DEPR pandas freqstr

Bump version

enlopy/__init__.py

@@ -4,4 +4,4 @@
 from .utils import *
 __author__ = 'Konstantinos Kavvadias'
 __license__ = 'BSD-3-Clause'
-__version__ = "0.1.dev12"
+__version__ = "0.1"

enlopy/analysis.py

@@ -1,4 +1,3 @@
-from __future__ import division
 import numpy as np
 import pandas as pd
 
@@ -120,7 +119,7 @@ def get_load_archetypes(Load, k=2, x='hour', y='dayofyear', plot_diagnostics=Fal
     return clusters_center_dewhitened
 
 
-def get_load_stats(Load, per='a') -> pd.DataFrame:
+def get_load_stats(Load, per='Y') -> pd.DataFrame:
     """Find load profile characteristics. Among other it estimates: peak, load factor, base load factor, operating hours,
 
     Arguments:

enlopy/generate.py

@@ -2,7 +2,6 @@
 """
 Methods that generate or adjusted energy related timeseries based on given assumptions/input
 """
-from __future__ import absolute_import, division, print_function
 
 import numpy as np
 import pandas as pd
@@ -395,11 +394,11 @@ def gen_demand_response(Load, percent_peak_hrs_month=0.03, percent_shifted=0.05,
     def hours_per_month(demand):
         """Assign to each row hours per month"""
         dic_hours_per_month = demand.groupby(demand.index.month).count().to_dict()
-        return demand.resample('m').transform(lambda x: list(map(dic_hours_per_month.get, x.index.month)))
+        return demand.resample('M').transform(lambda x: list(map(dic_hours_per_month.get, x.index.month)))
 
     # Monthly demand rank
     # TODO: parametrize: we can check peaks on a weekly or daily basis
-    demand_m_rank = demand.resample('m').transform(lambda x: x.rank(method='min', ascending=False))
+    demand_m_rank = demand.resample('M').transform(lambda x: x.rank(method='min', ascending=False))
 
     # find which hours are going to be shifted
     bool_shift_from = demand_m_rank <= np.round(hours_per_month(demand) * percent_peak_hrs_month)

enlopy/plot.py

@@ -1,4 +1,3 @@
-from __future__ import division
 import matplotlib.pyplot as plt
 import matplotlib.cm as cm
 import pandas as pd

enlopy/utils.py

@@ -34,11 +34,11 @@ def make_timeseries(x=None, year=None, length=None, startdate=None, freq=None):
         if freq is None:
             # Shortcuts: Commonly used frequencies are automatically assigned
             if len(x) == 8760 or len(x) == 8784:
-                freq = 'H'
+                freq = 'h'
             elif len(x) == 35040:
                 freq = '15min'
             elif len(x) == 12:
-                freq = 'm'
+                freq = 'M'
             else:
                 raise ValueError('Input vector length must be 12, 8760 or 35040. Otherwise freq has to be defined')
 

tests/test_analysis.py

@@ -25,7 +25,7 @@ def test_reshape_multiannual_dataframe():
     assert np.isclose(b.sum().sum(), a.sum())
 
 def test_reshape_monthly_multiannual():
-    ind = pd.DatetimeIndex(pd.date_range(start='Jan-1990', end='Jan-2017', freq='m'))
+    ind = pd.DatetimeIndex(pd.date_range(start='Jan-1990', end='Jan-2017', freq='M'))
     a = pd.DataFrame(np.random.rand(324), index=ind)
     b = reshape_timeseries(a, x='month', y='year', aggfunc='sum')
     assert b.shape == (27, 12)  # 27 years, 12 months

tests/test_generate.py

@@ -95,7 +95,7 @@ def test_disag_hourly_to_minutes(self):
         y = make_timeseries(y, freq='h')
 
         disag_profile = np.random.rand(60)
-        y_disag = disag_upsample(y, disag_profile, to_offset='t')
+        y_disag = disag_upsample(y, disag_profile, to_offset='min')
         assert np.isclose(np.sum(y_disag), np.sum(y) ) # <= 0.001 #FIXME: np test equality
         assert len(y_disag) == 8760*60