datacard making update

[ywan2]

Two features are added in the datacard making step using python make_cards.py pkl _np.pkl.gz:

• We can now set the wcs to certain values while making asimov datacards (excluding --unblind option), by including --wc-vals wc1={number1}, wc2={number2} ... An example code to run is:

  time python make_cards.py {your_file}_np.pkl.gz --do-nuisance --var-lst lj0pt ptz -d {your_datacard_directory} --do-mc-stat --skip-selected-wcs-check --wc-vals ctlTi=0.32, ctlSi=3.1

• There is an extra step of systematics check during the binning and zeroing process. When the nominal values are zeros, FFUp and FFDown are supposed to be zero. If any non-zero value occurs, the systematics is invalid thus an error would be raised. If this error shows up, it would be very likely that the datacards made won't pass text2workspace.py later in combine.

[codecov[bot]]

Codecov Report

Attention: Patch coverage is 85.71429% with 4 lines in your changes are missing coverage. Please review.

Project coverage is 27.44%. Comparing base (8095829) to head (e96fedb).

:exclamation: Current head e96fedb differs from pull request most recent head 36b68d4

Please upload reports for the commit 36b68d4 to get more accurate results.

Files	Patch %	Lines
topeft/modules/datacard_tools.py	80.95%	4 Missing :warning:

Additional details and impacted files

```diff @@ Coverage Diff @@ ## master #406 +/- ## ========================================== + Coverage 26.27% 27.44% +1.16% ========================================== Files 31 28 -3 Lines 4552 4245 -307 ========================================== - Hits 1196 1165 -31 + Misses 3356 3080 -276 ``` | [Flag](https://app.codecov.io/gh/TopEFT/topeft/pull/406/flags?src=pr&el=flags&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=TopEFT) | Coverage Δ | | |---|---|---| | [unittests](https://app.codecov.io/gh/TopEFT/topeft/pull/406/flags?src=pr&el=flag&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=TopEFT) | `27.44% <85.71%> (+1.16%)` | :arrow_up: | Flags with carried forward coverage won't be shown. [Click here](https://docs.codecov.io/docs/carryforward-flags?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=TopEFT#carryforward-flags-in-the-pull-request-comment) to find out more.

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.

[ywan2]

Thank you @bryates for your comments! I made the changes you suggested, except for the last one of "checking non-zero nominal for zero FFUp / FFDown ". There are some channels that have these values during this step but don't raise errors in Combine. Also, I think that keeping "print()" statement might be better than "raise Warning", in case people want to still have the rest of the datacards made to test some things. What do you think?

[bryates]

Thank you @bryates for your comments! I made the changes you suggested, except for the last one of "checking non-zero nominal for zero FFUp / FFDown ". There are some channels that have these values during this step but don't raise errors in Combine. Also, I think that keeping "print()" statement might be better than "raise Warning", in case people want to still have the rest of the datacards made to test some things. What do you think?

@ywan2 Raising a warning should still just print a message and continue. Raising an exception will for sure terminate the script. Are you seeing otherwise?

[bryates]

Thank you @bryates for your comments! I made the changes you suggested, except for the last one of "checking non-zero nominal for zero FFUp / FFDown ". There are some channels that have these values during this step but don't raise errors in Combine. Also, I think that keeping "print()" statement might be better than "raise Warning", in case people want to still have the rest of the datacards made to test some things. What do you think?

@ywan2 Raising a warning should still just print a message and continue. Raising an exception will for sure terminate the script. Are you seeing otherwise?

Ah maybe I was thinking of this (https://stackoverflow.com/questions/3891804/raise-warning-in-python-without-interrupting-program). You can switch back to print statements instead of warnings.

[ywan2]

@bryates, yes I tested the the raise Warning statement, it does terminate the code. So I switched back to print statement.

[bryates]

Merging this PR now. I've created issues for the remaining items so we can address them later.