def test_append_time_domain(test_pd_df, test_df_mixed, other, time, inplace):
df_year = IamDataFrame(test_pd_df[IAMC_IDX + [2005]],
meta=test_df_mixed.meta)
df_time = IamDataFrame(test_pd_df[IAMC_IDX + [2010]].rename(
{2010: time}, axis="columns"))
# append `df_time` to `df_year`
if other == "time":
if inplace:
obs = df_year.copy()
obs.append(df_time, inplace=True)
else:
obs = df_year.append(df_time)
# assert that original object was not modified
assert df_year.year == [2005]
# append `df_year` to `df_time`
else:
if inplace:
obs = df_time.copy()
obs.append(df_year, inplace=True)
else:
obs = df_time.append(df_year)
# assert that original object was not modified
assert df_time.time == pd.Index([datetime(2010, 7, 21)])
assert_iamframe_equal(obs, test_df_mixed)
def test_interpolate_extra_cols():
# check that interpolation with non-matching extra_cols has no effect
# (#351)
EXTRA_COL_DF = pd.DataFrame(
[
["foo", 2005, 1],
["foo", 2010, 2],
["bar", 2005, 2],
["bar", 2010, 3],
],
columns=["extra_col", "year", "value"],
)
df = IamDataFrame(
EXTRA_COL_DF,
model="model_a",
scenario="scen_a",
region="World",
variable="Primary Energy",
unit="EJ/yr",
)
# create a copy, interpolate
df2 = df.copy()
df2.interpolate(2007)
# interpolate should work as if extra_cols is in the _data index
assert_iamframe_equal(df, df2.filter(year=2007, keep=False))
obs = df2.filter(year=2007)["value"]
exp = pd.Series([2.4, 1.4], name="value")
pd.testing.assert_series_equal(obs, exp)
def test_interpolate_extra_cols():
# check that interpolation with non-matching extra_cols has no effect
# (#351)
EXTRA_COL_DF = pd.DataFrame(
[
['foo', 2005, 1],
['foo', 2010, 2],
['bar', 2005, 2],
['bar', 2010, 3],
],
columns=['extra_col', 'year', 'value'],
)
df = IamDataFrame(EXTRA_COL_DF,
model='model_a',
scenario='scen_a',
region='World',
variable='Primary Energy',
unit='EJ/yr')
# create a copy, interpolate
df2 = df.copy()
df2.interpolate(2007)
# interpolate should work as if extra_cols is in the _data index
assert_iamframe_equal(df, df2.filter(year=2007, keep=False))
obs = df2.filter(year=2007)['value']
exp = pd.Series([2.4, 1.4], name='value')
pd.testing.assert_series_equal(obs, exp)
def test_interpolate_extra_cols():
# check hat interpolation with non-matching extra_cols has no effect (#351)
EXTRA_COL_DF = pd.DataFrame([
['foo', 2005, 1],
['bar', 2010, 3],
],
columns=['extra_col', 'year', 'value'],
)
df = IamDataFrame(EXTRA_COL_DF, model='model_a', scenario='scen_a',
region='World', variable='Primary Energy', unit='EJ/yr')
# create a copy, interpolate
df2 = df.copy()
df2.interpolate(2007)
# assert that interpolation didn't change any data
assert_iamframe_equal(df, df2)
def test_limit_of_similar_xs(self, test_db):
# Check that the function returns the correct behaviour in the limit of similar
# lead values. We construct a db for exactly identical CH4 and for near-
# identical CH4 (one point is far away and in the wrong direction).
range_db = pd.DataFrame(
[[_ma, str(val), "World", _ech4, _gtc, val] for val in range(10)],
columns=_msrvu + [2010],
)
range_db = IamDataFrame(range_db)
range_db = self._adjust_time_style_to_match(range_db, test_db)
same_db = range_db.copy()
same_db["variable"] = _eco2
same_db["value"] = 1
same_db.append(range_db, inplace=True)
nearly_same_db = same_db.copy()
# We change the value of one point on the nearly_same and remove it on the same
nearly_same_db.data["value"].iloc[1] = 0
same_db.filter(scenario="0", keep=False, inplace=True)
same_db = IamDataFrame(same_db.data)
nearly_same_db = IamDataFrame(nearly_same_db.data)
# Derive crunchers and compare results from the two values
same_cruncher = self.tclass(same_db)
nearly_same_cruncher = self.tclass(nearly_same_db)
if test_db.time_col == "year":
single_date_df = test_db.filter(
year=int(same_db[same_db.time_col][0]))
else:
single_date_df = test_db.filter(
time=(same_db[same_db.time_col][0]))
# We choose the case model a, which has only values over 1.
single_date_df.filter(model=_mb,
scenario=_sa,
keep=False,
inplace=True)
same_res = same_cruncher.derive_relationship(_ech4,
[_eco2])(single_date_df)
nearly_same_res = nearly_same_cruncher.derive_relationship(
_ech4, [_eco2])(single_date_df)
assert np.allclose(same_res["value"],
nearly_same_res["value"],
rtol=5e-4)