 10 comments of KLH1472

svm.cpp and svm.h wrapper

Great work! I want to use thundersvm in my C++ project. I was eager to know when C++ interface will be available. Thanks!

image_data_format('channels_last')

Also want to know how to deal with the "channels_last" format~

Help! why there is a "np.square()" in the test.py?

I got it! We hope the predicted width equals the square root of the width of the ground truth box.

Retrain on voc2007+2012

I retrain the last layer using VOC2007+2012 trainval dataset. It worked fine and I got the mAP = 0.74. Personally, I recommend not to use the retrain.py, I think there...

Some questions about generateBoundingBox() function.

@DuinoDu I am wondered as well, do you have an answer?

Some questions about generateBoundingBox() function.

@DuinoDu That drives me crazy, do you know why??

Some questions about generateBoundingBox() function.

Python 3.6.4 |Anaconda, Inc.| (default, Jan 16 2018, 10:22:32) [MSC v.1900 64 bit (AMD64)] on win32 @DuinoDu

Some questions about generateBoundingBox() function.

Python 3.6.4 |Anaconda, Inc.| (default, Jan 16 2018, 10:22:32) [MSC v.1900 64 bit (AMD64)] on win32

当PNet检测到一个超过阈值的anchor时，为什么要做转置？

if y.shape[0] == 1: # only one point exceed threshold y = y.T x = x.T score = map[x,y].T dx1 = dx1.T dy1 = dy1.T dx2 = dx2.T dy2 =...

当PNet检测到一个超过阈值的anchor时，为什么要做转置？

我觉得这个转置操作是没必要的