def test_scalar(self):
"""
Test the various ways we handle scalars.
"""
plot_index = pd.date_range(start="2000-01-01", freq="D", periods=100)
plot_series = process_series(5, plot_index)
tm.assert_almost_equal(plot_series.index, plot_index) # index align
assert np.all(plot_series == 5)
# explicitly pass in the series index. Should have a plot_series with only iloc[10:20]
# equal to the scalar 5.
plot_series = process_series(5,
plot_index,
series_index=plot_index[10:20])
tm.assert_almost_equal(plot_series.index, plot_index) # index align
assert np.all(plot_series[10:20] == 5)
assert plot_series.isnull().sum() == 90
# no plot_index. This still works because we're passing in series_index
plot_series = process_series(5, None, series_index=plot_index[10:20])
correct = pd.Series(5, index=plot_index[10:20])
tm.assert_almost_equal(correct, plot_series)
# without any index, a scalar will error. Cannot plot a scalar on an
# empty plot without passing in an index
try:
plot_series = process_series(5, None)
except:
pass
else:
assert False, "scalar should fail without plot_index or series_index"
def test_scalar(self):
"""
Test the various ways we handle scalars.
"""
plot_index = pd.date_range(start="2000-01-01", freq="D", periods=100)
plot_series = process_series(5, plot_index)
tm.assert_almost_equal(plot_series.index, plot_index) # index align
assert np.all(plot_series == 5)
# explicitly pass in the series index. Should have a plot_series with only iloc[10:20]
# equal to the scalar 5.
plot_series = process_series(5, plot_index, series_index=plot_index[10:20])
tm.assert_almost_equal(plot_series.index, plot_index) # index align
assert np.all(plot_series[10:20] == 5)
assert plot_series.isnull().sum() == 90
# no plot_index. This still works because we're passing in series_index
plot_series = process_series(5, None, series_index=plot_index[10:20])
correct = pd.Series(5, index=plot_index[10:20])
tm.assert_almost_equal(correct, plot_series)
# without any index, a scalar will error. Cannot plot a scalar on an
# empty plot without passing in an index
try:
plot_series = process_series(5, None)
except:
pass
else:
assert False, "scalar should fail without plot_index or series_index"
def test_different_freqs(self):
"""
Tests indexes of differeing frequencies. This is more of repeat
test of test_partial_plot but with many holes instead of one half missing
value.
"""
plot_index = pd.date_range(start="2000-01-01", freq="D", periods=100)
series = pd.Series(range(100), index=plot_index)
grouped_series = series.resample('MS', 'max')
plot_series = process_series(grouped_series, plot_index)
tm.assert_almost_equal(plot_series.index, plot_index) # index align
# method=None, dropna should give back same series
tm.assert_almost_equal(plot_series.dropna(), grouped_series)
plot_series = process_series(grouped_series, plot_index, method='ffill')
tm.assert_almost_equal(plot_series.index, plot_index) # index align
assert plot_series.isnull().sum() == 0
month_ind = plot_series.index.month - 1
# assert that each value corresponds to its month in grouped_series
assert np.all(grouped_series[month_ind] == plot_series)
def test_unaligned_indexes(self):
"""
Test when series.index and plot_index have no common datetimes
"""
plot_index = pd.date_range(start="2000", freq="D", periods=100)
series = pd.Series(range(100), index=plot_index)
# move days to 11 PM the night before
shift_series = series.tshift(-1, '1h')
plot_series = process_series(shift_series, plot_index)
# without method, data doesn't align and we nothing but nans
tm.assert_almost_equal(plot_series.index, plot_index) # index aligh properly
assert np.all(plot_series.isnull()) # no data
# method = 'ffill'
plot_series = process_series(shift_series, plot_index, method='ffill')
# without method, data doesn't align and we nothing but nans
tm.assert_almost_equal(plot_series.index, plot_index) # index align
# since we're forward filling a series we tshifted into past
# plot_series should just equal the original series
tm.assert_almost_equal(plot_series, series)
def test_unaligned_indexes(self):
"""
Test when series.index and plot_index have no common datetimes
"""
plot_index = pd.date_range(start="2000", freq="D", periods=100)
series = pd.Series(range(100), index=plot_index)
# move days to 11 PM the night before
shift_series = series.tshift(-1, '1h')
plot_series = process_series(shift_series, plot_index)
# without method, data doesn't align and we nothing but nans
tm.assert_almost_equal(plot_series.index,
plot_index) # index aligh properly
assert np.all(plot_series.isnull()) # no data
# method = 'ffill'
plot_series = process_series(shift_series, plot_index, method='ffill')
# without method, data doesn't align and we nothing but nans
tm.assert_almost_equal(plot_series.index, plot_index) # index align
# since we're forward filling a series we tshifted into past
# plot_series should just equal the original series
tm.assert_almost_equal(plot_series, series)
def test_different_freqs(self):
"""
Tests indexes of differeing frequencies. This is more of repeat
test of test_partial_plot but with many holes instead of one half missing
value.
"""
plot_index = pd.date_range(start="2000-01-01", freq="D", periods=100)
series = pd.Series(range(100), index=plot_index)
grouped_series = series.resample('MS', 'max')
plot_series = process_series(grouped_series, plot_index)
tm.assert_almost_equal(plot_series.index, plot_index) # index align
# method=None, dropna should give back same series
tm.assert_almost_equal(plot_series.dropna(), grouped_series)
plot_series = process_series(grouped_series,
plot_index,
method='ffill')
tm.assert_almost_equal(plot_series.index, plot_index) # index align
assert plot_series.isnull().sum() == 0
month_ind = plot_series.index.month - 1
# assert that each value corresponds to its month in grouped_series
assert np.all(grouped_series[month_ind] == plot_series)
def test_iterable(self):
"""
Non pd.Series iterables require an equal length series_index or
plot_index.
"""
try:
plot_series = process_series(range(10), None)
except:
pass
else:
assert False, "iterable should fail without plot_index or series_index"
plot_index = pd.date_range(start="2000-01-01", freq="D", periods=100)
try:
plot_series = process_series(range(10), plot_index)
except:
pass
else:
assert False, "iterable requires an index of same length"
# equal length, good times
plot_series = process_series(range(10), plot_index[:10])
correct = pd.Series(range(10), index=plot_index[:10])
tm.assert_almost_equal(correct, plot_series)
def test_partial_plot(self):
"""
Test plotting series that is a subset of plot_index.
Should align and fill with nans
"""
plot_index = pd.date_range(start="2000", freq="D", periods=100)
series = pd.Series(range(100), index=plot_index)
series = series[:50] # only first 50
plot_series = process_series(series, plot_index)
# have same index
tm.assert_almost_equal(plot_series.index, plot_index)
assert plot_series.count() == 50
assert np.all(plot_series[50:].isnull()) # method=None so fill with nan
assert np.all(plot_series[:50] == series[:50])
def test_iterable(self):
"""
Non pd.Series iterables require an equal length series_index or
plot_index.
"""
try:
plot_series = process_series(range(10), None)
except:
pass
else:
assert False, "iterable should fail without plot_index or series_index"
plot_index = pd.date_range(start="2000-01-01", freq="D", periods=100)
try:
plot_series = process_series(range(10), plot_index)
except:
pass
else:
assert False, "iterable requires an index of same length"
# equal length, good times
plot_series = process_series(range(10), plot_index[:10])
correct = pd.Series(range(10), index=plot_index[:10])
tm.assert_almost_equal(correct, plot_series)
def test_partial_plot(self):
"""
Test plotting series that is a subset of plot_index.
Should align and fill with nans
"""
plot_index = pd.date_range(start="2000", freq="D", periods=100)
series = pd.Series(range(100), index=plot_index)
series = series[:50] # only first 50
plot_series = process_series(series, plot_index)
# have same index
tm.assert_almost_equal(plot_series.index, plot_index)
assert plot_series.count() == 50
assert np.all(
plot_series[50:].isnull()) # method=None so fill with nan
assert np.all(plot_series[:50] == series[:50])
def test_already_aligned(self):
plot_index = pd.date_range(start="2000", freq="D", periods=100)
series = pd.Series(range(100), index=plot_index)
plot_series = process_series(series, plot_index)
tm.assert_almost_equal(series, plot_series)
tm.assert_almost_equal(plot_series.index, plot_index)
def test_already_aligned(self):
plot_index = pd.date_range(start="2000", freq="D", periods=100)
series = pd.Series(range(100), index=plot_index)
plot_series = process_series(series, plot_index)
tm.assert_almost_equal(series, plot_series)
tm.assert_almost_equal(plot_series.index, plot_index)