Tutorial 1.

Linear Regression

January 11, 2017

1 Tutorial: Linear Regression

Agenda: 1. Spyder interface 2. Linear regression running example: boston data 3. Vectorize cost
function 4. Closed form solution 5. Gradient descent

In [1]: import matplotlib

import numpy as np
import matplotlib.pyplot as plt
%matplotlib inline

In [2]: from sklearn.datasets import load_boston

boston_data = load_boston()
print(boston_data['DESCR'])

Boston House Prices dataset

Notes
------
Data Set Characteristics:

:Number of Instances: 506

:Number of Attributes: 13 numeric/categorical predictive

:Median Value (attribute 14) is usually the target

:Attribute Information (in order):

- CRIM per capita crime rate by town
- ZN proportion of residential land zoned for lots over 25,000 sq.ft.
- INDUS proportion of non-retail business acres per town
- CHAS Charles River dummy variable (= 1 if tract bounds river; 0 other
- NOX nitric oxides concentration (parts per 10 million)
- RM average number of rooms per dwelling
- AGE proportion of owner-occupied units built prior to 1940
- DIS weighted distances to five Boston employment centres
- RAD index of accessibility to radial highways
- TAX full-value property-tax rate per $10,000

1
 - PTRATIO pupil-teacher ratio by town
- B 1000(Bk - 0.63)ˆ2 where Bk is the proportion of blacks by town
- LSTAT % lower status of the population
- MEDV Median value of owner-occupied homes in $1000's

:Missing Attribute Values: None

:Creator: Harrison, D. and Rubinfeld, D.L.

This is a copy of UCI ML housing dataset.

http://archive.ics.uci.edu/ml/datasets/Housing

This dataset was taken from the StatLib library which is maintained at Carnegie Mel

The Boston house-price data of Harrison, D. and Rubinfeld, D.L. 'Hedonic

prices and the demand for clean air', J. Environ. Economics & Management,
vol.5, 81-102, 1978. Used in Belsley, Kuh & Welsch, 'Regression diagnostics
...', Wiley, 1980. N.B. Various transformations are used in the table on
pages 244-261 of the latter.

The Boston house-price data has been used in many machine learning papers that addr
problems.

**References**

- Belsley, Kuh & Welsch, 'Regression diagnostics: Identifying Influential Data a

- Quinlan,R. (1993). Combining Instance-Based and Model-Based Learning. In Proce
- many more! (see http://archive.ics.uci.edu/ml/datasets/Housing)

In [3]: # take the boston data

data = boston_data['data']
# we will only work with two of the features: INDUS and RM
x_input = data[:, [2,5]]
y_target = boston_data['target']

In [5]: # Individual plots for the two features:

plt.title('Industrialness vs Med House Price')
plt.scatter(x_input[:, 0], y_target)
plt.xlabel('Industrialness')
plt.ylabel('Med House Price')
plt.show()

plt.title('Avg Num Rooms vs Med House Price')

plt.scatter(x_input[:, 1], y_target)
plt.xlabel('Avg Num Rooms')

2
plt.ylabel('Med House Price')
plt.show()

3
1.1 Define cost function
N
1 X (i)
E(y, t) = (y − t(i) )2
2N
i=1
N
1 X (i) (i)
E(y, t) = (w1 x1 + w2 x2 + b − t(i) )2
2N
i=1

In [6]: def cost(w1, w2, b, X, t):

'''
Evaluate the cost function in a non-vectorized manner for
inputs `X` and targets `t`, at weights `w1`, `w2` and `b`.
'''
costs = 0
for i in range(len(t)):
y_i = w1 * X[i, 0] + w2 * X[i, 1] + b
t_i = t[i]
costs += 0.5 * (y_i - t_i) ** 2
return costs / len(t)
In [7]: cost(3, 5, 20, x_input, y_target)
Out[7]: 2241.1239166749006
In [8]: cost(3, 5, 0, x_input, y_target)
Out[8]: 1195.1098850543478

1.2 Vectorizing the cost function:

1
kXw + b1 − tk2
E(y, t) =
2N
In [9]: def cost_vectorized(w1, w2, b, X, t):
'''
Evaluate the cost function in a vectorized manner for
inputs `X` and targets `t`, at weights `w1`, `w2` and `b`.
'''
N = len(y_target)
w = np.array([w1, w2])
y = np.dot(X, w) + b * np.ones(N)
return np.sum((y - t)**2) / (2.0 * N)
In [10]: cost_vectorized(3, 5, 20, x_input, y_target)
Out[10]: 2241.1239166749015
In [11]: cost(3, 5, 0, x_input, y_target)
Out[11]: 1195.1098850543478

4
1.3 Comparing speed of the vectorized vs unvectorized code
We’ll see below that the vectorized code already runs ~2x faster than the non-vectorized code!
Hopefully this will convince you to always vectorized your code whenever possible

In [12]: import time

t0 = time.time()
print cost(4, 5, 20, x_input, y_target)
t1 = time.time()
print t1 - t0

3182.40634167
0.00229597091675

In [13]: t0 = time.time()
print cost_vectorized(4, 5, 20, x_input, y_target)
t1 = time.time()
print t1 - t0

3182.40634167
0.000537872314453

1.4 Plotting cost in weight space

We’ll plot the cost for two of our weights, assuming that bias = -22.89831573.
We’ll see where that number comes from later.
Notice the shape of the contours are ovals.

In [15]: w1s = np.arange(-1.0, 0.0, 0.01)

w2s = np.arange(6.0, 10.0, 0.1)
z_cost = []
for w2 in w2s:
z_cost.append([cost_vectorized(w1, w2, -22.89831573, x_input, y_target
z_cost = np.array(z_cost)
np.shape(z_cost)
W1, W2 = np.meshgrid(w1s, w2s)
CS = plt.contour(W1, W2, z_cost, 25)
plt.clabel(CS, inline=1, fontsize=10)
plt.title('Costs for various values of w1 and w2 for b=0')
plt.xlabel("w1")
plt.ylabel("w2")
plt.plot([-0.33471389], [7.82205511], 'o') # this will be the minima that
plt.show()

5
2 Exact Solution
Work this out on the board:

1. ignore biases (add an extra feature & weight instead)

2. get equations from partial derivative
3. vectorize
4. write code.

In [16]: # add an extra feature (column in the input) that are just all ones
x_in = np.concatenate([x_input, np.ones([np.shape(x_input)[0], 1])], axis=
x_in

Out[16]: array([[ 2.31 , 6.575, 1. ],

[ 7.07 , 6.421, 1. ],
[ 7.07 , 7.185, 1. ],
...,
[ 11.93 , 6.976, 1. ],
[ 11.93 , 6.794, 1. ],
[ 11.93 , 6.03 , 1. ]])

In [17]: def solve_exactly(X, t):

'''

6
 Solve linear regression exactly. (fully vectorized)

Given `X` - NxD matrix of inputs

`t` - target outputs
Returns the optimal weights as a D-dimensional vector
'''
N, D = np.shape(X)
A = np.matmul(X.T, X)
c = np.dot(X.T, t)
return np.matmul(np.linalg.inv(A), c)

In [18]: solve_exactly(x_in, y_target)

Out[18]: array([ -0.33471389, 7.82205511, -22.89831573])

In [19]: # In real life we don't want to code it directly

np.linalg.lstsq(x_in, y_target)

Out[19]: (array([ -0.33471389, 7.82205511, -22.89831573]),

array([ 19807.614505]),
3,
array([ 318.75354429, 75.21961717, 2.10127199]))

2.1 Implement Gradient Function

∂E 1 X (i) (i)
= xj (y − t(i) )
∂wj N
i

In [20]: # Vectorized gradient function

def gradfn(weights, X, t):
'''
Given `weights` - a current "Guess" of what our weights should be
`X` - matrix of shape (N,D) of input features
`t` - target y values
Return gradient of each weight evaluated at the current value
'''
N, D = np.shape(X)
y_pred = np.matmul(X, weights)
error = y_pred - t
return np.matmul(np.transpose(x_in), error) / float(N)

In [23]: def solve_via_gradient_descent(X, t, print_every=5000,

niter=100000, alpha=0.005):
'''
Given `X` - matrix of shape (N,D) of input features
`t` - target y values
Solves for linear regression weights.
Return weights after `niter` iterations.
'''

7
 N, D = np.shape(X)
# initialize all the weights to zeros
w = np.zeros([D])
for k in range(niter):
dw = gradfn(w, X, t)
w = w - alpha*dw
if k % print_every == 0:
print 'Weight after %d iteration: %s' % (k, str(w))
return w

In [24]: solve_via_gradient_descent( X=x_in, t=y_target)

Weight after 0 iteration: [ 1.10241186 0.73047508 0.11266403]

Weight after 5000 iteration: [-0.48304613 5.10076868 -3.97899253]
Weight after 10000 iteration: [-0.45397323 5.63413678 -7.6871518 ]
Weight after 15000 iteration: [ -0.43059857 6.06296553 -10.66851736]
Weight after 20000 iteration: [ -0.41180532 6.40774447 -13.06553969]
Weight after 25000 iteration: [ -0.39669551 6.68494726 -14.9927492 ]
Weight after 30000 iteration: [ -0.38454721 6.90781871 -16.54222851]
Weight after 35000 iteration: [ -0.37477995 7.08700769 -17.78801217]
Weight after 40000 iteration: [ -0.36692706 7.23107589 -18.78962409]
Weight after 45000 iteration: [ -0.36061333 7.34690694 -19.59492155]
Weight after 50000 iteration: [ -0.35553708 7.44003528 -20.24238191]
Weight after 55000 iteration: [ -0.35145576 7.5149106 -20.762941 ]
Weight after 60000 iteration: [ -0.34817438 7.57511047 -21.18147127]
Weight after 65000 iteration: [ -0.34553614 7.62351125 -21.51797024]
Weight after 70000 iteration: [ -0.343415 7.66242555 -21.78851591]
Weight after 75000 iteration: [ -0.34170959 7.69371271 -22.00603503]
Weight after 80000 iteration: [ -0.34033844 7.71886763 -22.18092072]
Weight after 85000 iteration: [ -0.33923604 7.73909222 -22.32152908]
Weight after 90000 iteration: [ -0.3383497 7.75535283 -22.4345784 ]
Weight after 95000 iteration: [ -0.33763709 7.76842638 -22.52547023]

Out[24]: array([ -0.33706425, 7.77893565, -22.59853432])

In [25]: # For comparison, this was the exact result:

np.linalg.lstsq(x_in, y_target)

Out[25]: (array([ -0.33471389, 7.82205511, -22.89831573]),

array([ 19807.614505]),
3,
array([ 318.75354429, 75.21961717, 2.10127199]))