Details
-
Type:
Bug
-
Status: To Do
-
Resolution: Unresolved
-
Fix Version/s: None
-
Component/s: display_firefly, Firefly
-
Labels:None
-
Team:Portal
-
Urgent?:No
Description
When plotting calexp mask and variance using Firefly, the WCS is not being inherited. The calexp image has the correct coordinates when you mouse over the images in Firefly, but the mask and the variance appear to be using a default ra and dec, where the bottom left corner is displayed as around ~23h58m00s and ~0d29m00s
And example notebook that shows the issue can be found at:
(This is on the branch u/bsmart/Firefly-bugs of https://github.com/lsst-dm/ap_pipe-notebooks .)
Attachments
Issue Links
- relates to
-
- To Do
Activity
| Field | Original Value | New Value |
|---|---|---|
| Description |
When plotting calexp mask and variance using Firefly, the WCS is not being inherited. The calexp image has the correct coordinates when you mouse over the images in Firefly, but the mask and the variance appear to be using a default ra and dec, where the bottom left corner is displayed as around ~23h58m00s and ~0d29m00s
And example notebook that shows the issue can be found at: https://github.com/lsst-dm/ap_pipe-notebooks/blob/873176ea1d87870c1678037b13e7ff48bd4c56fb/notebooks/Firefly-Bug-Notebook.ipynb |
When plotting calexp mask and variance using Firefly, the WCS is not being inherited. The calexp image has the correct coordinates when you mouse over the images in Firefly, but the mask and the variance appear to be using a default ra and dec, where the bottom left corner is displayed as around ~23h58m00s and ~0d29m00s
And example notebook that shows the issue can be found at: https://github.com/lsst-dm/ap_pipe-notebooks/blob/873176ea1d87870c1678037b13e7ff48bd4c56fb/notebooks/Firefly-Bug-Notebook.ipynb (This in on the branch {{u/bsmart/Firefly-bugs}} of https://github.com/lsst-dm/ap_pipe-notebooks .) |
| Description |
When plotting calexp mask and variance using Firefly, the WCS is not being inherited. The calexp image has the correct coordinates when you mouse over the images in Firefly, but the mask and the variance appear to be using a default ra and dec, where the bottom left corner is displayed as around ~23h58m00s and ~0d29m00s
And example notebook that shows the issue can be found at: https://github.com/lsst-dm/ap_pipe-notebooks/blob/873176ea1d87870c1678037b13e7ff48bd4c56fb/notebooks/Firefly-Bug-Notebook.ipynb (This in on the branch {{u/bsmart/Firefly-bugs}} of https://github.com/lsst-dm/ap_pipe-notebooks .) |
When plotting calexp mask and variance using Firefly, the WCS is not being inherited. The calexp image has the correct coordinates when you mouse over the images in Firefly, but the mask and the variance appear to be using a default ra and dec, where the bottom left corner is displayed as around ~23h58m00s and ~0d29m00s
And example notebook that shows the issue can be found at: https://github.com/lsst-dm/ap_pipe-notebooks/blob/873176ea1d87870c1678037b13e7ff48bd4c56fb/notebooks/Firefly-Bug-Notebook.ipynb (This is on the branch {{u/bsmart/Firefly-bugs}} of https://github.com/lsst-dm/ap_pipe-notebooks .) |
I'm going to try to get this to run on the IDF / DP0.2, as not everyone on our team has their USDF account set up yet, but in the mean time, here's one other thing I noticed right away that needs a solution:
Something about how afw.display works with different back ends has led you to have a big branch in your code on Firefly vs. matplotlib, and that was definitely not the goal for how afw.display should work:
if matplot != None: |
afwDisplay.setDefaultBackend('matplotlib') |
|
for i, ref in enumerate(datasetRefs): |
|
|
calexp = butler.getDirect(ref) |
print('Visit: ', calexp.visitInfo.getId(), ', Detector: ',calexp.detector.getId()) |
fig, ax = plt.subplots(1, 3, figsize=(14, 7)) |
plt.sca(ax[0]) # set the first axis as current |
plt.xlabel('Image with Mask Overlay') |
display1 = afwDisplay.Display(frame=fig) |
display1.scale('asinh', -1, 30) |
display1.mtv(calexp)
|
plt.sca(ax[1]) # set the second axis as current |
display2 = afwDisplay.Display(frame=fig) |
display2.mtv(calexp.mask)
|
plt.tight_layout()
|
plt.sca(ax[2]) |
display3 = afwDisplay.Display(frame=fig) |
display3.scale('asinh', 'zscale') |
display3.mtv(calexp.variance)
|
plt.tight_layout()
|
plt.show()
|
remove_figure(fig)
|
|
if i == display_num: |
print('...') |
break |
|
|
else: |
afwDisplay.setDefaultBackend('firefly') |
for i, ref in enumerate(datasetRefs): |
|
|
calexp = butler.getDirect(ref) |
print('Visit: ', calexp.visitInfo.getId(), ', Detector: ',calexp.detector.getId()) |
display = afwDisplay.Display(frame=3*i) |
display.scale('asinh', -1, 30) |
display.mtv(calexp)
|
display = afwDisplay.Display(frame=3*i+1) |
display.setMaskTransparency(90) |
display.mtv(calexp.mask)
|
display = afwDisplay.Display(frame=3*i+2) |
display.scale('asinh', 'zscale') |
display.mtv(calexp.variance)
|
|
|
if i == display_num: |
print('...') |
break |
The goal is for this sort of thing to be necessary only rarely, even in a case like this where you are laying out a grid. So I hope after we get the actual bugs fixed we can look at this.
| Team | Portal [ 16312 ] |
| Component/s | display_firefly [ 15115 ] |
I think in
butler = lsst.daf.butler.Butler("/repo/main", collections=collection) |
because of the way the imports were done it needs to be just butler = Butler( ... ), without the "lsst" etc. fully-qualified name components?
Hi, Brianna Smart, thank you for reporting this.
The code sample has
I'm assuming this collection is on the USDF?