def test_empty_xlsx(self):
to_cero = ToCERO(conf=(TestToCERO._dd + r"test_empty_xlsx.yaml"))
cero = to_cero.create_cero()
with self.assertRaises(CERO.EmptyCERO):
self.assertTrue(CERO.is_cero(cero, empty_ok=False))
self.assertTrue(CERO.is_cero(cero))
def test_fillna(self):
df = pd.DataFrame.from_dict(
{
"A": [1, 2, 3],
"B": [3, 4, 5],
"C": [6, 7, 8]
}, orient="index")
df.columns = pd.DatetimeIndex(
pd.to_datetime([2017, 2018, 2019], format="%Y"))
df.sort_index(inplace=True)
df.iloc[1, 1] = pd.np.nan
df = df.astype(pd.np.float32)
self.assertTrue(CERO.is_cero(df))
libfuncs.fillna(df, value=0.0)
self.assertTrue(df.iloc[1, 1] == 0.0)
df.iloc[1, 1] = pd.np.nan
libfuncs.fillna(df, method="bfill")
self.assertTrue(df.iloc[1, 1] == 5.0)
df.iloc[1, 1] = pd.np.nan
libfuncs.fillna(df)
self.assertTrue(df.iloc[1, 1] == 3.0)
self.assertTrue(CERO.is_cero(df))
def test_rename_2(self):
to_cero = ToCERO(conf=(TestToCERO._dd + r"test_rename_2.yaml"))
cero = to_cero.create_cero()
CERO.is_cero(cero)
test_idx = ["PROFESSIONALS", ("1", "MANAGERS")
] # Rename operation always moves series to the end
self.assertTrue(
all([x == y for (x, y) in zip(test_idx, cero.index.tolist())]))
def test_apply_func(self):
df = pd.DataFrame.from_dict(
{
"A": [1, 2, 3],
"B": [3, 4, 5],
"C": [6, 7, 8]
},
orient="index",
dtype=pd.np.float32)
df.columns = pd.DatetimeIndex(
pd.to_datetime([2017, 2018, 2019], format="%Y"))
df.sort_index(inplace=True)
self.assertTrue(CERO.is_cero(df))
libfuncs.apply_func(df, numpy_func="square")
test_df = pd.DataFrame.from_dict(
{
"A": [1, 4, 9],
"B": [9, 16, 25],
"C": [36, 49, 64]
},
orient="index",
dtype=pd.np.float32)
test_df.columns = pd.DatetimeIndex(
pd.to_datetime([2017, 2018, 2019], format="%Y"))
test_df.sort_index(inplace=True)
self.assertTrue(df.equals(test_df))
def exec_procedures(self, cero):
""" Execute all the procedures of the FromCERO object
.
:param pandas.DataFrame cero: A CERO to serve as input for the procedures. The argument is not mutated/modified.
"""
CERO.is_cero(cero, raise_exception=True, empty_ok=False)
CERO.rename_index_values(cero, self.get("map", {}))
self.output_procedures = OrderedDict()
for procedure in self["procedures"]:
try:
ret = procedure.exec_ops(cero)
# if ret is not None, should be dict with key: procedure["name"], value: resultant CERO
except Exception as e:
raise e.__class__(e.__str__() + " Error in procedure '%s'." % (procedure["name"]))
if ret is None:
ret = {}
self.output_procedures.update(ret)
else:
if not self["procedures"]: # If empty list
self.output_procedures["default_output"] = cero
if any([not procedure.get("file") for procedure in self["procedures"]]):
msg = "It has been detected that not all procedures direct output to file. Therefore some output will go to \'%s\'." % self["file"]
print(msg)
FromCERO._logger.info(msg)
if self.output_procedures != {}:
file_ext = os.path.splitext(self["file"])[1][1:]
if file_ext in FromCERO.sup_output_types:
out_df = CERO.combine_ceros(list(self.output_procedures.values()))
FromCERO.dataframe_out(out_df, self["file"], output_type=file_ext)
elif file_ext in FromCERO._Procedure.sup_output_types:
raise ValueError("This data type is not supported for general export, because it probably has a more than 2 dimensions - export using 'procedures' instead.")
else:
raise ValueError("Unsupported data type detected for general export.")
def test_complex_xlsx(self):
to_cero = ToCERO(conf=(TestToCERO._dd +
r'test_complex_xlsx_import.yaml'))
cero = to_cero.create_cero()
df = DataTools.get_test_data(TestToCERO._dd +
"test_complex_xlsx_result.pickle")
self.assertTrue(CERO.is_cero(cero))
self.assertTrue(cero.equals(df))
def run(self) -> None:
"""
Execute a scenario run.
"""
self.cero = CERO.create_empty()
ceros = [in_conf.create_cero() for in_conf in self["input_conf"]]
if ceros:
self.cero = CERO.combine_ceros(ceros)
print("Successfully loaded scenario inputs as CERO.")
FromCERO.dataframe_out(self.cero,
(self.get_name() + "_%03d_step_%02d.xlsx" %
(self["run_no"], 0)), "xlsx")
for idx, model in enumerate(self["models"]):
m_cero = model.run(self.cero)
print(
"Completed run of model (%s) at %s." %
(model["name"], dt.datetime.now().strftime('%Y-%m-%d %H:%M')))
# If ouput_conf is not defined for a model, then None is returned...
if m_cero is None:
continue
if not CERO.is_cero(m_cero):
raise TypeError(
"Object returned from model run is *not* of CERO format.")
if model.get("export_mod_xlsx", self.get("export_mod_xlsx", True)):
# By default, export model outputs automatically to xlsx files
model_out_file = (self.get_name() + "_%03d_%s.xlsx" %
(self["run_no"], model["name"]))
print("Exporting output of %s to %s." %
(model["name"], model_out_file))
m_cero.to_excel(model_out_file)
self.cero = CERO.combine_ceros([self.cero, m_cero])
if self.get("export_int_xlsx", True):
# If true (default), export the intermediate steps to xlsx files
isfn = (self.get_name() + "_%03d_step_%02d.xlsx" %
(self["run_no"], idx + 1))
print("Exporting updated CERO to %s." % (isfn))
self.cero.to_excel(isfn)
for out_conf in self["output_conf"]:
out_conf.exec_procedures(self.cero)
else:
print("Completed generation of scenario outputs.")
def init_df():
df = pd.DataFrame.from_dict(
{
"A": [1, 2, 3],
"B": [3, 4, 5],
"C": [6, 7, 8]
},
orient="index")
df.columns = pd.DatetimeIndex(
pd.to_datetime([2017, 2018, 2019], format="%Y"))
df.sort_index(inplace=True)
df = df.astype(pd.np.float32)
self.assertTrue(CERO.is_cero(df))
return df
def test_export_to_csv(self):
cero = pd.DataFrame.from_dict({"A": [1], "B": [2], "C": [3], "D": [4], "E": [5], "F": [6], }, orient='index',
dtype=pd.np.float32)
cero.columns = pd.DatetimeIndex(data=pd.to_datetime([2018], format="%Y"))
cero.sort_index(inplace=True)
self.assertTrue(CERO.is_cero(cero))
fc = FromCERO(cfg.d_td + "test_procedure_export_csv.yaml")
fc.exec_procedures(cero)
df1 = pd.read_csv("csv_export.csv", index_col=0)
test_list = [1, 2, 3]
df1_vals = [x[0] for x in df1.values.tolist()]
self.assertTrue(all([np.isclose(x, y) for (x, y) in zip(test_list, df1_vals)]))
test_list = ["A", "B", "C"]
self.assertTrue(all([x == y for (x, y) in zip(test_list, df1.index.tolist())]))
os.remove("csv_export.csv")
def test_sets_and_mapping(self):
cero = pd.DataFrame.from_dict(
{
"A": [1],
"B": [2],
"C": [3],
"D": [4],
"E": [5],
"F": [6],
},
orient='index',
dtype=pd.np.float32)
cero.sort_index(inplace=True)
cero.columns = pd.DatetimeIndex(
data=pd.to_datetime([2018], format="%Y"))
self.assertTrue(CERO.is_cero(cero))
fc = FromCERO(TestFromCERO._dd + "test_fromcero_mapping.yaml")
fc.exec_procedures(cero)
df1 = pd.read_csv("test_fromcero_mapping1.csv", index_col=0)
test_list = [1, 2, 3]
df1_vals = [x[0] for x in df1.values.tolist()]
self.assertTrue(
all([np.isclose(x, y) for (x, y) in zip(test_list, df1_vals)]))
test_list = ["A", "B", "C"]
self.assertTrue(
all([x == y for (x, y) in zip(test_list, df1.index.tolist())]))
df2 = pd.read_csv("test_fromcero_mapping2.csv", index_col=0)
test_list = [4, 5, 6]
df2_vals = [x[0] for x in df2.values.tolist()]
self.assertTrue(all([x == y for (x, y) in zip(test_list, df2_vals)]))
test_list = ["G", "H", "I"]
self.assertTrue(
all([x == y for (x, y) in zip(test_list, df2.index.tolist())]))
os.remove("test_fromcero_mapping1.csv")
os.remove("test_fromcero_mapping2.csv")
def test_sets_and_mapping2(self):
cero = pd.DataFrame.from_dict(
{
("A", "1"): [1],
("A", "2"): [2],
("A", "3"): [3],
("B", "1"): [4],
("B", "2"): [5],
("B", "3"): [6],
("C", "1"): [7],
("C", "2"): [8],
("C", "3"): [9],
},
orient='index',
dtype=pd.np.float32)
cero.sort_index(inplace=True)
cero.columns = pd.DatetimeIndex(
data=pd.to_datetime([2018], format="%Y"))
self.assertTrue(CERO.is_cero(cero))
fc = FromCERO(TestFromCERO._dd + "test_fromcero_mapping2.yaml")
fc.exec_procedures(cero)
tc = ToCERO({
"files": [{
"file": "test_fromcero_complexmapping1.xlsx",
"sheet": "CERO",
"index_col": [0, 1]
}]
})
df1 = tc.create_cero()
test_list = list(range(1, 10))
df1_vals = [x[0] for x in df1.values.tolist()]
self.assertTrue(
all([np.isclose(x, y) for (x, y) in zip(test_list, df1_vals)]))
test_list = [("G", "1"), ("G", "2"), ("G", "3"), ("H", "1"),
("H", "2"), ("H", "3"), ("I", "1"), ("I", "2"),
("I", "3")]
self.assertTrue(
all([x == y for (x, y) in zip(test_list, df1.index.tolist())]))
tc = ToCERO({
"files": [{
"file": "test_fromcero_complexmapping2.xlsx",
"sheet": "CERO",
"index_col": [0, 1]
}]
})
df1 = tc.create_cero()
test_list = list(range(1, 10))
df1_vals = [x[0] for x in df1.values.tolist()]
self.assertTrue(
all([np.isclose(x, y) for (x, y) in zip(test_list, df1_vals)]))
test_list = [("A", "G"), ("A", "H"), ("A", "I"), ("B", "G"),
("B", "H"), ("B", "I"), ("C", "G"), ("C", "H"),
("C", "I")]
self.assertTrue(
all([x == y for (x, y) in zip(test_list, df1.index.tolist())]))
os.remove("test_fromcero_complexmapping1.xlsx")
os.remove("test_fromcero_complexmapping2.xlsx")
def test_multiindex_xlsx(self):
to_cero = ToCERO(conf=(TestToCERO._dd + r'test_multiindex_xlsx.yaml'))
cero = to_cero.create_cero()
self.assertTrue(CERO.is_cero(cero))
def wrapper(df: pd.DataFrame,
*args,
locs: "List[Union[tuple, str]]" = None,
ilocs: "List[int]" = None,
start_year: "Union[pd.datetime, int]" = None,
end_year: "Union[pd.datetime, int]" = None,
**kwargs):
"""
:param df: An CERO, which may or may not be a strict superset of data to perform the operation on.
:param args: Passed to the encapsulated function as positional arguments, immediately after the restricted \
``df``.
:param locs: ``locs``, if provided, must be a list of identifiers that correspond to values of ``df.index``. \
It is ``df``, reduced to these specific indices, that a wrapped function will receive as an argument. An \
error is raised if both ``locs`` and ``ilocs`` is specified.
:param ilocs: Identical in nature to ``locs``, though instead a list of integers (zero-indexed) is \
provided (corresponding to the row number of ``df``). An \
error is raised if both ``locs`` and ``ilocs`` is specified.
:param start_year: Note that ``df`` is a CERO, and CEROs have a ``pandas.DatetimeIndex`` on columns. \
``start_year`` restricts the CERO to years after and including ``start_year``.
:param end_year: Note that ``df`` is a CERO, and CEROs have a ``pandas.DatetimeIndex`` on columns. \
``end_year`` restricts the CERO to years up to and including ``end_year``.
:param kwargs: Keyword arguments to be passed to the encapsulated function.
:return: The return value of the encapsulated function.
"""
try:
assert(isinstance(df, pd.DataFrame))
except AssertionError:
raise TypeError("First function argument must be of pandas.DataFrame type.")
# Convert integer to datetime type
if isinstance(start_year, int):
start_year = pd.datetime(start_year, 1, 1)
if isinstance(end_year, int):
end_year = pd.datetime(end_year, 1, 1)
# Get index locations
if start_year is not None:
start_year = df.columns.get_loc(start_year)
if end_year is not None:
end_year= df.columns.get_loc(end_year)
if (locs is not None) and (ilocs is not None):
raise TypeError("Only one of 'locs' or 'ilocs' can be provided (not both).")
if locs is not None:
ilocs = [df.index.get_loc(loc) for loc in locs]
if ilocs is None:
ilocs = pd.IndexSlice[0:]
df_cp = df.iloc[ilocs, start_year:end_year].copy(deep=False) # df_cp is always different object to df
ret = func(df_cp, *args, **kwargs)
if ret is None:
return ret
elif issubclass(type(ret), pd.Series):
# If series, convert to dataframe
ret = pd.DataFrame(data=[ret])
CERO.is_cero(ret) # Performs checks to ensure ret is a valid CERO
return ret
def test_create_empty(self):
empty_cero = CERO.create_empty()
self.assertTrue(CERO.is_cero(empty_cero))
def test_is_cero(self):
"""Tests the validation method by feeding it deliberately False data."""
df = None
with self.assertRaises(CERO.InvalidCERO, msg=CERO._msg_inv_type):
CERO.is_cero(df)
df = pd.DataFrame.from_dict({
"A": [1, 2, 3],
"B": [4, 5, 6]
},
orient="index",
dtype=int)
df.index = pd.MultiIndex.from_tuples([("A", 1), ("A", 2)])
# ^^ ``df`` is not even close to being a CERO...
with self.assertRaises(CERO.InvalidCERO, msg=CERO._msg_bad_idx):
CERO.is_cero(df)
df.index = pd.Index(["A", "A"])
with self.assertRaises(CERO.InvalidCERO, msg=CERO._msg_bad_col):
CERO.is_cero(df)
df.columns = pd.DatetimeIndex(
pd.to_datetime([2017, 2017, 2019], format="%Y"))
with self.assertRaises(CERO.InvalidCERO, msg=CERO._msg_idx_nunique):
CERO.is_cero(df)
df.index = pd.Index(["A", "B"])
with self.assertRaises(CERO.InvalidCERO, msg=CERO._msg_col_nunique):
CERO.is_cero(df)
df.columns = pd.DatetimeIndex(
pd.to_datetime([2017, 2018, 2019], format="%Y"))
with self.assertRaises(CERO.InvalidCERO, msg=CERO._msg_val_type):
CERO.is_cero(df)
df = df.astype(pd.np.float32, copy=False)
self.assertTrue(CERO.is_cero(df))
