def draw_scatterplot(x, y, metadata, output_type=SCREEN, output_file=None,
**kwargs):
# Unpack the metadata information
variable = metadata.field_name
short_desc = metadata.short_description
units = metadata.units
# Set up the output figure
if output_type == SCREEN:
pl.ion()
else:
pl.ioff()
pl.clf()
pl.gcf().set_figwidth(3.4)
pl.gcf().set_figheight(3.0)
pl.gcf().set_dpi(250)
# Find the min and max of both axes
if (x.min() < y.min()):
abs_min = x.min()
else:
abs_min = y.min()
if (x.max() > y.max()):
abs_max = x.max()
else:
abs_max = y.max()
# Draw the scatterplot data and title
pl.scatter(x, y, s=2, c='b', edgecolor='k', linewidth=0.25, **kwargs)
pl.title(variable + ' : ' + short_desc, size=4.5)
# Calculate correlation coefficient, normalized RMSE and r_square
this_corr = statistics.pearson_r(x, y)
this_rmse = statistics.rmse(x, y) / x.mean()
this_r2 = statistics.r2(x, y)
# Draw the annotation text on the figure
pl.text(0.89, 0.93,
'1:1', transform=pl.gca().transAxes, size=4.5, rotation=45)
pl.text(0.05, 0.93,
'Correlation coefficient: %.4f' % (this_corr),
transform=pl.gca().transAxes, size=4.5)
pl.text(0.05, 0.89,
'Normalized RMSE: %.4f' % (this_rmse),
transform=pl.gca().transAxes, size=4.5)
pl.text(0.05, 0.85,
'R-square: %.4f' % (this_r2),
transform=pl.gca().transAxes, size=4.5)
# Draw the 1:1 line and format the x and y axes
pl.plot([abs_min, abs_max], [abs_min, abs_max], 'k-', linewidth=0.5)
ylabel_str = 'Predicted ' + variable
xlabel_str = 'Observed ' + variable
if units != 'none':
ylabel_str += ' (' + units + ')'
xlabel_str += ' (' + units + ')'
pl.ylabel(ylabel_str, size=4.5)
pl.xlabel(xlabel_str, size=4.5)
import matplotlib.ticker as ticker
f = ticker.OldScalarFormatter()
# f.set_powerlimits((-3, 4))
pl.gca().xaxis.set_major_formatter(f)
pl.gca().xaxis.set_minor_formatter(f)
pl.gca().yaxis.set_major_formatter(f)
pl.gca().yaxis.set_minor_formatter(f)
pl.xticks(size=4)
pl.yticks(size=4)
range = abs_max - abs_min
pl.xlim(abs_min - (0.01 * range), abs_max + (0.01 * range))
pl.ylim(abs_min - (0.01 * range), abs_max + (0.01 * range))
# Position the main axis within the figure
frame_x = 0.125
frame_width = 0.855
frame_y = 0.100
frame_height = 0.830
pl.gca().set_position([frame_x, frame_y, frame_width, frame_height])
pl.gca().axesPatch.set_linewidth(0.2)
axis = pl.gca()
for spine in axis.spines:
axis.spines[spine].set_linewidth(0.2)
# Set fill and edge for the figure
pl.gcf().figurePatch.set_edgecolor('k')
pl.gcf().figurePatch.set_linewidth(2.0)
# Draw and output to file if requested
pl.draw()
if output_type == FILE:
pl.savefig(output_file, dpi=250, edgecolor='k')
def run_diagnostic(self):
# Open the stats file and print out the header line
stats_fh = open(self.statistics_file, 'w')
out_list = [
'VARIABLE',
'PEARSON_R',
'SPEARMAN_R',
'RMSE',
'NORMALIZED_RMSE',
'BIAS_PERCENTAGE',
'R_SQUARE',
]
stats_fh.write(','.join(out_list) + '\n')
# Read the observed and predicted files into numpy recarrays
obs = utilities.csv2rec(self.observed_file)
prd = utilities.csv2rec(self.predicted_file)
# Subset the observed data just to the IDs that are in the
# predicted file
obs_keep = np.in1d(getattr(obs, self.id_field),
getattr(prd, self.id_field))
obs = obs[obs_keep]
# Read in the stand attribute metadata
mp = xsmp.XMLStandMetadataParser(self.stand_metadata_file)
# For each variable, calculate the statistics
for v in obs.dtype.names:
# Get the metadata for this field
try:
fm = mp.get_attribute(v)
except:
err_msg = v + ' is missing metadata.'
print err_msg
continue
# Only continue if this is a continuous accuracy variable
if fm.field_type != 'CONTINUOUS' or fm.accuracy_attr == 0:
continue
obs_vals = getattr(obs, v)
prd_vals = getattr(prd, v)
if np.all(obs_vals == 0.0):
pearson_r = 0.0
spearman_r = 0.0
rmse = 0.0
std_rmse = 0.0
bias = 0.0
r2 = 0.0
else:
if np.all(prd_vals == 0.0):
pearson_r = 0.0
spearman_r = 0.0
else:
pearson_r = statistics.pearson_r(obs_vals, prd_vals)
spearman_r = statistics.spearman_r(obs_vals, prd_vals)
rmse = statistics.rmse(obs_vals, prd_vals)
std_rmse = rmse / obs_vals.mean()
bias = statistics.bias_percentage(obs_vals, prd_vals)
r2 = statistics.r2(obs_vals, prd_vals)
# Print this out to the stats file
out_list = [
v,
'%.6f' % pearson_r,
'%.6f' % spearman_r,
'%.6f' % rmse,
'%.6f' % std_rmse,
'%.6f' % bias,
'%.6f' % r2,
]
stats_fh.write(','.join(out_list) + '\n')
stats_fh.close()
def run_diagnostic(self):
# Open the stats file and print out the header line
stats_fh = open(self.statistics_file, 'w')
out_list = [
'VARIABLE',
'PEARSON_R',
'SPEARMAN_R',
'RMSE',
'NORMALIZED_RMSE',
'BIAS_PERCENTAGE',
'R_SQUARE',
]
stats_fh.write(','.join(out_list) + '\n')
# Read the observed and predicted files into numpy recarrays
obs = utilities.csv2rec(self.observed_file)
prd = utilities.csv2rec(self.predicted_file)
# Subset the observed data just to the IDs that are in the
# predicted file
obs_keep = np.in1d(
getattr(obs, self.id_field), getattr(prd, self.id_field))
obs = obs[obs_keep]
# Read in the stand attribute metadata
mp = xsmp.XMLStandMetadataParser(self.stand_metadata_file)
# For each variable, calculate the statistics
for v in obs.dtype.names:
# Get the metadata for this field
try:
fm = mp.get_attribute(v)
except:
err_msg = v + ' is missing metadata.'
print err_msg
continue
# Only continue if this is a continuous accuracy variable
if fm.field_type != 'CONTINUOUS' or fm.accuracy_attr == 0:
continue
obs_vals = getattr(obs, v)
prd_vals = getattr(prd, v)
if np.all(obs_vals == 0.0):
pearson_r = 0.0
spearman_r = 0.0
rmse = 0.0
std_rmse = 0.0
bias = 0.0
r2 = 0.0
else:
if np.all(prd_vals == 0.0):
pearson_r = 0.0
spearman_r = 0.0
else:
pearson_r = statistics.pearson_r(obs_vals, prd_vals)
spearman_r = statistics.spearman_r(obs_vals, prd_vals)
rmse = statistics.rmse(obs_vals, prd_vals)
std_rmse = rmse / obs_vals.mean()
bias = statistics.bias_percentage(obs_vals, prd_vals)
r2 = statistics.r2(obs_vals, prd_vals)
# Print this out to the stats file
out_list = [
v,
'%.6f' % pearson_r,
'%.6f' % spearman_r,
'%.6f' % rmse,
'%.6f' % std_rmse,
'%.6f' % bias,
'%.6f' % r2,
]
stats_fh.write(','.join(out_list) + '\n')
stats_fh.close()