Closed karllark closed 3 years ago
Merging #113 (6df4037) into master (e7d7f2a) will increase coverage by
21.08%
. The diff coverage is80.00%
.
@@ Coverage Diff @@
## master #113 +/- ##
===========================================
+ Coverage 54.71% 75.80% +21.08%
===========================================
Files 4 5 +1
Lines 265 310 +45
===========================================
+ Hits 145 235 +90
+ Misses 120 75 -45
Impacted Files | Coverage Δ | |
---|---|---|
pahfit/base.py | 60.89% <0.00%> (+16.96%) |
:arrow_up: |
pahfit/helpers.py | 92.30% <92.30%> (ø) |
|
pahfit/component_models.py | 100.00% <0.00%> (+56.75%) |
:arrow_up: |
Continue to review full report at Codecov.
Legend - Click here to learn more
Δ = absolute <relative> (impact)
,ø = not affected
,? = missing data
Powered by Codecov. Last update e7d7f2a...6df4037. Read the comment docs.
One question: for 'test_fitting_spitzer.py', you use maxiter=200 (ln 17) while the default is set to 1000. Is this to decrease the time for testing or is this a leftover from the previous setting?
For testing so it is fully reproducible regardless of the default maxiter setting. And to reduce the time needed.
And how do I show that part of the code in this comment (like Alexandros did)?
If you go to the "Files changed" tab, you can highlight specific lines and add a comment/review. Use your mouse and the "+" symbols.
And how do I show that part of the code in this comment (like Alexandros did)?
If you go to the "Files changed" tab, you can highlight specific lines and add a comment/review. Use your mouse and the "+" symbols.
Thanks.
Two reviews enough? I think it is ready to merge.
Two reviews enough? I think it is ready to merge.
Fine with me. But there is one check which was unsuccessful ...
But there is one check which was unsuccessful ...
This is codacy being over careful in the default value for a variable in a function definition. It can be ignored.
Add in a fitting test. Simply reads in the spitzer scipack and M101 nucleus spectrum, runs the fit, and then compares the fit parameters to a cached array of the parameters. Will fail if the parameters are not close (machine precision I think).
Refactored run_pahfit and plot_pahfit scripts to have common functions to avoid code duplication and allow for easier testing. This refactoring will also make it easier to support other fitters, units, etc. in the future.