configuration.py 13.8 KB
Newer Older
1
# -*- coding: utf-8 -*-
2 3
# Copyright (C) 2012, 2013 Centre de données Astrophysiques de Marseille
# Licensed under the CeCILL-v2 licence - see Licence_CeCILL_V2-en.txt
Yannick Roehlly's avatar
Yannick Roehlly committed
4
# Author: Yannick Roehlly
5 6

import pkgutil
7
from collections import Iterable, OrderedDict
8
import multiprocessing as mp
9
from textwrap import wrap
10 11 12 13 14 15

import configobj
from glob import glob  # To allow the use of glob() in "eval..."
import pkg_resources
import numpy as np

16
from ..data import Database
17
from ..utils import read_table
18 19
from .. import creation_modules
from .. import analysis_modules
20

21 22 23
# Limit the redshift to this number of decimals
REDSHIFT_DECIMALS = 2

24 25 26 27

def list_modules(package_name):
    """Lists the modules available in a package

Yannick Roehlly's avatar
Yannick Roehlly committed
28
    Parameters
29
    ----------
30
    package_name: string
31
        Name of the package (e.g. pcigale.creation_modules).
32 33 34

    Returns
    -------
35
    module_name: array of strings
36 37 38 39 40 41 42 43 44
        List of the available modules.

    """
    directory = pkg_resources.resource_filename(package_name, '')
    module_names = [name for _, name, _ in pkgutil.iter_modules([directory])]

    return module_names


Yannick Roehlly's avatar
Yannick Roehlly committed
45 46 47 48 49 50
def evaluate_description(description):
    """Evaluate a description from the config file as a list.

    The description is read from the config file by configobj that transforms
    coma separated value in a list. From this description, this function try
    to evaluate the desired list of values:
51 52 53
    - If the description is a string beginning with 'eval ', then its content
      (without 'eval ') is evaluated as Python code and its result returned.
      An array is expected.
54 55 56
    - If the description is a string beginning by 'range', the start, step and
      stop values are then expected and the range is evaluated (stop included
      if reached.
Yannick Roehlly's avatar
Yannick Roehlly committed
57 58
    - Then the function tries to evaluate the description as a Numpy array of
      float and returns the mere list if this fails.
59

Yannick Roehlly's avatar
Yannick Roehlly committed
60
    Parameters
61
    ----------
62
    description: string or list
63 64 65 66
        The description to be evaluated.

    Returns
    -------
67
     results: list
Yannick Roehlly's avatar
Yannick Roehlly committed
68
        The evaluated list of values.
69 70

    """
71 72 73 74 75
    results = description
    if type(description) == str:
        if description.startswith('eval '):
            results = eval(description[4:])
            # If the evaluation lead to a single value, we put it in a list.
76
            if not isinstance(results, Iterable):
77 78 79 80 81 82 83 84
                results = [results]
        elif description.startswith('range '):
            start, stop, step = [float(item) for item
                                 in description[5:].split()]
            results = np.arange(start, stop+step, step)
        else:
            # We need to return a list to combine the list of possible values
            # for each parameter.
85
            results = [results]
86 87 88 89 90 91 92

    # We prefer to evaluate the parameter as a numpy array of floats if
    # possible.
    try:
        results = np.array(results, float)
    except ValueError:
        pass
93

Yannick Roehlly's avatar
Yannick Roehlly committed
94
    return results
95 96 97 98 99 100 101 102 103


class Configuration(object):
    """This class manages the configuration of pcigale.
    """

    def __init__(self, filename="pcigale.ini"):
        """Initialise a pcigale configuration.

Yannick Roehlly's avatar
Yannick Roehlly committed
104
        Parameters
105
        ----------
106
        filename: string
107 108 109 110 111
            Name of the configuration file (pcigale.conf by default).

        """
        self.config = configobj.ConfigObj(filename,
                                          write_empty_values=True,
112 113
                                          indent_type='  ',
                                          encoding='UTF8')
114 115 116 117 118 119 120 121 122 123 124 125

    def create_blank_conf(self):
        """Create the initial configuration file

        Write the initial pcigale configuration file where the user can state
        which data file to use, which modules to use for the SED creation, as
        well as the method selected for statistical analysis.

        """

        self.config['data_file'] = ""
        self.config.comments['data_file'] = wrap(
126 127 128 129 130 131
            "File containing the input data. The columns are 'id' (name of the"
            " object), 'redshift' (if 0 the distance is assumed to be 10 pc), "
            "the filter names for the fluxes, and the filter names with the "
            "'_err' suffix for the uncertainties. The fluxes and the "
            "uncertainties must be in mJy. This file is optional to generate "
            "the configuration file, in particular for the savefluxes module.")
132

133
        self.config['creation_modules'] = []
134 135 136 137 138 139 140 141 142 143
        self.config.comments['creation_modules'] = ([""] +
            ["Order of the modules use for SED creation. Available modules:"] +
            ["SFH: sfh2exp, sfhdelayed, sfhfromfile, sfhperiodic"] +
            ["SSP: bc03, m2005"] +
            ["Nebular emission: nebular"] +
            ["Dust attenuation: dustatt_calzleit, dustatt_powerlaw"] +
            ["Dust emission: casey2012, dale2014, dl2007, dl2014"] +
            ["AGN: dale2014, fritz2006"] +
            ["Radio: radio"] +
            ["Redshift: redshifting (mandatory!)"])
144 145 146 147

        self.config['analysis_method'] = ""
        self.config.comments['analysis_method'] = [""] + wrap(
            "Method used for statistical analysis. Available methods: "
148
            "pdf_analysis, savefluxes.")
149

150 151 152 153 154
        self.config['cores'] = ""
        self.config.comments['cores'] = [""] + wrap(
            "Number of CPU cores available. This computer has {} cores."
            .format(mp.cpu_count()))

155 156 157 158 159 160 161 162 163 164 165 166
        self.config.write()

    def generate_conf(self):
        """Generate the full configuration file

        Reads the user entries in the initial configuration file and add the
        configuration options of all selected modules as well as the filter
        selection based on the filters identified in the data table file.

        """

        # Getting the list of the filters available in pcigale database
167
        with Database() as base:
168
            filter_list = base.get_filter_names()
169

170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189
        if self.config['data_file'] != '':
            obs_table = read_table(self.config['data_file'])

            # Check that the id and redshift columns are present in the input
            # file
            if 'id' not in obs_table.columns:
                raise Exception("Column id not present in input file")
            if 'redshift' not in obs_table.columns:
                raise Exception("Column redshift not present in input file")

            # Finding the known filters in the data table
            column_list = []
            for column in obs_table.columns:
                filter_name = column[:-4] if column.endswith('_err') else column
                if filter_name in filter_list:
                    column_list.append(column)

            # Check that we don't have an error column without the associated
            # flux
            for column in column_list:
Médéric Boquien's avatar
Médéric Boquien committed
190 191
                if column.endswith('_err') and (column[:-4]
                                                not in column_list):
192 193
                    raise Exception("The observation table as a {} column "
                                    "but no {} column.".format(column,
Médéric Boquien's avatar
Médéric Boquien committed
194
                                                               column[:-4]))
195 196 197 198

            self.config['column_list'] = column_list
        else:
            self.config['column_list'] = ''
199 200 201 202 203
        self.config.comments['column_list'] = [""] + wrap(
            "List of the columns in the observation data file to use for "
            "the fitting.")

        # SED creation modules configurations. For each module, we generate
Yannick Roehlly's avatar
Yannick Roehlly committed
204
        # the configuration section from its parameter list.
205 206 207 208
        self.config['sed_creation_modules'] = {}
        self.config.comments['sed_creation_modules'] = ["", ""] + wrap(
            "Configuration of the SED creation modules.")

209
        for module_name in self.config['creation_modules']:
210 211 212
            self.config["sed_creation_modules"][module_name] = {}
            sub_config = self.config["sed_creation_modules"][module_name]

213
            for name, (typ, description, default) in \
214 215 216
                    creation_modules.get_module(
                        module_name,
                        blank=True).parameter_list.items():
217 218 219 220 221 222
                if default is None:
                    default = ''
                sub_config[name] = default
                sub_config.comments[name] = wrap(description)

            self.config['sed_creation_modules'].comments[module_name] = [
223
                creation_modules.get_module(module_name, blank=True).comments]
224

225 226
        self.check_modules()

227 228 229 230 231
        # Configuration for the analysis method
        self.config['analysis_configuration'] = {}
        self.config.comments['analysis_configuration'] = ["", ""] + wrap(
            "Configuration of the statistical analysis method.")
        module_name = self.config['analysis_method']
232
        for name, (typ, desc, default) in \
233
                analysis_modules.get_module(module_name).parameter_list.items():
234 235 236 237 238 239
            if default is None:
                default = ''
            self.config['analysis_configuration'][name] = default
            self.config['analysis_configuration'].comments[name] = wrap(desc)

        self.config.write()
Yannick Roehlly's avatar
Yannick Roehlly committed
240

241 242
    @property
    def configuration(self):
Yannick Roehlly's avatar
Yannick Roehlly committed
243
        """Returns a dictionary for the session configuration.
Yannick Roehlly's avatar
Yannick Roehlly committed
244 245 246

        Returns
        -------
247
        configuration['data_file']: string
Yannick Roehlly's avatar
Yannick Roehlly committed
248
            File containing the observations to fit.
249
        configuration['column_list']: list of strings
Yannick Roehlly's avatar
Yannick Roehlly committed
250
            List of the columns of data_file to use in the fitting.
251
        configuration['creation_modules']: list of strings
Yannick Roehlly's avatar
Yannick Roehlly committed
252
            List of the modules (in the right order) used to create the SEDs.
253
        configuration['creation_modules_params']: list of dictionaries
Yannick Roehlly's avatar
Yannick Roehlly committed
254 255
            Configuration parameters for each module. To each parameter, the
            dictionary associates a list of possible values (possibly only
Yannick Roehlly's avatar
Yannick Roehlly committed
256
            one).
257
        configuration['analysis_method']: string
Yannick Roehlly's avatar
Yannick Roehlly committed
258
            Statistical analysis module used to fit the data.
259
        configuration['analysis_method_params']: dictionary
Yannick Roehlly's avatar
Yannick Roehlly committed
260
            Parameters for the statistical analysis module. To each parameter
Yannick Roehlly's avatar
Yannick Roehlly committed
261 262 263 264
            is associated a list of possible values.
        """
        configuration = {}

265 266 267 268
        # Before building the configuration dictionary, we ensure that all the
        # fields are filled
        self.complete_conf()

269
        for section in ['data_file', 'column_list', 'creation_modules',
270
                        'analysis_method']:
Yannick Roehlly's avatar
Yannick Roehlly committed
271
            configuration[section] = self.config[section]
272
        configuration['cores'] = int(self.config['cores'])
Yannick Roehlly's avatar
Yannick Roehlly committed
273

Yannick Roehlly's avatar
Yannick Roehlly committed
274
        # Parsing the SED modules parameters
275 276
        configuration['creation_modules_params'] = []
        for module in self.config['creation_modules']:
277
            module_params = {}
Yannick Roehlly's avatar
Yannick Roehlly committed
278 279
            for key, value in \
                    self.config['sed_creation_modules'][module].items():
Yannick Roehlly's avatar
Yannick Roehlly committed
280
                module_params[key] = evaluate_description(value)
281
            configuration['creation_modules_params'].append(module_params)
282 283

        # Analysis method parameters
284 285
        configuration['analysis_method_params'] = \
            self.config['analysis_configuration']
Yannick Roehlly's avatar
Yannick Roehlly committed
286 287

        return configuration
288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307

    def check_modules(self):
        """Make a basic check to ensure that some required modules are present.
        Otherwise we emit a warning so the user knows their list of modules is
        suspicious. We do not emit an exception as they may be using an
        unofficial module that is not in our list
        """

        modules = OrderedDict((('SFH', ['sfh2exp', 'sfhdelayed', 'sfhfromfile',
                                        'sfhperiodic']),
                               ('SSP', ['bc03', 'm2005']),
                               ('nebular', ['nebular']),
                               ('dust attenuation', ['dustatt_calzleit',
                                                     'dustatt_powerlaw']),
                               ('dust emission', ['casey2012', 'dale2014',
                                                  'dl2007', 'dl2014']),
                               ('AGN', ['dale2014', 'fritz2006']),
                               ('radio', ['radio']),
                               ('redshift', ['redshifting'])))

Médéric Boquien's avatar
Médéric Boquien committed
308 309
        comments = {'SFH': "ERROR! Choosing one SFH module is mandatory.",
                    'SSP': "ERROR! Choosing one SSP module is mandatory.",
310 311 312 313
                    'nebular': "WARNING! Choosing the nebular module is "
                               "recommended. Without it the Lyman continuum "
                               "is left untouched.",
                    'dust attenuation': "No dust attenuation module found.",
314
                    'dust emission': "No dust emission module found.",
315 316
                    'AGN': "No AGN module found.",
                    'radio': "No radio module found.",
Médéric Boquien's avatar
Médéric Boquien committed
317
                    'redshift': "ERROR! No redshifting module found."}
318 319 320 321 322 323

        for module in modules:
            if all([user_module not in modules[module] for user_module in
                    self.config['creation_modules']]):
                print("{} Options are: {}.".
                      format(comments[module], ', '.join(modules[module])))
324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340

    def complete_conf(self):
        """Complete the configuration when there is missing information that is
        to be extracted from other sources such as the input file (redshifts)
        or the output parameters (single run)
        """

        z_mod = self.config['sed_creation_modules']['redshifting']['redshift']
        if type(z_mod) is str and not z_mod:
            if self.config['data_file']:
                obs_table = read_table(self.config['data_file'])
                z = np.unique(np.around(obs_table['redshift'],
                                        decimals=REDSHIFT_DECIMALS))
                self.config['sed_creation_modules']['redshifting']['redshift'] = z
            else:
                raise Exception("No flux file and no redshift indicated. "
                                "The spectra cannot be computed. Aborting.")