def setUp(self): dindex = pd.date_range("2000-01-01T00:00:00", periods=24, freq="H") self.ats = np.arange(0, 360, 15) self.ats = np.sin(2 * np.pi * self.ats / 360) self.ats = pd.DataFrame(self.ats, index=dindex) self.ats = tsutils.memory_optimize(self.ats) self.compare = self.ats.copy() self.compare = self.compare.join( pd.Series( np.zeros(len(self.ats)).astype("f"), index=self.ats.index, name="0::peak", )) self.compare = self.compare.join( pd.Series( np.zeros(len(self.ats)).astype("f"), index=self.ats.index, name="0::valley", )) self.compare.index.name = "Datetime" self.compare["0::peak"] = np.nan self.compare.loc[self.compare[0] == 1, "0::peak"] = 1 self.compare["0::valley"] = np.nan self.compare.loc[self.compare[0] == -1, "0::valley"] = -1 self.compare = tsutils.memory_optimize(self.compare)
def setUp(self): self.data_0_to_1 = tstoolbox.read( 'tests/data_sunspot_normalized_0_to_1.csv') self.data_10_to_20 = tstoolbox.read( 'tests/data_sunspot_normalized_10_to_20.csv') self.data_zscore = tstoolbox.read( 'tests/data_sunspot_normalized_zscore.csv') self.data_zscore = tsutils.memory_optimize(self.data_zscore) self.data_pct_rank = tstoolbox.read( 'tests/data_sunspot_normalized_pct_rank.csv') self.data_pct_rank = tsutils.memory_optimize(self.data_pct_rank)
def setUp(self): dr = pandas.date_range("2000-01-01", periods=2, freq="D") ts = pandas.Series([4.5, 4.6], index=dr) self.round_index_direct = pandas.DataFrame(ts, columns=["Value"]) self.round_index_direct.index.name = "Datetime" self.round_index_direct = tsutils.memory_optimize( self.round_index_direct) self.round_index_multiple_direct = pandas.DataFrame(ts, columns=["Value"]) self.round_index_multiple_direct = pandas.concat( [ self.round_index_multiple_direct, pandas.Series(ts, name="Value") ], axis="columns", ) self.round_index_multiple_direct.index.name = "Datetime" self.round_index_multiple_direct = tsutils.memory_optimize( self.round_index_multiple_direct) self.round_index_cli = b"""Datetime,Value 2000-01-01,4.5 2000-01-02,4.6 """ self.round_index_multiple_cli = b"""Datetime,Value,Value 2000-01-01,4.5,4.5 2000-01-02,4.6,4.6 """ self.round_index_tsstep_2_daily_cli = b"""Datetime,Value,Value1 2000-01-01,4.5,45.6 2000-01-03,4.7,34.2 2000-01-05,4.5,7.2 """ self.round_index_tsstep_2_daily = pandas.DataFrame( [[4.5, 45.6], [4.7, 34.2], [4.5, 7.2]], columns=["Value", "Value1"], index=pandas.DatetimeIndex( ["2000-01-01", "2000-01-03", "2000-01-05"]), ) self.round_index_tsstep_2_daily = tsutils.memory_optimize( self.round_index_tsstep_2_daily) self.round_index_tsstep_2_daily.index.name = "Datetime" self.round_index_blanks = b"""Datetime,Value_mean,Unnamed: 2_mean,Unnamed: 3_mean,Unnamed: 4_mean,Unnamed: 5_mean,Unnamed: 6_mean,Unnamed: 7_mean,Unnamed: 8_mean,Unnamed: 9_mean
def setUp(self): self.data_0_to_1 = tstoolbox.read( "tests/data_sunspot_normalized_0_to_1.csv") self.data_0_to_1.columns = ["Area::minmax"] self.data_10_to_20 = tstoolbox.read( "tests/data_sunspot_normalized_10_to_20.csv") self.data_10_to_20.columns = ["Area::minmax"] self.data_zscore = tstoolbox.read( "tests/data_sunspot_normalized_zscore.csv") self.data_zscore.columns = ["Area::zscore"] self.data_zscore = tsutils.memory_optimize(self.data_zscore) self.data_pct_rank = tstoolbox.read( "tests/data_sunspot_normalized_pct_rank.csv") self.data_pct_rank.columns = ["Area::pct_rank"] self.data_pct_rank = tsutils.memory_optimize(self.data_pct_rank)
def setUp(self): dr = pd.date_range('2011-01-01T12:00:00', periods=3, freq='H') self.date_slice = pd.DataFrame([2, 2, 2], index=dr, columns=['Value']) self.date_slice = tsutils.memory_optimize(self.date_slice) self.date_slice_cli = capture.capture(tsutils._printiso, self.date_slice)
def setUp(self): try: self.read_direct = pd.read_csv('tests/data_sunspot_EST.csv', index_col=0, parse_dates=[0]).tz_localize('UTC').tz_convert('EST') except TypeError: self.read_direct = pd.read_csv('tests/data_sunspot_EST.csv', index_col=0, parse_dates=[0]).tz_convert('EST') self.read_direct = tsutils.memory_optimize(self.read_direct)
def setUp(self): dr = pandas.date_range('2000-01-01', periods=6, freq='D') ts1 = pandas.Series([4.5, 4.6, 4.7, 4.6, 4.5, 4.4], index=dr) ts2 = pandas.Series([45.6, 90.5, 34.2, 23.1, 7.2, 4.3], index=dr) self.pick_multiple_direct = pandas.DataFrame(ts2, columns=['Value1']) self.pick_multiple_direct = self.pick_multiple_direct.join(pandas.DataFrame(ts1, columns=['Value'])) self.pick_multiple_direct = tsutils.memory_optimize(self.pick_multiple_direct) self.pick_cli = capture.capture(tsutils._printiso, self.pick_multiple_direct)
def setUp(self): self.add_trend_cli = b"""Datetime,Value::trend 2011-01-01 00:00:00,1 2011-01-01 01:00:00,1.04255 2011-01-01 02:00:00,1.08511 2011-01-01 03:00:00,1.12766 2011-01-01 04:00:00,1.17021 2011-01-01 05:00:00,1.21277 2011-01-01 06:00:00,1.25532 2011-01-01 07:00:00,1.29787 2011-01-01 08:00:00,1.34043 2011-01-01 09:00:00,1.38298 2011-01-01 10:00:00,1.42553 2011-01-01 11:00:00,1.46809 2011-01-01 12:00:00,1.51064 2011-01-01 13:00:00,1.55319 2011-01-01 14:00:00,1.59574 2011-01-01 15:00:00,1.6383 2011-01-01 16:00:00,1.68085 2011-01-01 17:00:00,1.7234 2011-01-01 18:00:00,1.76596 2011-01-01 19:00:00,1.80851 2011-01-01 20:00:00,1.85106 2011-01-01 21:00:00,1.89362 2011-01-01 22:00:00,1.93617 2011-01-01 23:00:00,1.97872 2011-01-02 00:00:00,2.02128 2011-01-02 01:00:00,2.06383 2011-01-02 02:00:00,2.10638 2011-01-02 03:00:00,2.14894 2011-01-02 04:00:00,2.19149 2011-01-02 05:00:00,2.23404 2011-01-02 06:00:00,2.2766 2011-01-02 07:00:00,2.31915 2011-01-02 08:00:00,2.3617 2011-01-02 09:00:00,2.40426 2011-01-02 10:00:00,2.44681 2011-01-02 11:00:00,2.48936 2011-01-02 12:00:00,2.53191 2011-01-02 13:00:00,2.57447 2011-01-02 14:00:00,2.61702 2011-01-02 15:00:00,2.65957 2011-01-02 16:00:00,2.70213 2011-01-02 17:00:00,2.74468 2011-01-02 18:00:00,2.78723 2011-01-02 19:00:00,2.82979 2011-01-02 20:00:00,2.87234 2011-01-02 21:00:00,2.91489 2011-01-02 22:00:00,2.95745 2011-01-02 23:00:00,3 """ self.add_trend_direct = tstoolbox.date_slice( input_ts=self.add_trend_cli) self.add_trend_direct.index.name = 'Datetime' self.add_trend_direct = tsutils.memory_optimize(self.add_trend_direct)
def setUp(self): self.add_trend_cli = b"""Datetime,Value_trend 2011-01-01 00:00:00,1 2011-01-01 01:00:00,1.04255 2011-01-01 02:00:00,1.08511 2011-01-01 03:00:00,1.12766 2011-01-01 04:00:00,1.17021 2011-01-01 05:00:00,1.21277 2011-01-01 06:00:00,1.25532 2011-01-01 07:00:00,1.29787 2011-01-01 08:00:00,1.34043 2011-01-01 09:00:00,1.38298 2011-01-01 10:00:00,1.42553 2011-01-01 11:00:00,1.46809 2011-01-01 12:00:00,1.51064 2011-01-01 13:00:00,1.55319 2011-01-01 14:00:00,1.59574 2011-01-01 15:00:00,1.6383 2011-01-01 16:00:00,1.68085 2011-01-01 17:00:00,1.7234 2011-01-01 18:00:00,1.76596 2011-01-01 19:00:00,1.80851 2011-01-01 20:00:00,1.85106 2011-01-01 21:00:00,1.89362 2011-01-01 22:00:00,1.93617 2011-01-01 23:00:00,1.97872 2011-01-02 00:00:00,2.02128 2011-01-02 01:00:00,2.06383 2011-01-02 02:00:00,2.10638 2011-01-02 03:00:00,2.14894 2011-01-02 04:00:00,2.19149 2011-01-02 05:00:00,2.23404 2011-01-02 06:00:00,2.2766 2011-01-02 07:00:00,2.31915 2011-01-02 08:00:00,2.3617 2011-01-02 09:00:00,2.40426 2011-01-02 10:00:00,2.44681 2011-01-02 11:00:00,2.48936 2011-01-02 12:00:00,2.53191 2011-01-02 13:00:00,2.57447 2011-01-02 14:00:00,2.61702 2011-01-02 15:00:00,2.65957 2011-01-02 16:00:00,2.70213 2011-01-02 17:00:00,2.74468 2011-01-02 18:00:00,2.78723 2011-01-02 19:00:00,2.82979 2011-01-02 20:00:00,2.87234 2011-01-02 21:00:00,2.91489 2011-01-02 22:00:00,2.95745 2011-01-02 23:00:00,3 """ self.add_trend_direct = tstoolbox.date_slice( input_ts=self.add_trend_cli) self.add_trend_direct.index.name = 'Datetime' self.add_trend_direct = tsutils.memory_optimize(self.add_trend_direct)
def setUp(self): """Prepare in-memory files data_stacked.csv and data_unstacked.csv.""" self.stacked = pd.read_csv( "tests/data_stacked.csv", index_col=0, parse_dates=True ) self.stacked.index.name = "Datetime" self.stacked = tsutils.memory_optimize(self.stacked) self.stacked_1 = pd.read_csv( "tests/data_stacked_1.csv", index_col=0, parse_dates=True ) self.stacked_1.index.name = "Datetime" self.stacked_1 = tsutils.memory_optimize(self.stacked_1) self.unstacked = pd.read_csv( "tests/data_unstacked.csv", index_col=0, parse_dates=True ) self.unstacked.rename(columns=lambda x: x.strip("'\" ")) self.unstacked.index.name = "Datetime" self.unstacked = tsutils.memory_optimize(self.unstacked)
def setUp(self): """Prepare in-memory files data_stacked.csv and data_unstacked.csv.""" self.stacked = pd.read_csv('tests/data_stacked.csv', index_col=0, parse_dates=True) self.stacked.index.name = 'Datetime' self.stacked = tsutils.memory_optimize(self.stacked) self.stacked_1 = pd.read_csv('tests/data_stacked_1.csv', index_col=0, parse_dates=True) self.stacked_1.index.name = 'Datetime' self.stacked_1 = tsutils.memory_optimize(self.stacked_1) self.unstacked = pd.read_csv('tests/data_unstacked.csv', index_col=0, parse_dates=True) self.unstacked.rename(columns=lambda x: x.strip('\'\" ')) self.unstacked.index.name = 'Datetime' self.unstacked = tsutils.memory_optimize(self.unstacked)
def setUp(self): dr = pandas.date_range('2000-01-01', periods=2, freq='D') ts = pandas.Series([4.5, 4.6], index=dr) self.round_index_direct = pandas.DataFrame(ts, columns=['Value']) self.round_index_direct.index.name = 'Datetime' self.round_index_direct = tsutils.memory_optimize(self.round_index_direct) self.round_index_multiple_direct = pandas.DataFrame(ts, columns=['Value']) self.round_index_multiple_direct = pandas.concat([self.round_index_multiple_direct, pandas.Series(ts, name='Value')], axis='columns') self.round_index_multiple_direct.index.name = 'Datetime' self.round_index_multiple_direct = tsutils.memory_optimize(self.round_index_multiple_direct) self.round_index_cli = b"""Datetime,Value 2000-01-01,4.5 2000-01-02,4.6 """ self.round_index_multiple_cli = b"""Datetime,Value,Value 2000-01-01,4.5,4.5 2000-01-02,4.6,4.6 """ self.round_index_tsstep_2_daily_cli = b"""Datetime,Value,Value1 2000-01-01,4.5,45.6 2000-01-03,4.7,34.2 2000-01-05,4.5,7.2 """ self.round_index_tsstep_2_daily = pandas.DataFrame( [[4.5, 45.6], [4.7, 34.2], [4.5, 7.2]], columns=['Value', 'Value1'], index=pandas.DatetimeIndex( ['2000-01-01', '2000-01-03', '2000-01-05'])) self.round_index_tsstep_2_daily = tsutils.memory_optimize(self.round_index_tsstep_2_daily) self.round_index_tsstep_2_daily.index.name = 'Datetime' self.round_index_blanks = b"""Datetime,Value_mean,Unnamed: 2_mean,Unnamed: 3_mean,Unnamed: 4_mean,Unnamed: 5_mean,Unnamed: 6_mean,Unnamed: 7_mean,Unnamed: 8_mean,Unnamed: 9_mean
def setUp(self): dr = pandas.date_range('2000-01-01', periods=2, freq='D') ts = pandas.Series([4.5, 4.6], index=dr) self.read_direct = pandas.DataFrame(ts, columns=['Value']) self.read_direct.index.name = 'Datetime' self.read_direct = tsutils.memory_optimize(self.read_direct) self.read_multiple_direct = pandas.DataFrame(ts, columns=['Value']) self.read_multiple_direct = pandas.concat([self.read_multiple_direct, pandas.Series(ts, name='Value')], axis='columns') self.read_multiple_direct.index.name = 'Datetime' self.read_multiple_direct = tsutils.memory_optimize(self.read_multiple_direct) self.read_cli = b"""Datetime,Value 2000-01-01,4.5 2000-01-02,4.6 """ self.read_multiple_cli = b"""Datetime,Value,Value 2000-01-01,4.5,4.5 2000-01-02,4.6,4.6 """ self.read_tsstep_2_daily_cli = b"""Datetime,Value,Value1 2000-01-01,4.5,45.6 2000-01-03,4.7,34.2 2000-01-05,4.5,7.2 """ self.read_tsstep_2_daily = pandas.DataFrame( [[4.5, 45.6], [4.7, 34.2], [4.5, 7.2]], columns=['Value', 'Value1'], index=pandas.DatetimeIndex( ['2000-01-01', '2000-01-03', '2000-01-05'])) self.read_tsstep_2_daily = tsutils.memory_optimize(self.read_tsstep_2_daily) self.read_tsstep_2_daily.index.name = 'Datetime' self.read_blanks = b"""Datetime,Value_mean,Unnamed: 2_mean,Unnamed: 3_mean,Unnamed: 4_mean,Unnamed: 5_mean,Unnamed: 6_mean,Unnamed: 7_mean,Unnamed: 8_mean,Unnamed: 9_mean
def setUp(self): dr = pandas.date_range('2011-01-01', periods=2, freq='D') ts = pandas.Series([2, 2], index=dr) self.aggregate_direct_mean = pandas.DataFrame(ts, columns=['Value::mean']) self.aggregate_direct_mean.index.name = 'Datetime' self.aggregate_direct_mean = tsutils.memory_optimize(self.aggregate_direct_mean) ts = pandas.Series([48, 48], index=dr) self.aggregate_direct_sum = pandas.DataFrame(ts, columns=['Value::sum']) self.aggregate_direct_sum.index.name = 'Datetime' self.aggregate_direct_sum = tsutils.memory_optimize(self.aggregate_direct_sum) self.aggregate_cli_mean = b"""Datetime,Value::mean 2011-01-01,2 2011-01-02,2 """ self.aggregate_cli_sum = b"""Datetime,Value::sum
def setUp(self): try: self.read_direct = ( pd.read_csv("tests/data_sunspot_EST.csv", index_col=0, parse_dates=[0]) .tz_localize("UTC") .tz_convert("EST") ) except TypeError: self.read_direct = pd.read_csv( "tests/data_sunspot_EST.csv", index_col=0, parse_dates=[0] ).tz_convert("EST") self.read_direct = tsutils.memory_optimize(self.read_direct)
def setUp(self): dr = pandas.date_range("2000-01-01", periods=6, freq="D") ts1 = pandas.Series([4.5, 4.6, 4.7, 4.6, 4.5, 4.4], index=dr) ts2 = pandas.Series([45.6, 90.5, 34.2, 23.1, 7.2, 4.3], index=dr) self.pick_multiple_direct = pandas.DataFrame(ts2, columns=["Value1"]) self.pick_multiple_direct = self.pick_multiple_direct.join( pandas.DataFrame(ts1, columns=["Value"]) ) self.pick_multiple_direct = tsutils.memory_optimize(self.pick_multiple_direct) self.pick_cli = capture.capture(tsutils._printiso, self.pick_multiple_direct)
def setUp(self): ''' Prepare in-memory versions of the files data_stacked.csv and data_unstacked.csv. ''' self.stacked = pd.read_csv('tests/data_stacked.csv', index_col=0, parse_dates=True) self.stacked.index.name = 'Datetime' self.stacked = tsutils.memory_optimize(self.stacked) self.stacked_1 = pd.read_csv('tests/data_stacked_1.csv', index_col=0, parse_dates=True) self.stacked_1.index.name = 'Datetime' self.stacked_1 = tsutils.memory_optimize(self.stacked_1) self.unstacked = pd.read_csv('tests/data_unstacked.csv', index_col=0, parse_dates=True) self.unstacked.rename(columns=lambda x: x.strip('\'\" ')) self.unstacked.index.name = 'Datetime' self.unstacked = tsutils.memory_optimize(self.unstacked)
def setUp(self): dr = pandas.date_range("2000-01-01", periods=2, freq="D") ts = pandas.Series([4.5, 4.6], index=dr) self.compare_direct_01 = pandas.DataFrame(ts, columns=["Value::convert"]) self.compare_direct_01.index.name = "Datetime" self.compare_direct_01 = tsutils.memory_optimize( self.compare_direct_01) dr = pandas.date_range("2000-01-01", periods=2, freq="D") ts = pandas.Series([11.0, 11.2], index=dr) self.compare_direct_02 = pandas.DataFrame(ts, columns=["Value::convert"]) self.compare_direct_02.index.name = "Datetime" self.compare_direct_02 = tsutils.memory_optimize( self.compare_direct_02) self.compare_cli_01 = b"""Datetime,Value::convert 2000-01-01,4.5 2000-01-02,4.6 """ self.compare_cli_02 = b"""Datetime,Value::convert
def setUp(self): dindex = pd.date_range('2000-01-01T00:00:00', periods=24, freq='H') self.ats = pd.np.arange(0, 360, 15) self.ats = pd.np.sin(2 * pd.np.pi * self.ats / 360) self.ats = pd.DataFrame(self.ats, index=dindex) self.ats = tsutils.memory_optimize(self.ats) self.compare = self.ats.copy() self.compare = self.compare.join( pd.Series(pd.np.zeros(len(self.ats)).astype('f'), index=self.ats.index, name='0_peak')) self.compare = self.compare.join( pd.Series(pd.np.zeros(len(self.ats)).astype('f'), index=self.ats.index, name='0_valley')) self.compare.index.name = 'Datetime' self.compare['0_peak'] = pd.np.nan self.compare.loc[self.compare[0] == 1, '0_peak'] = 1 self.compare['0_valley'] = pd.np.nan self.compare.loc[self.compare[0] == -1, '0_valley'] = -1 self.compare = tsutils.memory_optimize(self.compare)
def setUp(self): dindex = pd.date_range('2000-01-01T00:00:00', periods=24, freq='H') self.ats = pd.np.arange(0, 360, 15) self.ats = pd.np.sin(2*pd.np.pi*self.ats/360) self.ats = pd.DataFrame(self.ats, index=dindex) self.ats = tsutils.memory_optimize(self.ats) self.compare = self.ats.copy() self.compare = self.compare.join( pd.Series(pd.np.zeros(len(self.ats)).astype('f'), index=self.ats.index, name='0::peak')) self.compare = self.compare.join( pd.Series(pd.np.zeros(len(self.ats)).astype('f'), index=self.ats.index, name='0::valley')) self.compare.index.name = 'Datetime' self.compare['0::peak'] = pd.np.nan self.compare.loc[self.compare[0] == 1, '0::peak'] = 1 self.compare['0::valley'] = pd.np.nan self.compare.loc[self.compare[0] == -1, '0::valley'] = -1 self.compare = tsutils.memory_optimize(self.compare)
def setUp(self): dr = pandas.date_range('2000-01-01', periods=2, freq='D') ts = pandas.Series([4.5, 4.6], index=dr) self.read_direct = pandas.DataFrame(ts, columns=['Value']) self.read_direct.index.name = 'Datetime' self.read_direct = tsutils.memory_optimize(self.read_direct) self.read_cli = b"""Datetime,Value 2000-01-01,4.5 2000-01-02,4.6 """ dr = pandas.date_range('2000-01-01', periods=5, freq='D') ts = pandas.Series([4.5, 4.6, 4.7, 4.8, 4.9], index=dr) self.read_direct_sparse = pandas.DataFrame(ts, columns=['Value']) self.read_direct_sparse.index.name = 'Datetime' self.read_direct_sparse = tsutils.memory_optimize(self.read_direct_sparse) self.read_cli_sparse = b"""Datetime,Value
def setUp(self): dr = pandas.date_range("2000-01-01", periods=2, freq="D") ts = pandas.Series([4.5, 4.6], index=dr) self.read_direct = pandas.DataFrame(ts, columns=["Value"]) self.read_direct.index.name = "Datetime" self.read_direct = tsutils.memory_optimize(self.read_direct) self.read_cli = b"""Datetime,Value 2000-01-01,4.5 2000-01-02,4.6 """ dr = pandas.date_range("2000-01-01", periods=5, freq="D") ts = pandas.Series([4.5, 4.6, 4.7, 4.8, 4.9], index=dr) self.read_direct_sparse = pandas.DataFrame(ts, columns=["Value"]) self.read_direct_sparse.index.name = "Datetime" self.read_direct_sparse = tsutils.memory_optimize(self.read_direct_sparse) self.read_cli_sparse = b"""Datetime,Value
def setUp(self): dr = pandas.date_range("2011-01-01", periods=2, freq="D") ts = pandas.Series([2, 2], index=dr) self.aggregate_direct_mean = pandas.DataFrame(ts, columns=["Value::mean"]) self.aggregate_direct_mean.index.name = "Datetime" self.aggregate_direct_mean = tsutils.memory_optimize( self.aggregate_direct_mean) ts = pandas.Series([48, 48], index=dr) self.aggregate_direct_sum = pandas.DataFrame(ts, columns=["Value::sum"]) self.aggregate_direct_sum.index.name = "Datetime" self.aggregate_direct_sum = tsutils.memory_optimize( self.aggregate_direct_sum) self.aggregate_cli_mean = b"""Datetime,Value::mean 2011-01-01,2 2011-01-02,2 """ self.aggregate_cli_sum = b"""Datetime,Value::sum
def setUp(self): ''' Setup ''' dr = pandas.date_range('2011-01-01', periods=2, freq='D') ts = pandas.Series([2, 2], index=dr) self.aggregate_direct_mean = pandas.DataFrame(ts, columns=['Value_mean']) self.aggregate_direct_mean.index.name = 'Datetime' self.aggregate_direct_mean = tsutils.memory_optimize( self.aggregate_direct_mean) ts = pandas.Series([48, 48], index=dr) self.aggregate_direct_sum = pandas.DataFrame(ts, columns=['Value_sum']) self.aggregate_direct_sum.index.name = 'Datetime' self.aggregate_direct_sum = tsutils.memory_optimize( self.aggregate_direct_sum) self.aggregate_cli_mean = b"""Datetime,Value_mean 2011-01-01,2 2011-01-02,2 """ self.aggregate_cli_sum = b"""Datetime,Value_sum
def setUp(self): dindex = pd.date_range("2011-01-01T00:00:00", periods=26, freq="H") self.ats = np.ones((26)) * 2 self.ats = pd.DataFrame(self.ats, index=dindex, columns=["Value_with_missing::fill"]) self.ats.index.name = "Datetime" self.ats = tsutils.memory_optimize(self.ats) self.ats_cli = capture.capture(tsutils._printiso, self.ats) self.ffill_compare = self.ats.copy() self.ffill_compare["Value_with_missing::fill"][ "2011-01-01T09:00:00":"2011-01-01T12:00:00"] = 3 self.ffill_compare["Value_with_missing::fill"][ "2011-01-01T13:00:00"] = 9 self.ffill_compare_cli = capture.capture(tsutils._printiso, self.ffill_compare) self.bfill_compare = self.ats.copy() self.bfill_compare["Value_with_missing::fill"][ "2011-01-01T09:00:00"] = 3 self.bfill_compare["Value_with_missing::fill"][ "2011-01-01T10:00:00":"2011-01-01T13:00:00"] = 9 self.bfill_compare_cli = capture.capture(tsutils._printiso, self.bfill_compare) self.linear_compare = self.ats.copy() self.linear_compare["Value_with_missing::fill"][ "2011-01-01T09:00:00"] = 3.0 self.linear_compare["Value_with_missing::fill"][ "2011-01-01T10:00:00"] = 4.5 self.linear_compare["Value_with_missing::fill"][ "2011-01-01T11:00:00"] = 6.0 self.linear_compare["Value_with_missing::fill"][ "2011-01-01T12:00:00"] = 7.5 self.linear_compare["Value_with_missing::fill"][ "2011-01-01T13:00:00"] = 9.0 self.linear_compare_cli = capture.capture(tsutils._printiso, self.linear_compare) self.nearest_compare = self.ats.copy() self.nearest_compare["Value_with_missing::fill"][ "2011-01-01T09:00:00":"2011-01-01T11:00:00"] = 3.0 self.nearest_compare["Value_with_missing::fill"][ "2011-01-01T12:00:00":"2011-01-01T13:00:00"] = 9.0 self.nearest_compare = tsutils.memory_optimize(self.nearest_compare) self.nearest_compare_cli = capture.capture(tsutils._printiso, self.nearest_compare) self.mean_compare = self.ats.copy() self.mean_compare["Value_with_missing::fill"][ "2011-01-01T01:00:00"] = 2.4210526315789473 self.mean_compare["Value_with_missing::fill"][ "2011-01-01T09:00:00"] = 3.0 self.mean_compare["Value_with_missing::fill"][ "2011-01-01T10:00:00"] = 2.4210526315789473 self.mean_compare["Value_with_missing::fill"][ "2011-01-01T11:00:00"] = 2.4210526315789473 self.mean_compare["Value_with_missing::fill"][ "2011-01-01T12:00:00"] = 2.4210526315789473 self.mean_compare["Value_with_missing::fill"][ "2011-01-01T13:00:00"] = 9.0 self.mean_compare["Value_with_missing::fill"][ "2011-01-01T16:00:00"] = 2.4210526315789473 self.mean_compare["Value_with_missing::fill"][ "2011-01-01T22:00:00"] = 2.4210526315789473 self.mean_compare["Value_with_missing::fill"][ "2011-01-01T23:00:00"] = 2.4210526315789473 self.mean_compare_cli = capture.capture(tsutils._printiso, self.mean_compare) self.median_compare = self.ats.copy() self.median_compare["Value_with_missing::fill"][ "2011-01-01T01:00:00"] = 2.0 self.median_compare["Value_with_missing::fill"][ "2011-01-01T09:00:00"] = 3.0 self.median_compare["Value_with_missing::fill"][ "2011-01-01T10:00:00"] = 2.0 self.median_compare["Value_with_missing::fill"][ "2011-01-01T11:00:00"] = 2.0 self.median_compare["Value_with_missing::fill"][ "2011-01-01T12:00:00"] = 2.0 self.median_compare["Value_with_missing::fill"][ "2011-01-01T13:00:00"] = 9.0 self.median_compare["Value_with_missing::fill"][ "2011-01-01T16:00:00"] = 2.0 self.median_compare["Value_with_missing::fill"][ "2011-01-01T22:00:00"] = 2.0 self.median_compare["Value_with_missing::fill"][ "2011-01-01T23:00:00"] = 2.0 self.median_compare = tsutils.memory_optimize(self.median_compare) self.median_compare_cli = capture.capture(tsutils._printiso, self.median_compare) self.max_compare = self.ats.copy() self.max_compare["Value_with_missing::fill"][ "2011-01-01T01:00:00"] = 9.0 self.max_compare["Value_with_missing::fill"][ "2011-01-01T09:00:00"] = 3.0 self.max_compare["Value_with_missing::fill"][ "2011-01-01T10:00:00"] = 9.0 self.max_compare["Value_with_missing::fill"][ "2011-01-01T11:00:00"] = 9.0 self.max_compare["Value_with_missing::fill"][ "2011-01-01T12:00:00"] = 9.0 self.max_compare["Value_with_missing::fill"][ "2011-01-01T13:00:00"] = 9.0 self.max_compare["Value_with_missing::fill"][ "2011-01-01T16:00:00"] = 9.0 self.max_compare["Value_with_missing::fill"][ "2011-01-01T22:00:00"] = 9.0 self.max_compare["Value_with_missing::fill"][ "2011-01-01T23:00:00"] = 9.0 self.max_compare = tsutils.memory_optimize(self.max_compare) self.max_compare_cli = capture.capture(tsutils._printiso, self.max_compare) self.min_compare = self.ats.copy() self.min_compare["Value_with_missing::fill"][ "2011-01-01T01:00:00"] = 2.0 self.min_compare["Value_with_missing::fill"][ "2011-01-01T09:00:00"] = 3.0 self.min_compare["Value_with_missing::fill"][ "2011-01-01T10:00:00"] = 2.0 self.min_compare["Value_with_missing::fill"][ "2011-01-01T11:00:00"] = 2.0 self.min_compare["Value_with_missing::fill"][ "2011-01-01T12:00:00"] = 2.0 self.min_compare["Value_with_missing::fill"][ "2011-01-01T13:00:00"] = 9.0 self.min_compare["Value_with_missing::fill"][ "2011-01-01T16:00:00"] = 2.0 self.min_compare["Value_with_missing::fill"][ "2011-01-01T22:00:00"] = 2.0 self.min_compare["Value_with_missing::fill"][ "2011-01-01T23:00:00"] = 2.0 self.min_compare = tsutils.memory_optimize(self.min_compare) self.min_compare_cli = capture.capture(tsutils._printiso, self.min_compare) self.con_compare = self.ats.copy() self.con_compare["Value_with_missing::fill"][ "2011-01-01T01:00:00"] = 2.42 self.con_compare["Value_with_missing::fill"][ "2011-01-01T09:00:00"] = 3.0 self.con_compare["Value_with_missing::fill"][ "2011-01-01T10:00:00"] = 2.42 self.con_compare["Value_with_missing::fill"][ "2011-01-01T11:00:00"] = 2.42 self.con_compare["Value_with_missing::fill"][ "2011-01-01T12:00:00"] = 2.42 self.con_compare["Value_with_missing::fill"][ "2011-01-01T13:00:00"] = 9.0 self.con_compare["Value_with_missing::fill"][ "2011-01-01T16:00:00"] = 2.42 self.con_compare["Value_with_missing::fill"][ "2011-01-01T22:00:00"] = 2.42 self.con_compare["Value_with_missing::fill"][ "2011-01-01T23:00:00"] = 2.42 self.con_compare_cli = capture.capture(tsutils._printiso, self.con_compare)
def setUp(self): dindex = pd.date_range("2000-01-01T00:00:00", periods=2, freq="D") ts1 = pd.Series([4.5, 4.6], index=dindex) ts2 = pd.Series([20.0, 20.4], index=dindex) self.equation = pd.DataFrame( {"Value": ts1, "Value::equation": ts2}, index=dindex, dtype="float64" ) self.equation.index.name = "Datetime" self.equation = tsutils.memory_optimize(self.equation) self.equation_cli = capture.capture(tsutils._printiso, self.equation) dindex = pd.date_range("2000-01-01T00:00:00", periods=6, freq="D") ts1 = [4.50, 4.60, 4.70, 4.60, 4.50, 4.40] ts2 = [45.60, 90.50, 34.20, 23.10, 7.20, 4.30] ts3 = [ -4.00389911875, -5.61841686236, -5.87322876716, -4.86614401542, -2.6934178416, -4.60800663972, ] self.equation_multiple_cols_01 = pd.DataFrame( {"Value": ts1, "Value1": ts2, "_::equation": ts3}, index=dindex, dtype="float64", ) self.equation_multiple_cols_01.index.name = "Datetime" self.equation_multiple_cols_01 = tsutils.memory_optimize( self.equation_multiple_cols_01 ) self.equation_result = pd.DataFrame( {"_::equation": pd.np.array(ts2) * 10}, index=dindex, dtype="float64" ) self.equation_result = tsutils.memory_optimize(self.equation_result) self.equation_multiple_cols_01_cli = capture.capture( tsutils._printiso, self.equation_multiple_cols_01 ) ts3 = [50.1, 95.1, 38.9, 27.7, 11.7, 8.7] self.equation_multiple_cols_02 = pd.DataFrame( {"Value": ts1, "Value1": ts2, "_::equation": ts3}, index=dindex, dtype="float64", ) self.equation_multiple_cols_02.index.name = "Datetime" self.equation_multiple_cols_02 = tsutils.memory_optimize( self.equation_multiple_cols_02 ) self.equation_multiple_cols_02_cli = capture.capture( tsutils._printiso, self.equation_multiple_cols_02 ) ts3 = [0, 97.92, 41.66, 30.52, 14.46, 0] ts3[0] = pd.np.nan ts3[-1] = pd.np.nan self.equation_multiple_cols_03 = pd.DataFrame( {"Value": ts1, "Value1": ts2, "_::equation": ts3}, index=dindex, dtype="float64", ) self.equation_multiple_cols_03.index.name = "Datetime" self.equation_multiple_cols_03 = tsutils.memory_optimize( self.equation_multiple_cols_03 ) self.equation_multiple_cols_03_cli = capture.capture( tsutils._printiso, self.equation_multiple_cols_03 ) dindex = pd.date_range("2011-01-01T00:00:00", periods=48, freq="H") ts1 = [2] * 48 ts2 = [5.2] * 48 ts2[0] = pd.np.nan ts2[-1] = pd.np.nan self.equation_multiple_cols_04 = pd.DataFrame( {"Value": ts1, "Value::equation": ts2}, index=dindex ) self.equation_multiple_cols_04.index.name = "Datetime" self.equation_multiple_cols_04 = tsutils.memory_optimize( self.equation_multiple_cols_04 ) self.equation_multiple_cols_04_cli = capture.capture( tsutils._printiso, self.equation_multiple_cols_04 )
def setUp(self): self.read_direct = pd.read_csv('tests/data_sunspot_index.csv', index_col=0) self.read_direct = tsutils.memory_optimize(self.read_direct)
def setUp(self): dr = pd.date_range('2011-01-01T12:00:00', periods=3, freq='H') self.date_slice = pd.DataFrame([2, 2, 2], index=dr, columns=['Value']) self.date_slice = tsutils.memory_optimize(self.date_slice) self.date_slice_cli = capture.capture(tsutils._printiso, self.date_slice)
def setUp(self): self.ats = pd.read_csv(StringIO(test_sinwave), parse_dates=True, index_col=[0]) self.ats = tsutils.memory_optimize(self.ats)
def setUp(self): dindex = pd.date_range('2000-01-01T00:00:00', periods=2, freq='D') ts1 = pd.Series([4.5, 4.6], index=dindex) ts2 = pd.Series([20.0, 20.4], index=dindex) self.equation = pd.DataFrame({'Value': ts1, 'Value::equation': ts2}, index=dindex, dtype='float64') self.equation.index.name = 'Datetime' self.equation = tsutils.memory_optimize(self.equation) self.equation_cli = capture.capture(tsutils._printiso, self.equation) dindex = pd.date_range('2000-01-01T00:00:00', periods=6, freq='D') ts1 = [4.50, 4.60, 4.70, 4.60, 4.50, 4.40] ts2 = [45.60, 90.50, 34.20, 23.10, 7.20, 4.30] ts3 = [-4.00389911875, -5.61841686236, -5.87322876716, -4.86614401542, -2.6934178416, -4.60800663972] self.equation_multiple_cols_01 = pd.DataFrame({'Value': ts1, 'Value1': ts2, '_::equation': ts3}, index=dindex, dtype='float64') self.equation_multiple_cols_01.index.name = 'Datetime' self.equation_multiple_cols_01 = tsutils.memory_optimize(self.equation_multiple_cols_01) self.equation_result = pd.DataFrame({'_::equation': pd.np.array(ts2)*10}, index=dindex, dtype='float64') self.equation_result = tsutils.memory_optimize(self.equation_result) self.equation_multiple_cols_01_cli = capture.capture(tsutils._printiso, self.equation_multiple_cols_01) ts3 = [50.1, 95.1, 38.9, 27.7, 11.7, 8.7] self.equation_multiple_cols_02 = pd.DataFrame({'Value': ts1, 'Value1': ts2, '_::equation': ts3}, index=dindex, dtype='float64') self.equation_multiple_cols_02.index.name = 'Datetime' self.equation_multiple_cols_02 = tsutils.memory_optimize(self.equation_multiple_cols_02) self.equation_multiple_cols_02_cli = capture.capture(tsutils._printiso, self.equation_multiple_cols_02) ts3 = [0, 97.92, 41.66, 30.52, 14.46, 0] ts3[0] = pd.np.nan ts3[-1] = pd.np.nan self.equation_multiple_cols_03 = pd.DataFrame({'Value': ts1, 'Value1': ts2, '_::equation': ts3}, index=dindex, dtype='float64') self.equation_multiple_cols_03.index.name = 'Datetime' self.equation_multiple_cols_03 = tsutils.memory_optimize(self.equation_multiple_cols_03) self.equation_multiple_cols_03_cli = capture.capture(tsutils._printiso, self.equation_multiple_cols_03) dindex = pd.date_range('2011-01-01T00:00:00', periods=48, freq='H') ts1 = [2]*48 ts2 = [5.2]*48 ts2[0] = pd.np.nan ts2[-1] = pd.np.nan self.equation_multiple_cols_04 = pd.DataFrame({'Value': ts1, 'Value::equation': ts2}, index=dindex) self.equation_multiple_cols_04.index.name = 'Datetime' self.equation_multiple_cols_04 = tsutils.memory_optimize(self.equation_multiple_cols_04) self.equation_multiple_cols_04_cli = capture.capture(tsutils._printiso, self.equation_multiple_cols_04)
def setUp(self): self.read_direct = pd.read_csv( 'tests/data_sunspot_EST.csv', index_col=0, parse_dates=[0]).tz_localize('UTC').tz_convert('EST') self.read_direct = tsutils.memory_optimize(self.read_direct)
def setUp(self): self.read_direct = pd.read_csv( "tests/data_gainesville_daily_precip_index.csv", index_col=0, header=0 ) self.read_direct = tsutils.memory_optimize(self.read_direct)