Producing source dependent MC production for zenith 20 deg

[chaimain]

Starting source-dependent MC production, after which we can produce with some MC tuning in a separate PR

[codecov]

Codecov Report

Merging #188 (fb99cdd) into master (a231233) will not change coverage.
The diff coverage is n/a.

@@           Coverage Diff           @@
##           master     #188   +/-   ##
=======================================
  Coverage   15.77%   15.77%           
=======================================
  Files          35       35           
  Lines        2117     2117           
=======================================
  Hits          334      334           
  Misses       1783     1783           

---

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 a231233...fb99cdd. Read the comment docs.

[chaimain]

@SeiyaNozaki Can you just confirm if the parameters are correct as we discussed?

[SeiyaNozaki]

Thank you for this request @chaimain !
You need to use input parameters for source-dependent analysis at the training step.
The other concerns are:

• For source-dep analysis, point gamma MC should be used for RF training. I'm not sure where it can be configured but should be used.
• when creating source-dep IRF, we need to use an option of --source-dep. Can this option used just by modifying the configuration file?
• I guess dl2_to_sensitivity will not work with source-dep analysis

[chaimain]

Thank you for this request @chaimain ! You need to use input parameters for source-dependent analysis at the training step.

Done, thanks for checking.

The other concerns are:

• For source-dep analysis, point gamma MC should be used for RF training. I'm not sure where it can be configured but should be used.

I have removed diffuse-gamma MC from the workflow. Will that be ok, or do you want it to be only restricted in the RF training?

• when creating source-dep IRF, we need to use an option of --source-dep. Can this option used just by modifying the configuration file?

I will raise an issue here to add such support.

• I guess dl2_to_sensitivity will not work with source-dep analysis

For this, I have commented it from the lstmcpipe stages to be run.

[SeiyaNozaki]

I have removed diffuse-gamma MC from the workflow. Will that be ok, or do you want it to be only restricted in the RF training?

There is no need to use diffuse gamma, so it would be fine to remove diffuse-gamma MC, but also need to modify codes to add an option to use point-gamma MC as RF training:
https://github.com/cta-observatory/lstmcpipe/blob/master/lstmcpipe/stages/mc_train.py#L52-L53

[chaimain]

We can probably close this one and the 40 deg one. Unless anyone has any objections.