def monovariate_plot_sorted(self, dataset):
g = self.monovariate_plot_session["g"]
options_tipo_var = self.monovariate_plot_session["options_tipo_var"]
options_categorical_list = self.monovariate_plot_session[
"option_categorical_list"]
dataset = dataset
unique_values = dataset[g[0]].unique()
data = table.dist_frequenza(dataset,
g[0],
save=False,
tipo="categoriale",
lista_ordinale=options_categorical_list)
data_non_tot = data.drop("Totale")
datatest = pd.DataFrame({
"X": data_non_tot.index.values,
"Frequency": data_non_tot["Frequenze"].values
})
datatest.dropna(inplace=True)
datatest.set_index(datatest["X"], inplace=True)
datatest.index = datatest.index.map(str)
datatest = datatest.loc[options_categorical_list]
equilibrium_values = tools.Sq_output(datatest["Frequency"])
if options_tipo_var == "cardinale":
characteristic_values = {
"mean": datatest["Frequency"].mean(),
"standard deviation": datatest["Frequency"].std(),
"gini index": tools.gini(datatest["Frequency"].values)
}
if options_tipo_var == "categoriale":
characteristic_values = {
"mode":
datatest[datatest["Frequency"] ==
datatest["Frequency"].max()]["X"].values[0],
"total equilibrium":
datatest["Frequency"].sum() /
len(datatest["Frequency"].unique()),
"Sq":
equilibrium_values["Sq"],
"Sq_norm":
equilibrium_values["Sq_Norm"],
"Eq":
equilibrium_values["Eq"]
}
chart = altair_monovariate(data=datatest,
options_tipo_var=options_tipo_var,
lista_ordinale=options_categorical_list)
return render_template(
"monovariate/monovariate_plot.html",
operation_form=g,
chart_json=chart.to_json(),
data=data.to_html(classes="table table-striped"),
unique_values=unique_values,
options_tipo_var=options_tipo_var,
characteristic_values=characteristic_values)
def recode_operation(self, request, obj_elab):
g = request.form.getlist('recode_var')
self.recode_session["recode_var"] = g
obj_elab.recode_var = g
dataset = obj_elab.dataset
unique_values = dataset[g[0]].unique()
data = table.dist_frequenza(dataset,
g[0],
save=False,
tipo="categoriale")
return render_template(
"recode/recode_operation.html",
operation_form=g,
data=data.to_html(classes="table table-striped"),
unique_values=unique_values,
)
def monovariate_plot(self, request, dataset):
r = request
g = r.form.getlist('monovariate')
options_tipo_var = r.form["var_type"]
#options_ordinal_list = r.form["ordinal_list"]
options_drop = r.form["drop_missing"]
self.monovariate_plot_session = {
"g": g,
"options_tipo_var": options_tipo_var,
#'options_categorical_list': options_ordinal_list
}
if options_drop == "drop":
dataset.dropna(subset=[g[0]], inplace=True)
elif options_drop == "no_drop":
pass
unique_values = dataset[g[0]].unique()
series = dataset[g[0]]
data = table.dist_frequenza(dataset,
g[0],
save=False,
tipo=options_tipo_var)
if options_tipo_var == "cardinale":
dataset_cardinale = dataset.round(2)
data = table.dist_frequenza(dataset_cardinale,
g[0],
save=False,
tipo=options_tipo_var)
data_non_tot = data.drop("Totale")
datatest = pd.DataFrame({
"X": data_non_tot.index.values,
"Frequency": data_non_tot["Frequenze"].values
})
if options_tipo_var == "likert":
options_categorical_list = ["1", "2", "3", "4", "5"]
data = table.dist_frequenza(dataset,
g[0],
save=False,
tipo="ordinale",
lista_ordinale=[1, 2, 3, 4, 5])
data_non_tot = data.drop("Totale")
dataset_dist_plot = pd.DataFrame({
"X":
data_non_tot.index.values,
"Frequency":
data_non_tot["Frequenze"].values
})
dataset_dist_plot.dropna(inplace=True)
dataset_dist_plot.set_index(dataset_dist_plot["X"], inplace=True)
dataset_dist_plot.index = dataset_dist_plot.index.map(str)
dataset_dist_plot = dataset_dist_plot.loc[options_categorical_list]
dataset_dist_plot.fillna(0, inplace=True)
dataset_dist_plot.X = dataset_dist_plot.index
print(dataset_dist_plot)
try:
gini = tools.gini(series)
except:
gini = np.nan
equilibrium_values = tools.Sq_output(datatest["Frequency"])
if options_tipo_var == "cardinale":
characteristic_values = {
"mean": series.mean(),
"standard deviation": series.std(),
"gini index": str(gini)
}
elif options_tipo_var == "categoriale":
characteristic_values = {
"mode":
datatest[datatest["Frequency"] ==
datatest["Frequency"].max()]["X"].values[0],
"total equilibrium":
datatest["Frequency"].sum() /
len(datatest["Frequency"].unique()),
"Sq":
equilibrium_values["Sq"],
"Sq_norm":
equilibrium_values["Sq_Norm"],
"Eq":
equilibrium_values["Eq"]
}
elif options_tipo_var == "likert":
characteristic_values = {
"mean":
series.mean(),
"standard deviation":
series.std(),
"gini index":
str(gini),
"mode":
datatest[datatest["Frequency"] ==
datatest["Frequency"].max()]["X"].values[0],
"total equilibrium":
datatest["Frequency"].sum() /
len(datatest["Frequency"].unique()),
"Sq":
equilibrium_values["Sq"],
"Sq_norm":
equilibrium_values["Sq_Norm"],
"Eq":
equilibrium_values["Eq"]
}
print(options_tipo_var)
if options_tipo_var == "categoriale" or options_tipo_var == "ordinale":
chart = altair_monovariate_bar(data=datatest,
options_tipo_var=options_tipo_var,
lista_ordinale=False)
print(chart.to_json())
elif options_tipo_var == "likert":
chart = altair_monovariate_bar(
data=dataset_dist_plot,
options_tipo_var=options_tipo_var,
lista_ordinale=options_categorical_list)
print(chart.to_json())
elif options_tipo_var == "cardinale":
print(dataset[g[0]])
dataset_hist = dataset.round(2)
chart = altair_monovariate_hist(dataset_hist, g[0])
print(chart.to_json())
return render_template(
"monovariate/monovariate_plot.html",
operation_form=g,
chart_json=chart.to_json(),
data=data.to_html(classes="table table-striped"),
unique_values=unique_values,
options_tipo_var=options_tipo_var,
characteristic_values=characteristic_values)