def test_file_name_to_time_segmotion(self):
"""Ensures correct output from file_name_to_time.
In this case, data source is segmotion.
"""
this_time_unix_sec = tracking_io.file_name_to_time(SEGMOTION_FILE_NAME)
self.assertTrue(this_time_unix_sec == VALID_TIME_UNIX_SEC)
def test_file_name_to_time_probsevere(self):
"""Ensures correct output from file_name_to_time.
In this case, data source is probSevere.
"""
this_time_unix_sec = tracking_io.file_name_to_time(
PROBSEVERE_FILE_NAME)
self.assertTrue(this_time_unix_sec == VALID_TIME_UNIX_SEC)
def _run(top_orig_tracking_dir_name, top_new_tracking_dir_name,
first_spc_date_string, last_spc_date_string, output_file_name):
"""Plots storms that were removed by remove_storms_outside_conus.py.
This is effectively the main method.
:param top_orig_tracking_dir_name: See documentation at top of file.
:param top_new_tracking_dir_name: Same.
:param first_spc_date_string: Same.
:param last_spc_date_string: Same.
:param output_file_name: Same.
"""
file_system_utils.mkdir_recursive_if_necessary(file_name=output_file_name)
spc_date_strings = time_conversion.get_spc_dates_in_range(
first_spc_date_string=first_spc_date_string,
last_spc_date_string=last_spc_date_string)
orig_tracking_file_names = []
for d in spc_date_strings:
orig_tracking_file_names += tracking_io.find_files_one_spc_date(
top_tracking_dir_name=top_orig_tracking_dir_name,
tracking_scale_metres2=DUMMY_TRACKING_SCALE_METRES2,
source_name=tracking_utils.SEGMOTION_NAME,
spc_date_string=d,
raise_error_if_missing=False)[0]
valid_times_unix_sec = numpy.array(
[tracking_io.file_name_to_time(f) for f in orig_tracking_file_names],
dtype=int)
new_tracking_file_names = [
tracking_io.find_file(
top_tracking_dir_name=top_new_tracking_dir_name,
tracking_scale_metres2=DUMMY_TRACKING_SCALE_METRES2,
source_name=tracking_utils.SEGMOTION_NAME,
valid_time_unix_sec=t,
spc_date_string=time_conversion.time_to_spc_date_string(t),
raise_error_if_missing=True) for t in valid_times_unix_sec
]
orig_storm_object_table = tracking_io.read_many_files(
orig_tracking_file_names)
print(SEPARATOR_STRING)
new_storm_object_table = tracking_io.read_many_files(
new_tracking_file_names)
print(SEPARATOR_STRING)
orig_storm_id_strings = (
orig_storm_object_table[tracking_utils.FULL_ID_COLUMN].values.tolist())
orig_storm_times_unix_sec = (
orig_storm_object_table[tracking_utils.VALID_TIME_COLUMN].values)
new_storm_id_strings = (
new_storm_object_table[tracking_utils.FULL_ID_COLUMN].values.tolist())
new_storm_times_unix_sec = (
new_storm_object_table[tracking_utils.VALID_TIME_COLUMN].values)
num_orig_storm_objects = len(orig_storm_object_table.index)
orig_kept_flags = numpy.full(num_orig_storm_objects, 0, dtype=bool)
these_indices = tracking_utils.find_storm_objects(
all_id_strings=orig_storm_id_strings,
all_times_unix_sec=orig_storm_times_unix_sec,
id_strings_to_keep=new_storm_id_strings,
times_to_keep_unix_sec=new_storm_times_unix_sec,
allow_missing=False)
orig_kept_flags[these_indices] = True
orig_removed_indices = numpy.where(numpy.invert(orig_kept_flags))[0]
print('{0:d} of {1:d} storm objects were outside CONUS.'.format(
len(orig_removed_indices), num_orig_storm_objects))
removed_storm_object_table = orig_storm_object_table.iloc[
orig_removed_indices]
removed_latitudes_deg = removed_storm_object_table[
tracking_utils.CENTROID_LATITUDE_COLUMN].values
removed_longitudes_deg = removed_storm_object_table[
tracking_utils.CENTROID_LONGITUDE_COLUMN].values
figure_object, axes_object, basemap_object = (
plotting_utils.create_equidist_cylindrical_map(
min_latitude_deg=numpy.min(removed_latitudes_deg) - 1.,
max_latitude_deg=numpy.max(removed_latitudes_deg) + 1.,
min_longitude_deg=numpy.min(removed_longitudes_deg) - 1.,
max_longitude_deg=numpy.max(removed_longitudes_deg) + 1.,
resolution_string='i'))
plotting_utils.plot_coastlines(basemap_object=basemap_object,
axes_object=axes_object,
line_colour=BORDER_COLOUR)
plotting_utils.plot_countries(basemap_object=basemap_object,
axes_object=axes_object,
line_colour=BORDER_COLOUR)
plotting_utils.plot_states_and_provinces(basemap_object=basemap_object,
axes_object=axes_object,
line_colour=BORDER_COLOUR)
plotting_utils.plot_parallels(basemap_object=basemap_object,
axes_object=axes_object,
num_parallels=NUM_PARALLELS)
plotting_utils.plot_meridians(basemap_object=basemap_object,
axes_object=axes_object,
num_meridians=NUM_MERIDIANS)
conus_latitudes_deg, conus_longitudes_deg = (
conus_boundary.read_from_netcdf())
conus_latitudes_deg, conus_longitudes_deg = conus_boundary.erode_boundary(
latitudes_deg=conus_latitudes_deg,
longitudes_deg=conus_longitudes_deg,
erosion_distance_metres=EROSION_DISTANCE_METRES)
axes_object.plot(conus_longitudes_deg,
conus_latitudes_deg,
color=LINE_COLOUR,
linestyle='solid',
linewidth=LINE_WIDTH)
axes_object.plot(removed_longitudes_deg,
removed_latitudes_deg,
linestyle='None',
marker=MARKER_TYPE,
markersize=MARKER_SIZE,
markeredgewidth=0,
markerfacecolor=MARKER_COLOUR,
markeredgecolor=MARKER_COLOUR)
print('Saving figure to: "{0:s}"...'.format(output_file_name))
figure_object.savefig(output_file_name,
dpi=FIGURE_RESOLUTION_DPI,
pad_inches=0,
bbox_inches='tight')
pyplot.close(figure_object)
def _find_tracking_gaps(first_spc_date_string, last_spc_date_string,
top_tracking_dir_name, tracking_scale_metres2,
source_name, min_time_diff_seconds):
"""Finds gaps (temporal discontinuities) between storm-tracking files.
:param first_spc_date_string: See documentation at top of file.
:param last_spc_date_string: Same.
:param top_tracking_dir_name: Same.
:param tracking_scale_metres2: Same.
:param source_name: Same.
:param min_time_diff_seconds: Same.
"""
spc_date_strings = time_conversion.get_spc_dates_in_range(
first_spc_date_string=first_spc_date_string,
last_spc_date_string=last_spc_date_string)
tracking_file_names = []
unix_times_sec = numpy.array([], dtype=int)
num_spc_dates = len(spc_date_strings)
for i in range(num_spc_dates):
print('Finding tracking files for SPC date "{0:s}"...'.format(
spc_date_strings[i]))
these_file_names = tracking_io.find_files_one_spc_date(
spc_date_string=spc_date_strings[i],
source_name=source_name,
top_tracking_dir_name=top_tracking_dir_name,
tracking_scale_metres2=tracking_scale_metres2,
raise_error_if_missing=False)[0]
print(len(these_file_names))
if not len(these_file_names):
continue
these_file_sizes_bytes = numpy.array(
[os.path.getsize(f) for f in these_file_names], dtype=int)
these_valid_indices = numpy.where(
these_file_sizes_bytes > FILE_SIZE_WITHOUT_STORMS_BYTES)[0]
these_file_names = [these_file_names[k] for k in these_valid_indices]
these_unix_times_sec = numpy.array(
[tracking_io.file_name_to_time(f) for f in these_file_names],
dtype=int)
these_sort_indices = numpy.argsort(these_unix_times_sec)
these_unix_times_sec = these_unix_times_sec[these_sort_indices]
these_file_names = [these_file_names[k] for k in these_sort_indices]
tracking_file_names += these_file_names
unix_times_sec = numpy.concatenate(
(unix_times_sec, these_unix_times_sec))
time_diffs_seconds = numpy.diff(unix_times_sec)
time_gap_indices = numpy.where(
time_diffs_seconds >= min_time_diff_seconds)[0]
num_time_gaps = len(time_gap_indices)
print((
'\nThere are {0:d} time gaps (successive files >= {1:d} seconds apart),'
' listed below:\n').format(num_time_gaps, min_time_diff_seconds))
for i in time_gap_indices:
this_start_time_string = time_conversion.unix_sec_to_string(
unix_times_sec[i], TIME_FORMAT)
this_end_time_string = time_conversion.unix_sec_to_string(
unix_times_sec[i + 1], TIME_FORMAT)
print('Gap between {0:s} and {1:s} = {2:d} seconds'.format(
this_start_time_string, this_end_time_string,
time_diffs_seconds[i]))
def _run(top_input_dir_name, tracking_scale_metres2, first_spc_date_string,
last_spc_date_string, min_distances_metres, max_distances_metres,
top_output_dir_name):
"""Creates one or more distance buffers around each storm object (polygon).
This is effectively the main method.
:param top_input_dir_name: See documentation at top of file.
:param tracking_scale_metres2: Same.
:param first_spc_date_string: Same.
:param last_spc_date_string: Same.
:param min_distances_metres: Same.
:param max_distances_metres: Same.
:param top_output_dir_name: Same.
"""
min_distances_metres[min_distances_metres < 0] = numpy.nan
spc_date_strings = time_conversion.get_spc_dates_in_range(
first_spc_date_string=first_spc_date_string,
last_spc_date_string=last_spc_date_string)
for this_spc_date_string in spc_date_strings:
these_input_file_names = tracking_io.find_files_one_spc_date(
top_tracking_dir_name=top_input_dir_name,
tracking_scale_metres2=tracking_scale_metres2,
source_name=tracking_utils.SEGMOTION_NAME,
spc_date_string=this_spc_date_string,
raise_error_if_missing=False)[0]
if len(these_input_file_names) == 0:
continue
for this_input_file_name in these_input_file_names:
print('Reading input tracks from: "{0:s}"...'.format(
this_input_file_name))
this_storm_object_table = tracking_io.read_file(
this_input_file_name)
this_storm_object_table = (tracking_utils.create_distance_buffers(
storm_object_table=this_storm_object_table,
min_distances_metres=min_distances_metres,
max_distances_metres=max_distances_metres))
this_output_file_name = tracking_io.find_file(
top_tracking_dir_name=top_output_dir_name,
tracking_scale_metres2=tracking_scale_metres2,
source_name=tracking_utils.SEGMOTION_NAME,
valid_time_unix_sec=tracking_io.file_name_to_time(
this_input_file_name),
spc_date_string=this_spc_date_string,
raise_error_if_missing=False)
print('Writing input tracks + buffers to: "{0:s}"...\n'.format(
this_output_file_name))
tracking_io.write_file(storm_object_table=this_storm_object_table,
pickle_file_name=this_output_file_name)
print(SEPARATOR_STRING)
def _plot_one_example(full_id_string,
storm_time_unix_sec,
target_name,
forecast_probability,
tornado_dir_name,
top_tracking_dir_name,
top_myrorss_dir_name,
radar_field_name,
radar_height_m_asl,
latitude_buffer_deg,
longitude_buffer_deg,
top_output_dir_name,
aux_forecast_probabilities=None,
aux_activation_dict=None):
"""Plots one example with surrounding context at several times.
N = number of storm objects read from auxiliary activation file
:param full_id_string: Full storm ID.
:param storm_time_unix_sec: Storm time.
:param target_name: Name of target variable.
:param forecast_probability: Forecast tornado probability for this example.
:param tornado_dir_name: See documentation at top of file.
:param top_tracking_dir_name: Same.
:param top_myrorss_dir_name: Same.
:param radar_field_name: Same.
:param radar_height_m_asl: Same.
:param latitude_buffer_deg: Same.
:param longitude_buffer_deg: Same.
:param top_output_dir_name: Same.
:param aux_forecast_probabilities: length-N numpy array of forecast
probabilities. If this is None, will not plot forecast probs in maps.
:param aux_activation_dict: Dictionary returned by
`model_activation.read_file` from auxiliary file. If this is None, will
not plot forecast probs in maps.
"""
storm_time_string = time_conversion.unix_sec_to_string(
storm_time_unix_sec, TIME_FORMAT)
# Create output directory for this example.
output_dir_name = '{0:s}/{1:s}_{2:s}'.format(top_output_dir_name,
full_id_string,
storm_time_string)
file_system_utils.mkdir_recursive_if_necessary(
directory_name=output_dir_name)
# Find tracking files.
tracking_file_names = _find_tracking_files_one_example(
valid_time_unix_sec=storm_time_unix_sec,
top_tracking_dir_name=top_tracking_dir_name,
target_name=target_name)
tracking_times_unix_sec = numpy.array(
[tracking_io.file_name_to_time(f) for f in tracking_file_names],
dtype=int)
tracking_time_strings = [
time_conversion.unix_sec_to_string(t, TIME_FORMAT)
for t in tracking_times_unix_sec
]
# Read tracking files.
storm_object_table = tracking_io.read_many_files(tracking_file_names)
print('\n')
if aux_activation_dict is not None:
these_indices = tracking_utils.find_storm_objects(
all_id_strings=aux_activation_dict[model_activation.FULL_IDS_KEY],
all_times_unix_sec=aux_activation_dict[
model_activation.STORM_TIMES_KEY],
id_strings_to_keep=storm_object_table[
tracking_utils.FULL_ID_COLUMN].values.tolist(),
times_to_keep_unix_sec=storm_object_table[
tracking_utils.VALID_TIME_COLUMN].values,
allow_missing=True)
storm_object_probs = numpy.array([
aux_forecast_probabilities[k] if k >= 0 else numpy.nan
for k in these_indices
])
storm_object_table = storm_object_table.assign(
**{FORECAST_PROBABILITY_COLUMN: storm_object_probs})
primary_id_string = temporal_tracking.full_to_partial_ids([full_id_string
])[0][0]
this_storm_object_table = storm_object_table.loc[storm_object_table[
tracking_utils.PRIMARY_ID_COLUMN] == primary_id_string]
latitude_limits_deg, longitude_limits_deg = _get_plotting_limits(
storm_object_table=this_storm_object_table,
latitude_buffer_deg=latitude_buffer_deg,
longitude_buffer_deg=longitude_buffer_deg)
storm_min_latitudes_deg = numpy.array([
numpy.min(numpy.array(p.exterior.xy[1])) for p in storm_object_table[
tracking_utils.LATLNG_POLYGON_COLUMN].values
])
storm_max_latitudes_deg = numpy.array([
numpy.max(numpy.array(p.exterior.xy[1])) for p in storm_object_table[
tracking_utils.LATLNG_POLYGON_COLUMN].values
])
storm_min_longitudes_deg = numpy.array([
numpy.min(numpy.array(p.exterior.xy[0])) for p in storm_object_table[
tracking_utils.LATLNG_POLYGON_COLUMN].values
])
storm_max_longitudes_deg = numpy.array([
numpy.max(numpy.array(p.exterior.xy[0])) for p in storm_object_table[
tracking_utils.LATLNG_POLYGON_COLUMN].values
])
min_latitude_flags = numpy.logical_and(
storm_min_latitudes_deg >= latitude_limits_deg[0],
storm_min_latitudes_deg <= latitude_limits_deg[1])
max_latitude_flags = numpy.logical_and(
storm_max_latitudes_deg >= latitude_limits_deg[0],
storm_max_latitudes_deg <= latitude_limits_deg[1])
latitude_flags = numpy.logical_or(min_latitude_flags, max_latitude_flags)
min_longitude_flags = numpy.logical_and(
storm_min_longitudes_deg >= longitude_limits_deg[0],
storm_min_longitudes_deg <= longitude_limits_deg[1])
max_longitude_flags = numpy.logical_and(
storm_max_longitudes_deg >= longitude_limits_deg[0],
storm_max_longitudes_deg <= longitude_limits_deg[1])
longitude_flags = numpy.logical_or(min_longitude_flags,
max_longitude_flags)
good_indices = numpy.where(
numpy.logical_and(latitude_flags, longitude_flags))[0]
storm_object_table = storm_object_table.iloc[good_indices]
# Read tornado reports.
target_param_dict = target_val_utils.target_name_to_params(target_name)
min_lead_time_seconds = target_param_dict[
target_val_utils.MIN_LEAD_TIME_KEY]
max_lead_time_seconds = target_param_dict[
target_val_utils.MAX_LEAD_TIME_KEY]
tornado_table = linkage._read_input_tornado_reports(
input_directory_name=tornado_dir_name,
storm_times_unix_sec=numpy.array([storm_time_unix_sec], dtype=int),
max_time_before_storm_start_sec=-1 * min_lead_time_seconds,
max_time_after_storm_end_sec=max_lead_time_seconds,
genesis_only=True)
tornado_table = tornado_table.loc[
(tornado_table[linkage.EVENT_LATITUDE_COLUMN] >= latitude_limits_deg[0]
)
& (tornado_table[linkage.EVENT_LATITUDE_COLUMN] <=
latitude_limits_deg[1])]
tornado_table = tornado_table.loc[
(tornado_table[linkage.EVENT_LONGITUDE_COLUMN] >=
longitude_limits_deg[0])
& (tornado_table[linkage.EVENT_LONGITUDE_COLUMN] <=
longitude_limits_deg[1])]
for i in range(len(tracking_file_names)):
this_storm_object_table = storm_object_table.loc[storm_object_table[
tracking_utils.VALID_TIME_COLUMN] == tracking_times_unix_sec[i]]
_plot_one_example_one_time(
storm_object_table=this_storm_object_table,
full_id_string=full_id_string,
valid_time_unix_sec=tracking_times_unix_sec[i],
tornado_table=copy.deepcopy(tornado_table),
top_myrorss_dir_name=top_myrorss_dir_name,
radar_field_name=radar_field_name,
radar_height_m_asl=radar_height_m_asl,
latitude_limits_deg=latitude_limits_deg,
longitude_limits_deg=longitude_limits_deg)
if aux_activation_dict is None:
this_title_string = (
'Valid time = {0:s} ... forecast prob at {1:s} = {2:.3f}'
).format(tracking_time_strings[i], storm_time_string,
forecast_probability)
pyplot.title(this_title_string, fontsize=TITLE_FONT_SIZE)
this_file_name = '{0:s}/{1:s}.jpg'.format(output_dir_name,
tracking_time_strings[i])
print('Saving figure to file: "{0:s}"...\n'.format(this_file_name))
pyplot.savefig(this_file_name, dpi=FIGURE_RESOLUTION_DPI)
pyplot.close()
imagemagick_utils.trim_whitespace(input_file_name=this_file_name,
output_file_name=this_file_name)
def _find_tracking_files_one_example(top_tracking_dir_name,
valid_time_unix_sec, target_name):
"""Finds tracking files needed to make plots for one example.
:param top_tracking_dir_name: See documentation at top of file.
:param valid_time_unix_sec: Valid time for example.
:param target_name: Name of target variable.
:return: tracking_file_names: 1-D list of paths to tracking files.
:raises: ValueError: if no tracking files are found.
"""
target_param_dict = target_val_utils.target_name_to_params(target_name)
min_lead_time_seconds = target_param_dict[
target_val_utils.MIN_LEAD_TIME_KEY]
max_lead_time_seconds = target_param_dict[
target_val_utils.MAX_LEAD_TIME_KEY]
first_time_unix_sec = valid_time_unix_sec + min_lead_time_seconds
last_time_unix_sec = valid_time_unix_sec + max_lead_time_seconds
first_spc_date_string = time_conversion.time_to_spc_date_string(
first_time_unix_sec - TIME_INTERVAL_SECONDS)
last_spc_date_string = time_conversion.time_to_spc_date_string(
last_time_unix_sec + TIME_INTERVAL_SECONDS)
spc_date_strings = time_conversion.get_spc_dates_in_range(
first_spc_date_string=first_spc_date_string,
last_spc_date_string=last_spc_date_string)
tracking_file_names = []
for this_spc_date_string in spc_date_strings:
these_file_names = tracking_io.find_files_one_spc_date(
top_tracking_dir_name=top_tracking_dir_name,
tracking_scale_metres2=echo_top_tracking.
DUMMY_TRACKING_SCALE_METRES2,
source_name=tracking_utils.SEGMOTION_NAME,
spc_date_string=this_spc_date_string,
raise_error_if_missing=False)[0]
tracking_file_names += these_file_names
if len(tracking_file_names) == 0:
error_string = (
'Cannot find any tracking files for SPC dates "{0:s}" to "{1:s}".'
).format(first_spc_date_string, last_spc_date_string)
raise ValueError(error_string)
tracking_times_unix_sec = numpy.array(
[tracking_io.file_name_to_time(f) for f in tracking_file_names],
dtype=int)
sort_indices = numpy.argsort(tracking_times_unix_sec)
tracking_times_unix_sec = tracking_times_unix_sec[sort_indices]
tracking_file_names = [tracking_file_names[k] for k in sort_indices]
these_indices = numpy.where(
tracking_times_unix_sec <= first_time_unix_sec)[0]
if len(these_indices) == 0:
first_index = 0
else:
first_index = these_indices[-1]
these_indices = numpy.where(
tracking_times_unix_sec >= last_time_unix_sec)[0]
if len(these_indices) == 0:
last_index = len(tracking_file_names) - 1
else:
last_index = these_indices[0]
return tracking_file_names[first_index:(last_index + 1)]
def _extract_storm_images(
num_image_rows, num_image_columns, rotate_grids,
rotated_grid_spacing_metres, radar_field_names, refl_heights_m_agl,
spc_date_string, first_time_string, last_time_string,
tarred_myrorss_dir_name, untarred_myrorss_dir_name,
top_tracking_dir_name, elevation_dir_name, tracking_scale_metres2,
target_name, top_target_dir_name, top_output_dir_name):
"""Extracts storm-centered img for each field/height pair and storm object.
:param num_image_rows: See documentation at top of file.
:param num_image_columns: Same.
:param rotate_grids: Same.
:param rotated_grid_spacing_metres: Same.
:param radar_field_names: Same.
:param refl_heights_m_agl: Same.
:param spc_date_string: Same.
:param first_time_string: Same.
:param last_time_string: Same.
:param tarred_myrorss_dir_name: Same.
:param untarred_myrorss_dir_name: Same.
:param top_tracking_dir_name: Same.
:param elevation_dir_name: Same.
:param tracking_scale_metres2: Same.
:param target_name: Same.
:param top_target_dir_name: Same.
:param top_output_dir_name: Same.
:raises: ValueError: if `first_time_string` and `last_time_string` have
different SPC dates.
"""
if elevation_dir_name in ['', 'None']:
elevation_dir_name = None
if elevation_dir_name is None:
host_name = socket.gethostname()
if 'casper' in host_name:
elevation_dir_name = '/glade/work/ryanlage/elevation'
else:
elevation_dir_name = '/condo/swatwork/ralager/elevation'
if spc_date_string in ['', 'None']:
first_time_unix_sec = time_conversion.string_to_unix_sec(
first_time_string, TIME_FORMAT)
last_time_unix_sec = time_conversion.string_to_unix_sec(
last_time_string, TIME_FORMAT)
first_spc_date_string = time_conversion.time_to_spc_date_string(
first_time_unix_sec)
last_spc_date_string = time_conversion.time_to_spc_date_string(
last_time_unix_sec)
if first_spc_date_string != last_spc_date_string:
error_string = (
'First ({0:s}) and last ({1:s}) times have different SPC dates.'
' This script can handle only one SPC date.'
).format(first_time_string, last_time_string)
raise ValueError(error_string)
spc_date_string = first_spc_date_string
else:
first_time_unix_sec = 0
last_time_unix_sec = int(1e12)
if tarred_myrorss_dir_name in ['', 'None']:
tarred_myrorss_dir_name = None
if target_name in ['', 'None']:
target_name = None
if target_name is not None:
target_param_dict = target_val_utils.target_name_to_params(target_name)
target_file_name = target_val_utils.find_target_file(
top_directory_name=top_target_dir_name,
event_type_string=target_param_dict[
target_val_utils.EVENT_TYPE_KEY],
spc_date_string=spc_date_string)
print('Reading data from: "{0:s}"...'.format(target_file_name))
target_dict = target_val_utils.read_target_values(
netcdf_file_name=target_file_name, target_names=[target_name]
)
print('\n')
refl_heights_m_asl = radar_utils.get_valid_heights(
data_source=radar_utils.MYRORSS_SOURCE_ID,
field_name=radar_utils.REFL_NAME)
# Untar files.
if tarred_myrorss_dir_name is not None:
az_shear_field_names = list(
set(radar_field_names) & set(ALL_AZ_SHEAR_FIELD_NAMES)
)
if len(az_shear_field_names) > 0:
az_shear_tar_file_name = (
'{0:s}/{1:s}/azimuthal_shear_only/{2:s}.tar'
).format(
tarred_myrorss_dir_name, spc_date_string[:4], spc_date_string
)
myrorss_io.unzip_1day_tar_file(
tar_file_name=az_shear_tar_file_name,
field_names=az_shear_field_names,
spc_date_string=spc_date_string,
top_target_directory_name=untarred_myrorss_dir_name)
print(SEPARATOR_STRING)
non_shear_field_names = list(
set(radar_field_names) - set(ALL_AZ_SHEAR_FIELD_NAMES)
)
if len(non_shear_field_names) > 0:
non_shear_tar_file_name = '{0:s}/{1:s}/{2:s}.tar'.format(
tarred_myrorss_dir_name, spc_date_string[:4], spc_date_string
)
myrorss_io.unzip_1day_tar_file(
tar_file_name=non_shear_tar_file_name,
field_names=non_shear_field_names,
spc_date_string=spc_date_string,
top_target_directory_name=untarred_myrorss_dir_name,
refl_heights_m_asl=refl_heights_m_asl)
print(SEPARATOR_STRING)
# Read storm tracks for the given SPC date.
tracking_file_names = tracking_io.find_files_one_spc_date(
spc_date_string=spc_date_string,
source_name=tracking_utils.SEGMOTION_NAME,
top_tracking_dir_name=top_tracking_dir_name,
tracking_scale_metres2=tracking_scale_metres2
)[0]
file_times_unix_sec = numpy.array(
[tracking_io.file_name_to_time(f) for f in tracking_file_names],
dtype=int
)
good_indices = numpy.where(numpy.logical_and(
file_times_unix_sec >= first_time_unix_sec,
file_times_unix_sec <= last_time_unix_sec
))[0]
tracking_file_names = [tracking_file_names[k] for k in good_indices]
storm_object_table = tracking_io.read_many_files(
tracking_file_names
)[storm_images.STORM_COLUMNS_NEEDED]
print(SEPARATOR_STRING)
if target_name is not None:
print((
'Removing storm objects without target values (variable = '
'"{0:s}")...'
).format(target_name))
these_indices = tracking_utils.find_storm_objects(
all_id_strings=storm_object_table[
tracking_utils.FULL_ID_COLUMN].values.tolist(),
all_times_unix_sec=storm_object_table[
tracking_utils.VALID_TIME_COLUMN].values.astype(int),
id_strings_to_keep=target_dict[target_val_utils.FULL_IDS_KEY],
times_to_keep_unix_sec=target_dict[
target_val_utils.VALID_TIMES_KEY],
allow_missing=False)
num_storm_objects_orig = len(storm_object_table.index)
storm_object_table = storm_object_table.iloc[these_indices]
num_storm_objects = len(storm_object_table.index)
print('Removed {0:d} of {1:d} storm objects!\n'.format(
num_storm_objects_orig - num_storm_objects, num_storm_objects_orig
))
# Extract storm-centered radar images.
storm_images.extract_storm_images_myrorss_or_mrms(
storm_object_table=storm_object_table,
radar_source=radar_utils.MYRORSS_SOURCE_ID,
top_radar_dir_name=untarred_myrorss_dir_name,
top_output_dir_name=top_output_dir_name,
elevation_dir_name=elevation_dir_name,
num_storm_image_rows=num_image_rows,
num_storm_image_columns=num_image_columns, rotate_grids=rotate_grids,
rotated_grid_spacing_metres=rotated_grid_spacing_metres,
radar_field_names=radar_field_names,
reflectivity_heights_m_agl=refl_heights_m_agl)
print(SEPARATOR_STRING)
# Remove untarred MYRORSS files.
if tarred_myrorss_dir_name is not None:
myrorss_io.remove_unzipped_data_1day(
spc_date_string=spc_date_string,
top_directory_name=untarred_myrorss_dir_name,
field_names=radar_field_names,
refl_heights_m_asl=refl_heights_m_asl)
def _run(myrorss_tracking_dir_name, gridrad_tracking_dir_name,
max_distance_metres, source_dataset_name, first_spc_date_string,
last_spc_date_string, output_dir_name):
"""Matches storm objects between MYRORSS and GridRad datasets.
This is effectively the main method.
:param myrorss_tracking_dir_name: See documentation at end of file.
:param gridrad_tracking_dir_name: Same.
:param max_distance_metres: Same.
:param source_dataset_name: Same.
:param first_spc_date_string: Same.
:param last_spc_date_string: Same.
:param output_dir_name: Same.
:raises: ValueError: if `source_dataset_name not in VALID_DATASET_NAMES`.
"""
if source_dataset_name not in VALID_DATASET_NAMES:
error_string = (
'\n{0:s}\nValid datasets (listed above) do not include "{1:s}".'
).format(str(VALID_DATASET_NAMES), source_dataset_name)
raise ValueError(error_string)
spc_date_strings = time_conversion.get_spc_dates_in_range(
first_spc_date_string=first_spc_date_string,
last_spc_date_string=last_spc_date_string)
if source_dataset_name == radar_utils.MYRORSS_SOURCE_ID:
source_tracking_dir_name = myrorss_tracking_dir_name
target_tracking_dir_name = gridrad_tracking_dir_name
target_dataset_name = radar_utils.GRIDRAD_SOURCE_ID
else:
source_tracking_dir_name = gridrad_tracking_dir_name
target_tracking_dir_name = myrorss_tracking_dir_name
target_dataset_name = radar_utils.MYRORSS_SOURCE_ID
source_tracking_file_names = []
target_tracking_file_names = []
for this_spc_date_string in spc_date_strings:
source_tracking_file_names += tracking_io.find_files_one_spc_date(
top_tracking_dir_name=source_tracking_dir_name,
tracking_scale_metres2=TRACKING_SCALE_METRES2,
source_name=tracking_utils.SEGMOTION_NAME,
spc_date_string=this_spc_date_string,
raise_error_if_missing=True)[0]
target_tracking_file_names += tracking_io.find_files_one_spc_date(
top_tracking_dir_name=target_tracking_dir_name,
tracking_scale_metres2=TRACKING_SCALE_METRES2,
source_name=tracking_utils.SEGMOTION_NAME,
spc_date_string=this_spc_date_string,
raise_error_if_missing=True)[0]
source_times_unix_sec = numpy.array(
[tracking_io.file_name_to_time(f) for f in source_tracking_file_names],
dtype=int)
target_times_unix_sec = numpy.array(
[tracking_io.file_name_to_time(f) for f in target_tracking_file_names],
dtype=int)
source_to_target_indices = _match_all_times(
source_times_unix_sec=source_times_unix_sec,
target_times_unix_sec=target_times_unix_sec,
max_diff_seconds=MAX_TIME_DIFF_SECONDS)
print(SEPARATOR_STRING)
del target_times_unix_sec
target_tracking_file_names = [
target_tracking_file_names[k] for k in source_to_target_indices
]
num_source_times = len(source_times_unix_sec)
for i in range(num_source_times):
print('Reading data from: "{0:s}"...'.format(
source_tracking_file_names[i]))
this_source_object_table = tracking_io.read_file(
source_tracking_file_names[i])
print('Reading data from: "{0:s}"...'.format(
target_tracking_file_names[i]))
this_target_object_table = tracking_io.read_file(
target_tracking_file_names[i])
this_source_to_target_dict = _match_locations_one_time(
source_object_table=this_source_object_table,
target_object_table=this_target_object_table,
max_distance_metres=max_distance_metres)
this_match_file_name = tracking_io.find_match_file(
top_directory_name=output_dir_name,
valid_time_unix_sec=source_times_unix_sec[i],
raise_error_if_missing=False)
print('Writing results to: "{0:s}"...\n'.format(this_match_file_name))
tracking_io.write_matches(
pickle_file_name=this_match_file_name,
source_to_target_dict=this_source_to_target_dict,
max_time_diff_seconds=MAX_TIME_DIFF_SECONDS,
max_distance_metres=max_distance_metres,
source_dataset_name=source_dataset_name,
source_tracking_dir_name=source_tracking_dir_name,
target_dataset_name=target_dataset_name,
target_tracking_dir_name=target_tracking_dir_name)
def _run(storm_metafile_name, top_tracking_dir_name, lead_time_seconds,
output_file_name):
"""Plots spatial distribution of examples (storm objects) in file.
This is effectively the main method.
:param storm_metafile_name: See documentation at top of file.
:param top_tracking_dir_name: Same.
:param lead_time_seconds: Same.
:param output_file_name: Same.
"""
file_system_utils.mkdir_recursive_if_necessary(file_name=output_file_name)
# Read storm metadata.
print(
'Reading storm metadata from: "{0:s}"...'.format(storm_metafile_name))
orig_full_id_strings, orig_times_unix_sec = (
tracking_io.read_ids_and_times(storm_metafile_name))
orig_primary_id_strings = temporal_tracking.full_to_partial_ids(
orig_full_id_strings)[0]
# Find relevant tracking files.
spc_date_strings = [
time_conversion.time_to_spc_date_string(t) for t in orig_times_unix_sec
]
spc_date_strings += [
time_conversion.time_to_spc_date_string(t + lead_time_seconds)
for t in orig_times_unix_sec
]
spc_date_strings = list(set(spc_date_strings))
tracking_file_names = []
for this_spc_date_string in spc_date_strings:
tracking_file_names += tracking_io.find_files_one_spc_date(
top_tracking_dir_name=top_tracking_dir_name,
tracking_scale_metres2=DUMMY_TRACKING_SCALE_METRES2,
source_name=tracking_utils.SEGMOTION_NAME,
spc_date_string=this_spc_date_string,
raise_error_if_missing=False)[0]
file_times_unix_sec = numpy.array(
[tracking_io.file_name_to_time(f) for f in tracking_file_names],
dtype=int)
num_orig_storm_objects = len(orig_full_id_strings)
num_files = len(file_times_unix_sec)
keep_file_flags = numpy.full(num_files, 0, dtype=bool)
for i in range(num_orig_storm_objects):
these_flags = numpy.logical_and(
file_times_unix_sec >= orig_times_unix_sec[i],
file_times_unix_sec <= orig_times_unix_sec[i] + lead_time_seconds)
keep_file_flags = numpy.logical_or(keep_file_flags, these_flags)
del file_times_unix_sec
keep_file_indices = numpy.where(keep_file_flags)[0]
tracking_file_names = [tracking_file_names[k] for k in keep_file_indices]
# Read relevant tracking files.
num_files = len(tracking_file_names)
storm_object_tables = [None] * num_files
print(SEPARATOR_STRING)
for i in range(num_files):
print('Reading data from: "{0:s}"...'.format(tracking_file_names[i]))
this_table = tracking_io.read_file(tracking_file_names[i])
storm_object_tables[i] = this_table.loc[this_table[
tracking_utils.PRIMARY_ID_COLUMN].isin(
numpy.array(orig_primary_id_strings))]
if i == 0:
continue
storm_object_tables[i] = storm_object_tables[i].align(
storm_object_tables[0], axis=1)[0]
storm_object_table = pandas.concat(storm_object_tables,
axis=0,
ignore_index=True)
print(SEPARATOR_STRING)
# Find relevant storm objects.
orig_object_rows = tracking_utils.find_storm_objects(
all_id_strings=storm_object_table[
tracking_utils.FULL_ID_COLUMN].values.tolist(),
all_times_unix_sec=storm_object_table[
tracking_utils.VALID_TIME_COLUMN].values,
id_strings_to_keep=orig_full_id_strings,
times_to_keep_unix_sec=orig_times_unix_sec)
good_object_rows = numpy.array([], dtype=int)
for i in range(num_orig_storm_objects):
# Non-merging successors only!
first_rows = temporal_tracking.find_successors(
storm_object_table=storm_object_table,
target_row=orig_object_rows[i],
num_seconds_forward=lead_time_seconds,
max_num_sec_id_changes=1,
change_type_string=temporal_tracking.SPLIT_STRING,
return_all_on_path=True)
second_rows = temporal_tracking.find_successors(
storm_object_table=storm_object_table,
target_row=orig_object_rows[i],
num_seconds_forward=lead_time_seconds,
max_num_sec_id_changes=0,
change_type_string=temporal_tracking.MERGER_STRING,
return_all_on_path=True)
first_rows = first_rows.tolist()
second_rows = second_rows.tolist()
these_rows = set(first_rows) & set(second_rows)
these_rows = numpy.array(list(these_rows), dtype=int)
good_object_rows = numpy.concatenate((good_object_rows, these_rows))
good_object_rows = numpy.unique(good_object_rows)
storm_object_table = storm_object_table.iloc[good_object_rows]
times_of_day_sec = numpy.mod(
storm_object_table[tracking_utils.VALID_TIME_COLUMN].values,
NUM_SECONDS_IN_DAY)
storm_object_table = storm_object_table.assign(
**{tracking_utils.VALID_TIME_COLUMN: times_of_day_sec})
min_plot_latitude_deg = -LATLNG_BUFFER_DEG + numpy.min(
storm_object_table[tracking_utils.CENTROID_LATITUDE_COLUMN].values)
max_plot_latitude_deg = LATLNG_BUFFER_DEG + numpy.max(
storm_object_table[tracking_utils.CENTROID_LATITUDE_COLUMN].values)
min_plot_longitude_deg = -LATLNG_BUFFER_DEG + numpy.min(
storm_object_table[tracking_utils.CENTROID_LONGITUDE_COLUMN].values)
max_plot_longitude_deg = LATLNG_BUFFER_DEG + numpy.max(
storm_object_table[tracking_utils.CENTROID_LONGITUDE_COLUMN].values)
_, axes_object, basemap_object = (
plotting_utils.create_equidist_cylindrical_map(
min_latitude_deg=min_plot_latitude_deg,
max_latitude_deg=max_plot_latitude_deg,
min_longitude_deg=min_plot_longitude_deg,
max_longitude_deg=max_plot_longitude_deg,
resolution_string='i'))
plotting_utils.plot_coastlines(basemap_object=basemap_object,
axes_object=axes_object,
line_colour=BORDER_COLOUR,
line_width=BORDER_WIDTH * 2)
plotting_utils.plot_countries(basemap_object=basemap_object,
axes_object=axes_object,
line_colour=BORDER_COLOUR,
line_width=BORDER_WIDTH)
plotting_utils.plot_states_and_provinces(basemap_object=basemap_object,
axes_object=axes_object,
line_colour=BORDER_COLOUR,
line_width=BORDER_WIDTH)
plotting_utils.plot_parallels(basemap_object=basemap_object,
axes_object=axes_object,
num_parallels=NUM_PARALLELS,
line_width=BORDER_WIDTH)
plotting_utils.plot_meridians(basemap_object=basemap_object,
axes_object=axes_object,
num_meridians=NUM_MERIDIANS,
line_width=BORDER_WIDTH)
# colour_bar_object = storm_plotting.plot_storm_tracks(
# storm_object_table=storm_object_table, axes_object=axes_object,
# basemap_object=basemap_object, colour_map_object=COLOUR_MAP_OBJECT,
# colour_min_unix_sec=0, colour_max_unix_sec=NUM_SECONDS_IN_DAY - 1,
# line_width=TRACK_LINE_WIDTH,
# start_marker_type=None, end_marker_type=None
# )
colour_bar_object = storm_plotting.plot_storm_centroids(
storm_object_table=storm_object_table,
axes_object=axes_object,
basemap_object=basemap_object,
colour_map_object=COLOUR_MAP_OBJECT,
colour_min_unix_sec=0,
colour_max_unix_sec=NUM_SECONDS_IN_DAY - 1)
tick_times_unix_sec = numpy.linspace(0,
NUM_SECONDS_IN_DAY,
num=NUM_HOURS_IN_DAY + 1,
dtype=int)
tick_times_unix_sec = tick_times_unix_sec[:-1]
tick_times_unix_sec = tick_times_unix_sec[::2]
tick_time_strings = [
time_conversion.unix_sec_to_string(t, COLOUR_BAR_TIME_FORMAT)
for t in tick_times_unix_sec
]
colour_bar_object.set_ticks(tick_times_unix_sec)
colour_bar_object.set_ticklabels(tick_time_strings)
print('Saving figure to: "{0:s}"...'.format(output_file_name))
pyplot.savefig(output_file_name,
dpi=FIGURE_RESOLUTION_DPI,
pad_inches=0,
bbox_inches='tight')
pyplot.close()
def _find_io_files_for_renaming(top_input_dir_name, first_date_unix_sec,
last_date_unix_sec, top_output_dir_name):
"""Finds input and output files for renaming storms.
N = number of dates
:param top_input_dir_name: See documentation for `rename_storms.`
:param first_date_unix_sec: Same.
:param last_date_unix_sec: Same.
:param top_output_dir_name: Same.
:return: input_file_names_by_date: length-N list, where the [i]th item is a
numpy array of paths to input files for the [i]th date.
:return: output_file_names_by_date: Same as above, but for output files.
:return: valid_times_by_date_unix_sec: Same as above, but for valid times.
All 3 arrays for the [i]th date have the same length.
"""
dates_unix_sec = time_periods.range_and_interval_to_list(
start_time_unix_sec=first_date_unix_sec,
end_time_unix_sec=last_date_unix_sec,
time_interval_sec=DAYS_TO_SECONDS,
include_endpoint=True)
date_strings = [
time_conversion.unix_sec_to_string(t, DATE_FORMAT)
for t in dates_unix_sec
]
num_dates = len(date_strings)
input_file_names_by_date = [numpy.array([], dtype=object)] * num_dates
output_file_names_by_date = [numpy.array([], dtype=object)] * num_dates
valid_times_by_date_unix_sec = [numpy.array([], dtype=int)] * num_dates
for i in range(num_dates):
print('Finding input files for date {0:s}...'.format(date_strings[i]))
these_input_file_names = tracking_io.find_files_one_spc_date(
spc_date_string=date_strings[i],
source_name=tracking_utils.PROBSEVERE_NAME,
top_tracking_dir_name=top_input_dir_name,
tracking_scale_metres2=DUMMY_TRACKING_SCALE_METRES2,
raise_error_if_missing=True)[0]
these_input_file_names.sort()
these_valid_times_unix_sec = numpy.array(
[tracking_io.file_name_to_time(f) for f in these_input_file_names],
dtype=int)
these_output_file_names = []
for t in these_valid_times_unix_sec:
these_output_file_names.append(
tracking_io.find_file(
valid_time_unix_sec=t,
source_name=tracking_utils.PROBSEVERE_NAME,
top_tracking_dir_name=top_output_dir_name,
tracking_scale_metres2=DUMMY_TRACKING_SCALE_METRES2,
raise_error_if_missing=False))
input_file_names_by_date[i] = numpy.array(these_input_file_names,
dtype=object)
output_file_names_by_date[i] = numpy.array(these_output_file_names,
dtype=object)
valid_times_by_date_unix_sec[i] = these_valid_times_unix_sec
print(SEPARATOR_STRING)
return (input_file_names_by_date, output_file_names_by_date,
valid_times_by_date_unix_sec)