Příprava tabulky grafu přehledu prediktoru

Introduction to Predictive Analytics in Python

Nele Verbiest, Ph.D

Data Scientist @PythonPredictions

Tabulka grafu přehledu prediktoru

disc_mean_gift	Incidence	Size
[2, 78]	0.013042	20013
(78, 87]	0.029554	19997
(87, 94]	0.040831	20034
(94, 103]	0.063563	20405
(103, 197]	0.103524	19551

Výpočet tabulky grafu přehledu prediktoru

# Load the numpy module
import numpy as np
# Function that calculates the predictor insight graph table
def create_pig_table(df, target, variable):

    # Group by the variable you want to plot
    groups = df[[target,variable]].groupby(variable)

    # Calculate the size and incidence of each group
    pig_table = groups[target].agg(Incidence = 'mean').reset_index()
    return pig_table

print(create_pig_table(basetable,"target","country")

country    Incidence    Size
India    0.050934    49849
UK    0.050512    10057
USA    0.048486    40094

Výpočet více tabulek grafu přehledu prediktoru

variables = ["country", "gender", "disc_mean_gift", "age"]

# Empty dictionary.
pig_tables = {}

# Loop over all variables
for variable in variables:

    # Create the predictor insight graph table
    pig_table = create_pig_table(basetable, "target", variable)

    # Store the table in the dictionary
    pig_tables[variable] = pig_table

print(create_pig_table(basetable,"target","country")

country    Incidence    Size
India    0.050934    49849
UK    0.050512    10057
USA    0.048486    40094

Pojďme si procvičit!

Introduction to Predictive Analytics in Python