search

tardis-sn / tardis

TARDIS - Temperature And Radiative Diffusion In Supernovae
https://tardis-sn.github.io/tardis
199 stars 404 forks source link

Missing NLTE atomic dataset #505

Closed pfreddy closed 6 years ago

pfreddy commented 8 years ago

Hi,

I've tried to run a calculation with some NLTE-species after the example configuration file.

plasma: nlte: species: ['Ca 2']

It works well for a few ions, but it cannot handle e.g. Si 2 or O 2, and returns with an error message.

wkerzendorf commented 8 years ago

@pfreddy currently very busy trying to dig out the old data set.

unoebauer commented 8 years ago

Hi @pfreddy - can provide more details, in particular concerning the error messages?

pfreddy commented 8 years ago

I copy below the whole error message:

tardis.atomic - INFO - Preparing the NLTE data
tardis.plasma.properties.partition_function - INFO - Calculating rates for species (8, 0)
Traceback (most recent call last):
  File "/home/barna/anaconda/envs/tardis/bin/tardis", line 64, in <module>
    radial1d_mdl = model.Radial1DModel(tardis_config)
  File "/home/barna/anaconda/envs/tardis/lib/python2.7/site-packages/tardis/model.py", line 118, in __init__
    v_outer=tardis_config.structure.v_outer)
  File "/home/barna/anaconda/envs/tardis/lib/python2.7/site-packages/tardis/plasma/standard_plasmas.py", line 131, in __init__
    j_blues=None, w=w, link_t_rad_t_electron=link_t_rad_t_electron)
  File "/home/barna/anaconda/envs/tardis/lib/python2.7/site-packages/tardis/plasma/base.py", line 22, in __init__
    self.update(**kwargs)
  File "/home/barna/anaconda/envs/tardis/lib/python2.7/site-packages/tardis/plasma/base.py", line 147, in update
    self.plasma_properties_dict[module_name].update()
  File "/home/barna/anaconda/envs/tardis/lib/python2.7/site-packages/tardis/plasma/properties/base.py", line 97, in update
    *self._get_input_values()))
  File "/home/barna/anaconda/envs/tardis/lib/python2.7/site-packages/tardis/plasma/properties/partition_function.py", line 216, in _calculate_general
    previous_electron_densities)
  File "/home/barna/anaconda/envs/tardis/lib/python2.7/site-packages/tardis/plasma/properties/partition_function.py", line 159, in _main_nlte_calculation
    np.linalg.solve(rates_matrix[:, :, i], x)
  File "/home/barna/anaconda/envs/tardis/lib/python2.7/site-packages/numpy/linalg/linalg.py", line 381, in solve
    r = gufunc(a, b, signature=signature, extobj=extobj)
  File "/home/barna/anaconda/envs/tardis/lib/python2.7/site-packages/numpy/linalg/linalg.py", line 90, in _raise_linalgerror_singular
    raise LinAlgError("Singular matrix")
numpy.linalg.linalg.LinAlgError: Singular matrix
unoebauer commented 8 years ago

@aoifeboyle any idea why it should fail specifically for O I? Or is this just coincidence?

pfreddy commented 8 years ago

No coincidence, I chose the neutral oxygen. :)

As I mentioned above, the NLTE treatment works fine for a few ions (e.g. Ca 2 or Si 3), but fails in other cases (like O 1).

unoebauer commented 7 years ago

@chvogl - please comment

livnehra commented 7 years ago

Hi,

Just ran into this issue. Confirming [Ca 2] works, but [Si 2] raises LinAlgError: Singular matrix.

chvogl commented 7 years ago

This problem is due to errors in the currently used atomic dataset. The dataset contains levels that do not have any allowed radiative transitions (upwards- or downwards). As a result the rate matrix can contain rows of zeros, which causes a Singular matrix error when trying to solve the linear system of equations. For species such as [Ca 2], which do not have such levels, the NLTE excitation solver works fine. The respective levels seem to be duplicate levels, which have been erroneously written to the atomic datafile.

chvogl commented 7 years ago

The first step to resolve this issue would be to check if atomic data files created with the new Carsus database also feature the above mentioned duplicate levels.

unoebauer commented 7 years ago

This is something for you @yeganer!!

livnehra commented 7 years ago

Hi, I recently updated my installation of Tardis and found that NLTE stopped working completely.... Before updating it did work for part of the ions.

Now I get two types of errors: If I list only one species (e.g. [Ca 2]), I get:

[tardis.model.base   ][WARNING]  Abundances have not been normalized to 1. - normalizing (base.py:348)
tardis.model.base - WARNING - Abundances have not been normalized to 1. - normalizing
[tardis.plasma.standard_plasmas][INFO   ]  Reading Atomic Data from kurucz_cd23_chianti_H_He.h5 (standard_plasmas.py:72)
tardis.plasma.standard_plasmas - INFO - Reading Atomic Data from kurucz_cd23_chianti_H_He.h5
tardis.atomic - INFO - Read Atom Data with UUID=5ca3035ca8b311e3bb684437e69d75d7 and MD5=21095dd25faa1683f4c90c911a00c3f8
tardis.atomic - INFO - Preparing the NLTE data
[tardis.plasma.base  ][DEBUG  ]  Updating modules in the following order: (base.py:195)
tardis.plasma.base - DEBUG - Updating modules in the following order:
[tardis.plasma.properties.atomic][WARNING]  Zeta_data missing - replaced with 1s. Missing ions: [(12, 13), (14, 15), (16, 17), (18, 19), (20, 21), (22, 23), (24, 25), (26, 27), (27, 28), (28, 29)] (atomic.py:184)
tardis.plasma.properties.atomic - WARNING - Zeta_data missing - replaced with 1s. Missing ions: [(12, 13), (14, 15), (16, 17), (18, 19), (20, 21), (22, 23), (24, 25), (26, 27), (27, 28), (28, 29)]
[tardis.plasma.properties.partition_function][INFO   ]  Calculating rates for species (16, 1) (partition_function.py:119)
tardis.plasma.properties.partition_function - INFO - Calculating rates for species (16, 1)
---------------------------------------------------------------------------
KeyError                                  Traceback (most recent call last)
<ipython-input-22-42401495d0e2> in <module>()
----> 1 model = run_tardis('M07_M07_11_18.5_tardis_Ni_nlte.yml')

/home/ran/anaconda2/envs/tardis/lib/python2.7/site-packages/tardis/base.pyc in run_tardis(config, atom_data)
     35         tardis_config = Configuration.from_config_dict(config)
     36 
---> 37     simulation = Simulation.from_config(tardis_config, atom_data=atom_data)
     38     simulation.run()
     39 

/home/ran/anaconda2/envs/tardis/lib/python2.7/site-packages/tardis/simulation/base.pyc in from_config(cls, config, **kwargs)
    377         else:
    378             plasma = assemble_plasma(config, model,
--> 379                                      atom_data=kwargs.get('atom_data', None))
    380         if 'runner' in kwargs:
    381             runner = kwargs['runner']

/home/ran/anaconda2/envs/tardis/lib/python2.7/site-packages/tardis/plasma/standard_plasmas.pyc in assemble_plasma(config, model, atom_data)
    151 
    152     plasma = BasePlasma(plasma_properties=plasma_modules,
--> 153                         property_kwargs=property_kwargs, **kwargs)
    154 
    155     return plasma

/home/ran/anaconda2/envs/tardis/lib/python2.7/site-packages/tardis/plasma/base.pyc in __init__(self, plasma_properties, property_kwargs, **kwargs)
     22         self._build_graph()
     23 #        self.write_to_tex('Plasma_Graph')
---> 24         self.update(**kwargs)
     25 
     26     def __getattr__(self, item):

/home/ran/anaconda2/envs/tardis/lib/python2.7/site-packages/tardis/plasma/base.pyc in update(self, **kwargs)
    156 
    157         for module_name in self._resolve_update_list(kwargs.keys()):
--> 158             self.plasma_properties_dict[module_name].update()
    159 
    160     def _update_module_type_str(self):

/home/ran/anaconda2/envs/tardis/lib/python2.7/site-packages/tardis/plasma/properties/base.pyc in update(self)
    124         if len(self.outputs) == 1:
    125             setattr(self, self.outputs[0], self.calculate(
--> 126                 *self._get_input_values()))
    127         else:
    128             new_values = self.calculate(*self._get_input_values())

/home/ran/anaconda2/envs/tardis/lib/python2.7/site-packages/tardis/plasma/properties/radiative_properties.pyc in calculate(self, g, level_number_density, lines_lower_level_index, lines_upper_level_index, metastability, lines)
     75                                    (lines.ion_number == species[1])
     76             stimulated_emission_factor[(stimulated_emission_factor < 0) &
---> 77                 nlte_lines_mask[np.newaxis].T] = 0.0
     78         return stimulated_emission_factor
     79 

/home/ran/anaconda2/envs/tardis/lib/python2.7/site-packages/pandas/core/series.pyc in __getitem__(self, key)
    599         key = com._apply_if_callable(key, self)
    600         try:
--> 601             result = self.index.get_value(self, key)
    602 
    603             if not is_scalar(result):

/home/ran/anaconda2/envs/tardis/lib/python2.7/site-packages/pandas/core/indexes/base.pyc in get_value(self, series, key)
   2475         try:
   2476             return self._engine.get_value(s, k,
-> 2477                                           tz=getattr(series.dtype, 'tz', None))
   2478         except KeyError as e1:
   2479             if len(self) > 0 and self.inferred_type in ['integer', 'boolean']:

pandas/_libs/index.pyx in pandas._libs.index.IndexEngine.get_value (pandas/_libs/index.c:4404)()

pandas/_libs/index.pyx in pandas._libs.index.IndexEngine.get_value (pandas/_libs/index.c:4087)()

pandas/_libs/index.pyx in pandas._libs.index.IndexEngine.get_loc (pandas/_libs/index.c:5210)()

KeyError: None

If I list 2 or more (e.g. ['S II','Ca II','Si II']), I get:

[tardis.model.base   ][WARNING]  Abundances have not been normalized to 1. - normalizing (base.py:348)
tardis.model.base - WARNING - Abundances have not been normalized to 1. - normalizing
[tardis.plasma.standard_plasmas][INFO   ]  Reading Atomic Data from kurucz_cd23_chianti_H_He.h5 (standard_plasmas.py:72)
tardis.plasma.standard_plasmas - INFO - Reading Atomic Data from kurucz_cd23_chianti_H_He.h5
tardis.atomic - INFO - Read Atom Data with UUID=5ca3035ca8b311e3bb684437e69d75d7 and MD5=21095dd25faa1683f4c90c911a00c3f8
tardis.atomic - INFO - Preparing the NLTE data
[tardis.plasma.base  ][DEBUG  ]  Updating modules in the following order: (base.py:195)
tardis.plasma.base - DEBUG - Updating modules in the following order:
[tardis.plasma.properties.atomic][WARNING]  Zeta_data missing - replaced with 1s. Missing ions: [(12, 13), (14, 15), (16, 17), (18, 19), (20, 21), (22, 23), (24, 25), (26, 27), (27, 28), (28, 29)] (atomic.py:184)
tardis.plasma.properties.atomic - WARNING - Zeta_data missing - replaced with 1s. Missing ions: [(12, 13), (14, 15), (16, 17), (18, 19), (20, 21), (22, 23), (24, 25), (26, 27), (27, 28), (28, 29)]
[tardis.plasma.properties.partition_function][INFO   ]  Calculating rates for species (16, 1) (partition_function.py:119)
tardis.plasma.properties.partition_function - INFO - Calculating rates for species (16, 1)
---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-23-42401495d0e2> in <module>()
----> 1 model = run_tardis('M07_M07_11_18.5_tardis_Ni_nlte.yml')

/home/ran/anaconda2/envs/tardis/lib/python2.7/site-packages/tardis/base.pyc in run_tardis(config, atom_data)
     35         tardis_config = Configuration.from_config_dict(config)
     36 
---> 37     simulation = Simulation.from_config(tardis_config, atom_data=atom_data)
     38     simulation.run()
     39 

/home/ran/anaconda2/envs/tardis/lib/python2.7/site-packages/tardis/simulation/base.pyc in from_config(cls, config, **kwargs)
    377         else:
    378             plasma = assemble_plasma(config, model,
--> 379                                      atom_data=kwargs.get('atom_data', None))
    380         if 'runner' in kwargs:
    381             runner = kwargs['runner']

/home/ran/anaconda2/envs/tardis/lib/python2.7/site-packages/tardis/plasma/standard_plasmas.pyc in assemble_plasma(config, model, atom_data)
    151 
    152     plasma = BasePlasma(plasma_properties=plasma_modules,
--> 153                         property_kwargs=property_kwargs, **kwargs)
    154 
    155     return plasma

/home/ran/anaconda2/envs/tardis/lib/python2.7/site-packages/tardis/plasma/base.pyc in __init__(self, plasma_properties, property_kwargs, **kwargs)
     22         self._build_graph()
     23 #        self.write_to_tex('Plasma_Graph')
---> 24         self.update(**kwargs)
     25 
     26     def __getattr__(self, item):

/home/ran/anaconda2/envs/tardis/lib/python2.7/site-packages/tardis/plasma/base.pyc in update(self, **kwargs)
    156 
    157         for module_name in self._resolve_update_list(kwargs.keys()):
--> 158             self.plasma_properties_dict[module_name].update()
    159 
    160     def _update_module_type_str(self):

/home/ran/anaconda2/envs/tardis/lib/python2.7/site-packages/tardis/plasma/properties/base.pyc in update(self)
    124         if len(self.outputs) == 1:
    125             setattr(self, self.outputs[0], self.calculate(
--> 126                 *self._get_input_values()))
    127         else:
    128             new_values = self.calculate(*self._get_input_values())

/home/ran/anaconda2/envs/tardis/lib/python2.7/site-packages/tardis/plasma/properties/partition_function.pyc in _calculate_general(self, t_electrons, lines, atomic_data, nlte_data, general_level_boltzmann_factor, j_blues, previous_beta_sobolev, previous_electron_densities)
    208             atomic_data, nlte_data, t_electrons, j_blues,
    209             beta_sobolevs, general_level_boltzmann_factor,
--> 210             previous_electron_densities)
    211         return general_level_boltzmann_factor
    212 

/home/ran/anaconda2/envs/tardis/lib/python2.7/site-packages/tardis/plasma/properties/partition_function.pyc in _main_nlte_calculation(self, atomic_data, nlte_data, t_electrons, j_blues, beta_sobolevs, general_level_boltzmann_factor, previous_electron_densities)
    116         """
    117         for species in nlte_data.nlte_species:
--> 118             j_blues = j_blues.values
    119             logger.info('Calculating rates for species %s', species)
    120             number_of_levels = atomic_data.levels.energy.ix[species].count()

AttributeError: 'numpy.ndarray' object has no attribute 'values'
yeganer commented 7 years ago

@livnehra Can you please open a new issue and add a zip file containing the database and all information needed to run your setup. I'll be working on the NLTE part of tardis in the next weeks and this issue has to be solved.

livnehra commented 7 years ago

Sure, opened issue #777 . Tell me if I can help in any way. Thanks!

unoebauer commented 6 years ago

@yeganer - this should work now, right?

unoebauer commented 6 years ago

@yeganer ensured me that with the new default atomic data set (which is in the tardis-refdata repo) this error should not occur any more. I am closing this one until further notice.