La funzione è stata aggiornata così:

def make_plot_updated(dataset, x, y, kind, comp_type):
  whole_dat = []
  for industry in comp_type:
    dat = dataset.loc[dataset["comp_type"]==industry]
    values = [x,y if y!="Year" else x]
    dat_avg = dat.pivot_table(index="Year",
                              values=values).reset_index()
    dat_avg = dat_avg.loc[:,
                          ~dat_avg.columns.duplicated()]
    dat_avg["company"] = f"Avg {type}"
    dat_avg["comp_type"] = industry
    whole_dat.append(pd.concat([dat, dat_avg]))

  plot_df = pd.concat(whole_dat).reset_index(drop=True)
  sns.relplot(data=plot_df,
              x=x, y=y,
              hue="company",
              col="comp_type",
              kind=kind)
  plt.show()
  plt.close()

Qual è lo scopo di (1) values = [x,y if y!="Year" else x] e (2) dat_avg = dat_avg.loc[:,~dat_avg.columns.duplicated()]?

Puoi incollare il seguente codice nella console e dare un'occhiata all'output per capirlo. ~ nega l'oggetto a cui è applicato. Quindi ~x significa non x.

dat = dataset.loc[dataset["comp_type"]=="fmcg"]

x,y = "operating_margin", "Year"
values = [x,y if y!="Year" else x]
dat_avg = dat.pivot_table(index="Year", values=values).reset_index()
print(dat_avg)
print(dat_avg.columns.duplicated())
dat_avg = dat_avg.loc[:,~dat_avg.columns.duplicated()]
print(values)
print(dat_avg)


x,y = "operating_margin", "debt_to_equity"
values = [x,y if y!="Year" else x]
dat_avg = dat.pivot_table(index="Year", values=values).reset_index()
print(dat_avg)
print(dat_avg.columns.duplicated())
dat_avg = dat_avg.loc[:,~dat_avg.columns.duplicated()]
print(values)
print(dat_avg)

Ben fatto!

Create Your Free Account