def common_ts_setup():
series2D = ts.time_series([np.random.rand(25).reshape(5,5),
np.random.rand(25).reshape(5,5),
np.random.rand(25).reshape(5,5),
np.random.rand(25).reshape(5,5),
np.random.rand(25).reshape(5,5),],
start_date=ts.now('M'),
mask=[np.random.rand(25).reshape(5,5)>.5,
np.random.rand(25).reshape(5,5)>.5,
np.random.rand(25).reshape(5,5)>.5,
np.random.rand(25).reshape(5,5)>.5,
np.random.rand(25).reshape(5,5)>.5,]
)
series1D = ts.time_series(np.random.rand(25),
mask=np.random.rand(25)>0.7,
start_date=ts.now('M'),
fill_value=-999)
series5V = ts.time_series(np.random.rand(25).reshape(5,5),
mask=np.random.rand(25).reshape(5,5)>0.7,
start_date=ts.now('M'))
series5N = ts.time_series(zip(np.random.rand(5),
np.random.rand(5),
np.arange(5)),
start_date=ts.now('M'),
dtype=[('a',float),('b',float),('c',int)]
)
return dict(series1D=series1D,
series5V=series5V,
series2D=series2D,
series5N=series5N)
def common_ts_setup():
series2D = ts.time_series([
np.random.rand(25).reshape(5, 5),
np.random.rand(25).reshape(5, 5),
np.random.rand(25).reshape(5, 5),
np.random.rand(25).reshape(5, 5),
np.random.rand(25).reshape(5, 5),
],
start_date=ts.now('M'),
mask=[
np.random.rand(25).reshape(5, 5) > .5,
np.random.rand(25).reshape(5, 5) > .5,
np.random.rand(25).reshape(5, 5) > .5,
np.random.rand(25).reshape(5, 5) > .5,
np.random.rand(25).reshape(5, 5) > .5,
])
series1D = ts.time_series(np.random.rand(25),
mask=np.random.rand(25) > 0.7,
start_date=ts.now('M'),
fill_value=-999)
series5V = ts.time_series(np.random.rand(25).reshape(5, 5),
mask=np.random.rand(25).reshape(5, 5) > 0.7,
start_date=ts.now('M'))
series5N = ts.time_series(zip(np.random.rand(5), np.random.rand(5),
np.arange(5)),
start_date=ts.now('M'),
dtype=[('a', float), ('b', float), ('c', int)])
return dict(series1D=series1D,
series5V=series5V,
series2D=series2D,
series5N=series5N)
def parse_instrument(instrument):
S = {}
t=ts.now('D')
S['START']=ts.Date(freq="D",datetime=datetime.date(t.year,1,1))
S['END']=ts.Date(freq='D',datetime=datetime.date(t.year,12,31))
S['FREQ']='D'
S['NAN']=True
m = re_instrument.match(instrument)
if m:
logger.debug('Match instrument |%s| %s',instrument,m.lastindex)
ticker,specs = m.groups()
logger.debug('|%s| -- %s', len(ticker),specs)
if len(ticker)==0:
ticker=u"E4T"
S['TICKER']=str(ticker)
logger.debug('Match ticker %s',ticker)
if specs:
parts = specs[1:].split('~')
for part in parts:
logger.debug('Match part %s',part)
if part[0]==':':
S['END']=match_date(part[1:])
elif re.match('\[([0-9]+)\]',part):
print "GOT",part
m=re.match('\[([0-9]+)\]',part)
S['START']=0
S['END']=int(m.group(1))
S['FREQ']='0'
elif re.match('!',part[0]):
S['NAN']=False
elif re.match('[0-9]',part[0]):
S['START']=match_date(part)
S['KIND']='T'
elif re.match('M|Q|A|D|Y',part[0]):
S['FREQ']=str(part[0])
logger.debug('parsed: %s',S)
if 'END' in S:
d = S['END']
df = d.asfreq(S['FREQ'],relation='START')
logger.debug('modify END: %s %s',d,df)
S['END']=S['END'].asfreq(S['FREQ'])
if 'START' in S:
logger.debug('modify START: %s',S['START'])
S['START']=S['START'].asfreq(S['FREQ'])
logger.debug('parsed: %s',S)
logger.debug('parsed')
else:
logger.debug('Cant match instrument |%s|',instrument)
logger.debug('parsed: %s',S)
if 'FREQ' not in S:
S['FREQ']='D'
if 'END' not in S:
S['END']=ts.now(str(S['FREQ']))
if 'START' not in S:
S['START']=ts.now(str(S['FREQ']))-(365 if S['FREQ']=='D' else 24 if S['FREQ']=='M' else 12 if S['FREQ']=='Q' else 10 )
logger.debug('parsed: %s',S)
return S
def setup(self):
a = time_series(np.random.rand(24),
start_date=ts.now('M'))
b = time_series(np.random.rand(24) * 100, dtype=int,
start_date=ts.now('M'),)
# c = time_series(["%02i" % _ for _ in np.arange(24)],
# start_date=ts.now('M'))
c = time_series(np.arange(24),
start_date=ts.now('M'))
trec = fromarrays([a, b, c], dates=a.dates, names='a,b,c')
self.info = (a, b, c, trec)
def setup(self):
a = time_series(np.random.rand(24), start_date=ts.now('M'))
b = time_series(
np.random.rand(24) * 100,
dtype=int,
start_date=ts.now('M'),
)
# c = time_series(["%02i" % _ for _ in np.arange(24)],
# start_date=ts.now('M'))
c = time_series(np.arange(24), start_date=ts.now('M'))
trec = fromarrays([a, b, c], dates=a.dates, names='a,b,c')
self.info = (a, b, c, trec)
def add_diurnal(tseries, sine_period, peak_mag):
"""
Scales a time series to a sine wave of peak_mag with sine_period.
Input: tseries, sine_period (float, hrs), peak_mag (float)
Output: scaled_data (array-like)
"""
# Convert sine_period to same frequency as tseries
# Create a time delta of magnitude sine_period
# Convert that time delta into frequency units same as tseries
zero_date = ts.now('H')
second_date = zero_date + sine_period
time_delta = ts.date_array([zero_date, second_date])
time_delta = time_delta.asfreq(tseries.freq)
sine_period = float(time_delta[1] - time_delta[0])
angular_freq = (2. * np.pi) / sine_period
for i in range(len(tseries)-1):
passed_time = float(tseries.dates[i]- tseries.start_date)
sine_factor = peak_mag * np.sin(angular_freq * passed_time)
tseries[i] = tseries[i] + tseries[i] * sine_factor
return tseries
# Generate power density function (pdf) to create synthetic TPM from
# mean, stdev, autocorr, npointsx
# def gen_pdf(desired_mean, desired_stdev, bin_width):
## TODO
return 0
def test_apply_on_fields_series(self):
"Test apply_on_fields w/ time_series"
adtype = [('fi', int), ('ff', float)]
a = ts.time_series([(1, 1.), (2, 2.), (3, 3.)],
mask=[(0, 0), (0, 1), (0, 0)],
dtype=adtype,
start_date=ts.now('M'))
func = ma.cumsum
test = apply_on_fields(a, func)
control = ts.time_series([(1, 1.), (3, -1), (6, 4.)],
mask=[(0, 0), (0, 1), (0, 0)],
dtype=adtype,
start_date=ts.now('M'))
assert_equal(test, control)
self.failUnless(isinstance(test, ts.TimeSeries))
assert_equal(test.dates, control.dates)
def setUp(self):
(a, b) = (np.arange(10), np.random.rand(10))
ndtype = [('a', np.float), ('b', np.float)]
tarr = ts.time_series(np.array(zip(a, b), dtype=ndtype),
start_date=ts.now('M'))
tarr.mask[3] = (False, True)
self.data = (tarr, a, b)
def setUp(self):
(a, b) = (np.arange(10), np.random.rand(10))
ndtype = [('a', np.float), ('b', np.float)]
tarr = ts.time_series(np.array(zip(a, b), dtype=ndtype),
start_date=ts.now('M'))
tarr.mask[3] = (False, True)
self.data = (tarr, a, b)
def test_apply_on_fields_series(self):
"Test apply_on_fields w/ time_series"
adtype = [('fi', int), ('ff', float)]
a = ts.time_series([(1, 1.), (2, 2.), (3, 3.)],
mask=[(0, 0), (0, 1), (0, 0)],
dtype=adtype,
start_date=ts.now('M'))
func = ma.cumsum
test = apply_on_fields(a, func)
control = ts.time_series([(1, 1.), (3, -1), (6, 4.)],
mask=[(0, 0), (0, 1), (0, 0)],
dtype=adtype,
start_date=ts.now('M'))
assert_equal(test, control)
self.failUnless(isinstance(test, ts.TimeSeries))
assert_equal(test.dates, control.dates)
def test_ontimeseries(self):
"Test cmov_average on a 1D TimeSeries."
data = ts.time_series(self.maskeddata, start_date=ts.now("D"))
for width in [3, 5, 7]:
k = (width - 1) / 2
ravg = mf.cmov_average(data, width)
self.failUnless(isinstance(ravg, MaskedArray))
assert_equal(ravg, data)
m = np.zeros(len(data), bool)
m[:k] = m[-k:] = m[10 - k : 10 + k + 1] = True
assert_equal(ravg._mask, m)
assert_equal(ravg._dates, data._dates)
def test_ontimeseries(self):
"Test cmov_average on a 1D TimeSeries."
data = ts.time_series(self.maskeddata, start_date=ts.now('D'))
for width in [3, 5, 7]:
k = (width - 1) / 2
ravg = mf.cmov_average(data, width)
self.failUnless(isinstance(ravg, MaskedArray))
assert_equal(ravg, data)
m = np.zeros(len(data), bool)
m[:k] = m[-k:] = m[10 - k:10 + k + 1] = True
assert_equal(ravg._mask, m)
assert_equal(ravg._dates, data._dates)
def __init__(self, *args, **kwds):
TestCase.__init__(self, *args, **kwds)
series = ts.time_series(zip(np.random.rand(10), np.arange(10)),
start_date=ts.now('M'),
dtype=[('a', float), ('b', int)])
series.mask[0] = (0, 1)
series.mask[-1] = (1, 0)
self.tseries = series
self.marray = series._series
self.file = tempfile.mktemp(".hdf5")
self.h5file = tables.openFile(self.file, 'a')
self.populate()
def __init__(self, *args, **kwds):
TestCase.__init__(self, *args, **kwds)
self.marray = ma.array(np.random.rand(100).reshape(10,10),
mask = (np.random.rand(100).reshape(10,10) > 0.7),
fill_value=-999,
hard_mask=True)
self.marray._optinfo['memo'] = "Some line of text"
self.tseries = ts.time_series(self.marray,
start_date=ts.now('D'))
self.file = tempfile.mktemp(".hdf5")
self.fileh = tables.openFile(self.file,'a')
self.populate()
def __init__(self, *args, **kwds):
TestCase.__init__(self, *args, **kwds)
self.marray = ma.array(
np.random.rand(100).reshape(10, 10),
mask=(np.random.rand(100).reshape(10, 10) > 0.7),
fill_value=-999,
hard_mask=True)
self.marray._optinfo['memo'] = "Some line of text"
self.tseries = ts.time_series(self.marray, start_date=ts.now('D'))
self.file = tempfile.mktemp(".hdf5")
self.fileh = tables.openFile(self.file, 'a')
self.populate()
def __init__(self, *args, **kwds):
TestCase.__init__(self, *args, **kwds)
series = ts.time_series(zip(np.random.rand(10),
np.arange(10)),
start_date=ts.now('M'),
dtype=[('a',float),('b',int)])
series.mask[0] = (0,1)
series.mask[-1] = (1,0)
self.tseries = series
self.marray = series._series
self.file = tempfile.mktemp(".hdf5")
self.h5file = tables.openFile(self.file,'a')
self.populate()
def test_ontimeseries(self):
"Test the moving functions on a 1D TimeSeries"
data = ts.time_series(self.maskeddata, start_date=ts.now("D"))
for Mfunc, Nfunc in self.func_pairs:
for k in [3, 4, 5]:
result = Mfunc(data, k)
self.failUnless(isinstance(result, MaskedArray))
for x in range(len(data) - k + 1):
if result[x + k - 1] is not ma.masked:
assert_almost_equal(np.asarray(result[x + k - 1]), np.asarray(Nfunc(data[x : x + k])))
result_mask = np.array([1] * (k - 1) + [0] * (len(data) - k + 1))
result_mask[10 : 10 + k] = 1
assert_equal(result._mask, result_mask)
assert_equal(result._dates, data._dates)
def test_ontimeseries(self):
"Test the moving functions on a 1D TimeSeries"
data = ts.time_series(self.maskeddata, start_date=ts.now('D'))
for Mfunc, Nfunc in self.func_pairs:
for k in [3, 4, 5]:
result = Mfunc(data, k)
self.failUnless(isinstance(result, MaskedArray))
for x in range(len(data) - k + 1):
if result[x + k - 1] is not ma.masked:
assert_almost_equal(np.asarray(result[x + k - 1]),
np.asarray(Nfunc(data[x:x + k])))
result_mask = np.array([1] * (k - 1) + [0] *
(len(data) - k + 1))
result_mask[10:10 + k] = 1
assert_equal(result._mask, result_mask)
assert_equal(result._dates, data._dates)
def setUp(self):
"Initializes"
ndtype = [('lin',float),('rand',float)]
dlin = np.linspace(0,10,120)
drnd = np.random.rand(120)
data = np.array(zip(dlin, drnd), dtype=ndtype)
dates = ts.date_array(start_date=ts.now('M')-120, length=120, freq='M')
enso = ENSOIndicator(np.random.rand(120) + np.linspace(-1,1,120),
dates=dates,
thresholds=(-0.5,0.5),
full_year='False', refseason='NDH', minsize=5)
cdat = data.view(ClimateRecords)
cdat._dates = dates
cdat.ensoindicator = enso
self.dlin = dlin
self.cdat=cdat
self.enso=enso
def tests_onmultitimeseries(self):
"Test cmov_average on a nD TimeSeries."
maskeddata = MaskedArray(np.random.random(75).reshape(25, 3))
maskeddata[10] = masked
data = ts.time_series(maskeddata, start_date=ts.now("D"))
for width in [3, 5, 7]:
k = (width - 1) / 2
ravg = mf.cmov_average(data, width)
self.failUnless(isinstance(ravg, MaskedArray))
#!!!: __getitem__ used to return a TimeSeries, now returns an array
# assert_almost_equal(ravg[18]._series.squeeze(),
# data[18-k:18+k+1]._series.mean(0))
assert_almost_equal(ravg[18].squeeze(), data[18 - k : 18 + k + 1]._series.mean(0))
m = np.zeros(data.shape, bool)
m[:k] = m[-k:] = m[10 - k : 10 + k + 1] = True
assert_equal(ravg._mask, m)
assert_equal(ravg._dates, data._dates)
def tests_onmultitimeseries(self):
"Test cmov_average on a nD TimeSeries."
maskeddata = MaskedArray(np.random.random(75).reshape(25, 3))
maskeddata[10] = masked
data = ts.time_series(maskeddata, start_date=ts.now('D'))
for width in [3, 5, 7]:
k = (width - 1) / 2
ravg = mf.cmov_average(data, width)
self.failUnless(isinstance(ravg, MaskedArray))
#!!!: __getitem__ used to return a TimeSeries, now returns an array
# assert_almost_equal(ravg[18]._series.squeeze(),
# data[18-k:18+k+1]._series.mean(0))
assert_almost_equal(ravg[18].squeeze(),
data[18 - k:18 + k + 1]._series.mean(0))
m = np.zeros(data.shape, bool)
m[:k] = m[-k:] = m[10 - k:10 + k + 1] = True
assert_equal(ravg._mask, m)
assert_equal(ravg._dates, data._dates)
def setUp(self):
"Initializes"
ndtype = [('lin', float), ('rand', float)]
dlin = np.linspace(0, 10, 120)
drnd = np.random.rand(120)
data = np.array(zip(dlin, drnd), dtype=ndtype)
dates = ts.date_array(start_date=ts.now('M') - 120,
length=120,
freq='M')
enso = ENSOIndicator(np.random.rand(120) + np.linspace(-1, 1, 120),
dates=dates,
thresholds=(-0.5, 0.5),
full_year='False',
refseason='NDH',
minsize=5)
cdat = data.view(ClimateRecords)
cdat._dates = dates
cdat.ensoindicator = enso
self.dlin = dlin
self.cdat = cdat
self.enso = enso
import numpy as np
import matplotlib.pyplot as plt
import scikits.timeseries as ts
import scikits.timeseries.lib.plotlib as tpl
# generate some random data
num_points = 12
data = np.cumprod(1 + np.random.normal(0, 1, num_points) / 100)
series = ts.time_series(data, start_date=ts.now('d') - num_points)
fig = tpl.tsfigure()
fsp = fig.add_tsplot(111)
fsp.tsplot(series, '-')
fsp.set_xlim(int(series.start_date), int(series.end_date))
fsp.set_title('%i days' % num_points)
plt.show()
def load_usgs_flows(site_no, cfgdict=usgs_config):
"""
Downloads hydrological data from the USGS site.
Parameters
----------
site_no : str
2- to 15-digit identification number of the site whose information must
be downloaded.
netfile : {string}, optional
URL of the USGS flow gage site.
By default, the value is read from the configuration file.
begin_date : {string}, optional
First date to be retrieved.
By default, the value is read from the configuration file.
end_date : {None, string}, optional
Last date to be retrieved.
If None, uses the current date.
datadir : {None, string}, optional
Directory where to store the output.
If None, uses the current directory.
commentchar : {string}, optional
Character corresponding to a comment.
headerlines : {int}, optional
Number of uncommented lines to skip from the beginning of the file.
archive : {None, string}, optional
Name of the ZIP archive where to read pre-stored data or when to store
data newly retrieved.
The default is read from the configuration file.
Notes
-----
The list of data codes valid for the ``data_type`` parameters are available
on the `USGS site <http://waterdata.usgs.gov/nwis/pmcodes/pmcodes?pm_group=All+--+include+all+parameter+groups&pm_search=&format=html_table&show=parameter_group_nm&show=parameter_nm>`_
"""
# Force the site_no to a 0-padded, 8-character string
site_no = "%08i" % int(site_no)
#data_type = cfgdict.get('data_type', '00060')
data_type = '00060'
data_letter = data_dict[data_type]
usgsarchv = _check_archive(cfgdict)
archfname = "%s%sD" % (data_letter, site_no)
#
try:
zipf = zipfile.ZipFile(usgsarchv, 'r')
series = cPickle.loads(zipf.read(archfname))
except IOError:
zipf = zipfile.ZipFile(usgsarchv, 'w')
except KeyError:
zipf = zipfile.ZipFile(usgsarchv, 'a')
else:
if series.size > 0:
return series
else:
zipf = zipfile.ZipFile(usgsarchv, 'a')
#
options = dict(usgs_site=cfgdict['netfile'],
begin_date=cfgdict['begin_date'],
data_type=data_type,
site_no=site_no)
end_date = cfgdict.get('end_date', None)
if end_date is None:
end_date = ts.now('D').strftime("%Y-%m-%d")
options['end_date'] = end_date
commd = "%(usgs_site)s/dv?format=rdb&cb_%(data_type)s=on&" + \
"begin_date=%(begin_date)s&end_date=%(end_date)s&" + \
"site_no=%(site_no)s"
datadir = cfgdict.get('datadir', os.getcwd())
# We can't use np.lib._datasource.DataSource, the downloaded file has always
# the same name and would be cached.
sourcedir = np.lib._datasource.DataSource(datadir)
dfile = urllib.urlopen(commd % options)
#
headers = int(cfgdict['headerlines'])
#
(datelist, datalist) = ([], [])
for line in dfile.readlines():
line = line.strip().split("\t")
if (not line) or (line[0][0] == '#'):
continue
try:
datalist.append(line[3])
except IndexError:
continue
else:
datelist.append(line[2])
series = ts.time_series(
[float(_) for _ in datalist[headers:]],
dates=[ts.Date('D', _) for _ in datelist[headers:]],
freq='D')
#
zipf.writestr(archfname, cPickle.dumps(series))
zipf.close()
return series
def load_usgs_flows(site_no, cfgdict=usgs_config):
"""
Downloads hydrological data from the USGS site.
Parameters
----------
site_no : str
2- to 15-digit identification number of the site whose information must
be downloaded.
netfile : {string}, optional
URL of the USGS flow gage site.
By default, the value is read from the configuration file.
begin_date : {string}, optional
First date to be retrieved.
By default, the value is read from the configuration file.
end_date : {None, string}, optional
Last date to be retrieved.
If None, uses the current date.
datadir : {None, string}, optional
Directory where to store the output.
If None, uses the current directory.
commentchar : {string}, optional
Character corresponding to a comment.
headerlines : {int}, optional
Number of uncommented lines to skip from the beginning of the file.
archive : {None, string}, optional
Name of the ZIP archive where to read pre-stored data or when to store
data newly retrieved.
The default is read from the configuration file.
Notes
-----
The list of data codes valid for the ``data_type`` parameters are available
on the `USGS site <http://waterdata.usgs.gov/nwis/pmcodes/pmcodes?pm_group=All+--+include+all+parameter+groups&pm_search=&format=html_table&show=parameter_group_nm&show=parameter_nm>`_
"""
# Force the site_no to a 0-padded, 8-character string
site_no = "%08i" % int(site_no)
#data_type = cfgdict.get('data_type', '00060')
data_type = '00060'
data_letter = data_dict[data_type]
usgsarchv = _check_archive(cfgdict)
archfname = "%s%sD" % (data_letter, site_no)
#
try:
zipf = zipfile.ZipFile(usgsarchv, 'r')
series = cPickle.loads(zipf.read(archfname))
except IOError:
zipf = zipfile.ZipFile(usgsarchv, 'w')
except KeyError:
zipf = zipfile.ZipFile(usgsarchv, 'a')
else:
if series.size > 0:
return series
else:
zipf = zipfile.ZipFile(usgsarchv, 'a')
#
options = dict(usgs_site=cfgdict['netfile'],
begin_date=cfgdict['begin_date'],
data_type=data_type,
site_no= site_no)
end_date = cfgdict.get('end_date', None)
if end_date is None:
end_date = ts.now('D').strftime("%Y-%m-%d")
options['end_date'] = end_date
commd = "%(usgs_site)s/dv?format=rdb&cb_%(data_type)s=on&" + \
"begin_date=%(begin_date)s&end_date=%(end_date)s&" + \
"site_no=%(site_no)s"
datadir = cfgdict.get('datadir', os.getcwd())
# We can't use np.lib._datasource.DataSource, the downloaded file has always
# the same name and would be cached.
sourcedir = np.lib._datasource.DataSource(datadir)
dfile = urllib.urlopen(commd % options)
#
headers = int(cfgdict['headerlines'])
#
(datelist, datalist) = ([], [])
for line in dfile.readlines():
line = line.strip().split("\t")
if (not line) or (line[0][0] == '#'):
continue
try:
datalist.append(line[3])
except IndexError:
continue
else:
datelist.append(line[2])
series = ts.time_series([float(_) for _ in datalist[headers:]],
dates=[ts.Date('D', _) for _ in datelist[headers:]],
freq='D')
#
zipf.writestr(archfname, cPickle.dumps(series))
zipf.close()
return series
def filterMngsByDate(mngs, date = 1):
date = ts.now('m') - date
def _filtermng(mng, date):
return len(mng.dates[mng.dates>date])
return [i for i in mngs if _filtermng(i[1], date)]
if hasattr(rand,generator):
f = getattr(rand,generator)
dd = f(**params)
## Add other distributions here
else:
raise TypeError, 'Unknown generator'
except TypeError, exc:
logging.error('%s generator - %s',rq['Source'],exc.args[0])
except:
raise
if s['FREQ']!='0':
data = ts.time_series(dd,
freq=s['FREQ'],
dates=dr)
if s['NAN']:
t=ts.now(data.freqstr)
if t<data.end_date:
da = ts.date_array(start_date=t,end_date=data.end_date)
data[da]=np.nan
rx[kw['NAME']]=Timeseries(data=data,name=name)
else: # for a vector
print type(dd),dd.shape
xx = dd.reshape((1,-1))
print type(xx),xx.shape
rx[kw['NAME']]=np.array(dd)
return rx
# random://normal/loc=10,scale=.2~:2012-12-31~2007-01-01~M?name=UNI
if __name__=="__main__":
parse_instrument('NAME=PIPPO')
import numpy as np
import matplotlib.pyplot as plt
import scikits.timeseries as ts
import scikits.timeseries.lib.plotlib as tpl
# generate some random data
num_points = 250
data = np.cumprod(1 + np.random.normal(0, 1, num_points)/100)
series = ts.time_series(data, start_date=ts.now('d')-num_points)
fig = tpl.tsfigure()
fsp = fig.add_tsplot(111)
fsp.tsplot(series, '-')
fsp.set_xlim(int(series.start_date), int(series.end_date))
fsp.set_title('%i days' % num_points)
plt.show()
def turningp(self,tsl,D,V=0.0,invert=False):
"""Turning Point - fill a series with a value from a turning point
:param tsl: the input time-series
:type tsl: a time-series
:param D: a date specification
:type D: a string representing a date (YYYY for A, YYYYMM for M, YYYYMMDD for D)
:param V: a value (default 0.0)
:type V: float
:return: the modified time-series
:rtype: a time-series
:Example:
dstream://Datastream/1.0$M_LIMIT?name=SOMEONE&proc=turningp(201101)
dstream://Datastream/0.0$M_LIMIT?name=SOMEZERO&proc=turningp(201101,1.0)
"""
freq = tsl._data.freqstr[0]
if re.match('^today$',D,re.I):
_tp = ts.now(freq)
else:
if freq == 'A':
_Y = int(D[0:4])
_tp = ts.Date(freq=freq,year=_Y)
elif freq == 'M':
_Y = int(D[0:4])
_M = int(D[4:6])
_tp = ts.Date(freq=freq,year=_Y,month=_M)
elif freq == 'Q':
_Y = int(D[0:4])
_Q = int(D[5:6])
_tp = ts.Date(freq=freq,year=_Y,quarter=_Q)
elif freq == 'D':
_Y = int(D[0:4])
_M = int(D[4:6])
_D = int(D[6:8])
_tp = ts.Date(freq=freq,year=_Y,month=_M,day=_D)
else:
raise OperationError, "turningp"
print "TURNINGP DATE",_tp
if re.match('^NAN$',V,re.I):
_V=np.nan
else:
_V = float(V)
_t1 = ts.time_series(tsl._data,copy=True)
_t1.fill(0.0)
if invert:
_tpd = ts.date_array(start_date=_t1.start_date,end_date=_tp)
else:
_tpd = ts.date_array(start_date=_tp,end_date=_t1.end_date)
_t1[_tpd]=1.0
_t = tsl._data * _t1
_t2 = ts.time_series(tsl._data,copy=True)
_t2.fill(_V)
if invert:
_tpd = ts.date_array(start_date=_t2.start_date,end_date=_tp)
else:
_tpd = ts.date_array(start_date=_tp,end_date=_t2.end_date)
_t2[_tpd]=0.0
_tx = _t + _t2
tsl._data = _tx
return tsl
def countCrossOver(self, nom = None):
if nom == None:
return len(self.crossRecord)
else:
temp = self.getCrossOverDate()
return len([i for i in (temp.dates > (ts.now('m') - nom)) if i])
