cpu - compare full extraction
- Dominant language
- Jupyter Notebook
- Stars
- 2
- Forks
- 3
- PR merge metrics
- No merged PRs in 30d
Description
I added a notebook in a new branch "comapre_full_extraction" to compare the output from `spex` and `desi_extract_spectra`. It's just a start but I've already identified two issues that seem like quick fixes.
* The size of the wavelength array in the outputs differs by 1. Looks like `desi_extract_spectra` interprets the wavelength range as a closed interval `[wmin, wmax]` and `spex` is interpreting it as half open `[wmin, wmax)`.
* The flux values in `desi_extract_spectra` are scaled by a factor of `dw`
After applying hotfixes for those issues directly on the output in the comparison notebook, I'm able to generate the following image comparing flux values between the two extractions:

The values are mostly within np.allclose of each other. They tend to disagree closer to the patch edges, especially on the the higher wavelength side.
@sbailey, let me know if you have any suggestions for additional comparisons.
Contributor guide
No contributing guide indexed for this repository
Assessment
This issue has not been assessed yet.