Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

ROIAlign bug #148

Open
eli2014 opened this issue Aug 30, 2018 · 0 comments
Open

ROIAlign bug #148

eli2014 opened this issue Aug 30, 2018 · 0 comments

Comments

@eli2014
Copy link

eli2014 commented Aug 30, 2018

I find that there maybe a bug in the ROIAlign operator. For example,
x = mx.nd.array([[[[ 0., 1., 2., 3., 4., 5.],
[ 6., 7., 8., 9., 10., 11.],
[ 12., 13., 14., 15., 16., 17.],
[ 18., 19., 20., 21., 22., 23.],
[ 24., 25., 26., 27., 28., 29.],
[ 30., 31., 32., 33., 34., 35.],
[ 36., 37., 38., 39., 40., 41.],
[ 42., 43., 44., 45., 46., 47.]]]])

// region of interest i.e. bounding box coordinates.
y = mx.nd.array([[0,0,0,4,4]])
pooled_size = (2, 2)
spatial_scale = 1.0
The result of mx.nd.ROIAlign(x, y, pooled_size, spatial_scale) is
[[[[-3.4028235e+38 -3.4028235e+38]
[-3.4028235e+38 -3.4028235e+38]]]].

When I use ROIAlign from MxNet(https://github.com/apache/incubator-mxnet/blob/master/src/operator/contrib/roi_align.cc), I get the same result as caffe2's implementation of ROIAlign(https://github.com/pytorch/pytorch/blob/master/caffe2/operators/roi_align_op.cc).

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant