def summarize_fd_by_subject(df):
return (
df
>> p.group_by("subject_id", "condition", "data_id", "headcase")
>> p.summarize(
fd_mean="mean(FramewiseDisplacement)",
fd_median="median(FramewiseDisplacement)",
fd_mean_filter="filter_mean(FramewiseDisplacement)",
fd_median_filter="filter_median(FramewiseDisplacement)",
perc_spikes="perc_high_motion(FramewiseDisplacement)",
)
>> p.do(
lambda df: df.melt(
id_vars=["subject_id", "data_id", "condition", "headcase"],
value_vars=[
"fd_mean",
"fd_median",
"fd_mean_filter",
"fd_median_filter",
"perc_spikes",
],
var_name="measure",
value_name="val",
)
)
>> p.arrange("subject_id")
>> p.call(".reset_index", drop=True)
)
def test_count():
df = pd.DataFrame({
'x': [1, 2, 3, 4, 5, 6],
'y': ['a', 'b', 'a', 'b', 'a', 'b'],
'w': [1, 2, 1, 2, 1, 2]
})
result = df >> count()
assert result.loc[0, 'n'] == 6
result = df >> count('y')
assert result.loc[:, 'n'].tolist() == [3, 3]
result = df >> count('y', 'w')
assert result.loc[:, 'n'].tolist() == [3, 3]
result = df >> count('y', weights='w')
assert result.loc[:, 'n'].tolist() == [3, 6]
result2 = df >> group_by('y') >> summarize(n='sum(w)')
assert result.equals(result2)
result = df >> count('w-1')
assert result.loc[:, 'w-1'].tolist() == [0, 1]
assert result.loc[:, 'n'].tolist() == [3, 3]
result1 = df >> group_by('y') >> count('w')
result2 = df >> count('y', 'w')
assert result1.plydata_groups == ['y']
assert pd.DataFrame(result1).equals(result2)
def test_tally():
df = pd.DataFrame({
'x': [1, 2, 3, 4, 5, 6],
'y': ['a', 'b', 'a', 'b', 'a', 'b'],
'w': [1, 2, 1, 2, 1, 2]
})
result = df >> tally()
assert result.loc[0, 'n'] == 6
result = df >> group_by('y') >> tally()
assert result.loc[:, 'n'].tolist() == [3, 3]
result = df >> group_by('y') >> tally('w')
assert result.loc[:, 'n'].tolist() == [3, 6]
result2 = df >> group_by('y') >> summarize(n='sum(w)')
assert result.equals(result2)
# External weights
result = df >> tally(range(5))
assert result.loc[0, 'n'] == 10
# Sort
result = df >> group_by('y') >> tally('w', sort=True)
assert result.loc[:, 'n'].tolist() == [6, 3]
def summarize_mpars_by_subject(df):
return (
df
>> p.group_by("subject_id", "condition", "data_id", "headcase")
>> p.summarize(
x_mean="mean(x)",
x_median="median(x)",
x_std="std(x)",
y_mean="mean(y)",
y_median="median(y)",
y_std="std(y)",
z_mean="mean(z)",
z_median="median(z)",
z_std="std(z)",
pitch_mean="mean(pitch)",
pitch_median="median(pitch)",
pitch_std="std(pitch)",
roll_mean="mean(roll)",
roll_median="median(roll)",
roll_std="std(roll)",
yaw_mean="mean(yaw)",
yaw_median="median(yaw)",
yaw_std="std(yaw)",
)
>> p.call(
".melt",
id_vars=["subject_id", "data_id", "condition", "headcase"],
value_vars=[
"x_mean",
"y_mean",
"z_mean",
"x_median",
"y_median",
"z_median",
"x_std",
"y_std",
"z_std",
"pitch_mean",
"roll_mean",
"yaw_mean",
"pitch_median",
"roll_median",
"yaw_median",
"pitch_std",
"roll_std",
"yaw_std",
],
var_name="measure",
value_name="val",
)
>> p.arrange("subject_id")
>> p.call(".reset_index", drop=True)
)
def combine_sherlock_view_runs_and_replace(df):
"""
Given a summarized dataframe (i.e. 1 datapoint per subject per condition/motion-direction),
average the summary params for runs view1 and view2 for sherlocks subs and reattach to
the original frame such that it only contains 'view' and 'recall' conditions rather than
'view1', 'view2', and 'recall' conditions. This is because realignment is computed on a per
run basis, of which sherlock subs have 2 'view' runs, but summary statistics (i.e. mean FD)
are computed as: (mean of run1 + mean of run2 / 2)
"""
sherlock_combined = (
df
>> p.query("data_id == 'sherlock' and condition != 'recall'")
>> p.group_by("subject_id", "measure", "data_id", "headcase")
>> p.summarize(val="mean(val)")
>> p.call(".assign", condition="view")
>> p.select("subject_id", "data_id", "condition", "headcase", "measure", "val")
)
df_no_sherlock = df.query("condition == 'view' or condition == 'recall'")
return pd.concat([df_no_sherlock, sherlock_combined], axis=0).reset_index(drop=True)
def test_no_groups(self):
result = self.df >> summarize('min(x)')
assert result.loc[0, 'min(x)'] == 0
result = self.df >> summarize('first(x)')
assert result.loc[0, 'first(x)'] == 0
result = self.df >> summarize('last(x)')
assert result.loc[0, 'last(x)'] == 5
result = self.df >> summarize('nth(y, 4)')
assert result.loc[0, 'nth(y, 4)'] == 2
result = self.df >> summarize('n_distinct(y)')
assert result.loc[0, 'n_distinct(y)'] == 4
result = self.df >> summarize('n()')
assert result.loc[0, 'n()'] == 6
result = self.df >> summarize(nth='nth(x, 100)')
assert np.isnan(result.loc[0, 'nth'])
def test_groups(self):
result = self.df >> group_by('y') >> summarize('mean(x)')
assert all(result['mean(x)'] == [0.5, 2.5, 4, 5])
result = self.df >> group_by('y') >> summarize('n()')
assert all(result['n()'] == [2, 2, 1, 1])