def add_paired_property(
dataset,
combined_data: CombinedData,
study_type: str,
design: Dict[str, str] = None): # check same sizes are identical
global paired
x = None
y = None
combined_data.properties[paired] = False
if isinstance(combined_data, BivariateData):
if study_type == experiment_identifier:
# Just need one variable to be Categorical and another to be Continuous (regardless of role)
x = combined_data.get_vars(iv_identifier)
y = combined_data.get_vars(dv_identifier)
else: # study_type == observational_identifier
x = combined_data.get_vars(contributor_identifier)
y = combined_data.get_vars(outcome_identifier)
if x and y:
assert (len(x) == len(y) == 1)
x = x[0]
y = y[0]
if x.is_categorical() and y.is_continuous():
if within_subj in design and design[within_subj] == x.metadata[
name]:
combined_data.properties[paired] = True
def add_eq_variance_property(dataset, combined_data: CombinedData,
study_type: str):
xs = None
ys = None
cat_xs = []
cont_ys = []
grouped_data = []
if study_type == experiment_identifier:
# Just need one variable to be Catogrical and another to be Continuous (regardless of role) -- both could be variable_identifier types
xs = combined_data.get_vars(iv_identifier)
ys = combined_data.get_vars(dv_identifier)
else: # study_type == observational_identifier
xs = combined_data.get_vars(contributor_identifier)
ys = combined_data.get_vars(outcome_identifier)
for x in xs:
if x.is_categorical():
cat_xs.append(x)
for y in ys:
if y.is_continuous():
cont_ys.append(y)
combined_data.properties[eq_variance] = None
if cat_xs and cont_ys:
for y in ys:
for x in xs:
cat = [k for k, v in x.metadata[categories].items()]
for c in cat:
data = dataset.select(
y.metadata[name],
where=[f"{x.metadata[name]} == '{c}'"])
grouped_data.append(data)
if isinstance(combined_data, BivariateData):
# Equal variance
eq_var = compute_eq_variance(grouped_data)
combined_data.properties[eq_variance] = eq_var
elif isinstance(combined_data, MultivariateData):
combined_data.properties[
eq_variance + '::' + x.metadata[name] + ':' +
y.metadata[name]] = compute_eq_variance(grouped_data)
else:
raise ValueError(
f"combined_data_data object is neither BivariateData nor MultivariateData: {type(combined_data)}"
)
def add_categories_normal(dataset,
combined_data: CombinedData,
study_type: str,
design: Dict[str, str] = None):
global cat_distribution
xs = None
ys = None
cat_xs = []
cont_ys = []
grouped_data = dict()
if study_type == experiment_identifier:
# Just need one variable to be Catogrical and another to be Continuous (regardless of role) -- both could be variable_identifier types
xs = combined_data.get_vars(iv_identifier)
ys = combined_data.get_vars(dv_identifier)
else: # study_type == observational_identifier
xs = combined_data.get_vars(contributor_identifier)
ys = combined_data.get_vars(outcome_identifier)
for x in xs:
if x.is_categorical():
cat_xs.append(x)
for y in ys:
if y.is_continuous():
cont_ys.append(y)
combined_data.properties[cat_distribution] = None
if cat_xs and cont_ys:
for y in ys:
for x in xs:
cat = [k for k, v in x.metadata[categories].items()]
for c in cat:
data = dataset.select(
y.metadata[name],
where=[f"{x.metadata[name]} == '{c}'"])
grouped_data_name = str(x.metadata[name] + ':' + c)
grouped_data[grouped_data_name] = compute_distribution(
data)
combined_data.properties[cat_distribution] = dict()
combined_data.properties[cat_distribution][
y.metadata[name] + '::' + x.metadata[name]] = grouped_data