Update .zattrs file and test result #20
Conversation
| ): | ||
| array[ | ||
| channel, idx, x : x + subdat_size[-2], y : y + subdat_size[-1] | ||
| ] = subdat[channel : channel + 1, ...][0] |
There was a problem hiding this comment.
Is the same loop repeated twice or am I missing something?
| (shape0[0] / shape[0] - 1) * vxh * 0.5, | ||
| (shape0[1] / shape[1] - 1) * vxw * 0.5, | ||
| float(2**n - 1) * vxh * 0.5, | ||
| float(2**n - 1) * vxw * 0.5, |
There was a problem hiding this comment.
Can you explain what the issue with webknossos is?
The thing it that you may be fixing things for webknossos, but it actually mis-registers the layer (if we're following the spec) and will break things down in our own code in the future. So I am a bit against this change.
There was a problem hiding this comment.
Hi Yael, I don't have the error message but it should have something to do with the multiscale part in .zattrs file that doesn't match with what webknossos expects. One example is that if there are both even and odd numbers for the size of image, as the resolution level increases, the difference between 2 axis gets larger, and causes a mismatch in x and y direction for the scale part in coordinateTransformation. I do agree we shouldn't modify the code as what webknossos accepts. I'll undo the change here. Thanks. @balbasty
There was a problem hiding this comment.
Hi @jingjingwu1225, I am a bit out of the loop, but could you please explain the motivation for these changes? Thank you.
tests/data/lsm.zarr.zip
Outdated
There was a problem hiding this comment.
It looks like this pull request is focused on changes to the dark-field images, so I would suggest that this light-sheet test file be updated in a separate pull request.
jingjingwu1225
force-pushed
the
update-zattrs
branch
2 times, most recently
from
938dc22
to
3ba6cd9
Compare
jingjingwu1225
force-pushed
the
update-zattrs
branch
from
3ba6cd9
to
5c67764
Compare
No description provided.