def sectorwise(self, column=None, sectors=12, plot='matplotlib', **kwargs):
'''Bin and plot the data sectorwise
Parameters:
___________
column: tuple, default None
Column to perform sectorwise analysis on
sectors: int, default 12
Number of sectors to bin
plot: string, default 'matplotlib'
Choose whether or not to plot your data, and what method.
Currently only supporting matplotlib, but hoping to add
Bokeh as that library evolves.
Returns:
________
DataFrame with sectorwise distribution
'''
cuts = 360 / sectors
bins = [0, cuts / 2]
bins.extend(np.arange(cuts * 1.5, 360 - cuts, cuts))
bins.extend([360 - cuts / 2, 360])
zeroed = lambda x: 0 if x == 360 else x
self.data[column] = self.data[column].apply(zeroed)
cats = pd.cut(self.data[column], bins, right=False)
array = pd.value_counts(cats).reindex(cats.levels).fillna(0)
wind_rose = pd.Series({
'[{0}, {1})'.format(360 - cuts / 2, 0 + cuts / 2):
array.ix[-1] + array.ix[0]
})
array = array.drop([array.index[0], array.index[-1]], axis=0)
wind_rose = wind_rose.append(array)
new_index = {
x: y
for x, y in zip(wind_rose.index, np.arange(0, 360, cuts))
}
wind_rose = wind_rose.rename(new_index)
freq_frame = pd.DataFrame(
{
'Counts': wind_rose,
'Frequencies': wind_rose / wind_rose.sum()
},
index=wind_rose.index)
if plot == 'matplotlib':
plottools.wind_rose(freq_frame['Frequencies'].values,
sectors=sectors,
**kwargs)
return freq_frame
def sectorwise(self, column=None, sectors=12, plot='matplotlib', **kwargs):
'''Bin and plot the data sectorwise
Parameters:
___________
column: tuple, default None
Column to perform sectorwise analysis on
sectors: int, default 12
Number of sectors to bin
plot: string, default 'matplotlib'
Choose whether or not to plot your data, and what method.
Currently only supporting matplotlib, but hoping to add
Bokeh as that library evolves.
Returns:
________
DataFrame with sectorwise distribution
'''
cuts = 360/sectors
bins = [0, cuts/2]
bins.extend(np.arange(cuts*1.5, 360-cuts, cuts))
bins.extend([360-cuts/2, 360])
zeroed = lambda x: 0 if x == 360 else x
self.data[column] = self.data[column].apply(zeroed)
cats = pd.cut(self.data[column], bins, right=False)
array = pd.value_counts(cats).reindex(cats.levels).fillna(0)
wind_rose = pd.Series({'[{0}, {1})'.format(360-cuts/2, 0+cuts/2):
array.ix[-1] + array.ix[0]})
array = array.drop([array.index[0], array.index[-1]], axis=0)
wind_rose = wind_rose.append(array)
new_index = {x: y for x, y in zip(wind_rose.index,
np.arange(0, 360, cuts))}
wind_rose = wind_rose.rename(new_index)
freq_frame = pd.DataFrame({'Counts': wind_rose,
'Frequencies': wind_rose/wind_rose.sum()},
index=wind_rose.index)
if plot == 'matplotlib':
plottools.wind_rose(freq_frame['Frequencies'].values,
sectors=sectors, **kwargs)
return freq_frame
def binned(self,
column=None,
bins=None,
stat='mean',
name=None,
plot=None):
'''Bin all data based on a single column.
Parameters:
___________
column: tuple, default None
Column on which to bin data
bins: array, default None
List or np.array with bins
stat: string, default 'mean'
Statistic you want to perform on binned data (mean, max, etc)
name: string, default None
Attribute name for binned data. Will create a new MetMast
attribute with binned data.
plot: tuple, default None
If you are binning by wind direction, plot=column_name will pass the
data to plottools.wind_rose
Returns:
________
self.data_binned_name, DataFrame with data summed by bins/stat
Examples:
_________
>>> mast.binned(column=('WS Mean 1', 56), bins=np.arange(0, 41, 1))
>>> mast.data_binned
>>> mast.binned(column=('WD Mean 1', 56), bins=np.arange(0, 375, 15),
stat='max', name='WD1_Max', plot=('WS Mean 1', 56))
>>> mast.data_binned_WD1_Max
'''
print('Mapping bins to data...')
def map_bin(x, bins):
kwargs = {}
if x == max(bins):
kwargs['right'] = True
bin = bins[np.digitize([x], bins, **kwargs)[0]]
bin_lower = bins[np.digitize([x], bins, **kwargs)[0] - 1]
return '[{0}-{1}]'.format(bin_lower, bin)
step = bins[1] - bins[0]
new_index = ['[{0}-{1}]'.format(x, x + step) for x in bins]
new_index.pop(-1)
temp_df = self.data.dropna()
temp_df['Binned'] = temp_df[column].apply(map_bin, bins=bins)
grouped = temp_df.groupby('Binned')
grouped_stat = getattr(grouped, stat)()
grouped_stat = grouped_stat.reindex(new_index)
if name is not None:
attr_name = 'data_binned_{0}'.format(name)
else:
attr_name = 'data_binned'
setattr(self, attr_name, grouped_stat)
if plot:
sect = len(new_index)
plottools.wind_rose(grouped_stat[plot].tolist(), sectors=sect)
def binned(self, column=None, bins=None, stat='mean', name=None,
plot=None):
'''Bin all data based on a single column.
Parameters:
___________
column: tuple, default None
Column on which to bin data
bins: array, default None
List or np.array with bins
stat: string, default 'mean'
Statistic you want to perform on binned data (mean, max, etc)
name: string, default None
Attribute name for binned data. Will create a new MetMast
attribute with binned data.
plot: tuple, default None
If you are binning by wind direction, plot=column_name will pass the
data to plottools.wind_rose
Returns:
________
self.data_binned_name, DataFrame with data summed by bins/stat
Examples:
_________
>>> mast.binned(column=('WS Mean 1', 56), bins=np.arange(0, 41, 1))
>>> mast.data_binned
>>> mast.binned(column=('WD Mean 1', 56), bins=np.arange(0, 375, 15),
stat='max', name='WD1_Max', plot=('WS Mean 1', 56))
>>> mast.data_binned_WD1_Max
'''
print('Mapping bins to data...')
def map_bin(x, bins):
kwargs = {}
if x == max(bins):
kwargs['right'] = True
bin = bins[np.digitize([x], bins, **kwargs)[0]]
bin_lower = bins[np.digitize([x], bins, **kwargs)[0]-1]
return '[{0}-{1}]'.format(bin_lower, bin)
step = bins[1]-bins[0]
new_index = ['[{0}-{1}]'.format(x, x+step) for x in bins]
new_index.pop(-1)
temp_df = self.data.dropna()
temp_df['Binned'] = temp_df[column].apply(map_bin, bins=bins)
grouped = temp_df.groupby('Binned')
grouped_stat = getattr(grouped, stat)()
grouped_stat = grouped_stat.reindex(new_index)
if name is not None:
attr_name = 'data_binned_{0}'.format(name)
else:
attr_name = 'data_binned'
setattr(self, attr_name, grouped_stat)
if plot:
sect = len(new_index)
plottools.wind_rose(grouped_stat[plot].tolist(), sectors=sect)