def test_assert_similar_hists():
""" Test assert on similarity of list of histograms
Check similarity of: type, n-dim, sub-hists, specific type attributes
"""
# dummy dataset with mixed types
# convert timestamp (col D) to nanosec since 1970-1-1
df = pd.util.testing.makeMixedDataFrame()
df['date'] = df['D'].apply(to_ns)
# building 1d-, 2d-, and 3d-histogram (iteratively)
hist0 = hg.Bin(5, 0, 5, unit('A'))
hist1 = hg.Categorize(unit('C'))
hist2 = hg.Bin(5, 0, 5, unit('A'), value=hist1)
hist3 = hg.Categorize(unit('C'), value=hist0)
hist4 = hg.SparselyBin(origin=pd.Timestamp('2009-01-01').value,
binWidth=pd.Timedelta(days=1).value,
quantity=unit('date'),
value=hist2)
hist5 = hg.SparselyBin(origin=pd.Timestamp('2009-01-01').value,
binWidth=pd.Timedelta(days=1).value,
quantity=unit('date'),
value=hist3)
# fill them
for hist in [hist0, hist1, hist2, hist3, hist4, hist5]:
hist.fill.numpy(df)
hc0 = HistogramContainer(hist0)
hc1 = HistogramContainer(hist1)
hc2 = HistogramContainer(hist2)
hc3 = HistogramContainer(hist3)
hc4 = HistogramContainer(hist4)
hc5 = HistogramContainer(hist5)
for hc in [hc0, hc1, hc2, hc3, hc4, hc5]:
assert check_similar_hists([hc, hc])
args01 = ['']
args23 = ['']
args45 = ['']
try:
assert_similar_hists([hc0, hc1])
except AssertionError as e:
args01 = e.args
try:
assert_similar_hists([hc2, hc3])
except AssertionError as e:
args23 = e.args
try:
assert_similar_hists([hc4, hc5])
except AssertionError as e:
args45 = e.args
assert args01[0] == 'Input histograms are not all similar.'
assert args23[0] == 'Input histograms are not all similar.'
assert args45[0] == 'Input histograms are not all similar.'
def test_get_consistent_numpy_entries():
""" Test extraction of number of entries
When first making bin_edges of input histograms consistent to each other.
"""
df1 = pd.DataFrame({
'A': [0, 1, 2, 3, 4, 3, 2, 1, 1, 1],
'C': ['f1', 'f3', 'f4', 'f3', 'f4', 'f2', 'f2', 'f1', 'f3', 'f4']
})
df2 = pd.DataFrame({
'A': [2, 3, 4, 5, 7, 4, 6, 5, 7, 8],
'C': ['f7', 'f3', 'f5', 'f8', 'f9', 'f2', 'f3', 'f6', 'f7', 'f7']
})
# building 1d-, 2d-, and 3d-histogram (iteratively)
hist0 = HistogramContainer(hg.Categorize(unit('C')))
hist1 = HistogramContainer(hg.Categorize(unit('C')))
hist2 = HistogramContainer(
hg.SparselyBin(origin=0.0, binWidth=1.0, quantity=unit('A')))
hist3 = HistogramContainer(
hg.SparselyBin(origin=0.0, binWidth=1.0, quantity=unit('A')))
# fill them
for hist, df in zip([hist0, hist1, hist2, hist3], [df1, df2, df1, df2]):
hist.hist.fill.numpy(df)
e0, e1 = get_consistent_numpy_entries([hist0, hist1], get_bin_labels=False)
_, labels01 = get_consistent_numpy_entries([hist0, hist1],
get_bin_labels=True)
e2, e3 = get_consistent_numpy_entries([hist2, hist3], get_bin_labels=False)
_, centers23 = get_consistent_numpy_entries([hist2, hist3],
get_bin_labels=True)
entries0 = [2., 2., 3., 3., 0., 0., 0., 0., 0.]
entries1 = [0., 1., 2., 0., 1., 1., 3., 1., 1.]
labels = ['f1', 'f2', 'f3', 'f4', 'f5', 'f6', 'f7', 'f8', 'f9']
entries2 = [1., 4., 2., 2., 1., 0., 0., 0., 0.]
entries3 = [0., 0., 1., 1., 2., 2., 1., 2., 1.]
centers = [0.5, 1.5, 2.5, 3.5, 4.5, 5.5, 6.5, 7.5, 8.5]
np.testing.assert_array_equal(e0, entries0)
np.testing.assert_array_equal(e1, entries1)
np.testing.assert_array_equal(labels01, labels)
np.testing.assert_array_equal(e2, entries2)
np.testing.assert_array_equal(e3, entries3)
np.testing.assert_array_equal(centers23, centers)
def test_get_consistent_numpy_entries():
"""Test extraction of number of entries
When first making bin_edges of input histograms consistent to each other.
"""
df1 = pd.DataFrame({
"A": [0, 1, 2, 3, 4, 3, 2, 1, 1, 1],
"C": ["f1", "f3", "f4", "f3", "f4", "f2", "f2", "f1", "f3", "f4"],
})
df2 = pd.DataFrame({
"A": [2, 3, 4, 5, 7, 4, 6, 5, 7, 8],
"C": ["f7", "f3", "f5", "f8", "f9", "f2", "f3", "f6", "f7", "f7"],
})
# building 1d-, 2d-, and 3d-histogram (iteratively)
hist0 = HistogramContainer(hg.Categorize(unit("C")))
hist1 = HistogramContainer(hg.Categorize(unit("C")))
hist2 = HistogramContainer(
hg.SparselyBin(origin=0.0, binWidth=1.0, quantity=unit("A")))
hist3 = HistogramContainer(
hg.SparselyBin(origin=0.0, binWidth=1.0, quantity=unit("A")))
# fill them
for hist, df in zip([hist0, hist1, hist2, hist3], [df1, df2, df1, df2]):
hist.hist.fill.numpy(df)
e0, e1 = get_consistent_numpy_entries([hist0, hist1], get_bin_labels=False)
_, labels01 = get_consistent_numpy_entries([hist0, hist1],
get_bin_labels=True)
e2, e3 = get_consistent_numpy_entries([hist2, hist3], get_bin_labels=False)
_, centers23 = get_consistent_numpy_entries([hist2, hist3],
get_bin_labels=True)
entries0 = [2.0, 2.0, 3.0, 3.0, 0.0, 0.0, 0.0, 0.0, 0.0]
entries1 = [0.0, 1.0, 2.0, 0.0, 1.0, 1.0, 3.0, 1.0, 1.0]
labels = ["f1", "f2", "f3", "f4", "f5", "f6", "f7", "f8", "f9"]
entries2 = [1.0, 4.0, 2.0, 2.0, 1.0, 0.0, 0.0, 0.0, 0.0]
entries3 = [0.0, 0.0, 1.0, 1.0, 2.0, 2.0, 1.0, 2.0, 1.0]
centers = [0.5, 1.5, 2.5, 3.5, 4.5, 5.5, 6.5, 7.5, 8.5]
np.testing.assert_array_equal(e0, entries0)
np.testing.assert_array_equal(e1, entries1)
np.testing.assert_array_equal(labels01, labels)
np.testing.assert_array_equal(e2, entries2)
np.testing.assert_array_equal(e3, entries3)
np.testing.assert_array_equal(centers23, centers)
def get_test_histograms1():
""" Get set 1 of test histograms
"""
# dummy dataset with mixed types
# convert timestamp (col D) to nanosec since 1970-1-1
df = pd.util.testing.makeMixedDataFrame()
df['date'] = df['D'].apply(to_ns)
df['boolT'] = True
df['boolF'] = False
# building 1d-, 2d-, and 3d-histogram (iteratively)
hist1 = hg.Categorize(unit('C'))
hist2 = hg.Bin(5, 0, 5, unit('A'), value=hist1)
hist3 = hg.SparselyBin(origin=pd.Timestamp('2009-01-01').value,
binWidth=pd.Timedelta(days=1).value,
quantity=unit('date'),
value=hist2)
# fill them
hist1.fill.numpy(df)
hist2.fill.numpy(df)
hist3.fill.numpy(df)
hc1 = HistogramContainer(hist1)
hc2 = HistogramContainer(hist2)
hc3 = HistogramContainer(hist3)
return df, hc1, hc2, hc3
def test_prepare_2dgrid():
""" Test preparation of grid for extraction of number of entries for 2d hists
"""
df, hc1, hc2, hc3 = get_test_histograms1()
# building 1d-, 2d-, and 3d-histogram (iteratively)
hist1 = hg.Categorize(unit('C'))
hist2 = hg.Bin(5, 0, 5, unit('A'), value=hist1)
hist3 = hg.SparselyBin(origin=pd.Timestamp('2009-01-01').value,
binWidth=pd.Timedelta(days=1).value,
quantity=unit('date'),
value=hist2)
# fill them
hist1.fill.numpy(df)
hist2.fill.numpy(df)
hist3.fill.numpy(df)
xkeys1, ykeys1 = prepare_2dgrid(hist1)
xkeys2, ykeys2 = prepare_2dgrid(hist2)
xkeys3, ykeys3 = prepare_2dgrid(hist3)
np.testing.assert_array_equal(xkeys1, [])
np.testing.assert_array_equal(ykeys1, [])
np.testing.assert_array_equal(xkeys2, [0, 1, 2, 3, 4])
np.testing.assert_array_equal(ykeys2,
['foo1', 'foo2', 'foo3', 'foo4', 'foo5'])
np.testing.assert_array_equal(xkeys3, [0, 1, 4, 5, 6])
np.testing.assert_array_equal(ykeys3, [0, 1, 2, 3, 4])
def test_check_similar_hists():
"""Test similarity of list of histograms
Check similarity of: type, n-dim, sub-hists, specific type attributes
"""
# dummy dataset with mixed types
# convert timestamp (col D) to nanosec since 1970-1-1
df = pd.util.testing.makeMixedDataFrame()
df["date"] = df["D"].apply(to_ns)
# building 1d-, 2d-, and 3d-histogram (iteratively)
hist0 = hg.Bin(5, 0, 5, unit("A"))
hist1 = hg.Categorize(unit("C"))
hist2 = hg.Bin(5, 0, 5, unit("A"), value=hist1)
hist3 = hg.Categorize(unit("C"), value=hist0)
hist4 = hg.SparselyBin(
origin=pd.Timestamp("2009-01-01").value,
binWidth=pd.Timedelta(days=1).value,
quantity=unit("date"),
value=hist2,
)
hist5 = hg.SparselyBin(
origin=pd.Timestamp("2009-01-01").value,
binWidth=pd.Timedelta(days=1).value,
quantity=unit("date"),
value=hist3,
)
# fill them
for hist in [hist0, hist1, hist2, hist3, hist4, hist5]:
hist.fill.numpy(df)
hc0 = HistogramContainer(hist0)
hc1 = HistogramContainer(hist1)
hc2 = HistogramContainer(hist2)
hc3 = HistogramContainer(hist3)
hc4 = HistogramContainer(hist4)
hc5 = HistogramContainer(hist5)
for hc in [hc0, hc1, hc2, hc3, hc4, hc5]:
assert check_similar_hists([hc, hc])
assert not check_similar_hists([hc0, hc1])
assert not check_similar_hists([hc2, hc3])
assert not check_similar_hists([hc4, hc5])
def test_get_consistent_numpy_1dhists():
""" Test extraction of number of entries and bin-edges/labels
When first making bin_edges/bin-labels of input histograms consistent to each other.
"""
df1 = pd.DataFrame({'A': [0, 1, 2, 3, 4, 3, 2, 1, 1, 1]})
df2 = pd.DataFrame({'A': [2, 3, 4, 5, 7, 4, 6, 5, 7, 8]})
# building 1d-, 2d-, and 3d-histogram (iteratively)
hist1 = hg.SparselyBin(origin=0.0, binWidth=1.0, quantity=unit('A'))
hist2 = hg.SparselyBin(origin=0.0, binWidth=1.0, quantity=unit('A'))
# fill them
hist1.fill.numpy(df1)
hist2.fill.numpy(df2)
hc1 = HistogramContainer(hist1)
hc2 = HistogramContainer(hist2)
nphist1, nphist2 = get_consistent_numpy_1dhists([hc1, hc2],
get_bin_labels=False)
nphist_list, centers = get_consistent_numpy_1dhists([hc1, hc2],
get_bin_labels=True)
entries1 = [1., 4., 2., 2., 1., 0., 0., 0., 0.]
entries2 = [0., 0., 1., 1., 2., 2., 1., 2., 1.]
bin_edges = [0., 1., 2., 3., 4., 5., 6., 7., 8., 9.]
bin_centers = [0.5, 1.5, 2.5, 3.5, 4.5, 5.5, 6.5, 7.5, 8.5]
np.testing.assert_array_equal(nphist1[0], entries1)
np.testing.assert_array_equal(nphist1[1], bin_edges)
np.testing.assert_array_equal(nphist2[0], entries2)
np.testing.assert_array_equal(nphist2[1], bin_edges)
np.testing.assert_array_equal(nphist_list[0][0], entries1)
np.testing.assert_array_equal(nphist_list[0][1], bin_edges)
np.testing.assert_array_equal(nphist_list[1][0], entries2)
np.testing.assert_array_equal(nphist_list[1][1], bin_edges)
np.testing.assert_array_equal(centers, bin_centers)
def get_histograms():
df = get_test_data()
hist1 = hg.Categorize(unit('C'))
hist2 = hg.Bin(5, 0, 5, unit('A'), value=hist1)
hist3 = hg.SparselyBin(origin=pd.Timestamp('2009-01-01').value,
binWidth=pd.Timedelta(days=1).value,
quantity=unit('date'),
value=hist2)
for hist in [hist1, hist2, hist3]:
hist.fill.numpy(df)
return hist1, hist2, hist3
def test_get_consistent_numpy_2dgrids():
""" Test extraction of number of entries for 2d hists
When first making bin_edges of input histograms consistent to each other.
"""
df1 = pd.DataFrame({
'A': [0, 1, 2, 3, 4, 3, 2, 1, 1, 1],
'C': ['f1', 'f3', 'f4', 'f3', 'f4', 'f2', 'f2', 'f1', 'f3', 'f4']
})
df2 = pd.DataFrame({
'A': [2, 3, 4, 5, 7, 4, 6, 5, 7, 8],
'C': ['f7', 'f3', 'f5', 'f8', 'f9', 'f2', 'f3', 'f6', 'f7', 'f7']
})
# building 1d-, 2d-, and 3d-histogram (iteratively)
hist0 = hg.Categorize(unit('C'))
hist1 = hg.SparselyBin(origin=0.0,
binWidth=1.0,
quantity=unit('A'),
value=hist0)
hist2 = hg.SparselyBin(origin=0.0,
binWidth=1.0,
quantity=unit('A'),
value=hist0)
# fill them
hist0.fill.numpy(df1)
hist1.fill.numpy(df1)
hist2.fill.numpy(df2)
hc0 = HistogramContainer(hist0)
hc1 = HistogramContainer(hist1)
hc2 = HistogramContainer(hist2)
args = ['']
try:
get_consistent_numpy_2dgrids([hc0, hc0])
except AssertionError as e:
args = e.args
grid2d_list = get_consistent_numpy_2dgrids([hc1, hc2])
g1 = np.asarray([[1., 1., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 1., 1., 0., 0., 0., 0., 0.],
[0., 2., 0., 1., 0., 0., 0., 0., 0.],
[0., 1., 1., 0., 1., 0., 0., 0., 0.],
[0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 0., 0., 0., 0., 0., 0., 0.]])
g2 = np.asarray([[0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 0., 0., 1., 0., 0., 0., 0.],
[0., 0., 0., 1., 0., 0., 1., 0., 0.],
[0., 0., 0., 0., 0., 0., 0., 0., 0.],
[0., 0., 0., 0., 1., 0., 0., 0., 0.],
[0., 0., 0., 0., 0., 1., 0., 0., 0.],
[0., 0., 1., 0., 0., 0., 0., 1., 1.],
[0., 0., 0., 0., 0., 1., 0., 0., 0.],
[0., 0., 0., 0., 0., 0., 0., 1., 0.]])
grid2d_comp = [g1, g2]
# MB 20190828: not sure if this is the right way to test for exceptions.
assert args[
0] == 'Input histogram only has 1 dimensions (<2). Cannot compute 2d-grid.'
for i in range(2):
assert (grid2d_list[i] == grid2d_comp[i]).all()
def test_project_split2dhist_on_axis():
df = get_test_data()
histA = hg.Bin(5, 0, 5, unit('A'))
histC = hg.Categorize(unit('C'))
hist1 = hg.Categorize(unit('C'), value=histA)
hist2 = hg.Bin(5, 0, 5, unit('A'), value=histC)
histDCA = hg.SparselyBin(origin=pd.Timestamp('2009-01-01').value,
binWidth=pd.Timedelta(days=1).value,
quantity=unit('date'),
value=hist1)
histDAC = hg.SparselyBin(origin=pd.Timestamp('2009-01-01').value,
binWidth=pd.Timedelta(days=1).value,
quantity=unit('date'),
value=hist2)
histDA = hg.SparselyBin(origin=pd.Timestamp('2009-01-01').value,
binWidth=pd.Timedelta(days=1).value,
quantity=unit('date'),
value=histA)
histDC = hg.SparselyBin(origin=pd.Timestamp('2009-01-01').value,
binWidth=pd.Timedelta(days=1).value,
quantity=unit('date'),
value=histC)
for hist in [histDA, histDC, histDCA, histDAC]:
hist.fill.numpy(df)
# split along date axis
splitAC = HistogramContainer(histDAC).split_hist_along_first_dimension(
xname='x', yname='y', short_keys=True, convert_time_index=True)
splitCA = HistogramContainer(histDCA).split_hist_along_first_dimension(
xname='x', yname='y', short_keys=True, convert_time_index=True)
splitA0 = HistogramContainer(histDA).split_hist_along_first_dimension(
xname='x', yname='y', short_keys=True, convert_time_index=True)
splitC0 = HistogramContainer(histDC).split_hist_along_first_dimension(
xname='x', yname='y', short_keys=True, convert_time_index=True)
splitA1 = project_split2dhist_on_axis(splitAC, 'x')
splitA2 = project_split2dhist_on_axis(splitCA, 'y')
splitC1 = project_split2dhist_on_axis(splitAC, 'y')
splitC2 = project_split2dhist_on_axis(splitCA, 'x')
assert len(splitA0) == len(splitA1)
assert len(splitA0) == len(splitA2)
for key, h0 in splitA0.items():
assert key in splitA1
assert key in splitA2
h1 = splitA1[key]
h2 = splitA2[key]
bin_edges0 = h0.bin_edges()
bin_edges1 = h1.bin_edges()
bin_edges2 = h2.bin_edges()
bin_entries0 = h0.bin_entries()
bin_entries1 = h1.bin_entries()
bin_entries2 = h2.bin_entries()
np.testing.assert_array_equal(bin_edges0, bin_edges1)
np.testing.assert_array_equal(bin_edges0, bin_edges2)
np.testing.assert_array_equal(bin_entries0, bin_entries1)
np.testing.assert_array_equal(bin_entries0, bin_entries2)
assert len(splitC0) == len(splitC1)
assert len(splitC0) == len(splitC2)
for key, h0 in splitC0.items():
assert key in splitC1
assert key in splitC2
h1 = splitC1[key]
h2 = splitC2[key]
bin_labels0 = h0.bin_labels()
bin_labels1 = h1.bin_labels()
bin_labels2 = h2.bin_labels()
bin_entries0 = h0.bin_entries()
bin_entries1 = h1.bin_entries(bin_labels0)
bin_entries2 = h2.bin_entries(bin_labels0)
np.testing.assert_array_equal(sorted(bin_labels0), sorted(bin_labels1))
np.testing.assert_array_equal(sorted(bin_labels0), sorted(bin_labels2))
np.testing.assert_array_equal(bin_entries0, bin_entries1)
np.testing.assert_array_equal(bin_entries0, bin_entries2)
def test_get_consistent_numpy_2dgrids():
"""Test extraction of number of entries for 2d hists
When first making bin_edges of input histograms consistent to each other.
"""
df1 = pd.DataFrame({
"A": [0, 1, 2, 3, 4, 3, 2, 1, 1, 1],
"C": ["f1", "f3", "f4", "f3", "f4", "f2", "f2", "f1", "f3", "f4"],
})
df2 = pd.DataFrame({
"A": [2, 3, 4, 5, 7, 4, 6, 5, 7, 8],
"C": ["f7", "f3", "f5", "f8", "f9", "f2", "f3", "f6", "f7", "f7"],
})
# building 1d-, 2d-, and 3d-histogram (iteratively)
hist0 = hg.Categorize(unit("C"))
hist1 = hg.SparselyBin(origin=0.0,
binWidth=1.0,
quantity=unit("A"),
value=hist0)
hist2 = hg.SparselyBin(origin=0.0,
binWidth=1.0,
quantity=unit("A"),
value=hist0)
# fill them
hist0.fill.numpy(df1)
hist1.fill.numpy(df1)
hist2.fill.numpy(df2)
hc0 = HistogramContainer(hist0)
hc1 = HistogramContainer(hist1)
hc2 = HistogramContainer(hist2)
args = [""]
try:
get_consistent_numpy_2dgrids([hc0, hc0])
except ValueError as e:
args = e.args
grid2d_list = get_consistent_numpy_2dgrids([hc1, hc2])
g1 = np.asarray([
[1.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0],
[0.0, 0.0, 1.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0],
[0.0, 2.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0],
[0.0, 1.0, 1.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0],
[0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0],
[0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0],
[0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0],
[0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0],
[0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0],
])
g2 = np.asarray([
[0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0],
[0.0, 0.0, 0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0],
[0.0, 0.0, 0.0, 1.0, 0.0, 0.0, 1.0, 0.0, 0.0],
[0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0],
[0.0, 0.0, 0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0],
[0.0, 0.0, 0.0, 0.0, 0.0, 1.0, 0.0, 0.0, 0.0],
[0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 1.0, 1.0],
[0.0, 0.0, 0.0, 0.0, 0.0, 1.0, 0.0, 0.0, 0.0],
[0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 1.0, 0.0],
])
grid2d_comp = [g1, g2]
# MB 20190828: not sure if this is the right way to test for exceptions.
assert (
args[0] ==
"Input histogram only has 1 dimensions (<2). Cannot compute 2d-grid.")
for i in range(2):
assert (grid2d_list[i] == grid2d_comp[i]).all()
