def read_mat(self, file_name, var_names=None):
"""Read MATLAB file and store variables in exposures.
Parameters:
file_name (str): absolute path file
var_names (dict, optional): dictionary containing the name of the
MATLAB variables. Default: DEF_VAR_MAT.
"""
LOGGER.info('Reading %s', file_name)
if not var_names:
var_names = DEF_VAR_MAT
data = u_hdf5.read(file_name)
try:
data = data[var_names['sup_field_name']]
except KeyError:
pass
try:
data = data[var_names['field_name']]
exposures = dict()
_read_mat_obligatory(exposures, data, var_names)
_read_mat_optional(exposures, data, var_names)
except KeyError as var_err:
LOGGER.error("Not existing variable: %s", str(var_err))
raise var_err
self.gdf = GeoDataFrame(data=exposures, crs=self.crs)
_read_mat_metadata(self, data, file_name, var_names)
def read_mat(self, file_name, var_names=DEF_VAR_MAT):
"""Read MATLAB file and store variables in exposures.
Parameters:
file_name (str): absolute path file
var_names (dict, optional): dictionary containing the name of the
MATLAB variables. Default: DEF_VAR_MAT.
"""
if var_names is None:
var_names = DEF_VAR_MAT
data = hdf5.read(file_name)
try:
data = data[var_names['sup_field_name']]
except KeyError:
pass
try:
data = data[var_names['field_name']]
exposures = dict()
_read_mat_obligatory(exposures, data, var_names)
_read_mat_optional(exposures, data, var_names)
except KeyError as var_err:
LOGGER.error("Not existing variable: %s", str(var_err))
raise var_err
Exposures.__init__(self, data=exposures)
_read_mat_metadata(self, data, file_name, var_names)
def read_mat(self, file_name, description='', var_names=DEF_VAR_MAT):
"""Read MATLAB file generated with previous MATLAB CLIMADA version.
Parameters:
file_name (str): absolute file name
description (str, optional): description of the data
var_names (dict, optional): name of the variables in the file
"""
disc = hdf5.read(file_name)
self.clear()
self.tag.file_name = file_name
self.tag.description = description
try:
disc = disc[var_names['sup_field_name']]
except KeyError:
pass
try:
disc = disc[var_names['field_name']]
self.years = np.squeeze(disc[var_names['var_name']['year']]). \
astype(int, copy=False)
self.rates = np.squeeze(disc[var_names['var_name']['disc']])
except KeyError as err:
LOGGER.error("Not existing variable: %s", str(err))
raise err
def read_mat(self, file_name, description='', var_names=DEF_VAR_MAT):
"""
Read MATLAB file generated with previous MATLAB CLIMADA version.
Parameters
----------
file_name: str
filename including path and extension
description: str, optional
description of the data. The default is ''
var_names: dict, optional
name of the variables in the file. The Default is
DEF_VAR_MAT = {'sup_field_name': 'entity', 'field_name': 'discount',
'var_name': {'year': 'year', 'disc': 'discount_rate'}}
"""
disc = u_hdf5.read(file_name)
self.clear()
self.tag.file_name = str(file_name)
self.tag.description = description
try:
disc = disc[var_names['sup_field_name']]
except KeyError:
pass
try:
disc = disc[var_names['field_name']]
self.years = np.squeeze(disc[var_names['var_name']['year']]). \
astype(int, copy=False)
self.rates = np.squeeze(disc[var_names['var_name']['disc']])
except KeyError as err:
raise KeyError("Not existing variable: %s" % str(err)) from err
def read_mat(self, file_name, description='', centroids=None,
var_names=DEF_VAR_MAT):
"""Read climada hazard generate with the MATLAB code.
Parameters:
file_name (str): absolute file name
description (str, optional): description of the data
centroids (Centroids, optional): provide centroids if not contained
in the file
var_names (dict, default): name of the variables in the file,
default: DEF_VAR_MAT constant
Raises:
KeyError
"""
self.clear()
self.tag.file_name = file_name
self.tag.description = description
try:
data = hdf5.read(file_name)
try:
data = data[var_names['field_name']]
except KeyError:
pass
haz_type = hdf5.get_string(data[var_names['var_name']['per_id']])
self.tag.haz_type = haz_type
self._read_centroids(centroids=centroids, var_names=var_names['var_cent'])
self._read_att_mat(data, file_name, var_names)
except KeyError as var_err:
LOGGER.error("Not existing variable: %s", str(var_err))
raise var_err
def read_mat(self, file_name, var_names=None):
"""Read MATLAB file and store variables in exposures.
Parameters:
file_name (str): absolute path file
var_names (dict, optional): dictionary containing the name of the
MATLAB variables. Default: DEF_VAR_MAT.
"""
LOGGER.info('Reading %s', file_name)
if not var_names:
var_names = DEF_VAR_MAT
data = u_hdf5.read(file_name)
try:
data = data[var_names['sup_field_name']]
except KeyError:
pass
try:
data = data[var_names['field_name']]
exposures = dict()
_read_mat_obligatory(exposures, data, var_names)
_read_mat_optional(exposures, data, var_names)
except KeyError as var_err:
raise KeyError(f"Variable not in MAT file: {var_names.get('field_name')}")\
from var_err
self.set_gdf(GeoDataFrame(data=exposures))
_read_mat_metadata(self, data, file_name, var_names)
def test_with_refs_pass(self):
'''Allow to load references of the matlab file'''
# Load input
refs = True
contents = hdf5.read(HAZ_TEST_MAT, refs)
# Check read contents
self.assertEqual(2, len(contents))
self.assertTrue('hazard' in contents.keys())
self.assertTrue('#refs#' in contents.keys())
def test_hazard_pass(self):
'''Checking result against matlab atl_prob.mat file'''
# Load input
contents = hdf5.read(HAZ_TEST_MAT)
# Check read contents
self.assertEqual(1, len(contents))
self.assertTrue('hazard' in contents.keys())
self.assertEqual(False, '#refs#' in contents.keys())
hazard = contents['hazard']
self.assertTrue('reference_year' in hazard.keys())
self.assertTrue('lon' in hazard.keys())
self.assertTrue('lat' in hazard.keys())
self.assertTrue('centroid_ID' in hazard.keys())
self.assertTrue('orig_years' in hazard.keys())
self.assertTrue('orig_event_count' in hazard.keys())
self.assertTrue('event_count' in hazard.keys())
self.assertTrue('event_ID' in hazard.keys())
self.assertTrue('category' in hazard.keys())
self.assertTrue('orig_event_flag' in hazard.keys())
self.assertTrue('yyyy' in hazard.keys())
self.assertTrue('mm' in hazard.keys())
self.assertTrue('dd' in hazard.keys())
self.assertTrue('datenum' in hazard.keys())
self.assertTrue('scenario' in hazard.keys())
self.assertTrue('intensity' in hazard.keys())
self.assertTrue('name' in hazard.keys())
self.assertTrue('frequency' in hazard.keys())
self.assertTrue('matrix_density' in hazard.keys())
self.assertTrue('windfield_comment' in hazard.keys())
self.assertTrue('peril_ID' in hazard.keys())
self.assertTrue('filename' in hazard.keys())
self.assertTrue('comment' in hazard.keys())
self.assertTrue('date' in hazard.keys())
self.assertTrue('units' in hazard.keys())
self.assertTrue('orig_yearset' in hazard.keys())
self.assertTrue('fraction' in hazard.keys())
self.assertEqual(27, len(hazard.keys()))
# Check some random values
mat_shape = (len(contents['hazard']['event_ID']), \
len(contents['hazard']['centroid_ID']))
sp_mat = hdf5.get_sparse_csr_mat(hazard['intensity'], mat_shape)
self.assertTrue(np.array_equal(np.array([[84], [67]]), \
hazard['peril_ID']))
self.assertEqual(34.537289477809473, sp_mat[2862, 97])
self.assertEqual(-80, hazard['lon'][46])
self.assertEqual(28, hazard['lat'][87])
self.assertEqual(2016, hazard['reference_year'])
def read_mat(self, file_name, var_names=DEF_VAR_MAT):
"""Read centroids from CLIMADA's MATLAB version.
Parameters
----------
file_name : str
absolute or relative file name
var_names : dict, default
name of the variables
Raises
------
KeyError
"""
LOGGER.info('Reading %s', file_name)
if var_names is None:
var_names = DEF_VAR_MAT
cent = u_hdf5.read(file_name)
# Try open encapsulating variable FIELD_NAMES
num_try = 0
for field in var_names['field_names']:
try:
cent = cent[field]
break
except KeyError:
num_try += 1
if num_try == len(var_names['field_names']):
LOGGER.warning("Variables are not under: %s.",
var_names['field_names'])
try:
cen_lat = np.squeeze(cent[var_names['var_name']['lat']])
cen_lon = np.squeeze(cent[var_names['var_name']['lon']])
self.set_lat_lon(cen_lat, cen_lon)
try:
self.dist_coast = np.squeeze(
cent[var_names['var_name']['dist_coast']])
except KeyError:
pass
try:
self.region_id = np.squeeze(
cent[var_names['var_name']['region_id']])
except KeyError:
pass
except KeyError as err:
LOGGER.error("Not existing variable: %s", str(err))
raise err
def test_get_string_pass(self):
'''Check function to get a string from input integer array'''
# Load input
contents = hdf5.read(HAZ_TEST_MAT)
# Convert several strings
str_date = hdf5.get_string(contents['hazard']['date'])
str_comment = hdf5.get_string(contents['hazard']['comment'])
str_wf = hdf5.get_string(contents['hazard']['windfield_comment'])
str_fn = hdf5.get_string(contents['hazard']['filename'])
# Check results
self.assertEqual('14-Nov-2017 10:09:05', str_date)
self.assertEqual(
'TC hazard event set, generated 14-Nov-2017 10:09:05', \
str_comment)
self.assertEqual(
'generating 14450 windfields took 0.25 min ' + \
'(0.0010 sec/event)', str_wf)
self.assertEqual('/Users/aznarsig/Documents/MATLAB/climada_data/' + \
'hazards/atl_prob.mat', str_fn)
def from_mat(cls, file_name, description='', var_names=None):
"""
Read MATLAB file generated with previous MATLAB CLIMADA version.
Parameters
----------
file_name: str
filename including path and extension
description: str, optional
description of the data. The default is ''
var_names: dict, optional
name of the variables in the file. The Default is
DEF_VAR_MAT = {'sup_field_name': 'entity', 'field_name': 'discount',
'var_name': {'year': 'year', 'disc': 'discount_rate'}}
Returns
-------
climada.entity.DiscRates :
The disc rates from matlab
"""
if var_names is None:
var_names = DEF_VAR_MAT
disc = u_hdf5.read(file_name)
tag = Tag(file_name=str(file_name), description=description)
try:
disc = disc[var_names['sup_field_name']]
except KeyError:
pass
try:
disc = disc[var_names['field_name']]
years = np.squeeze(disc[var_names['var_name']['year']]). \
astype(int, copy=False)
rates = np.squeeze(disc[var_names['var_name']['disc']])
except KeyError as err:
raise KeyError("Not existing variable: %s" % str(err)) from err
return cls(years=years, rates=rates, tag=tag)
def test_get_sparse_mat_pass(self):
'''Check contents of imported sparse matrix, using the function \
to build a sparse matrix from the read HDF5 variable'''
# Load input
contents = hdf5.read(HAZ_TEST_MAT)
# get matrix size
mat_shape = (len(contents['hazard']['event_ID']), \
len(contents['hazard']['centroid_ID']))
spr_mat = hdf5.get_sparse_csr_mat(contents['hazard']['intensity'], \
mat_shape)
self.assertEqual(mat_shape[0], spr_mat.shape[0])
self.assertEqual(mat_shape[1], spr_mat.shape[1])
self.assertEqual(0, spr_mat[0, 0])
self.assertEqual(9.2029355562400745, spr_mat[7916, 98])
self.assertEqual(34.426829019435729, spr_mat[12839, 96])
self.assertEqual(61.81217342446773, spr_mat[1557, 97])
self.assertEqual(6.4470644550625753, spr_mat[6658, 97])
self.assertEqual(0, spr_mat[298, 9])
self.assertEqual(19.385821399329995, spr_mat[15, 0])
self.assertEqual(52.253444444444447, spr_mat[76, 95])
self.assertEqual(0, spr_mat[126, 86])
def read_mat(self, file_name, description='', var_names=DEF_VAR_MAT):
"""Read MATLAB file generated with previous MATLAB CLIMADA version.
Parameters:
file_name (str): absolute file name
description (str, optional): description of the data
var_names (dict, optional): name of the variables in the file
"""
def read_att_mat(measures, data, file_name, var_names):
"""Read MATLAB measures attributes"""
num_mes = len(data[var_names['var_name']['name']])
for idx in range(0, num_mes):
meas = Measure()
meas.name = hdf5.get_str_from_ref(
file_name, data[var_names['var_name']['name']][idx][0])
color_str = hdf5.get_str_from_ref(
file_name, data[var_names['var_name']['color']][idx][0])
meas.color_rgb = np.fromstring(color_str, dtype=float, sep=' ')
meas.cost = data[var_names['var_name']['cost']][idx][0]
meas.haz_type = hdf5.get_str_from_ref(
file_name, data[var_names['var_name']['haz']][idx][0])
meas.hazard_freq_cutoff = data[var_names['var_name']
['haz_frq']][idx][0]
meas.hazard_set = hdf5.get_str_from_ref(
file_name, data[var_names['var_name']['haz_set']][idx][0])
try:
meas.hazard_inten_imp = (
data[var_names['var_name']['haz_int_a']][idx][0],
data[var_names['var_name']['haz_int_b']][0][idx])
except KeyError:
meas.hazard_inten_imp = (
data[var_names['var_name']['haz_int_a'][:-2]][idx][0],
0)
# different convention of signes followed in MATLAB!
meas.mdd_impact = (
data[var_names['var_name']['mdd_a']][idx][0],
data[var_names['var_name']['mdd_b']][idx][0])
meas.paa_impact = (
data[var_names['var_name']['paa_a']][idx][0],
data[var_names['var_name']['paa_b']][idx][0])
meas.imp_fun_map = hdf5.get_str_from_ref(
file_name, data[var_names['var_name']['fun_map']][idx][0])
meas.exposures_set = hdf5.get_str_from_ref(
file_name, data[var_names['var_name']['exp_set']][idx][0])
exp_region_id = data[var_names['var_name']['exp_reg']][idx][0]
if exp_region_id:
meas.exp_region_id = [exp_region_id]
meas.risk_transf_attach = data[var_names['var_name']
['risk_att']][idx][0]
meas.risk_transf_cover = data[var_names['var_name']
['risk_cov']][idx][0]
measures.append(meas)
data = hdf5.read(file_name)
self.clear()
self.tag.file_name = file_name
self.tag.description = description
try:
data = data[var_names['sup_field_name']]
except KeyError:
pass
try:
data = data[var_names['field_name']]
read_att_mat(self, data, file_name, var_names)
except KeyError as var_err:
LOGGER.error("Not existing variable %s", str(var_err))
raise var_err
def read_mat(self, file_name, description='', var_names=DEF_VAR_MAT):
"""Read MATLAB file generated with previous MATLAB CLIMADA version.
Parameters:
file_name (str): absolute file name
description (str, optional): description of the data
var_names (dict, optional): name of the variables in the file
"""
def _get_hdf5_funcs(imp, file_name, var_names):
"""Get rows that fill every impact function and its name."""
func_pos = dict()
for row, (fun_id, fun_type) in enumerate(
zip(imp[var_names['var_name']['fun_id']].squeeze(),
imp[var_names['var_name']['peril']].squeeze())):
type_str = u_hdf5.get_str_from_ref(file_name, fun_type)
key = (type_str, int(fun_id))
if key not in func_pos:
func_pos[key] = list()
func_pos[key].append(row)
return func_pos
def _get_hdf5_str(imp, idxs, file_name, var_name):
"""Get rows with same string in var_name."""
prev_str = ""
for row in idxs:
cur_str = u_hdf5.get_str_from_ref(file_name, imp[var_name][row][0])
if prev_str == "":
prev_str = cur_str
elif prev_str != cur_str:
LOGGER.error("Impact function with two different %s.", var_name)
raise ValueError
return prev_str
imp = u_hdf5.read(file_name)
self.clear()
self.tag.file_name = str(file_name)
self.tag.description = description
try:
imp = imp[var_names['sup_field_name']]
except KeyError:
pass
try:
imp = imp[var_names['field_name']]
funcs_idx = _get_hdf5_funcs(imp, file_name, var_names)
for imp_key, imp_rows in funcs_idx.items():
func = ImpactFunc()
func.haz_type = imp_key[0]
func.id = imp_key[1]
# check that this function only has one intensity unit, if provided
try:
func.intensity_unit = _get_hdf5_str(imp, imp_rows,
file_name,
var_names['var_name']['unit'])
except KeyError:
pass
# check that this function only has one name
try:
func.name = _get_hdf5_str(imp, imp_rows, file_name,
var_names['var_name']['name'])
except KeyError:
func.name = str(func.id)
func.intensity = np.take(imp[var_names['var_name']['inten']], imp_rows)
func.mdd = np.take(imp[var_names['var_name']['mdd']], imp_rows)
func.paa = np.take(imp[var_names['var_name']['paa']], imp_rows)
self.append(func)
except KeyError as err:
LOGGER.error("Not existing variable: %s", str(err))
raise err
def from_mat(cls, file_name, description='', var_names=DEF_VAR_MAT):
"""Read MATLAB file generated with previous MATLAB CLIMADA version.
Parameters
----------
file_name : str
absolute file name
description : str, optional
description of the data
var_names : dict, optional
name of the variables in the file
Returns
-------
meas_set: climada.entity.MeasureSet()
Measure Set from matlab file
"""
def read_att_mat(measures, data, file_name, var_names):
"""Read MATLAB measures attributes"""
num_mes = len(data[var_names['var_name']['name']])
for idx in range(0, num_mes):
meas = Measure()
meas.name = u_hdf5.get_str_from_ref(
file_name, data[var_names['var_name']['name']][idx][0])
color_str = u_hdf5.get_str_from_ref(
file_name, data[var_names['var_name']['color']][idx][0])
meas.color_rgb = np.fromstring(color_str, dtype=float, sep=' ')
meas.cost = data[var_names['var_name']['cost']][idx][0]
meas.haz_type = u_hdf5.get_str_from_ref(
file_name, data[var_names['var_name']['haz']][idx][0])
meas.hazard_freq_cutoff = data[var_names['var_name']
['haz_frq']][idx][0]
meas.hazard_set = u_hdf5.get_str_from_ref(
file_name, data[var_names['var_name']['haz_set']][idx][0])
try:
meas.hazard_inten_imp = (
data[var_names['var_name']['haz_int_a']][idx][0],
data[var_names['var_name']['haz_int_b']][0][idx])
except KeyError:
meas.hazard_inten_imp = (
data[var_names['var_name']['haz_int_a'][:-2]][idx][0],
0)
# different convention of signes followed in MATLAB!
meas.mdd_impact = (
data[var_names['var_name']['mdd_a']][idx][0],
data[var_names['var_name']['mdd_b']][idx][0])
meas.paa_impact = (
data[var_names['var_name']['paa_a']][idx][0],
data[var_names['var_name']['paa_b']][idx][0])
meas.imp_fun_map = u_hdf5.get_str_from_ref(
file_name, data[var_names['var_name']['fun_map']][idx][0])
meas.exposures_set = u_hdf5.get_str_from_ref(
file_name, data[var_names['var_name']['exp_set']][idx][0])
exp_region_id = data[var_names['var_name']['exp_reg']][idx][0]
if exp_region_id:
meas.exp_region_id = [exp_region_id]
meas.risk_transf_attach = data[var_names['var_name']
['risk_att']][idx][0]
meas.risk_transf_cover = data[var_names['var_name']
['risk_cov']][idx][0]
measures.append(meas)
data = u_hdf5.read(file_name)
meas_set = cls()
meas_set.tag.file_name = str(file_name)
meas_set.tag.description = description
try:
data = data[var_names['sup_field_name']]
except KeyError:
pass
try:
data = data[var_names['field_name']]
read_att_mat(meas_set, data, file_name, var_names)
except KeyError as var_err:
raise KeyError("Variable not in MAT file: " +
str(var_err)) from var_err
return meas_set
def from_mat(cls, file_name, description='', var_names=None):
"""Read MATLAB file generated with previous MATLAB CLIMADA version.
Parameters
----------
file_name : str
absolute file name
description : str, optional
description of the data
var_names : dict, optional
name of the variables in the file
Return
------
impf_set : climada.entity.impact_func_set.ImpactFuncSet
Impact func set as defined in matlab file.
"""
if var_names is None:
var_names = DEF_VAR_MAT
def _get_hdf5_funcs(imp, file_name, var_names):
"""Get rows that fill every impact function and its name."""
func_pos = dict()
for row, (fun_id, fun_type) in enumerate(
zip(imp[var_names['var_name']['fun_id']].squeeze(),
imp[var_names['var_name']['peril']].squeeze())):
type_str = u_hdf5.get_str_from_ref(file_name, fun_type)
key = (type_str, int(fun_id))
if key not in func_pos:
func_pos[key] = list()
func_pos[key].append(row)
return func_pos
def _get_hdf5_str(imp, idxs, file_name, var_name):
"""Get rows with same string in var_name."""
prev_str = ""
for row in idxs:
cur_str = u_hdf5.get_str_from_ref(file_name,
imp[var_name][row][0])
if prev_str == "":
prev_str = cur_str
elif prev_str != cur_str:
raise ValueError("Impact function with two different %s." %
var_name)
return prev_str
imp = u_hdf5.read(file_name)
impf_set = cls()
impf_set.tag.file_name = str(file_name)
impf_set.tag.description = description
try:
imp = imp[var_names['sup_field_name']]
except KeyError:
pass
try:
imp = imp[var_names['field_name']]
funcs_idx = _get_hdf5_funcs(imp, file_name, var_names)
for imp_key, imp_rows in funcs_idx.items():
func = ImpactFunc()
func.haz_type = imp_key[0]
func.id = imp_key[1]
# check that this function only has one intensity unit, if provided
try:
func.intensity_unit = _get_hdf5_str(
imp, imp_rows, file_name,
var_names['var_name']['unit'])
except KeyError:
pass
# check that this function only has one name
try:
func.name = _get_hdf5_str(imp, imp_rows, file_name,
var_names['var_name']['name'])
except KeyError:
func.name = str(func.id)
func.intensity = np.take(imp[var_names['var_name']['inten']],
imp_rows)
func.mdd = np.take(imp[var_names['var_name']['mdd']], imp_rows)
func.paa = np.take(imp[var_names['var_name']['paa']], imp_rows)
impf_set.append(func)
except KeyError as err:
raise KeyError("Not existing variable: %s" % str(err)) from err
return impf_set