shuttle.py

import pandas as pd
import scipy.stats as st
import pylab as pl
import math

# Download dataframe
url = "https://archive.ics.uci.edu/ml/machine-learning-databases/statlog/shuttle/shuttle.tst"
df = pd.read_csv(url, header=None, delimiter=' ')

# From http://odds.cs.stonybrook.edu/shuttle-dataset/
# The smallest five classes, i.e. 2, 3, 5, 6, 7 are combined to form the outliers class,
# while class 1 forms the inlier class. Data for class 4 is discarded.
df = df.loc[df[9] != 4]

# Plot Distribution and Kurtosis of Columns
df_norm = (df - df.mean()) / (df.max() - df.min())

def plot_normal(data, mu, var):
    sigma = math.sqrt(var)
    pl.plot(data,st.norm.pdf(data, mu, sigma), "-o")
    pl.hist(data, normed=True)
    print("Kurtosis: " + str(st.kurtosis(data)))
    pl.show()

for col in range(1,len(df_norm.iloc[0])-1):
    print("Column: " + str(col))
    values = df_norm.iloc[:, col].values
    plot_normal(sorted(values), values.mean(), values.var())