churn.py

from __future__ import division
import pandas as pd
import numpy as np

# get churn.csv from here: https://raw.githubusercontent.com/EricChiang/churn/master/data/churn.csv
churn_df = pd.read_csv('churn.csv')
col_names = churn_df.columns.tolist()

print "Column names:"
print col_names

to_show = col_names[:6] + col_names[-6:]

print "\nSample data:"
churn_df[to_show].head(6)
      


# Isolate target data
churn_result = churn_df['Churn?']
y = np.where(churn_result == 'True.',1,0)

# We don't need these columns
to_drop = ['State','Area Code','Phone','Churn?']
churn_feat_space = churn_df.drop(to_drop,axis=1)

# 'yes'/'no' has to be converted to boolean values
# NumPy converts these from boolean to 1. and 0. later
yes_no_cols = ["Int'l Plan","VMail Plan"]
churn_feat_space[yes_no_cols] = churn_feat_space[yes_no_cols] == 'yes'

# Pull out features for future use
features = churn_feat_space.columns

X = churn_feat_space.as_matrix().astype(np.float)

# This is important
from sklearn.preprocessing import StandardScaler
scaler = StandardScaler()
X = scaler.fit_transform(X)

print "Feature space holds %d observations and %d features" % X.shape
print "Unique target labels:", np.unique(y)
      

log_func.py

import numpy as np
import pylab as pl

x = np.random.uniform(1, 100, 1000)
y = np.log(x) + np.random.normal(0, .3, 1000)

pl.scatter(x, y, s=1, label="log(x) with noise")
pl.plot(np.arange(1, 100), np.log(np.arange(1, 100)), c="b", label="log(x) true function")
pl.xlabel("x")
pl.ylabel("f(x) = log(x)")
pl.legend(loc="best")
pl.title("A Basic Log Function")
pl.show()

rf.py

from sklearn.datasets import load_iris
from sklearn.ensemble import RandomForestClassifier
import pandas as pd
import numpy as np

iris = load_iris()
df = pd.DataFrame(iris.data, columns=iris.feature_names)
df['is_train'] = np.random.uniform(0, 1, len(df)) <= .75
df['species'] = pd.Factor(iris.target, iris.target_names)
df.head()

train, test = df[df['is_train']==True], df[df['is_train']==False]

features = df.columns[:4]
clf = RandomForestClassifier(n_jobs=2)
y, _ = pd.factorize(train['species'])
clf.fit(train[features], y)

preds = iris.target_names[clf.predict(test[features])]
pd.crosstab(test['species'], preds, rownames=['actual'], colnames=['preds'])

time-series-agg.py

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from ggplot import *

meat = meat.dropna(thresh=800, axis=1) # drop columns that have fewer than 800 observations
ts = meat.set_index(['date'])

print ts.groupby(ts.index.year).sum().head(10)

the1940s = ts.groupby(ts.index.year).sum().ix['1940-01-01':'1949-12-31']
the1940s

def floor_decade(date_value):
    "Takes a date. Returns the decade."
    return (date_value.year // 10) * 10

pd.to_datetime('2013-10-09')
floor_decade(_)

ts.groupby(floor_decade).sum()

the1940s.sum().reset_index(name='meat sums in the 1940s')

by_decade = ts.groupby(floor_decade).sum()

by_decade.index.name = 'year'

by_decade = by_decade.reset_index()

print ggplot(by_decade, aes('year', weight='beef')) + \
    geom_bar() + \
    scale_y_continuous(labels='comma') + \
    ggtitle('Head of Cattle Slaughtered by Decade')


by_decade_long = pd.melt(by_decade, id_vars="year")

print ggplot(aes(x='year', weight='value', colour='variable'), data=by_decade_long) + \
geom_bar() + \
ggtitle("Meat Production by Decade")


from ggplot import meat
meat_lng = pd.melt(meat, id_vars=['date'])
print ggplot(aes(x='date', y='value', colour='variable'), data=meat_lng) + geom_line()

print ggplot(aes(x='date', y='value', colour='variable'), data=meat_lng) + \
    stat_smooth(span=0.10) + \
    ggtitle("Smoothed Livestock Production")

https://raw.githubusercontent.com/EricChiang/churn/master/data/churn.csv is a broken link. Do you have a live link to the data?