JP-2326: Fixed incorrect slope calculation #68
Conversation
Codecov Report
@@ Coverage Diff @@
## main #68 +/- ##
==========================================
+ Coverage 94.51% 94.53% +0.02%
==========================================
Files 14 14
Lines 1624 1630 +6
==========================================
+ Hits 1535 1541 +6
Misses 89 89
Continue to review full report at Codecov.
|
src/stcal/ramp_fitting/ols_fit.py
Outdated
| @@ -1695,6 +1698,25 @@ def calc_slope(data_sect, gdq_sect, frame_time, opt_res, save_opt, rn_sect, | |||
|
|
|||
| mask_2d[gdq_sect_r != 0] = False # RE-exclude bad group dq values | |||
|
|
|||
| # Ensure that the first group to be fit is the cosmic-ray-affected | |||
There was a problem hiding this comment.
This comment is a bit confusing, and could use a little clarification.
There was a problem hiding this comment.
I changed the comment here. Please take a look at it to see if my edit makes it more clear.
| @@ -290,6 +290,79 @@ def test_utils_dq_compress_final(): | |||
| assert(not(dq[0, 2] & dqflags["DO_NOT_USE"])) | |||
|
|
|||
|
|
|||
| def jp_2326_test_setup(): | |||
| # Set up ramp data | |||
| ramp = np.array([120.133545, 117.85222, 87.38832, 66.90588, 51.392555, | |||
There was a problem hiding this comment.
Is there a reason a ramp with a negative slope is created ?
There was a problem hiding this comment.
Instead of generating my own test data for the ramp, I used a ramp for a pixel that was failing in a regression test. I used the same science data, DQ data, etc., changing only groupdq[0, 1, 0, 0] to be set to DO_NOT_USE in one test and JUMP_DET in the second test. The test with the group flagged as DO_NOT_USE should not be used in calculations, while the one flagged as JUMP_DET should.
There was a problem hiding this comment.
Yeah, I figured it was from a failing test.
src/stcal/ramp_fitting/ols_fit.py
Outdated
| @@ -1695,6 +1698,25 @@ def calc_slope(data_sect, gdq_sect, frame_time, opt_res, save_opt, rn_sect, | |||
|
|
|||
| mask_2d[gdq_sect_r != 0] = False # RE-exclude bad group dq values | |||
|
|
|||
| # Ensure that the first group to be fit is the cosmic-ray-affected | |||
| @@ -290,6 +290,79 @@ def test_utils_dq_compress_final(): | |||
| assert(not(dq[0, 2] & dqflags["DO_NOT_USE"])) | |||
|
|
|||
|
|
|||
| def jp_2326_test_setup(): | |||
| # Set up ramp data | |||
| ramp = np.array([120.133545, 117.85222, 87.38832, 66.90588, 51.392555, | |||
There was a problem hiding this comment.
Yeah, I figured it was from a failing test.
|
@kmacdonald-stsci Don't forget to enter a log entry in the jwst CHANGES file. |
When computing a segment for a ramp if a group exists just before the zeroth group of the current segment and it is flagged only
JUMP_DETit should be added to the beginning of the ramp segment for computation. Previously any flagged groups got added back to the beginning of the ramp segment. The segment mask now excludes any group not flagged with onlyJUMP_DET.Two tests have been added to ramp fitting unit tests to test the above conditions.
#67