Kurucz spectrum #601

menasrac · 2023-08-02T02:04:12Z

Description

PR for the 2023 GSOC project : Common API for large molecular databases.
This PR includes the implementation of the Kurucz database to Radis with an example to plot spectrum using Spectrum Factory and all the related changes to the API or other sections of the code
Follows
PR #579 and adds broadening, a spectrum factory based example for Kurucz, also fixes minor requested changes.

Fixes #

erwanp

See comments

environment.yml

examples/calc_spectrumkurucz.py

radis/lbl/loader.py

erwanp · 2023-08-02T07:56:46Z

examples/calc_spectrumkurucz.py

+ionization_state = '00'  # adjust as needed
+
+# Create a spectrum factory
+sf = SpectrumFactory(wavenum_min, wavenum_max, atom=atom, ionization_state=ionization_state, wstep="auto")


There is a lineshape_optimization / optimization parameter in SpectrumFactory. It is designed for molecular spectra with millions of lines. I think it will slow your calculations with Kuruzc. Set it to None, compare calculation times.

In calc_spectrum you'll set the default based on the input parameters (I.e None for atomic spectra)

I will try with a None input and compare the runtime

It didn't change much, the runtime is sometimes even longer @erwanp

radis/lbl/factory.py

radis/lbl/loader.py

radis/test/io/test_kurucz.py

erwanp

More comments.

I also cannot find the changes you must have made to the Partition functions to include the ones from Kuruzc

radis/lbl/base.py

examples/calc_spectrumkurucz.py

radis/lbl/broadening.py

minouHub

Please clean this PR. Notably:

remove all prints
delete pfdat.txt that is unnecessary. You can use git rm the_file_to_remove then git commit -am "removed duplicate pfdat"
move NIST_iE.txt to radis/db. You can use git mm old_loc new_loc then git commit -am "moved NIST_iE"
git revert to the files that you should not have modified
I proposed very minor commits so that it's easy to you to apply them ==> see printscreen below

NIST_iE.txt

pfdat.txt

examples/calc_spectrumkurucz.py

examples/plot_kuruczspec.py

radis/lbl/base.py

radis/lbl/broadening.py

minouHub · 2023-08-04T09:03:46Z

radis/lbl/broadening.py

@@ -989,6 +1041,8 @@ def _check_accuracy(self, wstep):

        # TODO: thresholds depend whether we're computing Transmittance/optically thin emission,
        # for a homogeneous slab, or self-absorbed radiance combined with other slabs.
+        min_width = None


Suggested change

min_width = None

radis/lbl/broadening.py

radis/lbl/loader.py

Delete file to only keep the example using SpectrumFactory

Was renamed kuruczpartfn.txt and put in db

into KuruczSpectrum

Replaced by plot_kurucz.py in the examples

…cz partition functions

into KuruczSpectrum

radis/lbl/base.py

erwanp · 2023-08-22T05:39:15Z

radis/api/kuruczapi.py

@@ -98,7 +98,7 @@ def load_pf_Barklem2016(self):
        `Barklem & Collet (2016), Table 8 <https://doi.org/10.1051/0004-6361/201526961>`_

        """
-        file_path="radis/levels/pfTKurucz_values.txt"
+        file_path=r"radis/levels/pfTKurucz_values.txt"


Make it an absolute path. There are some functions in Radis/dB and Radis/test/utils to get proper maths independently of 1) the working directory 2) your OS

erwanp · 2023-08-22T05:42:34Z

radis/lbl/factory.py

-                        )
-                        + "\n- under nonequilibrium: {0}".format(
-                            MOLECULES_LIST_NONEQUILIBRIUM
+        if molecule is not None and species is not None:


Fix molecule and/or species when reading the function parameters.

Then use one only through the whole code. Here you're still using both it makes the code hard to read.

You can rename the internal name to species to be very clear.

radis/lbl/loader.py

into KuruczSpectrum

dcmvdbekerom · 2023-08-27T22:18:36Z

@menasrac I believe the issue with test_gascomp.py is on Cantera's end, see also this issue I opened there: Cantera/cantera#1607
For the moment, I would suggest to set - cantera>=2.5.1, <3.0.0 in environment.yml

…oadening.py and calc.py

menasrac · 2023-08-27T23:08:26Z

@menasrac I believe the issue with test_gascomp.py is on Cantera's end, see also this issue I opened there: Cantera/cantera#1607 For the moment, I would suggest to set - cantera>=2.5.1, <3.0.0 in environment.yml

I tried your suggestion, the test passed indeed but other tests in calc.py and broadening.py failed after this modification @dcmvdbekerom

requirements.txt

minouHub · 2023-08-28T10:01:45Z

A few issues remaining, sorted in order of priority

the file plot_Kurucz.py is still not running (see below)
There is no test of SpectrumFactory using atoms
fetch_databank() is not implemented for atoms (see below)
the calc_spectrum function is not compatible with atomic spectra

---- fetchdatabank() ----

from radis import SpectrumFactory
from astropy import units as u
sf = SpectrumFactory(wavelength_min=4165 * u.nm,
                     wavelength_max=4200 * u.nm,
                     species='Fe_I',
                     medium='air',
                     verbose=1,    # more for more details
                     )
sf.fetch_databank()

---- plot_kurucz.py ----
output

Traceback (most recent call last):

  File C:\Anaconda\envs\radis-Kurucz\lib\site-packages\spyder_kernels\py3compat.py:356 in compat_exec
    exec(code, globals, locals)

  File c:\users\nicolas minesi\python\examples\plot_kurucz.py:6
    hdf5_path = fetch_kurucz('Fe_I')[0] # replace 'Fe_I' with atom and ionization_state

  File c:\users\nicolas minesi\python\radis\io\kurucz.py:21 in fetch_kurucz
    kurucz = AdBKurucz(species)

  File c:\users\nicolas minesi\python\radis\api\kuruczapi.py:36 in __init__
    self.pfTdat, self.pfdat = self.load_pf_Barklem2016()

  File c:\users\nicolas minesi\python\radis\api\kuruczapi.py:99 in load_pf_Barklem2016
    with open(file_path, "r") as file:

FileNotFoundError: [Errno 2] No such file or directory: 'radis/levels/pfTKurucz_values.txt'

menasrac · 2023-08-28T17:46:55Z

A few issues remaining, sorted in order of priority

the file plot_Kurucz.py is still not running (see below)
There is no test of SpectrumFactory using atoms
fetch_databank() is not implemented for atoms (see below)
the calc_spectrum function is not compatible with atomic spectra

---- fetchdatabank() ----

from radis import SpectrumFactory
from astropy import units as u
sf = SpectrumFactory(wavelength_min=4165 * u.nm,
                     wavelength_max=4200 * u.nm,
                     species='Fe_I',
                     medium='air',
                     verbose=1,    # more for more details
                     )
sf.fetch_databank()

---- plot_kurucz.py ---- output

Traceback (most recent call last):

  File C:\Anaconda\envs\radis-Kurucz\lib\site-packages\spyder_kernels\py3compat.py:356 in compat_exec
    exec(code, globals, locals)

  File c:\users\nicolas minesi\python\examples\plot_kurucz.py:6
    hdf5_path = fetch_kurucz('Fe_I')[0] # replace 'Fe_I' with atom and ionization_state

  File c:\users\nicolas minesi\python\radis\io\kurucz.py:21 in fetch_kurucz
    kurucz = AdBKurucz(species)

  File c:\users\nicolas minesi\python\radis\api\kuruczapi.py:36 in __init__
    self.pfTdat, self.pfdat = self.load_pf_Barklem2016()

  File c:\users\nicolas minesi\python\radis\api\kuruczapi.py:99 in load_pf_Barklem2016
    with open(file_path, "r") as file:

FileNotFoundError: [Errno 2] No such file or directory: 'radis/levels/pfTKurucz_values.txt'

Fetch databank is not implemented for atoms, however load_databank works with atoms as in the plot_kurucz examples. You have to mention sf.load_databank(path=hdf5_path, format='kurucz',parfuncfmt='kurucz') @minouHub

menasrac · 2023-08-28T18:57:30Z

A few issues remaining, sorted in order of priority

the file plot_Kurucz.py is still not running (see below)
There is no test of SpectrumFactory using atoms
fetch_databank() is not implemented for atoms (see below)
the calc_spectrum function is not compatible with atomic spectra

---- fetchdatabank() ----

from radis import SpectrumFactory
from astropy import units as u
sf = SpectrumFactory(wavelength_min=4165 * u.nm,
                     wavelength_max=4200 * u.nm,
                     species='Fe_I',
                     medium='air',
                     verbose=1,    # more for more details
                     )
sf.fetch_databank()

---- plot_kurucz.py ---- output

Traceback (most recent call last):

  File C:\Anaconda\envs\radis-Kurucz\lib\site-packages\spyder_kernels\py3compat.py:356 in compat_exec
    exec(code, globals, locals)

  File c:\users\nicolas minesi\python\examples\plot_kurucz.py:6
    hdf5_path = fetch_kurucz('Fe_I')[0] # replace 'Fe_I' with atom and ionization_state

  File c:\users\nicolas minesi\python\radis\io\kurucz.py:21 in fetch_kurucz
    kurucz = AdBKurucz(species)

  File c:\users\nicolas minesi\python\radis\api\kuruczapi.py:36 in __init__
    self.pfTdat, self.pfdat = self.load_pf_Barklem2016()

  File c:\users\nicolas minesi\python\radis\api\kuruczapi.py:99 in load_pf_Barklem2016
    with open(file_path, "r") as file:

FileNotFoundError: [Errno 2] No such file or directory: 'radis/levels/pfTKurucz_values.txt'

Fetch databank is not implemented for atoms, however load_databank works with atoms as in the plot_kurucz examples. You have to mention sf.load_databank(path=hdf5_path, format='kurucz',parfuncfmt='kurucz') @minouHub

I have also added a test for fetch_kurucz. Though it is not directly a test for load_databank, it still asserts that the df which will be used by calc_spectrum contains all the required columns

erwanp · 2023-08-28T20:50:35Z

@menasrac please add a runnable test code in the PR main text so we can test it

menasrac added 10 commits June 28, 2023 17:28

Update develop from upstream

ab48aa0

Merge remote-tracking branch 'upstream/develop' into develop

6155439

Kurucz implementation and tests

c1a6bcd

periodictable added to environnement

d4910ce

Added local pfdat.txt file

1b3afd5

Remove trailing whitespaces

f67c110

Save KuruczSpectrumV1

133f981

Kurucz+SpectrumFactory

5e3511d

Fixed Loader.py

6489bd0

Reverting unuseful changes in Broadening.py and factory.py

e554810

erwanp requested changes Aug 2, 2023

View reviewed changes

menasrac added 2 commits August 2, 2023 20:43

Minor fixes

777e2cd

Use fetch_kurucz in loader.py

49381c5

minouHub self-requested a review August 3, 2023 09:07

minouHub added this to the 0.16 milestone Aug 3, 2023

minouHub requested changes Aug 4, 2023

View reviewed changes

minouHub marked this pull request as draft August 4, 2023 09:23

menasrac and others added 12 commits August 4, 2023 15:08

Broadening.py adjustments

ef77085

Delete plot_kuruczspec.py

82e4304

Delete file to only keep the example using SpectrumFactory

Delete pfdat.txt

0305d79

Was renamed kuruczpartfn.txt and put in db

Removed unused methods from the API

c7289dd

Merge branch 'KuruczSpectrum' of https://github.com/menasrac/radis_racim

4efb579

into KuruczSpectrum

Removed prints for release

ff6a81d

Removed prints and adjusted imports in loader.py

c383f6c

Added partition function format kurucz

7c33699

Added a new Class for kurucz partition functions

85c9927

Moved NIST_iE.txt to new directory

ae1e9ba

Delete calc_spectrumkurucz.py

1cb3639

Replaced by plot_kurucz.py in the examples

Adjusted the temperatures for interpolation and added a test for kuru…

81407c7

…cz partition functions

menasrac added 2 commits August 22, 2023 00:57

Fixing tests ?

b469ee0

Merge branch 'KuruczSpectrum' of https://github.com/menasrac/radis_racim

8ae795a

into KuruczSpectrum

erwanp requested changes Aug 22, 2023

View reviewed changes

menasrac and others added 8 commits August 22, 2023 23:02

Merge branch 'develop' into KuruczSpectrum

d86e516

Removed prints

f3468e4

Merge branch 'KuruczSpectrum' of https://github.com/menasrac/radis_racim

6acccaa

into KuruczSpectrum

Whitespace minor correction

f2395c4

Fixed most of the tests

8e148da

Fixed diluent molefraction test

cb9d8f1

Replaced mendeleev by periodictable due to dependencies issues

6ea2bff

Fixed the test_molecule_lists test

380cadb

menasrac added 2 commits August 28, 2023 00:44

Updated cantera in dependencies

57a84bf

Reverted the cantera modification because it broke other tests in br…

3b99539

…oadening.py and calc.py

minouHub reviewed Aug 28, 2023

View reviewed changes

requirements.txt Outdated Show resolved Hide resolved

menasrac added 2 commits August 28, 2023 12:08

fix radis#614

c30a5f3

Fixed plot_kurucz links for examples

7cc7c02

Added a test for fetch_kurucz

3470bcd

Removed mendeleev from requirements.txt

70d82b1

menasrac added 6 commits August 31, 2023 00:27

Kurucz fetch databank + atomic_calc_spectrum+example

e9ebd19

Fixed calc.py test

3eb2289

Added comment about species naming for calc_spectrum

9ea433b

Modified environment.yml to solve cantera issue

a4333fc

Linting

f9df931

Linting part 2

4a08a8b

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Kurucz spectrum #601

Kurucz spectrum #601

menasrac commented Aug 2, 2023 •

edited

erwanp left a comment

erwanp Aug 2, 2023

menasrac Aug 2, 2023

menasrac Aug 3, 2023

erwanp left a comment

minouHub left a comment •

edited

minouHub Aug 4, 2023

erwanp Aug 22, 2023

erwanp Aug 22, 2023

dcmvdbekerom commented Aug 27, 2023

menasrac commented Aug 27, 2023

minouHub commented Aug 28, 2023

menasrac commented Aug 28, 2023

menasrac commented Aug 28, 2023

erwanp commented Aug 28, 2023 •

edited

Kurucz spectrum #601

Are you sure you want to change the base?

Kurucz spectrum #601

Conversation

menasrac commented Aug 2, 2023 • edited

Description

erwanp left a comment

Choose a reason for hiding this comment

erwanp Aug 2, 2023

Choose a reason for hiding this comment

menasrac Aug 2, 2023

Choose a reason for hiding this comment

menasrac Aug 3, 2023

Choose a reason for hiding this comment

erwanp left a comment

Choose a reason for hiding this comment

minouHub left a comment • edited

Choose a reason for hiding this comment

minouHub Aug 4, 2023

Choose a reason for hiding this comment

erwanp Aug 22, 2023

Choose a reason for hiding this comment

erwanp Aug 22, 2023

Choose a reason for hiding this comment

dcmvdbekerom commented Aug 27, 2023

menasrac commented Aug 27, 2023

minouHub commented Aug 28, 2023

menasrac commented Aug 28, 2023

menasrac commented Aug 28, 2023

erwanp commented Aug 28, 2023 • edited

menasrac commented Aug 2, 2023 •

edited

minouHub left a comment •

edited

erwanp commented Aug 28, 2023 •

edited