spacetelescope / acs-notebook Goto Github PK

Tyler Desjardins mentions that we should consider moving emails from help[at]stsci.edu to point to the web portal where possible and appropriate. For HST (or any non-JWST), it is https://hsthelp.stsci.edu . For JWST, it is https://jwsthelp.stsci.edu . Please update info in setup.py, setup.cfg, documentation, etc as appropriate.

Please close this issue if it is irrelevant to your repository. This is an automated issue. If this is opened in error, please let pllim know!

xref spacetelescope/hstcal#317

Provide a conda environment file that has all of the stuff necessary for running the examples (astroquery, nodejs, etc.). I'm not sure if the jupyterlab extensions can also be included this way (I hope so!).

“This tutorial will show you how to…” in the intro of this notebook seems to be talking about the dark subtraction of SBC data.

Replace http://www.stsci.edu/hst/acs/analysis/throughputs with a notebook.
Add new page with throughput curve example shown. Make sure it re-creates all of the plots on the page dynamically, so they are the most up to date possible.

If we do this, we may want to move the downloadable throughput files to another page similar to the ACS aperture corrections page and then just link to it. Unless we want the notebook to make these files on the fly or replace them in some other manner. @nmiles2718 any thoughts?

I suggest you give more details about the 0.2% difference.
(Also I think it would be better to quote the ratio rather than the difference... making sure your PAM is generated using a dataset taken in 2004 in the F606W filter. That would tell you how much impact the improved distortion solution makes).

In 2004, a single PAM was created using the F606W distortion solution, based on polynomial fits applied by the IDCTAB reference file. The PAM did not account for filter-dependent polynomial solutions, higher order corrections applied now via the D2IMFILE and the NPOLFILE, a time-dependent skew corrections, or the effect of velocity aberration at the time of observation.

I'm not sure which of these would be the biggest effect, but you could determine it empirically by making different PAMs... one with the NPOLFILE set to N/A, one with the VAFACTOR set to 1.0, etc.

This is what I sent to WFC3 team, which puts the responsibility on the user for determining how accurate the PAM needs to be.
"These static files have not been updated to reflect improvements to the distortion model, and as the model has become more complex, a ‘one-size-fits-all’ PAM may no longer provide the accuracy required for some science goals."

The zeropoints notebook needs a few updates:

• The pivot wavelength is incorrect. It should be band.pivot(), not obs.pivot().
• The link to the zeropoints calculator webpage should be included since we are keeping it. Users should be told that both the notebook and the webpage tool use the same method and give the same answers.
• In the photometric systems section, the equation that defines vegamag says F = ... instead of M = ...
• All links that are not anchors to other parts of the page should open in a new tab

Some of the formatting in the subarray CTE notebook needs to be altered. In particular, div alert boxes should be used to warnings and important notes. Also emphasize more strongly to users that they will need to set up JREF to run the example. Add in a table of contents.

Create a new notebook detailing the behavior of calacs. This does not need to have comprehensive examples as there will be a CSI notebook with that, but it can have the examples presented in the Data Handbook. It should also mention acs_destripe_plus and link to the subarray CTE correction page.

Under "contents", notebook names are highlighted but then I have to dig through the different directories to get to them. It would be easier if I could click on a link directly.

I was re-running acs_destripe_plus notebook to test the changes, but it crashed when it attempted to overwrite existing FITS file. Either clarify in the notebook that user need to remove existing files or gracefully handle the situation.

There is a stale calib_example branch that one inherits when forking this repo. Deleting the branch would cause less confusion.

Many of the links in the notebooks are now broken due to Jahia and HDox launches.

According to Erik, os.rename will get around the error raised by shutil.move if the file already exists. We can then simplify all of the cells using astroquery that currently have try except blocks in them.

I see __pycache__ directory in this repo. It is not supposed to be version controlled as it is generated by Python when you run the stuff locally in your source checkout.

Also have git ignore it permanently in .gitignore file in the root directory.