def _prepare_datasets_for_evaluation(reference, targets, config_data):
""""""
subset = config_data['evaluation'].get('subset', None)
temporal_time_delta = config_data['evaluation'].get(
'temporal_time_delta', None)
spatial_regrid_lats = config_data['evaluation'].get(
'spatial_regrid_lats', None)
spatial_regrid_lons = config_data['evaluation'].get(
'spatial_regrid_lons', None)
# If we have a temporal time delta and it's daily (i.e., 1) we will
# normalize the data as daily data (which means we adjust the start times
# for each bucket of data to be consistent). By default we will normalize
# the data as monthly. Note that this will not break yearly data so it's
# safer to do this no matter what. This keeps us from ending up with 1-off
# errors in the resulting dataset shape post-temporal/spatial adjustments
# that break evaluations.
string_time_delta = 'monthly'
if temporal_time_delta and temporal_time_delta == 1:
string_time_delta = 'daily'
reference = dsp.normalize_dataset_datetimes(reference, string_time_delta)
targets = [
dsp.normalize_dataset_datetimes(t, string_time_delta) for t in targets
]
if subset:
start = dateutil.parser.parse(subset[4])
end = dateutil.parser.parse(subset[5])
bounds = Bounds(subset[0], subset[1], subset[2], subset[3], start, end)
if reference:
reference = dsp.safe_subset(bounds, reference)
if targets:
targets = [dsp.safe_subset(bounds, t) for t in targets]
if temporal_time_delta:
resolution = timedelta(temporal_time_delta)
if reference:
reference = dsp.temporal_rebin(reference, resolution)
if targets:
targets = [dsp.temporal_rebin(t, resolution) for t in targets]
if spatial_regrid_lats and spatial_regrid_lons:
lats = np.arange(spatial_regrid_lats[0], spatial_regrid_lats[1],
spatial_regrid_lats[2])
lons = np.arange(spatial_regrid_lons[0], spatial_regrid_lons[1],
spatial_regrid_lons[2])
if reference:
reference = dsp.spatial_regrid(reference, lats, lons)
if targets:
targets = [dsp.spatial_regrid(t, lats, lons) for t in targets]
return reference, targets
def _prepare_datasets_for_evaluation(reference, targets, config_data):
""""""
subset = config_data['evaluation'].get('subset', None)
temporal_time_delta = config_data['evaluation'].get('temporal_time_delta', None)
spatial_regrid_lats = config_data['evaluation'].get('spatial_regrid_lats', None)
spatial_regrid_lons = config_data['evaluation'].get('spatial_regrid_lons', None)
# If we have a temporal time delta and it's daily (i.e., 1) we will
# normalize the data as daily data (which means we adjust the start times
# for each bucket of data to be consistent). By default we will normalize
# the data as monthly. Note that this will not break yearly data so it's
# safer to do this no matter what. This keeps us from ending up with 1-off
# errors in the resulting dataset shape post-temporal/spatial adjustments
# that break evaluations.
string_time_delta = 'monthly'
if temporal_time_delta and temporal_time_delta == 1:
string_time_delta = 'daily'
reference = dsp.normalize_dataset_datetimes(reference, string_time_delta)
targets = [dsp.normalize_dataset_datetimes(t, string_time_delta) for t in targets]
if subset:
start = dateutil.parser.parse(subset[4])
end = dateutil.parser.parse(subset[5])
bounds = Bounds(subset[0], subset[1], subset[2], subset[3], start, end)
if reference:
reference = dsp.safe_subset(bounds, reference)
if targets:
targets = [dsp.safe_subset(bounds, t) for t in targets]
if temporal_time_delta:
resolution = timedelta(temporal_time_delta)
if reference:
reference = dsp.temporal_rebin(reference, resolution)
if targets:
targets = [dsp.temporal_rebin(t, resolution) for t in targets]
if spatial_regrid_lats and spatial_regrid_lons:
lats = np.arange(spatial_regrid_lats[0], spatial_regrid_lats[1], spatial_regrid_lats[2])
lons = np.arange(spatial_regrid_lons[0], spatial_regrid_lons[1], spatial_regrid_lons[2])
if reference:
reference = dsp.spatial_regrid(reference, lats, lons)
if targets:
targets = [dsp.spatial_regrid(t, lats, lons) for t in targets]
return reference, targets
def setUp(self):
self.input_dataset = ten_year_monthly_dataset()
self.new_lats = np.array(range(-89, 90, 4))
self.new_lons = np.array(range(-179, 180, 4))
self.regridded_dataset = dp.spatial_regrid(self.input_dataset,
self.new_lats,
self.new_lons)
def test_two_dimensional_lats_lons(self):
self.input_dataset.lats = np.array(range(-89, 90, 2))
self.input_dataset.lons = np.array(range(-179, 180, 4))
self.input_dataset.lats = self.input_dataset.lats.reshape(2, 45)
self.input_dataset.lons = self.input_dataset.lons.reshape(2, 45)
new_dataset = dp.spatial_regrid(
self.input_dataset, self.new_lats, self.new_lons)
np.testing.assert_array_equal(new_dataset.lats, self.new_lats)
def test_two_dimensional_lats_lons(self):
self.input_dataset.lats = np.array(range(-89, 90, 2))
self.input_dataset.lons = np.array(range(-179, 180, 4))
self.input_dataset.lats = self.input_dataset.lats.reshape(2, 45)
self.input_dataset.lons = self.input_dataset.lons.reshape(2, 45)
new_dataset = dp.spatial_regrid(
self.input_dataset, self.new_lats, self.new_lons)
np.testing.assert_array_equal(new_dataset.lats, self.new_lats)
def _prepare_datasets_for_evaluation(reference, targets, config_data):
""""""
subset = config_data['evaluation'].get('subset', None)
temporal_time_delta = config_data['evaluation'].get('temporal_time_delta', None)
spatial_regrid_lats = config_data['evaluation'].get('spatial_regrid_lats', None)
spatial_regrid_lons = config_data['evaluation'].get('spatial_regrid_lons', None)
if subset:
start = dateutil.parser.parse(subset[4])
end = dateutil.parser.parse(subset[5])
bounds = Bounds(subset[0], subset[1], subset[2], subset[3], start, end)
if reference:
reference = dsp.safe_subset(bounds, reference)
if targets:
targets = [dsp.safe_subset(bounds, t) for t in targets]
if temporal_time_delta:
resolution = timedelta(temporal_time_delta)
if reference:
reference = dsp.temporal_rebin(reference, resolution)
if targets:
targets = [dsp.temporal_rebin(t, resolution) for t in targets]
if spatial_regrid_lats and spatial_regrid_lons:
lats = np.arange(spatial_regrid_lats[0], spatial_regrid_lats[1], spatial_regrid_lats[2])
lons = np.arange(spatial_regrid_lons[0], spatial_regrid_lons[1], spatial_regrid_lons[2])
if reference:
reference = dsp.spatial_regrid(reference, lats, lons)
if targets:
targets = [dsp.spatial_regrid(t, lats, lons) for t in targets]
return reference, targets
def run_evaluation():
''' Run an OCW Evaluation.
*run_evaluation* expects the Evaluation parameters to be POSTed in
the following format.
.. sourcecode:: javascript
{
reference_dataset: {
// Id that tells us how we need to load this dataset.
'data_source_id': 1 == local, 2 == rcmed,
// Dict of data_source specific identifying information.
//
// if data_source_id == 1 == local:
// {
// 'id': The path to the local file on the server for loading.
// 'var_name': The variable data to pull from the file.
// 'lat_name': The latitude variable name.
// 'lon_name': The longitude variable name.
// 'time_name': The time variable name
// 'name': Optional dataset name
// }
//
// if data_source_id == 2 == rcmed:
// {
// 'dataset_id': The dataset id to grab from RCMED.
// 'parameter_id': The variable id value used by RCMED.
// 'name': Optional dataset name
// }
'dataset_info': {..}
},
// The list of target datasets to use in the Evaluation. The data
// format for the dataset objects should be the same as the
// reference_dataset above.
'target_datasets': [{...}, {...}, ...],
// All the datasets are re-binned to the reference dataset
// before being added to an experiment. This step (in degrees)
// is used when re-binning both the reference and target datasets.
'spatial_rebin_lat_step': The lat degree step. Integer > 0,
// Same as above, but for lon
'spatial_rebin_lon_step': The lon degree step. Integer > 0,
// The temporal resolution to use when doing a temporal re-bin
// This is a timedelta of days to use so daily == 1, monthly is
// (1, 31], annual/yearly is (31, 366], and full is anything > 366.
'temporal_resolution': Integer in range(1, 999),
// A list of the metric class names to use in the evaluation. The
// names must match the class name exactly.
'metrics': [Bias, TemporalStdDev, ...]
// The bounding values used in the Evaluation. Note that lat values
// should range from -180 to 180 and lon values from -90 to 90.
'start_time': start time value in the format '%Y-%m-%d %H:%M:%S',
'end_time': end time value in the format '%Y-%m-%d %H:%M:%S',
'lat_min': The minimum latitude value,
'lat_max': The maximum latitude value,
'lon_min': The minimum longitude value,
'lon_max': The maximum longitude value,
// NOTE: At the moment, subregion support is fairly minimal. This
// will be addressed in the future. Ideally, the user should be able
// to load a file that they have locally. That would change the
// format that this data is passed.
'subregion_information': Path to a subregion file on the server.
}
'''
# TODO: validate input parameters and return an error if not valid
eval_time_stamp = datetime.now().strftime('%Y-%m-%d_%H-%M-%S')
data = request.json
eval_bounds = {
'start_time': datetime.strptime(data['start_time'], '%Y-%m-%d %H:%M:%S'),
'end_time': datetime.strptime(data['end_time'], '%Y-%m-%d %H:%M:%S'),
'lat_min': float(data['lat_min']),
'lat_max': float(data['lat_max']),
'lon_min': float(data['lon_min']),
'lon_max': float(data['lon_max'])
}
# Load all the datasets
ref_dataset = _process_dataset_object(data['reference_dataset'], eval_bounds)
target_datasets = [_process_dataset_object(obj, eval_bounds)
for obj
in data['target_datasets']]
# Normalize the dataset time values so they break on consistent days of the
# month or time of the day, depending on how they will be rebinned.
resolution = data['temporal_resolution']
time_delta = timedelta(days=resolution)
time_step = 'daily' if resolution == 1 else 'monthly'
ref_dataset = dsp.normalize_dataset_datetimes(ref_dataset, time_step)
target_datasets = [dsp.normalize_dataset_datetimes(ds, time_step)
for ds in target_datasets]
# Subset the datasets
start = eval_bounds['start_time']
end = eval_bounds['end_time']
# Normalize all the values to the first of the month if we're not
# dealing with daily data. This will ensure that a valid subregion
# isn't considered out of bounds do to a dataset's time values
# being shifted to the first of the month.
if time_step != 'daily':
if start.day != 1:
day_offset = start.day - 1
start -= timedelta(days=day_offset)
if end.day != 1:
day_offset = end.day - 1
end -= timedelta(days=day_offset)
subset = Bounds(eval_bounds['lat_min'],
eval_bounds['lat_max'],
eval_bounds['lon_min'],
eval_bounds['lon_max'],
start,
end)
ref_dataset = dsp.safe_subset(subset, ref_dataset)
target_datasets = [dsp.safe_subset(subset, ds)
for ds
in target_datasets]
# Do temporal re-bin based off of passed resolution
ref_dataset = dsp.temporal_rebin(ref_dataset, time_delta)
target_datasets = [dsp.temporal_rebin(ds, time_delta)
for ds
in target_datasets]
# Do spatial re=bin based off of reference dataset + lat/lon steps
lat_step = data['spatial_rebin_lat_step']
lon_step = data['spatial_rebin_lon_step']
lat_bins, lon_bins = _calculate_new_latlon_bins(eval_bounds,
lat_step,
lon_step)
ref_dataset = dsp.spatial_regrid(ref_dataset, lat_bins, lon_bins)
target_datasets = [dsp.spatial_regrid(ds, lat_bins, lon_bins)
for ds
in target_datasets]
# Load metrics
loaded_metrics = _load_metrics(data['metrics'])
# Prime evaluation object with data
evaluation = Evaluation(ref_dataset, target_datasets, loaded_metrics)
# Run evaluation
evaluation.run()
# Plot
_generate_evaluation_plots(evaluation, lat_bins, lon_bins, eval_time_stamp)
return json.dumps({'eval_work_dir': eval_time_stamp})
ref_dataset = dsp.normalize_dataset_datetimes(ref_dataset, "monthly")
target_dataset = dsp.normalize_dataset_datetimes(target_dataset, "monthly")
# Subset down the evaluation datasets to our selected evaluation bounds.
target_dataset = dsp.subset(EVAL_BOUNDS, target_dataset)
ref_dataset = dsp.subset(EVAL_BOUNDS, ref_dataset)
# Do a monthly temporal rebin of the evaluation datasets.
target_dataset = dsp.temporal_rebin(target_dataset,
datetime.timedelta(days=30))
ref_dataset = dsp.temporal_rebin(ref_dataset, datetime.timedelta(days=30))
# Spatially regrid onto a 1 degree lat/lon grid within our evaluation bounds.
new_lats = np.arange(LAT_MIN, LAT_MAX, 1.0)
new_lons = np.arange(LON_MIN, LON_MAX, 1.0)
target_dataset = dsp.spatial_regrid(target_dataset, new_lats, new_lons)
ref_dataset = dsp.spatial_regrid(ref_dataset, new_lats, new_lons)
# Load the datasets for the evaluation.
mean_bias = metrics.MeanBias()
# These versions of the metrics require seasonal bounds prior to running
# the metrics. You should set these values above in the evaluation
# configuration section.
spatial_std_dev_ratio = metrics.SeasonalSpatialStdDevRatio(
month_start=SEASON_MONTH_START, month_end=SEASON_MONTH_END)
pattern_correlation = metrics.SeasonalPatternCorrelation(
month_start=SEASON_MONTH_START, month_end=SEASON_MONTH_END)
# Create our example evaluation.
example_eval = evaluation.Evaluation(
ref_dataset, # Reference dataset for the evaluation
def setUp(self):
self.input_dataset = ten_year_monthly_dataset()
self.new_lats = np.array(range(-89, 90, 4))
self.new_lons = np.array(range(-179, 180, 4))
self.regridded_dataset = dp.spatial_regrid(
self.input_dataset, self.new_lats, self.new_lons)
nlon = (max_lon - min_lon)/delta_lon+1
new_lat = np.linspace(min_lat, max_lat, nlat)
new_lon = np.linspace(min_lon, max_lon, nlon)
# number of models
nmodel = len(model_datasets)
print 'Dataset loading completed'
print 'Observation data:', ref_name
print 'Number of model datasets:',nmodel
for model_name in model_names:
print model_name
""" Step 4: Spatial regriding of the reference datasets """
print 'Regridding datasets: ', config['regrid']
if not config['regrid']['regrid_on_reference']:
ref_dataset = dsp.spatial_regrid(ref_dataset, new_lat, new_lon)
print 'Reference dataset has been regridded'
for idata,dataset in enumerate(model_datasets):
model_datasets[idata] = dsp.spatial_regrid(dataset, new_lat, new_lon, boundary_check = boundary_check_model)
print model_names[idata]+' has been regridded'
print 'Propagating missing data information'
ref_dataset = dsp.mask_missing_data([ref_dataset]+model_datasets)[0]
model_datasets = dsp.mask_missing_data([ref_dataset]+model_datasets)[1:]
""" Step 5: Checking and converting variable units """
print 'Checking and converting variable units'
ref_dataset = dsp.variable_unit_conversion(ref_dataset)
for idata,dataset in enumerate(model_datasets):
model_datasets[idata] = dsp.variable_unit_conversion(dataset)
# Temporally re-bin the data into a monthly timestep. ################################################################################ knmi_dataset = dsp.temporal_rebin(knmi_dataset, datetime.timedelta(days=30)) wrf_dataset = dsp.temporal_rebin(wrf_dataset, datetime.timedelta(days=30)) # Spatially regrid the datasets onto a 1 degree grid. ################################################################################ # Get the bounds of the reference dataset and use it to create a new # set of lat/lon values on a 1 degree step # Using the bounds we will create a new set of lats and lons on 1 degree step min_lat, max_lat, min_lon, max_lon = knmi_dataset.spatial_boundaries() new_lons = numpy.arange(min_lon, max_lon, 1) new_lats = numpy.arange(min_lat, max_lat, 1) # Spatially regrid datasets using the new_lats, new_lons numpy arrays knmi_dataset = dsp.spatial_regrid(knmi_dataset, new_lats, new_lons) wrf_dataset = dsp.spatial_regrid(wrf_dataset, new_lats, new_lons) # Load the metrics that we want to use for the evaluation. ################################################################################ taylor_diagram = metrics.SpatialPatternTaylorDiagram() # Create our new evaluation object. The knmi dataset is the evaluations # reference dataset. We then provide a list of 1 or more target datasets # to use for the evaluation. In this case, we only want to use the wrf dataset. # Then we pass a list of all the metrics that we want to use in the evaluation. ################################################################################ test_evaluation = evaluation.Evaluation(knmi_dataset, [wrf_dataset], [taylor_diagram]) test_evaluation.run() # Pull our the evaluation results and prepare them for drawing a Taylor diagram.
# 15 Jan 2014, 15 Feb 2014 => 1 Jan 2014, 1 Feb 2014
ref_dataset = dsp.normalize_dataset_datetimes(ref_dataset, "monthly")
target_dataset = dsp.normalize_dataset_datetimes(target_dataset, "monthly")
# Subset down the evaluation datasets to our selected evaluation bounds.
target_dataset = dsp.subset(EVAL_BOUNDS, target_dataset)
ref_dataset = dsp.subset(EVAL_BOUNDS, ref_dataset)
# Do a monthly temporal rebin of the evaluation datasets.
target_dataset = dsp.temporal_rebin(target_dataset, datetime.timedelta(days=30))
ref_dataset = dsp.temporal_rebin(ref_dataset, datetime.timedelta(days=30))
# Spatially regrid onto a 1 degree lat/lon grid within our evaluation bounds.
new_lats = np.arange(LAT_MIN, LAT_MAX, 1.0)
new_lons = np.arange(LON_MIN, LON_MAX, 1.0)
target_dataset = dsp.spatial_regrid(target_dataset, new_lats, new_lons)
ref_dataset = dsp.spatial_regrid(ref_dataset, new_lats, new_lons)
# Load the datasets for the evaluation.
mean_bias = metrics.MeanBias()
# These versions of the metrics require seasonal bounds prior to running
# the metrics. You should set these values above in the evaluation
# configuration section.
spatial_std_dev_ratio = metrics.SeasonalSpatialStdDevRatio(month_start=SEASON_MONTH_START, month_end=SEASON_MONTH_END)
pattern_correlation = metrics.SeasonalPatternCorrelation(month_start=SEASON_MONTH_START, month_end=SEASON_MONTH_END)
# Create our example evaluation.
example_eval = evaluation.Evaluation(ref_dataset, # Reference dataset for the evaluation
# 1 or more target datasets for the evaluation
[target_dataset],
# 1 ore more metrics to use in the evaluation
print("Processing datasets ...")
CRU31 = dsp.normalize_dataset_datetimes(CRU31, 'monthly')
print("... on units")
CRU31 = dsp.water_flux_unit_conversion(CRU31)
for member, each_target_dataset in enumerate(target_datasets):
target_datasets[member] = dsp.subset(target_datasets[member], EVAL_BOUNDS)
target_datasets[member] = dsp.water_flux_unit_conversion(
target_datasets[member])
target_datasets[member] = dsp.normalize_dataset_datetimes(
target_datasets[member], 'monthly')
print("... spatial regridding")
new_lats = np.arange(LAT_MIN, LAT_MAX, gridLatStep)
new_lons = np.arange(LON_MIN, LON_MAX, gridLonStep)
CRU31 = dsp.spatial_regrid(CRU31, new_lats, new_lons)
for member, each_target_dataset in enumerate(target_datasets):
target_datasets[member] = dsp.spatial_regrid(target_datasets[member],
new_lats, new_lons)
# find the total annual mean. Note the function exists in util.py as def
# calc_climatology_year(dataset):
_, CRU31.values = utils.calc_climatology_year(CRU31)
for member, each_target_dataset in enumerate(target_datasets):
_, target_datasets[member].values = utils.calc_climatology_year(
target_datasets[member])
# make the model ensemble
target_datasets_ensemble = dsp.ensemble(target_datasets)
def run_evaluation():
''' Run an OCW Evaluation.
*run_evaluation* expects the Evaluation parameters to be POSTed in
the following format.
.. sourcecode:: javascript
{
reference_dataset: {
// Id that tells us how we need to load this dataset.
'data_source_id': 1 == local, 2 == rcmed,
// Dict of data_source specific identifying information.
//
// if data_source_id == 1 == local:
// {
// 'id': The path to the local file on the server for loading.
// 'var_name': The variable data to pull from the file.
// 'lat_name': The latitude variable name.
// 'lon_name': The longitude variable name.
// 'time_name': The time variable name
// 'name': Optional dataset name
// }
//
// if data_source_id == 2 == rcmed:
// {
// 'dataset_id': The dataset id to grab from RCMED.
// 'parameter_id': The variable id value used by RCMED.
// 'name': Optional dataset name
// }
'dataset_info': {..}
},
// The list of target datasets to use in the Evaluation. The data
// format for the dataset objects should be the same as the
// reference_dataset above.
'target_datasets': [{...}, {...}, ...],
// All the datasets are re-binned to the reference dataset
// before being added to an experiment. This step (in degrees)
// is used when re-binning both the reference and target datasets.
'spatial_rebin_lat_step': The lat degree step. Integer > 0,
// Same as above, but for lon
'spatial_rebin_lon_step': The lon degree step. Integer > 0,
// The temporal resolution to use when doing a temporal re-bin
// This is a timedelta of days to use so daily == 1, monthly is
// (1, 31], annual/yearly is (31, 366], and full is anything > 366.
'temporal_resolution': Integer in range(1, 999),
// A list of the metric class names to use in the evaluation. The
// names must match the class name exactly.
'metrics': [Bias, TemporalStdDev, ...]
// The bounding values used in the Evaluation. Note that lat values
// should range from -180 to 180 and lon values from -90 to 90.
'start_time': start time value in the format '%Y-%m-%d %H:%M:%S',
'end_time': end time value in the format '%Y-%m-%d %H:%M:%S',
'lat_min': The minimum latitude value,
'lat_max': The maximum latitude value,
'lon_min': The minimum longitude value,
'lon_max': The maximum longitude value,
// NOTE: At the moment, subregion support is fairly minimal. This
// will be addressed in the future. Ideally, the user should be able
// to load a file that they have locally. That would change the
// format that this data is passed.
'subregion_information': Path to a subregion file on the server.
}
'''
# TODO: validate input parameters and return an error if not valid
eval_time_stamp = datetime.now().strftime('%Y-%m-%d_%H-%M-%S')
data = request.json
eval_bounds = {
'start_time': datetime.strptime(data['start_time'], '%Y-%m-%d %H:%M:%S'),
'end_time': datetime.strptime(data['end_time'], '%Y-%m-%d %H:%M:%S'),
'lat_min': float(data['lat_min']),
'lat_max': float(data['lat_max']),
'lon_min': float(data['lon_min']),
'lon_max': float(data['lon_max'])
}
# Load all the datasets
ref_dataset = _process_dataset_object(data['reference_dataset'], eval_bounds)
target_datasets = [_process_dataset_object(obj, eval_bounds)
for obj
in data['target_datasets']]
# Normalize the dataset time values so they break on consistent days of the
# month or time of the day, depending on how they will be rebinned.
resolution = data['temporal_resolution']
time_delta = timedelta(days=resolution)
time_step = 'daily' if resolution == 1 else 'monthly'
ref_dataset = dsp.normalize_dataset_datetimes(ref_dataset, time_step)
target_datasets = [dsp.normalize_dataset_datetimes(ds, time_step)
for ds in target_datasets]
# Subset the datasets
start = eval_bounds['start_time']
end = eval_bounds['end_time']
# Normalize all the values to the first of the month if we're not
# dealing with daily data. This will ensure that a valid subregion
# isn't considered out of bounds do to a dataset's time values
# being shifted to the first of the month.
if time_step != 'daily':
if start.day != 1:
day_offset = start.day - 1
start -= timedelta(days=day_offset)
if end.day != 1:
day_offset = end.day - 1
end -= timedelta(days=day_offset)
subset = Bounds(eval_bounds['lat_min'],
eval_bounds['lat_max'],
eval_bounds['lon_min'],
eval_bounds['lon_max'],
start,
end)
ref_dataset = dsp.safe_subset(ref_dataset, subset)
target_datasets = [dsp.safe_subset(ds, subset)
for ds
in target_datasets]
# Do temporal re-bin based off of passed resolution
ref_dataset = dsp.temporal_rebin(ref_dataset, time_delta)
target_datasets = [dsp.temporal_rebin(ds, time_delta)
for ds
in target_datasets]
# Do spatial re=bin based off of reference dataset + lat/lon steps
lat_step = data['spatial_rebin_lat_step']
lon_step = data['spatial_rebin_lon_step']
lat_bins, lon_bins = _calculate_new_latlon_bins(eval_bounds,
lat_step,
lon_step)
ref_dataset = dsp.spatial_regrid(ref_dataset, lat_bins, lon_bins)
target_datasets = [dsp.spatial_regrid(ds, lat_bins, lon_bins)
for ds
in target_datasets]
# Load metrics
loaded_metrics = _load_metrics(data['metrics'])
# Prime evaluation object with data
evaluation = Evaluation(ref_dataset, target_datasets, loaded_metrics)
# Run evaluation
evaluation.run()
# Plot
_generate_evaluation_plots(evaluation, lat_bins, lon_bins, eval_time_stamp)
return json.dumps({'eval_work_dir': eval_time_stamp})
print("Temporally Rebinning the Datasets to a Single Timestep")
# To run FULL temporal Rebinning
knmi_dataset = dsp.temporal_rebin(knmi_dataset, temporal_resolution='full')
cru31_dataset = dsp.temporal_rebin(cru31_dataset, temporal_resolution='full')
print("KNMI_Dataset.values shape: %s" % (knmi_dataset.values.shape, ))
print("CRU31_Dataset.values shape: %s \n\n" % (cru31_dataset.values.shape, ))
""" Spatially Regrid the Dataset Objects to a 1/2 degree grid """
# Using the bounds we will create a new set of lats and lons on 0.5 degree step
new_lons = np.arange(min_lon, max_lon, 0.5)
new_lats = np.arange(min_lat, max_lat, 0.5)
# Spatially regrid datasets using the new_lats, new_lons numpy arrays
print("Spatially Regridding the KNMI_Dataset...")
knmi_dataset = dsp.spatial_regrid(knmi_dataset, new_lats, new_lons)
print("Spatially Regridding the CRU31_Dataset...")
cru31_dataset = dsp.spatial_regrid(cru31_dataset, new_lats, new_lons)
print("Final shape of the KNMI_Dataset:%s" % (knmi_dataset.values.shape, ))
print("Final shape of the CRU31_Dataset:%s" % (cru31_dataset.values.shape, ))
""" Step 4: Build a Metric to use for Evaluation - Bias for this example """
# You can build your own metrics, but OCW also ships with some common metrics
print("Setting up a Bias metric to use for evaluation")
bias = metrics.Bias()
""" Step 5: Create an Evaluation Object using Datasets and our Metric """
# The Evaluation Class Signature is:
# Evaluation(reference, targets, metrics, subregions=None)
# Evaluation can take in multiple targets and metrics, so we need to convert
# our examples into Python lists. Evaluation will iterate over the lists
print("Making the Evaluation definition")
bias_evaluation = evaluation.Evaluation(knmi_dataset, [cru31_dataset], [bias])
def spatialRegrid(dataset, lats, lons):
'''Spatially regrid dataset variable to a new grid with specified resolution, where lats & lons
are the new coordinate vectors.
'''
return dsp.spatial_regrid(dataset, lats, lons)
# since the main intent of this program is to evaluate RCMs. However, it can be
# used for GCMs in which case it should be set to False to save time.
boundary_check = config['regrid'].get('boundary_check', True)
# number of target datasets (usually models, but can also be obs / reanalysis)
ntarget = len(target_datasets)
print('Dataset loading completed')
print('Reference data: {}'.format(reference_name))
print('Number of target datasets: {}'.format(ntarget))
for target_name in target_names:
print(target_name)
""" Step 3: Spatial regriding of the datasets """
print('Regridding datasets: {}'.format(config['regrid']))
if not config['regrid']['regrid_on_reference']:
reference_dataset = dsp.spatial_regrid(reference_dataset, new_lat, new_lon)
print('Reference dataset has been regridded')
for i, dataset in enumerate(target_datasets):
target_datasets[i] = dsp.spatial_regrid(dataset, new_lat, new_lon,
boundary_check=boundary_check)
print('{} has been regridded'.format(target_names[i]))
print('Propagating missing data information')
datasets = dsp.mask_missing_data([reference_dataset]+target_datasets)
reference_dataset = datasets[0]
target_datasets = datasets[1:]
""" Step 4: Checking and converting variable units """
print('Checking and converting variable units')
reference_dataset = dsp.variable_unit_conversion(reference_dataset)
for i, dataset in enumerate(target_datasets):
target_datasets[i] = dsp.variable_unit_conversion(dataset)
nlat = (max_lat - min_lat) / delta_lat + 1
nlon = (max_lon - min_lon) / delta_lon + 1
new_lat = np.linspace(min_lat, max_lat, nlat)
new_lon = np.linspace(min_lon, max_lon, nlon)
# number of models
nmodel = len(model_datasets)
print 'Dataset loading completed'
print 'Observation data:', ref_name
print 'Number of model datasets:', nmodel
for model_name in model_names:
print model_name
""" Step 4: Spatial regriding of the reference datasets """
print 'Regridding datasets: ', config['regrid']
if not config['regrid']['regrid_on_reference']:
ref_dataset = dsp.spatial_regrid(ref_dataset, new_lat, new_lon)
for idata, dataset in enumerate(model_datasets):
model_datasets[idata] = dsp.spatial_regrid(dataset, new_lat, new_lon)
print 'Propagating missing data information'
ref_dataset = dsp.mask_missing_data([ref_dataset] + model_datasets)[0]
model_datasets = dsp.mask_missing_data([ref_dataset] + model_datasets)[1:]
""" Step 5: Checking and converting variable units """
print 'Checking and converting variable units'
ref_dataset = dsp.variable_unit_conversion(ref_dataset)
for idata, dataset in enumerate(model_datasets):
model_datasets[idata] = dsp.variable_unit_conversion(dataset)
print 'Generating multi-model ensemble'
if len(model_datasets) >= 2.:
model_datasets.append(dsp.ensemble(model_datasets))
nlon = (max_lon - min_lon)/delta_lon+1
new_lat = np.linspace(min_lat, max_lat, nlat)
new_lon = np.linspace(min_lon, max_lon, nlon)
# number of models
nmodel = len(model_datasets)
print 'Dataset loading completed'
print 'Observation data:', ref_name
print 'Number of model datasets:',nmodel
for model_name in model_names:
print model_name
""" Step 4: Spatial regriding of the reference datasets """
print 'Regridding datasets: ', config['regrid']
if not config['regrid']['regrid_on_reference']:
ref_dataset = dsp.spatial_regrid(ref_dataset, new_lat, new_lon)
print 'Reference dataset has been regridded'
for idata,dataset in enumerate(model_datasets):
model_datasets[idata] = dsp.spatial_regrid(dataset, new_lat, new_lon)
print model_names[idata]+' has been regridded'
print 'Propagating missing data information'
ref_dataset = dsp.mask_missing_data([ref_dataset]+model_datasets)[0]
model_datasets = dsp.mask_missing_data([ref_dataset]+model_datasets)[1:]
""" Step 5: Checking and converting variable units """
print 'Checking and converting variable units'
ref_dataset = dsp.variable_unit_conversion(ref_dataset)
for idata,dataset in enumerate(model_datasets):
model_datasets[idata] = dsp.variable_unit_conversion(dataset)
# Get flag for boundary checking for regridding. By default, this is set to True
# since the main intent of this program is to evaluate RCMs. However, it can be
# used for GCMs in which case it should be set to False to save time.
boundary_check = config['regrid'].get('boundary_check', True)
# number of target datasets (usually models, but can also be obs / reanalysis)
ntarget = len(target_datasets)
print('Dataset loading completed')
print('Reference data: {}'.format(reference_name))
print('Number of target datasets: {}'.format(ntarget))
for target_name in target_names:
print(target_name)
""" Step 3: Spatial regriding of the datasets """
print('Regridding datasets: {}'.format(config['regrid']))
if not config['regrid']['regrid_on_reference']:
reference_dataset = dsp.spatial_regrid(reference_dataset, new_lat, new_lon)
print('Reference dataset has been regridded')
for i, dataset in enumerate(target_datasets):
target_datasets[i] = dsp.spatial_regrid(dataset,
new_lat,
new_lon,
boundary_check=boundary_check)
print('{} has been regridded'.format(target_names[i]))
print('Propagating missing data information')
datasets = dsp.mask_missing_data([reference_dataset] + target_datasets)
reference_dataset = datasets[0]
target_datasets = datasets[1:]
""" Step 4: Checking and converting variable units """
print('Checking and converting variable units')
reference_dataset = dsp.variable_unit_conversion(reference_dataset)
for i, dataset in enumerate(target_datasets):
nlon = (max_lon - min_lon)/delta_lon+1
new_lat = np.linspace(min_lat, max_lat, nlat)
new_lon = np.linspace(min_lon, max_lon, nlon)
# number of models
nmodel = len(model_datasets)
print 'Dataset loading completed'
print 'Observation data:', ref_name
print 'Number of model datasets:',nmodel
for model_name in model_names:
print model_name
""" Step 4: Spatial regriding of the reference datasets """
print 'Regridding datasets: ', config['regrid']
if not config['regrid']['regrid_on_reference']:
ref_dataset = dsp.spatial_regrid(ref_dataset, new_lat, new_lon)
print 'Reference dataset has been regridded'
for idata,dataset in enumerate(model_datasets):
model_datasets[idata] = dsp.spatial_regrid(dataset, new_lat, new_lon, boundary_check = boundary_check_model)
print model_names[idata]+' has been regridded'
print 'Propagating missing data information'
ref_dataset = dsp.mask_missing_data([ref_dataset]+model_datasets)[0]
model_datasets = dsp.mask_missing_data([ref_dataset]+model_datasets)[1:]
""" Step 5: Checking and converting variable units """
print 'Checking and converting variable units'
ref_dataset = dsp.variable_unit_conversion(ref_dataset)
for idata,dataset in enumerate(model_datasets):
model_datasets[idata] = dsp.variable_unit_conversion(dataset)
def spatialRegrid(dataset, lats, lons):
'''Spatially regrid dataset variable to a new grid with specified resolution, where lats & lons
are the new coordinate vectors.
'''
return dsp.spatial_regrid(dataset, lats, lons)
nlat = (max_lat - min_lat) / delta_lat + 1
nlon = (max_lon - min_lon) / delta_lon + 1
new_lat = np.linspace(min_lat, max_lat, nlat)
new_lon = np.linspace(min_lon, max_lon, nlon)
# number of models
nmodel = len(model_datasets)
print 'Dataset loading completed'
print 'Observation data:', obs_name
print 'Number of model datasets:', nmodel
for model_name in model_names:
print model_name
""" Step 4: Spatial regriding of the reference datasets """
print 'Regridding datasets: ', config['regrid']
if not config['regrid']['regrid_on_reference']:
obs_dataset = dsp.spatial_regrid(obs_dataset, new_lat, new_lon)
print 'Reference dataset has been regridded'
for i, dataset in enumerate(model_datasets):
model_datasets[i] = dsp.spatial_regrid(dataset,
new_lat,
new_lon,
boundary_check=boundary_check)
print model_names[i] + ' has been regridded'
print 'Propagating missing data information'
obs_dataset = dsp.mask_missing_data([obs_dataset] + model_datasets)[0]
model_datasets = dsp.mask_missing_data([obs_dataset] + model_datasets)[1:]
""" Step 5: Checking and converting variable units """
print 'Checking and converting variable units'
obs_dataset = dsp.variable_unit_conversion(obs_dataset)
for idata, dataset in enumerate(model_datasets):
model_datasets[idata] = dsp.variable_unit_conversion(dataset)
datasets = local.load_multiple_files(paths, 'clt')
with time_block(ocw_results, 'Domain Subsetting'):
for i, ds in enumerate(datasets):
datasets[i] = dsp.subset(ds, bnds)
with time_block(ocw_results, 'Seasonal Subsetting'):
for i, ds in enumerate(datasets):
datasets[i] = dsp.temporal_subset(ds, 9, 11)
with time_block(ocw_results, 'Resampling'):
for i, ds in enumerate(datasets):
datasets[i] = dsp.temporal_rebin(ds, 'annual')
with time_block(ocw_results, 'Regridding'):
for i, ds in enumerate(datasets):
datasets[i] = dsp.spatial_regrid(ds, new_lats, new_lons)
print(f'OCW Results: {ocw_results}')
# Plot results
matplotlib.style.use('ggplot')
df = pd.DataFrame({'OCW': ocw_results, 'BCDP': bcdp_results})
df.plot.bar(logy=True, rot=12)
for p in ax.patches:
val = np.round(p.get_height(), decimals=2)
ax.annotate(str(val), (p.get_x() + .02, p.get_height()), size=9.5)
plt.ylabel('Running Time [s]')
plt.savefig('bcdp_ocw_benchmarks.png')
# Running Temporal Rebin early helps negate the issue of datasets being on different
# days of the month (1st vs. 15th)
# Create a Bounds object to use for subsetting
new_bounds = Bounds(min_lat, max_lat, min_lon, max_lon, start_time, end_time)
# Subset our model datasets so they are the same size
knmi_dataset = dsp.subset(knmi_dataset, new_bounds)
wrf311_dataset = dsp.subset(wrf311_dataset, new_bounds)
""" Spatially Regrid the Dataset Objects to a 1/2 degree grid """
# Using the bounds we will create a new set of lats and lons on 1/2 degree step
new_lons = np.arange(min_lon, max_lon, 0.5)
new_lats = np.arange(min_lat, max_lat, 0.5)
# Spatially regrid datasets using the new_lats, new_lons numpy arrays
knmi_dataset = dsp.spatial_regrid(knmi_dataset, new_lats, new_lons)
wrf311_dataset = dsp.spatial_regrid(wrf311_dataset, new_lats, new_lons)
cru31_dataset = dsp.spatial_regrid(cru31_dataset, new_lats, new_lons)
# Generate an ensemble dataset from knmi and wrf models
ensemble_dataset = dsp.ensemble([knmi_dataset, wrf311_dataset])
""" Step 4: Build a Metric to use for Evaluation - Bias for this example """
print("Setting up a Bias metric to use for evaluation")
bias = metrics.Bias()
""" Step 5: Create an Evaluation Object using Datasets and our Metric """
# The Evaluation Class Signature is:
# Evaluation(reference, targets, metrics, subregions=None)
# Evaluation can take in multiple targets and metrics, so we need to convert
# our examples into Python lists. Evaluation will iterate over the lists
""" Step 3: Processing datasets so they are the same shape ... """
print("Processing datasets so they are the same shape")
CRU31 = dsp.water_flux_unit_conversion(CRU31)
CRU31 = dsp.normalize_dataset_datetimes(CRU31, 'monthly')
for member, each_target_dataset in enumerate(target_datasets):
target_datasets[member] = dsp.subset(target_datasets[member], EVAL_BOUNDS)
target_datasets[member] = dsp.water_flux_unit_conversion(target_datasets[
member])
target_datasets[member] = dsp.normalize_dataset_datetimes(
target_datasets[member], 'monthly')
print("... spatial regridding")
new_lats = np.arange(LAT_MIN, LAT_MAX, gridLatStep)
new_lons = np.arange(LON_MIN, LON_MAX, gridLonStep)
CRU31 = dsp.spatial_regrid(CRU31, new_lats, new_lons)
for member, each_target_dataset in enumerate(target_datasets):
target_datasets[member] = dsp.spatial_regrid(
target_datasets[member], new_lats, new_lons)
# find climatology monthly for obs and models
CRU31.values, CRU31.times = utils.calc_climatology_monthly(CRU31)
for member, each_target_dataset in enumerate(target_datasets):
target_datasets[member].values, target_datasets[
member].times = utils.calc_climatology_monthly(target_datasets[member])
# make the model ensemble
target_datasets_ensemble = dsp.ensemble(target_datasets)
# Temporally re-bin the data into a monthly timestep.
##########################################################################
knmi_dataset = dsp.temporal_rebin(knmi_dataset, temporal_resolution='monthly')
wrf_dataset = dsp.temporal_rebin(wrf_dataset, temporal_resolution='monthly')
# Spatially regrid the datasets onto a 1 degree grid.
##########################################################################
# Get the bounds of the reference dataset and use it to create a new
# set of lat/lon values on a 1 degree step
# Using the bounds we will create a new set of lats and lons on 1 degree step
min_lat, max_lat, min_lon, max_lon = knmi_dataset.spatial_boundaries()
new_lons = numpy.arange(min_lon, max_lon, 1)
new_lats = numpy.arange(min_lat, max_lat, 1)
# Spatially regrid datasets using the new_lats, new_lons numpy arrays
knmi_dataset = dsp.spatial_regrid(knmi_dataset, new_lats, new_lons)
wrf_dataset = dsp.spatial_regrid(wrf_dataset, new_lats, new_lons)
# Load the metrics that we want to use for the evaluation.
##########################################################################
taylor_diagram = metrics.SpatialPatternTaylorDiagram()
# Create our new evaluation object. The knmi dataset is the evaluations
# reference dataset. We then provide a list of 1 or more target datasets
# to use for the evaluation. In this case, we only want to use the wrf dataset.
# Then we pass a list of all the metrics that we want to use in the evaluation.
##########################################################################
test_evaluation = evaluation.Evaluation(knmi_dataset, [wrf_dataset],
[taylor_diagram])
test_evaluation.run()
def run_screen(model_datasets, models_info, observations_info,
overlap_start_time, overlap_end_time, overlap_min_lat,
overlap_max_lat, overlap_min_lon, overlap_max_lon,
temp_grid_setting, spatial_grid_setting, working_directory, plot_title):
'''Generates screen to show running evaluation process.
:param model_datasets: list of model dataset objects
:type model_datasets: list
:param models_info: list of dictionaries that contain information for each model
:type models_info: list
:param observations_info: list of dictionaries that contain information for each observation
:type observations_info: list
:param overlap_start_time: overlap start time between model and obs start time
:type overlap_start_time: datetime
:param overlap_end_time: overlap end time between model and obs end time
:type overlap_end_time: float
:param overlap_min_lat: overlap minimum lat between model and obs minimum lat
:type overlap_min_lat: float
:param overlap_max_lat: overlap maximum lat between model and obs maximum lat
:type overlap_max_lat: float
:param overlap_min_lon: overlap minimum lon between model and obs minimum lon
:type overlap_min_lon: float
:param overlap_max_lon: overlap maximum lon between model and obs maximum lon
:type overlap_max_lon: float
:param temp_grid_setting: temporal grid option such as hourly, daily, monthly and annually
:type temp_grid_setting: string
:param spatial_grid_setting:
:type spatial_grid_setting: string
:param working_directory: path to a directory for storring outputs
:type working_directory: string
:param plot_title: Title for plot
:type plot_title: string
'''
option = None
if option != "0":
ready_screen("manage_obs_screen")
y = screen.getmaxyx()[0]
screen.addstr(2, 2, "Evaluation started....")
screen.refresh()
OUTPUT_PLOT = "plot"
dataset_id = int(observations_info[0]['dataset_id']) #just accepts one dataset at this time
parameter_id = int(observations_info[0]['parameter_id']) #just accepts one dataset at this time
new_bounds = Bounds(overlap_min_lat, overlap_max_lat, overlap_min_lon, overlap_max_lon, overlap_start_time, overlap_end_time)
model_dataset = dsp.subset(new_bounds, model_datasets[0]) #just accepts one model at this time
#Getting bound info of subseted model file to retrive obs data with same bound as subseted model
new_model_spatial_bounds = model_dataset.spatial_boundaries()
new_model_temp_bounds = model_dataset.time_range()
new_min_lat = new_model_spatial_bounds[0]
new_max_lat = new_model_spatial_bounds[1]
new_min_lon = new_model_spatial_bounds[2]
new_max_lon = new_model_spatial_bounds[3]
new_start_time = new_model_temp_bounds[0]
new_end_time = new_model_temp_bounds[1]
screen.addstr(4, 4, "Retrieving data...")
screen.refresh()
#Retrieve obs data
obs_dataset = rcmed.parameter_dataset(
dataset_id,
parameter_id,
new_min_lat,
new_max_lat,
new_min_lon,
new_max_lon,
new_start_time,
new_end_time)
screen.addstr(4, 4, "--> Data retrieved.")
screen.refresh()
screen.addstr(5, 4, "Temporally regridding...")
screen.refresh()
if temp_grid_setting.lower() == 'hourly':
days = 0.5
elif temp_grid_setting.lower() == 'daily':
days = 1
elif temp_grid_setting.lower() == 'monthly':
days = 31
else:
days = 365
model_dataset = dsp.temporal_rebin(model_dataset, timedelta(days))
obs_dataset = dsp.temporal_rebin(obs_dataset, timedelta(days))
screen.addstr(5, 4, "--> Temporally regridded.")
screen.refresh()
new_lats = np.arange(new_min_lat, new_max_lat, spatial_grid_setting)
new_lons = np.arange(new_min_lon, new_max_lon, spatial_grid_setting)
screen.addstr(6, 4, "Spatially regridding...")
screen.refresh()
spatial_gridded_model = dsp.spatial_regrid(model_dataset, new_lats, new_lons)
spatial_gridded_obs = dsp.spatial_regrid(obs_dataset, new_lats, new_lons)
screen.addstr(6, 4, "--> Spatially regridded.")
screen.refresh()
screen.addstr(7, 4, "Setting up metrics...")
screen.refresh()
bias = metrics.Bias()
bias_evaluation = evaluation.Evaluation(spatial_gridded_model, [spatial_gridded_obs], [bias])
screen.addstr(7, 4, "--> Metrics setting done.")
screen.refresh()
screen.addstr(8, 4, "Running evaluation.....")
screen.refresh()
bias_evaluation.run()
results = bias_evaluation.results[0][0]
screen.addstr(8, 4, "--> Evaluation Finished.")
screen.refresh()
screen.addstr(9, 4, "Generating plots....")
screen.refresh()
lats = new_lats
lons = new_lons
gridshape = (1, 1)
sub_titles = [""] #No subtitle set for now
if not os.path.exists(working_directory):
os.makedirs(working_directory)
for i in range(len(results)):
fname = working_directory + OUTPUT_PLOT + str(i)
plotter.draw_contour_map(results[i], lats, lons, fname,
gridshape=gridshape, ptitle=plot_title,
subtitles=sub_titles)
screen.addstr(9, 4, "--> Plots generated.")
screen.refresh()
screen.addstr(y-2, 1, "Press 'enter' to Exit: ")
option = screen.getstr()