def objective_function(self, subset, **kwargs):
r"""Evaluates objective function on specified subset. Returns distance
between self.goal and the series in subset
Notes
-----
self.objective must match the name of a function in
objective_functions.py
Parameters
----------
subset : list
subset of data dictionaries from data
Returns
-------
return : float
a measure of distance
"""
subset = dm.listify(subset)
if hasattr(objective_functions, self.objective):
return eval("objective_functions." + self.
objective)(self, subset, **kwargs)
elif hasattr(early_detection, self.objective):
return eval("early_detection." + self.
objective)(self, subset, **kwargs)
else:
print(self.objective + " is not an objective function")
sys.exit()
def graph_data_sources(self, sources):
r"""Plots the data sources from sources and prints a legend
Parameters
----------
sources : list
subset of data dictionaries from data
"""
sources = dm.listify(sources)
for source in sources:
source = dm.normalize(source)
time = source['data']['times']
values = source['data']['values']
name = source['metadata']['name']
subname = source['metadata']['subname']
plt.plot(time, values, label = name + " | " + subname)
plt.legend(loc=2, prop={'size': 10})
plt.show()
def test_OOS(self, subset, **kwargs):
r"""Evaluates objective function on specified subset. Returns distance
between self.goal and the series in subset
Notes
-----
Parameters
----------
subset : list
subset of data dictionaries from data
Returns
-------
return : float
a measure of distance
"""
subset = dm.listify(subset)
return objective_functions.R_squared_OOS(self, subset, **kwargs)