# Python for Data Science¶

Joe McCarthy, Director, Analytics & Data Science, Atigeo, LLC

In [1]:
from IPython.display import display, Image, HTML


## 1. Introduction¶

This short primer on Python is designed to provide a rapid "on-ramp" to enable computer programmers who are already familiar with concepts and constructs in other programming languages learn enough about Python to facilitate the effective use of open-source and proprietary Python-based machine learning and data science tools.

The primer is motivated, in part, by the approach taken in the Natural Language Toolkit (NLTK) book, which provides a rapid on-ramp for using Python and the open-source NLTK library to develop programs using natural language processing techniques (many of which involve machine learning).

The Python Tutorial offers a more comprehensive primer, and opens with an excellent - if biased - overview of some of the general strengths of the Python programming language:

Python is an easy to learn, powerful programming language. It has efficient high-level data structures and a simple but effective approach to object-oriented programming. Python’s elegant syntax and dynamic typing, together with its interpreted nature, make it an ideal language for scripting and rapid application development in many areas on most platforms.

Hans Petter Langtangen, author of Python Scripting for Computational Science, emphasizes the utility of Python for many of the common tasks in all areas of computational science:

Very often programming is about shuffling data in and out of different tools, converting one data format to another, extracting numerical data from a text, and administering numerical experiments involving a large number of data files and directories. Such tasks are much faster to accomplish in a language like Python than in Fortran, C, C++, C#, or Java

Foster Provost, co-author of Data Science for Business, describes why Python is such a useful programming language for practical data science in Python: A Practical Tool for Data Science, :

The practice of data science involves many interrelated but different activities, including accessing data, manipulating data, computing statistics about data, plotting/graphing/visualizing data, building predictive and explanatory models from data, evaluating those models on yet more data, integrating models into production systems, etc. One option for the data scientist is to learn several different software packages that each specialize in one or two of these things, but don’t do them all well, plus learn a programming language to tie them together. (Or do a lot of manual work.)

An alternative is to use a general-purpose, high-level programming language that provides libraries to do all these things. Python is an excellent choice for this. It has a diverse range of open source libraries for just about everything the data scientist will do. It is available everywhere; high performance python interpreters exist for running your code on almost any operating system or architecture. Python and most of its libraries are both open source and free. Contrast this with common software packages that are available in a course via an academic license, yet are extremely expensive to license and use in industry.

The goal of this primer is to provide efficient and sufficient scaffolding for software engineers with no prior knowledge of Python to be able to effectively use Python-based tools for data science research and development, such as the open-source library scikit-learn. There is another, more comprehensive tutorial for scikit-learn, Python Scientific Lecture Notes, that includes coverage of a number of other useful Python open-source libraries used by scikit-learn (numpy, scipy and matplotlib) - all highly recommended ... and, to keep things simple, all beyond the scope of this primer.

The initial motivation for this primer was a 2-hour training session for a group of experienced software engineers to learn enough Python to utilize the Atigeo xPatterns analytics framework API in their software development work. I am grateful to the company for affording me the opportunity to develop this educational tool, and to make it freely available to others who might be looking for a fast on-ramp to Python for data science.

Using an IPython Notebook as a delivery vehicle for this primer was motivated by Brian Granger's inspiring tutorial, The IPython Notebook: Get Close to Your Data with Python and JavaScript, one of the highlights from my Strata 2014 conference experience.

One final note on external resources: the Python Style Guide (PEP-0008) offers helpful tips on how best to format Python code. Code like a Pythonista offers a number of additional tips on Python programming style and philosophy, several of which are incorporated into this primer.

We will focus entirely on using Python within the interpreter environment (as supported within an IPython Notebook). Python scripts - files containing definitions of functions and variables, and typically including code invoking some of those functions - can also be run from a command line. Using Python scripts from the command line may be the subject of a future primer.

To help motivate the data science-oriented Python programming examples provided in this primer, we will start off with a brief overview of basic concepts and terminology in data science.

## 2. Data Science: Basic Concepts¶

### Data Science and Data Mining¶

Foster Provost and Tom Fawcett offer succinct descriptions of data science and data mining in Data Science for Business:

Data science involves principles, processes and techniques for understanding phenomena via the (automated) analysis of data.

Data mining is the extraction of knowledge from data, via technologies that incorporate these principles.

### Knowledge Discovery, Data Mining and Machine Learning¶

Provost & Fawcett also offer some history and insights into the relationship between data mining and machine learning, terms which are often used somewhat interchangeably:

The field of Data Mining (or KDD: Knowledge Discovery and Data Mining) started as an offshoot of Machine Learning, and they remain closely linked. Both fields are concerned with the analysis of data to find useful or informative patterns. Techniques and algorithms are shared between the two; indeed, the areas are so closely related that researchers commonly participate in both communities and transition between them seamlessly. Nevertheless, it is worth pointing out some of the differences to give perspective.

Speaking generally, because Machine Learning is concerned with many types of performance improvement, it includes subfields such as robotics and computer vision that are not part of KDD. It also is concerned with issues of agency and cognition — how will an intelligent agent use learned knowledge to reason and act in its environment — which are not concerns of Data Mining.

Historically, KDD spun off from Machine Learning as a research field focused on concerns raised by examining real-world applications, and a decade and a half later the KDD community remains more concerned with applications than Machine Learning is. As such, research focused on commercial applications and business issues of data analysis tends to gravitate toward the KDD community rather than to Machine Learning. KDD also tends to be more concerned with the entire process of data analytics: data preparation, model learning, evaluation, and so on.

### Cross Industry Standard Process for Data Mining (CRISP-DM)¶

The Cross Industry Standard Process for Data Mining introduced a process model for data mining in 2000 that has become widely adopted.

The model emphasizes the iterative nature of the data mining process, distinguishing several different stages that are regularly revisited in the course of developing and deploying data-driven solutions to business problems:

• Data understanding
• Data preparation
• Modeling
• Deployment

We will be focusing primarily on using Python for data preparation and modeling.

### Data Science Workflow¶

Philip Guo presents a Data Science Workflow offering a slightly different process model emhasizing the importance of reflection and some of the meta-data, data management and bookkeeping challenges that typically arise in the data science process. His 2012 PhD thesis, Software Tools to Facilitate Research Programming, offers an insightful and more comprehensive description of many of these challenges.

Provost & Fawcett list a number of different tasks in which data science techniques are employed:

• Classification and class probability estimation
• Regression (aka value estimation)
• Similarity matching
• Clustering
• Co-occurrence grouping (aka frequent itemset mining, association rule discovery, market-basket analysis)
• Profiling (aka behavior description, fraud / anomaly detection)
• Data reduction
• Causal modeling

We will be focusing primarily on classification and class probability estimation tasks, which are defined by Provost & Fawcett as follows:

Classification and class probability estimation attempt to predict, for each individual in a population, which of a (small) set of classes this individual belongs to. Usually the classes are mutually exclusive. An example classification question would be: “Among all the customers of MegaTelCo, which are likely to respond to a given offer?” In this example the two classes could be called will respond and will not respond.

To further simplify this primer, we will focus exclusively on supervised methods, in which the data is explicitly labeled with classes. There are also unsupervised methods that involve working with data in which there are no pre-specified class labels.

### Supervised Classification¶

The Natural Language Toolkit (NLTK) book provides a diagram and succinct description (below, with italics and bold added for emphasis) of supervised classification:

Supervised Classification. (a) During training, a feature extractor is used to convert each input value to a feature set. These feature sets, which capture the basic information about each input that should be used to classify it, are discussed in the next section. Pairs of feature sets and labels are fed into the machine learning algorithm to generate a model. (b) During prediction, the same feature extractor is used to convert unseen inputs to feature sets. These feature sets are then fed into the model, which generates predicted labels.

### Data Mining Terminology¶

• Structured data has simple, well-defined patterns (e.g., a table or graph)
• Unstructured data has less well-defined patterns (e.g., text, images)
• Model: a pattern that captures / generalizes regularities in data (e.g., an equation, set of rules, decision tree)
• Attribute (aka variable, feature, signal, column): an element used in a model
• Example (aka instance, feature vector, row): a representation of an entity being modeled
• Target attribute (aka dependent variable, class label): the class / type / category of an entity being modeled

### Data Mining Example: UCI Mushroom dataset¶

The Center for Machine Learning and Intelligent Systems at the University of California, Irvine (UCI), hosts a Machine Learning Repository containing over 200 publicly available data sets.

We will use the mushroom data set, which forms the basis of several examples in Chapter 3 of the Provost & Fawcett data science book.

The following description of the dataset is provided at the UCI repository:

This data set includes descriptions of hypothetical samples corresponding to 23 species of gilled mushrooms in the Agaricus and Lepiota Family (pp. 500-525 [The Audubon Society Field Guide to North American Mushrooms, 1981]). Each species is identified as definitely edible, definitely poisonous, or of unknown edibility and not recommended. This latter class was combined with the poisonous one. The Guide clearly states that there is no simple rule for determining the edibility of a mushroom; no rule like leaflets three, let it be'' for Poisonous Oak and Ivy.

Number of Instances: 8124

Number of Attributes: 22 (all nominally valued)

Attribute Information: (classes: edible=e, poisonous=p)

1. cap-shape: bell=b, conical=c, convex=x, flat=f, knobbed=k, sunken=s
2. cap-surface: fibrous=f, grooves=g, scaly=y, smooth=s
3. cap-color: brown=n ,buff=b, cinnamon=c, gray=g, green=r, pink=p, purple=u, red=e, white=w, yellow=y
4. bruises?: bruises=t, no=f
5. odor: almond=a, anise=l, creosote=c, fishy=y, foul=f, musty=m, none=n, pungent=p, spicy=s
6. gill-attachment: attached=a, descending=d, free=f, notched=n
7. gill-spacing: close=c, crowded=w, distant=d
9. gill-color: black=k, brown=n, buff=b, chocolate=h, gray=g, green=r, orange=o, pink=p, purple=u, red=e, white=w, yellow=y
10. stalk-shape: enlarging=e, tapering=t
11. stalk-root: bulbous=b, club=c, cup=u, equal=e, rhizomorphs=z, rooted=r, missing=?
12. stalk-surface-above-ring: fibrous=f, scaly=y, silky=k, smooth=s
13. stalk-surface-below-ring: fibrous=f, scaly=y, silky=k, smooth=s
14. stalk-color-above-ring: brown=n, buff=b, cinnamon=c, gray=g, orange=o, pink=p, red=e, white=w, yellow=y
15. stalk-color-below-ring: brown=n, buff=b, cinnamon=c, gray=g, orange=o, pink=p, red=e, white=w, yellow=y
16. veil-type: partial=p, universal=u
17. veil-color: brown=n, orange=o, white=w, yellow=y
18. ring-number: none=n, one=o, two=t
19. ring-type: cobwebby=c, evanescent=e, flaring=f, large=l, none=n, pendant=p, sheathing=s, zone=z
20. spore-print-color: black=k, brown=n, buff=b, chocolate=h, green=r, orange=o, purple=u, white=w, yellow=y
21. population: abundant=a, clustered=c, numerous=n, scattered=s, several=v, solitary=y
22. habitat: grasses=g, leaves=l, meadows=m, paths=p, urban=u, waste=w, woods=d

Missing Attribute Values: 2480 of them (denoted by "?"), all for attribute #11.

Class Distribution: -- edible: 4208 (51.8%) -- poisonous: 3916 (48.2%) -- total: 8124 instances

The data file associated with this dataset has one instance of a hypothetical mushroom per line, with abbreviations for the values of the class and each of the other 22 attributes separated by commas.

Here is a sample line from the data file:

p,k,f,n,f,n,f,c,n,w,e,?,k,y,w,n,p,w,o,e,w,v,d

This instance represents a mushroom with the following attribute values:

class: edible=e, poisonous=p

1. cap-shape: bell=b, conical=c, convex=x, flat=f, knobbed=k, sunken=s
2. cap-surface: fibrous=f, grooves=g, scaly=y, smooth=s
3. cap-color: brown=n ,buff=b, cinnamon=c, gray=g, green=r, pink=p, purple=u, red=e, white=w, yellow=y
4. bruises?: bruises=t, no=f
5. odor: almond=a, anise=l, creosote=c, fishy=y, foul=f, musty=m, none=n, pungent=p, spicy=s
6. gill-attachment: attached=a, descending=d, free=f, notched=n
7. gill-spacing: close=c, crowded=w, distant=d
9. gill-color: black=k, brown=n, buff=b, chocolate=h, gray=g, green=r, orange=o, pink=p, purple=u, red=e, white=w, yellow=y
10. stalk-shape: enlarging=e, tapering=t
11. stalk-root: bulbous=b, club=c, cup=u, equal=e, rhizomorphs=z, rooted=r, missing=?
12. stalk-surface-above-ring: fibrous=f, scaly=y, silky=k, smooth=s
13. stalk-surface-below-ring: fibrous=f, scaly=y, silky=k, smooth=s
14. stalk-color-above-ring: brown=n, buff=b, cinnamon=c, gray=g, orange=o, pink=p, red=e, white=w, yellow=y
15. stalk-color-below-ring: brown=n, buff=b, cinnamon=c, gray=g, orange=o, pink=p, red=e, white=w, yellow=y
16. veil-type: partial=p, universal=u
17. veil-color: brown=n, orange=o, white=w, yellow=y
18. ring-number: none=n, one=o, two=t
19. ring-type: cobwebby=c, evanescent=e, flaring=f, large=l, none=n, pendant=p, sheathing=s, zone=z
20. spore-print-color: black=k, brown=n, buff=b, chocolate=h, green=r, orange=o, purple=u, white=w, yellow=y
21. population: abundant=a, clustered=c, numerous=n, scattered=s, several=v, solitary=y
22. habitat: grasses=g, leaves=l, meadows=m, paths=p, urban=u, waste=w, woods=d

Building a model with this data set will serve as a motivating example throughout much of this primer.

## 3. Python: Basic Concepts¶

### A note on Python 2 vs. Python 3¶

There are 2 major versions of Python in widespread use: Python 2 and Python 3. Python 3 has some features that are not backward compatible with Python 2, and some Python 2 libraries have not been updated to work with Python 3. I have been using Python 2, primarily because I use some of those Python 2[-only] libraries, but an increasing proportion of them are migrating to Python 3, and I anticipate shifting to Python 3 in the near future.

For more on the topic, I recommend a very well documented IPython Notebook, which includes numerous helpful examples and links, by Sebastian Raschka, Key differences between Python 2.7.x and Python 3.x, the Cheat Sheet: Writing Python 2-3 compatible code by Ed Schofield ... or googling Python 2 vs 3.

Nick Coghlan, a CPython core developer, sent me an email suggesting that relatively minor changes in this notebook would enable it to run with Python 2 or Python 3: importing the print_function from the __future__ module, and changing my print statements (Python 2) to print function calls (Python 3). Although a relatively minor conceptual change, it necessitated the changing of many individual cells to reflect the Python 3 print syntax. I also needed to replace a few functions that are no longer available in Python 3 with related functions that are available in both versions; I've added notes in nearby cells where the incompatible functions were removed explaining why they are related ... and no longer available.

In [1]:
from __future__ import print_function, division


### Identifiers, strings, lists and tuples¶

The sample instance of a mushroom shown above can be represented as a string. A Python string (str) is a sequence of 0 or more characters enclosed within a pair of single quotes (') or a pair double quotes (").

In [3]:
'p,k,f,n,f,n,f,c,n,w,e,?,k,y,w,n,p,w,o,e,w,v,d'

Out[3]:
'p,k,f,n,f,n,f,c,n,w,e,?,k,y,w,n,p,w,o,e,w,v,d'

Python identifiers (or names) are composed of letters, numbers and/or underscores ('_'), starting with a letter or underscore. Python identifiers are case sensitive. Although camelCase identifiers can be used, it is generally considered more pythonic to use underscores. Python variables and functions typically start with lowercase letters; Python classes start with uppercase letters.

The following assignment statement binds the value of the string shown above to the name single_instance_str.

In [4]:
single_instance_str = 'p,k,f,n,f,n,f,c,n,w,e,?,k,y,w,n,p,w,o,e,w,v,d'


The print function writes the value of its comma-delimited arguments to sys.stdout (typically the console). Each value in the output is separated by a single blank space. If an end argument that does not include \n (newline character) is supplied, the output cursor will not move to the next line.

In [5]:
print('Instance 1:', single_instance_str)
print('A', 'B', end=' ') # use a space rather than newline at the end of the line
print('C') # will appear on same line

Instance 1: p,k,f,n,f,n,f,c,n,w,e,?,k,y,w,n,p,w,o,e,w,v,d
A B C


The Python comment character is '#': anything after '#' on the line is ignored by the Python interpreter.

Pairs of triple quotes (''' or """) can be used to delimit multi-line comments.

In [6]:
'''
A multi-line
comment
'''
print('no comment')

no comment


A list is an ordered sequence of 0 or more comma-delimited elements enclosed within square brackets ('[', ']'). The Python str.split(sep) method can be used to split a sep-delimited string into a corresponding list of elements.

In [7]:
single_instance_list = single_instance_str.split(',')
print(single_instance_list)

['p', 'k', 'f', 'n', 'f', 'n', 'f', 'c', 'n', 'w', 'e', '?', 'k', 'y', 'w', 'n', 'p', 'w', 'o', 'e', 'w', 'v', 'd']


Python sequences are heterogeneous, i.e., they can contain elements of different types.

In [8]:
mixed_list = ['a', 1, 2.3, True, [1, 'b']]
print(mixed_list)

['a', 1, 2.3, True, [1, 'b']]


The Python + operator can be used to concatenate lists.

In [9]:
concatenated_list = ['a', 1] + [2.3, True] + [[1, 'b']]
print(concatenated_list)

['a', 1, 2.3, True, [1, 'b']]


Individual elements of sequences (lists, strings and other data structures) can be accessed by specifying their zero-based index position within square brackets ('[', ']').

In [10]:
print(single_instance_str[2], single_instance_list[2])

k f


Negative index values can be used to specify a position offset from the end of the sequence. It is often useful to use a -1 index value to access the last element of a sequence.

In [11]:
print(single_instance_str[-1], single_instance_list[-1])

d d


The Python slice notation can be used to access subsequences by specifying two index positions separated by a colon (':'); seq[start:stop] returns all the elements in seq between start and stop - 1 (inclusive).

In [12]:
print(single_instance_str[2:4])
print(single_instance_list[2:4])

k,
['f', 'n']


Slices indices can be negative values.

In [13]:
print(single_instance_str[-4:-2])
print(single_instance_list[-4:-2])

,v
['e', 'w']


The start and/or stop index can be omitted. A common use of slices with a single index value is to access all but the first element or all but the last element of a sequence.

In [14]:
print(single_instance_str[:-1]) # all but the last
print(single_instance_list[:-1])
print(single_instance_str[1:]) # all but the first
print(single_instance_list[1:])

p,k,f,n,f,n,f,c,n,w,e,?,k,y,w,n,p,w,o,e,w,v,
['p', 'k', 'f', 'n', 'f', 'n', 'f', 'c', 'n', 'w', 'e', '?', 'k', 'y', 'w', 'n', 'p', 'w', 'o', 'e', 'w', 'v']
,k,f,n,f,n,f,c,n,w,e,?,k,y,w,n,p,w,o,e,w,v,d
['k', 'f', 'n', 'f', 'n', 'f', 'c', 'n', 'w', 'e', '?', 'k', 'y', 'w', 'n', 'p', 'w', 'o', 'e', 'w', 'v', 'd']


Slice notation includes an optional third element, step, as in seq[start:stop:step], that specifies the steps or increments by which elements are retrieved from seq between start and step - 1:

In [15]:
print(single_instance_str)
print(single_instance_str[::2]) # print elements in even-numbered positions (the values, in this case)
print(single_instance_str[1::2]) # print elements in odd-numbered positions (the commas, in this case)
print(single_instance_str[::-1]) # reverse the string

p,k,f,n,f,n,f,c,n,w,e,?,k,y,w,n,p,w,o,e,w,v,d
pkfnfnfcnwe?kywnpwoewvd
,,,,,,,,,,,,,,,,,,,,,,
d,v,w,e,o,w,p,n,w,y,k,?,e,w,n,c,f,n,f,n,f,k,p


The Python tutorial offers a helpful ASCII art representation to show how positive and negative indexes are interpreted:

 +---+---+---+---+---+
| H | e | l | p | A |
+---+---+---+---+---+
0   1   2   3   4   5
-5  -4  -3  -2  -1


Python statements are typically separated by newlines (rather than, say, the semi-colon in Java). Statements can extend over more than one line; it is generally best to break the lines after commas within parentheses, braces or brackets. Inserting a backslash character ('\') at the end of a line will also enable continuation of the statement on the next line, but it is generally best to look for other alternatives.

In [16]:
attribute_names = ['class',
'cap-shape', 'cap-surface', 'cap-color',
'bruises?',
'odor',
'gill-attachment', 'gill-spacing', 'gill-size', 'gill-color',
'stalk-shape', 'stalk-root',
'stalk-surface-above-ring', 'stalk-surface-below-ring',
'stalk-color-above-ring', 'stalk-color-below-ring',
'veil-type', 'veil-color',
'ring-number', 'ring-type',
'spore-print-color',
'population',
'habitat']
print(attribute_names)

['class', 'cap-shape', 'cap-surface', 'cap-color', 'bruises?', 'odor', 'gill-attachment', 'gill-spacing', 'gill-size', 'gill-color', 'stalk-shape', 'stalk-root', 'stalk-surface-above-ring', 'stalk-surface-below-ring', 'stalk-color-above-ring', 'stalk-color-below-ring', 'veil-type', 'veil-color', 'ring-number', 'ring-type', 'spore-print-color', 'population', 'habitat']


The str.strip([chars]) method returns a copy of str in which any leading or trailing chars are removed. If no chars are specified, it removes all leading and trailing whitespace. [Whitespace is any sequence of spaces, tabs ('\t') and/or newline ('\n') characters.]

Note that since a blank space is inserted in the output after every item in a comma-delimited list, the last asterisk is printed after a leading blank space is inserted on the new line.

In [17]:
print('*', '\tp,k,f,n,f,n,f,c,n,w,e,?,k,y,w,n,p,w,o,e,w,v,d\n', '*')

* 	p,k,f,n,f,n,f,c,n,w,e,?,k,y,w,n,p,w,o,e,w,v,d
*

In [18]:
print('*', '\tp,k,f,n,f,n,f,c,n,w,e,?,k,y,w,n,p,w,o,e,w,v,d\n'.strip(), '*')

* p,k,f,n,f,n,f,c,n,w,e,?,k,y,w,n,p,w,o,e,w,v,d *


A common programming pattern when dealing with CSV (comma-separated values) data files containing is to repeatedly

1. read a line from a file
2. strip off any leading and trailing whitespace
3. split the values separated by commas into a list

We will get to repetition control structures (loops) and file input and output shortly, but here is an example of how str.strip() and str.split() be chained together in a single instruction:

In [19]:
single_instance_str = '\tp,k,f,n,f,n,f,c,n,w,e,?,k,y,w,n,p,w,o,e,w,v,d\n'
single_instance_list = single_instance_str.strip().split(',') # first strip leading & trailing whitespace, then split on commas
print(single_instance_list)

['p', 'k', 'f', 'n', 'f', 'n', 'f', 'c', 'n', 'w', 'e', '?', 'k', 'y', 'w', 'n', 'p', 'w', 'o', 'e', 'w', 'v', 'd']


The str.join(words) method is the inverse of str.split(), returning a single string in which each string in the sequence of words is separated by str.

In [20]:
print('*', ','.join(single_instance_list), '*')

* p,k,f,n,f,n,f,c,n,w,e,?,k,y,w,n,p,w,o,e,w,v,d *


A number of Python methods can be used on strings, lists and other sequences.

The len(s) function can be used to find the length of (number of items in) a sequence s. It will also return the number of items in a dictionary, a data structure we will cover further below.

In [21]:
print(len(single_instance_str), len(single_instance_list))

47 23


The in operator can be used to determine whether a sequence contains a value.

Boolean values in Python are True and False (note the capitalization).

In [22]:
print(',' in single_instance_str, ',' in single_instance_list)

True False


The s.count(x) ormethod can be used to count the number of occurrences of item x in sequence s.

In [23]:
print(single_instance_str.count(','), single_instance_list.count('f'))

22 3


The s.index(x) method can be used to find the first 0-based index of item x in sequence s.

In [24]:
print(single_instance_str.index(','), single_instance_list.index('f'))

2 2


One important distinction between strings and lists has to do with their mutability.

Python strings are immutable, i.e., they cannot be modified. Most string methods (like str.strip()) return modified copies of the strings on which they are used.

Python lists are mutable, i.e., they can be modified.

The examples below illustrate a number of list methods that modify lists.

In [25]:
list_1 = [4, 2, 3, 5, 1]
list_2 = list_1 # list_2 now references the same object as list_1
print('list_1:          ', list_1)
print('list_2:          ', list_2)
list_1.remove(1)
print('list_1.remove(1):', list_1)
list_1.append(6)
print('list_1.append(6):', list_1)
list_1.sort()
print('list_1.sort():   ', list_1)
list_1.reverse()
print('list_1.reverse():', list_1)

list_1:           [4, 2, 3, 5, 1]
list_2:           [4, 2, 3, 5, 1]
list_1.remove(1): [4, 2, 3, 5]
list_1.append(6): [4, 2, 3, 5, 6]
list_1.sort():    [2, 3, 4, 5, 6]
list_1.reverse(): [6, 5, 4, 3, 2]


When more than one name (e.g., a variable) is bound to the same mutable object, changes made to that object are reflected in all names bound to that object. For example, in the second statement above, list_2 is bound to the same object that is bound to list_1, namely, the list [4, 2, 3, 5 1]. All changes made to the object bound to list_1 will thus be reflected in list_2 (since they both reference the same object).

In [26]:
print('list_1:          ', list_1)
print('list_2:          ', list_2)

list_1:           [6, 5, 4, 3, 2]
list_2:           [6, 5, 4, 3, 2]


There are sorting and reversing functions, sorted() and reversed(), that do not modify their arguments, and can thus be used on mutable or immutable objects. We will elaborate on each of these functions further below, but here are a couple of examples of how sorted() returns a sorted list of each element in its argument.

In [27]:
print('sorted(list_1):', sorted(list_1)) # return a copy of list_1 in sorted order
print('list_1:        ', list_1)
print('sorted(single_instance_str):', sorted(single_instance_str)) # returns a list of sorted elements in the string
print('single_instance_str:        ', single_instance_str)

sorted(list_1): [2, 3, 4, 5, 6]
list_1:         [6, 5, 4, 3, 2]
sorted(single_instance_str): ['\t', '\n', ',', ',', ',', ',', ',', ',', ',', ',', ',', ',', ',', ',', ',', ',', ',', ',', ',', ',', ',', ',', ',', ',', '?', 'c', 'd', 'e', 'e', 'f', 'f', 'f', 'k', 'k', 'n', 'n', 'n', 'n', 'o', 'p', 'p', 'v', 'w', 'w', 'w', 'w', 'y']
single_instance_str:         	p,k,f,n,f,n,f,c,n,w,e,?,k,y,w,n,p,w,o,e,w,v,d



A tuple is an ordered, immutable sequence of 0 or more comma-delimited values enclosed in parentheses ('(', ')'). Many of the functions that operate on strings and lists also operate on tuples.

In [28]:
x = (1, 2, 3, 4, 5) # a tuple
print('x =', x, ', len(x) =', len(x), ', x.index(3) =', x.index(3), ', x[4:2:-1] = ', x[4:2:-1])
print('sorted(x, reverse=True):', sorted(x, reverse=True)) # sorted always returns a list; reverse=True specifies reverse sort order

x = (1, 2, 3, 4, 5) , len(x) = 5 , x.index(3) = 2 , x[4:2:-1] =  (5, 4)
sorted(x, reverse=True): [5, 4, 3, 2, 1]


If the s.index(x) or list.remove(x) method is used on a sequence s or list that does not contain the value x, a ValueError exception is raised.

In [29]:
print(x.index(6)) # a ValueError will be raised

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-29-553ddc490596> in <module>()
----> 1 print(x.index(6)) # a ValueError will be raised

ValueError: tuple.index(x): x not in tuple

### Conditionals¶

One common approach to handling errors is to look before you leap (LBYL), i.e., test for potential exceptions before executing instructions that might raise those exceptions.

This approach can be implemented using the if statement (which may optionally include an else and any number of elif clauses).

The following is a simple example of an if statement:

In [30]:
class_value = 'e' # try changing this to 'p' or 'x'

if class_value == 'e':
print('edible')
elif class_value == 'p':
print('poisonous')
else:
print('unknown')

edible


Note that

• a colon (':') is used at the end of the lines with if, else or elif
• no parentheses are required to enclose the boolean condition (it is presumed to include everything between if or elif and the colon)
• the statements below each if, elif and else line are all indented

Python does not have special characters to delimit statement blocks (like the '{' and '}' delimiters in Java); instead, sequences of statements with the same indentation level are treated as a statement block. The Python Style Guide recommends using 4 spaces for each indentation level.

An if statement can be used to follow the LBYL paradigm in preventing the ValueError that occured in an earlier example:

In [31]:
attribute = 'bruises' # try substituting 'bruises?' for 'bruises' and re-running this code

if attribute in attribute_names:
i = attribute_names.index(attribute)
print(attribute, 'is in position', i)
else:
print(attribute, 'is not in', attribute_names)

bruises is not in ['class', 'cap-shape', 'cap-surface', 'cap-color', 'bruises?', 'odor', 'gill-attachment', 'gill-spacing', 'gill-size', 'gill-color', 'stalk-shape', 'stalk-root', 'stalk-surface-above-ring', 'stalk-surface-below-ring', 'stalk-color-above-ring', 'stalk-color-below-ring', 'veil-type', 'veil-color', 'ring-number', 'ring-type', 'spore-print-color', 'population', 'habitat']


Another perspective on handling errors championed by some pythonistas is that it is easier to ask forgiveness than permission (EAFP).

As in many practical applications of philosophy, religion or dogma, it is helpful to think before you choose (TBYC). There are a number of factors to consider in deciding whether to follow the EAFP or LBYL paradigm, including code readability and the anticipated likelihood and relative severity of encountering an exception. Oran Looney wrote a blog post providing a nice overview of the debate over LBYL vs. EAFP.

We will follow the LBYL paradigm throughout most of this primer. However, as an illustration of EAFP in Python, here is an alternate implementation of the functionality of the code above, using a try/except statement.

In [32]:
attribute = 'bruises' # try substituting 'bruises?' for 'bruises' and re-running this code

try:
i = attribute_names.index(attribute)
print(attribute, 'is in position', i)
except ValueError:

bruises is not found


The Python null object is None (note the capitalization).

In [33]:
attribute = 'bruises?'

if attribute not in attribute_names: # equivalent to 'not attribute in attribute_names'
value = None
else:
i = attribute_names.index(attribute)
value = single_instance_list[i]

print(attribute, '=', value)

bruises? = f


### Defining and calling functions¶

Python function definitions start with the def keyword followed by a function name, a list of 0 or more comma-delimited parameters (aka 'formal parameters') enclosed within parentheses, and then a colon (':').

A function definition may include one or more return statemens to indicate the value(s) returned to where the function is called. It is good practice to include a short docstring to briefly describe the behavior of the function and the value(s) it returns.

In [34]:
def attribute_value(instance, attribute, attribute_names):
'''Returns the value of attribute in instance, based on the position of attribute in the list of attribute_names'''
if attribute not in attribute_names:
return None
else:
i = attribute_names.index(attribute)
return instance[i] # using the parameter name here


A function call starts with the function name, followed by a list of 0 or more comma-delimited arguments (aka 'actual parameters') enclosed within parentheses. A function call can be used as a statement or within an expression.

In [35]:
attribute = 'cap-shape' # try substituting any of the other attribute names shown above
print(attribute, '=', attribute_value(single_instance_list, attribute, attribute_names))

cap-shape = k


Note that Python does not distinguish between names used for variables and names used for functions. An assignment statement binds a value to a name; a function definition also binds a value to a name. At any given time, the value most recently bound to a name is the one that is used.

The type(object) function returns the type of object.

In [36]:
x = 0
print('x used as a variable:', x, type(x))

def x():
print('x')

print('x used as a function:', x, type(x))

x used as a variable: 0 <type 'int'>
x used as a function: <function x at 0x1050076e0> <type 'function'>


Another way to determine the type of an object is to use isinstance(object, class). This is generally preferable, as it takes into account class inheritance. There is a larger issue of duck typing, and whether code should ever explicitly check for the type of an object, but we will omit further discussion of the topic in this notebook.

Checking whether an object is a function type will require the use of the types library ... and more thorough checking could be done if one wants to include built-in as well as user-defined functions.

In [37]:
import types

x = 0
print('Is x an int?', isinstance(x, int))
print('Is x a function?', isinstance(x, types.FunctionType))

def x():
print('x')

print('Is x an int?', isinstance(x, int))
print('Is x a function?', isinstance(x, types.FunctionType))

Is x an int? True
Is x a function? False
Is x an int? False
Is x a function? True


Another important feature of Python functions is that arguments are passed using call by sharing. If a mutable object is passed as an argument to a function parameter, assignment statements using that parameter do not affect the passed argument, however mutations to the parameter do affect the passed argument.

Not being aware of - or forgetting - this important distinction can lead to challenging debugging sessions.

In [38]:
def modify_parameters(parameter1, parameter2):
'''Inserts "x" at the head of parameter1, assigns "x" to parameter2'''
parameter1.insert(0, 'x')
print('parameter1, after inserting "x":', parameter1)
parameter2 = 'x'
print('parameter2, after assigning "x"', parameter2)
return

argument1 = [1, 2, 3] # passing a named object will affect the object bound to that name
argument2 = 4
print('argument1, before calling modify_parameters:', argument1)
print('argument2, before calling modify_parameters:', argument2)
print()
modify_parameters(argument1, argument2)
print()
print('argument1, after calling modify_parameters:', argument1)
print('argument2, after calling modify_parameters:', argument2)

argument1, before calling modify_parameters: [1, 2, 3]
argument2, before calling modify_parameters: 4

parameter1, after inserting "x": ['x', 1, 2, 3]
parameter2, after assigning "x" x

argument1, after calling modify_parameters: ['x', 1, 2, 3]
argument2, after calling modify_parameters: 4


One way of preventing functions from modifying mutable objects passed as parameters is to make a copy of those objects inside the function. Here is another version of the function above that makes a shallow copy of the list_parameter using the slice operator.

[Note: the Python copy module provides both [shallow] copy() and deepcopy() methods; we will cover modules further below.]

In [39]:
def modify_parameter_copy(parameter_1):
'''Inserts "x" at the head of parameter_1, without modifying the list argument'''
parameter_1_copy = parameter_1[:]
parameter_1_copy.insert(0, 'x')
print('Inserted x:', parameter_1_copy)
return

argument_1 = [1, 2, 3] # passing a named object will not affect the object bound to that name
print('Before:', argument_1)
modify_parameter_copy(argument_1)
print('After:', argument_1)

Before: [1, 2, 3]
Inserted x: ['x', 1, 2, 3]
After: [1, 2, 3]


Python functions can return more than one value, by separating those return values with commas in the return statement. Multiple values are returned as a tuple. If the function-invoking expression is an assignment statement, multiple variables can be assigned the multiple values returned by the function in a single statement. This combining of values and subsequent separation is known as tuple packing and unpacking.

In [40]:
def min_and_max(list_of_values):
'''Returns a tuple containing the min and max values in the list_of_values'''
return min(list_of_values), max(list_of_values)

list_1 = [3, 1, 4, 2, 5]
print('min and max of', list_1, ':', min_and_max(list_1))

min_and_max_list_1 = min_and_max(list_1) # a single variable is assigned the two-element tuple
print('min and max of', list_1, ':', min_and_max_list_1)

min_list_1, max_list_1 = min_and_max(list_1) # the 1st variable is assigned the 1st value, the 2nd variable is assigned the 2nd value
print('min and max of', list_1, ':', min_list_1, ',', max_list_1)

min and max of [3, 1, 4, 2, 5] : (1, 5)
min and max of [3, 1, 4, 2, 5] : (1, 5)
min and max of [3, 1, 4, 2, 5] : 1 , 5


### Iteration: for, range¶

The for statement iterates over the elements of a sequence.

In [41]:
for i in [0, 1, 2]:
print(i)

0
1
2

In [42]:
for c in 'abc':
print(c)

a
b
c


In Python 2, the range(stop) function returns a list of values from 0 up to stop - 1 (inclusive). It is often used in the context of a for loop that iterates over the list of values.

In [43]:
print('Values for the', len(attribute_names), 'attributes:', end='\n\n')
for i in range(len(attribute_names)):
print(attribute_names[i], '=',
attribute_value(single_instance_list, attribute_names[i], attribute_names))

Values for the 23 attributes:

class = p
cap-shape = k
cap-surface = f
cap-color = n
bruises? = f
odor = n
gill-attachment = f
gill-spacing = c
gill-size = n
gill-color = w
stalk-shape = e
stalk-root = ?
stalk-surface-above-ring = k
stalk-surface-below-ring = y
stalk-color-above-ring = w
stalk-color-below-ring = n
veil-type = p
veil-color = w
ring-number = o
ring-type = e
spore-print-color = w
population = v
habitat = d


The more general form of the function, range(start, stop[, step]), returns a list of values from start to stop - 1 (inclusive) increasing by step (which defaults to 1), or from start down to stop + 1 (inclusive) decreasing by step if step is negative.

In [44]:
for i in range(3, 0, -1):
print(i)

3
2
1


In Python 2, the xrange(stop[, stop[, step]]) function is an iterable version of the range() function. In the context of a for loop, it returns the next item of the sequence for each iteration of the loop rather than creating all the elements of the sequence before the first iteration. This can reduce memory consumption in cases where iteration over all the items is not required.

In Python 3, the range() function behaves the same way as the xrange() function does in Python 2, and so the xrange() function is not defined in Python 3.

To maximize compatibility, we will use range() throughout this notebook; however, note that it is generally advisable to use xrange() rather than range() wherever possible in Python 2.

### Modules, namespaces and dotted notation¶

A Python module is a file containing related definitions (e.g., of functions and variables). Modules are used to help organize the Python namespaces, the set of identifiers accessible in a particular contexts. All of the functions and variables we define in this IPython Notebook are in the __main__ namespace, so accessing them does not require any specification of a module.

A Python module named simple_ml (in the file simple_ml.py), contains a set of solutions to the exercises in this IPython Notebook. Accessing functions in that module requires that we first import the module, and then prefix the function names with the module name followed by a dot (this is known as dotted notation).

For example, the following function call Exercise 1 below:

simple_ml.print_attribute_names_and_values(single_instance_list, attribute_names)

uses dotted notation to reference the print_attribute_names_and_values() function in the simple_ml module.

After you have defined your function in Exercise 1, you can test it by deleting the simple_ml module specification, so that the statement becomes

print_attribute_names_and_values(single_instance_list, attribute_names)

This will reference the print_attribute_names_and_values() function in the current namespace (__main__), i.e., the top-level interpreter environment. The simple_ml.print_attribute_names_and_values() function will still be accessible in the simple_ml namespace by using the "simple_ml." prefix.

### Exercise 1: define print_attribute_names_and_values()¶

Complete the following function definition, print_attribute_names_and_values(instance, attribute_names), so that it generates exactly the same output as the code above.

In [45]:
def print_attribute_names_and_values(instance, attribute_names):
'''Prints the attribute names and values for an instance'''
return

import simple_ml # this module contains my solutions to exercises

# to test your function, delete the 'simple_ml.' module specification in the call to print_attribute_names_and_values() below
simple_ml.print_attribute_names_and_values(single_instance_list, attribute_names)

Values for the 23 attributes:

class = p
cap-shape = k
cap-surface = f
cap-color = n
bruises? = f
odor = n
gill-attachment = f
gill-spacing = c
gill-size = n
gill-color = w
stalk-shape = e
stalk-root = ?
stalk-surface-above-ring = k
stalk-surface-below-ring = y
stalk-color-above-ring = w
stalk-color-below-ring = n
veil-type = p
veil-color = w
ring-number = o
ring-type = e
spore-print-color = w
population = v
habitat = d


### File I/O¶

Python file input and output is done through file objects. A file object is created with the open(name[, mode]) statement, where name is a string representing the name of the file, and mode is 'r' (read), 'w' (write) or 'a' (append); if no second argument is provided, the mode defaults to 'r'.

A common Python programming pattern for processing an input text file is to

The following code creates a list of instances, where each instance is a list of attribute values (like instance_1_str above).

In [46]:
all_instances = [] # initialize instances to an empty list
data_filename = 'agaricus-lepiota.data'

with open(data_filename, 'r') as f:
for line in f:
all_instances.append(line.strip().split(','))

print('First instance:', all_instances[0]) # we don't want to print all the instances, so let's just print one to verify

Read 8124 instances from agaricus-lepiota.data
First instance: ['p', 'x', 's', 'n', 't', 'p', 'f', 'c', 'n', 'k', 'e', 'e', 's', 's', 'w', 'w', 'p', 'w', 'o', 'p', 'k', 's', 'u']


Define a function, load_instances(filename), that returns a list of instances in a text file. The function definition is started for you below. The function should exhibit the same behavior as the code above.

In [47]:
def load_instances(filename):
'''Returns a list of instances stored in a file.

filename is expected to have a series of comma-separated attribute values per line, e.g.,
p,k,f,n,f,n,f,c,n,w,e,?,k,y,w,n,p,w,o,e,w,v,d'''
instances = []
return instances

data_filename = 'agaricus-lepiota.data'
# to test your function, delete the 'simple_ml.' module specification in the call to load_instances() below
print('First instance:', all_instances_2[0])

Read 8124 instances from agaricus-lepiota.data
First instance: ['p', 'x', 's', 'n', 't', 'p', 'f', 'c', 'n', 'k', 'e', 'e', 's', 's', 'w', 'w', 'p', 'w', 'o', 'p', 'k', 's', 'u']


Output to text file is usually done via file.write(str) method.

As we saw earlier, the str.join(words) method returns a single str-delimited string containing each of the strings in the list words.

SQL and Hive database tables often use the pipe ('|') delimiter to separate column values for each row when they are stored as flat files. The following code creates a new data file using pipes rather than commas to separate the attribute values.

To help maintain internal consistency, it is generally a good practice to define a variable such as delimiter or separator and bind it to the intended delimiter string, and then use the variable throughout.

In [48]:
delimiter = '|'

print('Converting to {}-delimited strings, e.g.,'.format(delimiter), delimiter.join(all_instances[0]))

datafile2 = 'agaricus-lepiota-2.data'
with open(datafile2, 'w') as f:
for instance in all_instances:
f.write('|'.join(instance) + '\n') # '+' is the concatenation operator when used with strings

all_instances_3 = []
with open(datafile2, 'r') as f:
for line in f:
all_instances_3.append(line.strip().split(delimiter)) # note: changed ',' to '|'
print('First instance:', all_instances_3[0]) # we don't want to print all the instances, so let's just print one to verify

Converting to |-delimited strings, e.g., p|x|s|n|t|p|f|c|n|k|e|e|s|s|w|w|p|w|o|p|k|s|u
First instance: ['p', 'x', 's', 'n', 't', 'p', 'f', 'c', 'n', 'k', 'e', 'e', 's', 's', 'w', 'w', 'p', 'w', 'o', 'p', 'k', 's', 'u']


### List comprehensions¶

Python provides a powerful list comprehension construct to simplify the creation of a list by specifying a formula in a single expression.

Some programmers find list comprehensions confusing, and avoid their use. We won't rely on list comprehensions here, but will show examples with and without list comprehensions below.

One common use of list comprehensions is in the context of the str.join(words) method we saw earlier.

If we wanted to construct a pipe-delimited string containing elements of the list, we could use a for loop to iteratively add list elements and pipe delimiters to a string. We would thereby add one pipe delimiter too many, and would thus have to shave that off at the end.

In [49]:
delimited_string = ''
for x in [1, 2, 3]:
delimited_string += str(x) + delimiter
delimited_string = delimited_string[:-1]
delimited_string

Out[49]:
'1|2|3'

This process is much simpler using a list comprehension.

In [50]:
delimiter.join([str(x) for x in [1, 2, 3]])

Out[50]:
'1|2|3'

As noted in the initial description of the UCI mushroom set above, 2480 of the 8124 instances have missing values (denoted by '?') for an attribute. There are several techniques for dealing with instances that include missing values, but to simplify things in the context of this primer - and following the example in the Data Science for Business book - we will restrict our focus to only those clean instances that have no missing values.

We could use several lines of code - with an if statement inside a for loop - to create a clean_instances list from the all_instances list. Or we could use a list comprehension.

We will show both approaches to creating clean_instances below.

In [51]:
# version 1: using an if statement nested within a for statement
unknown_value = '?'

clean_instances = []
for instance in all_instances:
if unknown_value not in instance:
clean_instances.append(instance)

print(len(clean_instances), 'clean instances')

5644 clean instances

In [52]:
# version 2: using an equivalent list comprehension
clean_instances_2 = [instance for instance in all_instances if unknown_value not in instance]

print(len(clean_instances_2), 'clean instances')

5644 clean instances


### Dictionaries (dicts)¶

Although single character abbreviations of attribute values (e.g., 'x') allow for more compact data files, they are not as easy to understand by human readers as the longer attribute value descriptions (e.g., 'convex').

A Python dictionary (or dict) is an unordered, comma-delimited collection of key, value pairs, serving a siimilar function as a hash table or hashmap in other programming languages.

We could create a dictionary for the cap-type attribute values shown above:

bell=b, conical=c, convex=x, flat=f, knobbed=k, sunken=s

Since we will want to look up the value using the abbreviation (which is the representation of the value stored in the file), we will use the abbreviations as keys and the descriptions as values.

A Python dictionary can be created by specifying all key: value pairs (with colons separating each key and value), or by adding them iteratively. We will use the first method below, and use the second method further below. A value in a Python dictionary (dict) is accessed by specifying its key using the general form dict[key].

In [53]:
attribute_values_cap_type = {'b': 'bell',
'c': 'conical',
'x': 'convex',
'f': 'flat',
'k': 'knobbed',
's': 'sunken'}

attribute_value_abbrev = 'x'
print(attribute_value_abbrev, '=', attribute_values_cap_type[attribute_value_abbrev])

x = convex


A Python dictionary is an iterable container, so we can iterate over the keys in a dictionary using a for loop.

Note that since a dictionary is an unordered collection, the sequence of abbreviations and associated values is not guaranteed to appear in any particular order.

In [54]:
for attribute_value_abbrev in attribute_values_cap_type:
print(attribute_value_abbrev, '=', attribute_values_cap_type[attribute_value_abbrev])

c = conical
b = bell
f = flat
k = knobbed
s = sunken
x = convex


Python supports dictionary comprehensions, which have a similar form as the list comprehensions described above.

For example, if we provisionally omit the 'convex' cap-type (whose abbreviation is the last letter rather than first letter in the attribute name), we could construct a dictionary of abbreviations and descriptions using the following expression.

In [55]:
attribute_values_cap_type_2 = {x[0]: x for x in ['bell', 'conical', 'flat', 'knobbed', 'sunken']}
print(attribute_values_cap_type_2)

{'s': 'sunken', 'c': 'conical', 'b': 'bell', 'k': 'knobbed', 'f': 'flat'}


While it's useful to have a dictionary of values for the cap-type attribute, it would be even more useful to have a dictionary of values for every attribute. Earlier, we created a list of attribute_names; let's expand this to create a list of attribute_values wherein each list element is a dictionary.

Rather than explicitly type in each dictionary entry in the Python interpreter, we'll define a function to read a file containing the list of attribute names, values and value abbreviations in the format shown above:

• class: edible=e, poisonous=p
• cap-shape: bell=b, conical=c, convex=x, flat=f, knobbed=k, sunken=s
• cap-surface: fibrous=f, grooves=g, scaly=y, smooth=s
• ...
In [56]:
def load_attribute_values(filename):
'''Returns a list of attribute values in a file.

The attribute values are represented as dictionaries, wherein the keys are abbreviations and the values are descriptions.
filename is expected to have one attribute name and set of values per line, with the following format:
name: value_description=value_abbreviation[,value_description=value_abbreviation]*
for example
cap-shape: bell=b, conical=c, convex=x, flat=f, knobbed=k, sunken=s
The attribute value description dictionary created from this line would be the following:
{'c': 'conical', 'b': 'bell', 'f': 'flat', 'k': 'knobbed', 's': 'sunken', 'x': 'convex'}'''
attribute_values = []
with open(filename) as f:
for line in f:
attribute_name_and_value_string_list = line.strip().split(':')
attribute_name = attribute_name_and_value_string_list[0]
if len(attribute_name_and_value_string_list) < 2:
attribute_values.append({}) # no values for this attribute
else:
value_abbreviation_description_dict = {}
description_and_abbreviation_string_list = attribute_name_and_value_string_list[1].strip().split(',')
for description_and_abbreviation_string in description_and_abbreviation_string_list:
description_and_abbreviation = description_and_abbreviation_string.strip().split('=')
description = description_and_abbreviation[0]
if len(description_and_abbreviation) < 2: # assumption: no more than 1 value is missing an abbreviation
value_abbreviation_description_dict[None] = description
else:
abbreviation = description_and_abbreviation[1]
value_abbreviation_description_dict[abbreviation] = description
attribute_values.append(value_abbreviation_description_dict)
return attribute_values

attribute_filename = 'agaricus-lepiota.attributes'
print('Read', len(attribute_values), 'attribute values from', attribute_filename)
print('First attribute values list:', attribute_values[0])

Read 23 attribute values from agaricus-lepiota.attributes
First attribute values list: {'p': 'poisonous', 'e': 'edible'}


We earlier created the attribute_names list manually. The load_attribute_values() function above creates the attribute_values list automatically from the contents of a file ... each line of which starts with the name of each attribute ... which we discard.

Complete the following function definition so that the code implements the functionality described in the docstring.

In [57]:
def load_attribute_names_and_values(filename):
'''Returns a list of attribute names and values in a file.

This list contains dictionaries wherein the keys are names
and the values are value description dictionariess.

Each value description sub-dictionary will use the attribute value abbreviations as its keys
and the attribute descriptions as the values.

filename is expected to have one attribute name and set of values per line, with the following format:
name: value_description=value_abbreviation[,value_description=value_abbreviation]*
for example
cap-shape: bell=b, conical=c, convex=x, flat=f, knobbed=k, sunken=s
The attribute name and values dictionary created from this line would be the following:
{'name': 'cap-shape', 'values': {'c': 'conical', 'b': 'bell', 'f': 'flat', 'k': 'knobbed', 's': 'sunken', 'x': 'convex'}}'''
attribute_names_and_values = [] # this will be a list of dicts
return attribute_names_and_values

attribute_filename = 'agaricus-lepiota.attributes'
# to test your function, delete the 'simple_ml.' module specification in the call to load_attribute_names_and_values() below
print('Read', len(attribute_names_and_values), 'attribute values from', attribute_filename)
print('First attribute name:', attribute_names_and_values[0]['name'],
'; values:', attribute_names_and_values[0]['values'])

Read 23 attribute values from agaricus-lepiota.attributes
First attribute name: class ; values: {'p': 'poisonous', 'e': 'edible'}


### Counting¶

Data scientists often need to count things. For example, we might want to count the numbers of edible and poisonous mushrooms in the clean_instances list we created earlier.

In [58]:
edible_count = 0
for instance in clean_instances:
if instance[0] == 'e':
edible_count += 1 # this is shorthand for edible_count = edible_count + 1

print('There are', edible_count, 'edible mushrooms among the',
len(clean_instances), 'clean instances')

There are 3488 edible mushrooms among the 5644 clean instances


More generally, we often want to count the number of occurrences (frequencies) of each possible value for an attribute. One way to do so is to create a dictionary where each dictionary key is an attribute value and each dictionary value is the count of instances with that attribute value.

Using an ordinary dictionary, we must be careful to create a new dictionary entry the first time we see a new attribute value (that is not already contained in the dictionary).

In [59]:
cap_state_value_counts = {}
for instance in clean_instances:
cap_state_value = instance[1] # cap-state is the 2nd attribute
if cap_state_value not in cap_state_value_counts:
cap_state_value_counts[cap_state_value] = 0
cap_state_value_counts[cap_state_value] += 1

print('Counts for each value of cap-state:')
for value in cap_state_value_counts:
print(value, ':', cap_state_value_counts[value])

Counts for each value of cap-state:
c : 4
b : 300
f : 2432
k : 36
s : 32
x : 2840


The Python collections module provides a number of high performance container datatypes. A frequently useful datatype is a defaultdict, which automatically creates an appropriate default value for a new key. For example, a defaultdict(int) automatically initializes a new dictionary entry to 0 (zero); a defaultdict(list) automatically initializes a new dictionary entry to the empty list ([]).

After first importing defaultdict from collections, we can use defaultdict(int) to simplify the code above:

In [60]:
from collections import defaultdict # don't need to use collections.defaultdict() below

cap_state_value_counts = defaultdict(int)
for instance in clean_instances:
cap_state_value = instance[1]
cap_state_value_counts[cap_state_value] += 1

print('Counts for each value of cap-state:')
for value in cap_state_value_counts:
print(value, ':', cap_state_value_counts[value])

Counts for each value of cap-state:
c : 4
b : 300
f : 2432
k : 36
s : 32
x : 2840


### Exercise 4: define attribute_value_counts()¶

Define a function, attribute_value_counts(instances, attribute, attribute_names), that returns a defaultdict containing the counts of occurrences of each value of attribute in the list of instances. attribute_names is the list we created above, where each element is the name of an attribute.

In [61]:
# your definition goes here

attribute = 'cap-shape'
# remove 'simple_ml.' below to test your function definition
attribute_value_counts = simple_ml.attribute_value_counts(clean_instances, attribute, attribute_names)

print('Counts for each value of', attribute, ':')
for value in attribute_value_counts:
print(value, ':', attribute_value_counts[value])

Counts for each value of cap-shape :
c : 4
b : 300
f : 2432
k : 36
s : 32
x : 2840


### Sorting¶

Earlier, we saw that there is a list.sort() method that will sort a list in-place, i.e., by replacing the original value of list with a sorted version of the elements in list.

The Python sorted(iterable[, cmp[, key[, reverse]]]) function can be used to return a copy of a list, dictionary or any other iterable container it is passed, in ascending order.

In [62]:
original_list = [3, 1, 4, 2, 5]
sorted_list = sorted(original_list)
print(original_list)
print(sorted_list)

[3, 1, 4, 2, 5]
[1, 2, 3, 4, 5]


Since it returns a copy, sorted() can be used with strings.

In [63]:
print(sorted('python'))

['h', 'n', 'o', 'p', 't', 'y']


sorted() can also be used with dictionaries (it returns a sorted list of the dictionary keys).

In [64]:
print(sorted(attribute_values_cap_type)) # returns a list of sorted keys (but not values) in the dictionary

['b', 'c', 'f', 'k', 's', 'x']


However, we can use the sorted keys to access the values of a dictionary.

In [65]:
for attribute_value_abbrev in sorted(attribute_values_cap_type):
print(attribute_value_abbrev, '=', attribute_values_cap_type[attribute_value_abbrev])

b = bell
c = conical
f = flat
k = knobbed
s = sunken
x = convex


An optional keyword argument, reverse, can be used to reverse the order of the sorted list returned by the function. The default value of this optional parameter is False, to get non-default behavior, we must specify the name and value of the argument: reverse=True.

In [66]:
print(sorted([3, 1, 4, 2, 5], reverse=True))

[5, 4, 3, 2, 1]

In [67]:
print(sorted(attribute_values_cap_type, reverse=True))

['x', 's', 'k', 'f', 'c', 'b']

In [68]:
attribute = 'cap-shape'
attribute_value_counts = simple_ml.attribute_value_counts(clean_instances, attribute, attribute_names)

print('Counts for each value of', attribute, ':')
for value in sorted(attribute_value_counts):
print(value, ':', attribute_value_counts[value])

Counts for each value of cap-shape :
b : 300
c : 4
f : 2432
k : 36
s : 32
x : 2840


### Sorting a dictionary by values¶

We often want to sort a dictionary by its values rather than its keys.

For example, when we printed out the counts of the attribute values for cap-shape above, the counts appeared in an ascending alphabetic order of their attribute names. It is often more helpful to show the attribute value counts in descending order of the counts (which are the values in that dictionary).

There are a variety of ways to sort a dictionary by values, but the approach described in PEP-256 is generally considered the most efficient.

In order to understand the components used in this approach, we will revisit and elaborate on a few concepts involving dictionaries, iterators and modules.

The dict.items() method returns an unordered list of (key, value) tuples in dict.

In [69]:
attribute_values_cap_type.items()

Out[69]:
[('c', 'conical'),
('b', 'bell'),
('f', 'flat'),
('k', 'knobbed'),
('s', 'sunken'),
('x', 'convex')]

In Python 2, a related method, dict.iteritems(), returns an iterator: a callable object that returns the next item in a sequence each time it is referenced (e.g., during each iteration of a for loop), which can be more efficient than generating all the items in the sequence before any are used ... and so should be used rather than items() wherever possible

This is similar to the distinction between xrange() and range() described above ... and, also similarly, dict.items() is an iterator in Python 3 and so dict.iteritems() is no longer needed (nor defined) ... and further similarly, we will use only dict.items() in this notebook, but dict.iteritems() should be used wherever possible in Python 2.

In [70]:
for key, value in attribute_values_cap_type.items():
print(key, ':', value)

c : conical
b : bell
f : flat
k : knobbed
s : sunken
x : convex


The Python operator module contains a number of functions that perform object comparisons, logical operations, mathematical operations, sequence operations, and abstract type tests.

To facilitate sorting a dictionary by values, we will use the operator.itemgetter(item) function that can be used to retrieve an indexed value (item) in a tuple (such as a (key, value) pair returned by [iter]items()).

We can use operator.itemgetter(1)) to reference the value - the 2nd item in each (key, value) tuple, (at zero-based index position 1) - rather than the key - the first item in each (key, value) tuple (at index position 0).

We will use the optional keyword argument key in sorted(iterable[, cmp[, key[, reverse]]]) to specify a sorting key that is not the same as the dict key (the dict key is the default sorting key)

In [71]:
import operator

sorted(attribute_values_cap_type.items(), key=operator.itemgetter(1))

Out[71]:
[('b', 'bell'),
('c', 'conical'),
('x', 'convex'),
('f', 'flat'),
('k', 'knobbed'),
('s', 'sunken')]

We can now sort the counts of attribute values in descending frequency of occurrence, and print them out using tuple unpacking.

In [72]:
attribute = 'cap-shape'
value_counts = simple_ml.attribute_value_counts(clean_instances, attribute, attribute_names)

print('Counts for each value of', attribute, '(sorted by count):')
for value, count in sorted(value_counts.items(), key=operator.itemgetter(1), reverse=True):
print(value, ':', count)

Counts for each value of cap-shape (sorted by count):
x : 2840
f : 2432
b : 300
k : 36
s : 32
c : 4


### Exercise 5: define print_all_attribute_value_counts()¶

Define a function, print_all_attribute_value_counts(instances, attribute_names), that prints each attribute name in attribute_names, and then for each attribute value, prints the value abbreviation, the count of occurrences of that value and the proportion of instances that have that attribute value.

You may find it helpful to use fancier output formatting. More details can be found in the Python documentation on format string syntax.

Examples of the str.format() function used in conjunction with print statements is shown below, followed by sample output of the simple_ml version of print_all_attribute_value_counts() (which uses similar formatting, but without hard-coded values).

In [73]:
print('Output of a sample line using str.format():')
print('class:', end=' ') # keeps cursor on the same line for subsequent print statements
print('{} = {} ({:5.3f}),'.format('e', 3488, 3488 / 5644.0), end=' ')
print('{} = {} ({:5.3f}),'.format('p', 2156, 2156 / 5644.0), end=' ')
print() # a print statement with no arguments will advance the cursor to the beginning of the next line
print('End of sample line')

Output of a sample line using str.format():
class: e = 3488 (0.618), p = 2156 (0.382),
End of sample line


Define your version of print_all_attribute_value_counts(instances, attribute_names) below, deleting the simple_ml. module specification when you are ready to test your function.

In [74]:
# your function definition goes here

print('\nCounts for all attributes and values:\n')
simple_ml.print_all_attribute_value_counts(clean_instances, attribute_names)

Counts for all attributes and values:

class: e = 3488 (0.618), p = 2156 (0.382),
cap-shape: x = 2840 (0.503), f = 2432 (0.431), b = 300 (0.053), k = 36 (0.006), s = 32 (0.006), c = 4 (0.001),
cap-surface: y = 2220 (0.393), f = 2160 (0.383), s = 1260 (0.223), g = 4 (0.001),
cap-color: g = 1696 (0.300), n = 1164 (0.206), y = 1056 (0.187), w = 880 (0.156), e = 588 (0.104), b = 120 (0.021), p = 96 (0.017), c = 44 (0.008),
bruises?: t = 3184 (0.564), f = 2460 (0.436),
odor: n = 2776 (0.492), f = 1584 (0.281), a = 400 (0.071), l = 400 (0.071), p = 256 (0.045), c = 192 (0.034), m = 36 (0.006),
gill-attachment: f = 5626 (0.997), a = 18 (0.003),
gill-spacing: c = 4620 (0.819), w = 1024 (0.181),
gill-size: b = 4940 (0.875), n = 704 (0.125),
gill-color: p = 1384 (0.245), n = 984 (0.174), w = 966 (0.171), h = 720 (0.128), g = 656 (0.116), u = 480 (0.085), k = 408 (0.072), r = 24 (0.004), y = 22 (0.004),
stalk-shape: t = 2880 (0.510), e = 2764 (0.490),
stalk-root: b = 3776 (0.669), e = 1120 (0.198), c = 556 (0.099), r = 192 (0.034),
stalk-surface-above-ring: s = 3736 (0.662), k = 1332 (0.236), f = 552 (0.098), y = 24 (0.004),
stalk-surface-below-ring: s = 3544 (0.628), k = 1296 (0.230), f = 552 (0.098), y = 252 (0.045),
stalk-color-above-ring: w = 3136 (0.556), p = 1008 (0.179), g = 576 (0.102), n = 448 (0.079), b = 432 (0.077), c = 36 (0.006), y = 8 (0.001),
stalk-color-below-ring: w = 3088 (0.547), p = 1008 (0.179), g = 576 (0.102), n = 496 (0.088), b = 432 (0.077), c = 36 (0.006), y = 8 (0.001),
veil-type: p = 5644 (1.000),
veil-color: w = 5636 (0.999), y = 8 (0.001),
ring-number: o = 5488 (0.972), t = 120 (0.021), n = 36 (0.006),
ring-type: p = 3488 (0.618), l = 1296 (0.230), e = 824 (0.146), n = 36 (0.006),
spore-print-color: n = 1920 (0.340), k = 1872 (0.332), h = 1584 (0.281), w = 148 (0.026), r = 72 (0.013), u = 48 (0.009),
population: v = 2160 (0.383), y = 1688 (0.299), s = 1104 (0.196), a = 384 (0.068), n = 256 (0.045), c = 52 (0.009),
habitat: d = 2492 (0.442), g = 1860 (0.330), p = 568 (0.101), u = 368 (0.065), m = 292 (0.052), l = 64 (0.011),


## 4. Using Python to Build and Use a Simple Decision Tree Classifier¶

### Decision Trees¶

Wikipedia offers the following description of a decision tree (with italics added to emphasize terms that will be elaborated below):

A decision tree is a flowchart-like structure in which each internal node represents a test of an attribute, each branch represents an outcome of that test and each leaf node represents class label (a decision taken after testing all attributes in the path from the root to the leaf). Each path from the root to a leaf can also be represented as a classification rule.

The image below depicts a decision tree created from the UCI mushroom dataset that appears on Andy G's blog post about Decision Tree Learning, where

• a white box represents an internal node (and the label represents the attribute being tested)
• a blue box represents an attribute value (an outcome of the test of that attribute)
• a green box represents a leaf node with a class label of edible
• a red box represents a leaf node with a class label of poisonous

It is important to note that the UCI mushroom dataset consists entirely of categorical variables, i.e., every variable (or attribute) has an enumerated set of possible values. Many datasets include numeric variables that can take on int or float values. Tests for such variables typically use comparison operators, e.g., $age < 65$ or $36,250 < adjusted\_gross\_income <= 87,850$. [Aside: Python supports boolean expressions containing multiple comparison operators, such as the expression comparing adjusted_gross_income in the preceding example.]

Our simple decision tree will only accommodate categorical variables. We will closely follow a version of the decision tree learning algorithm implementation offered by Chris Roach.

Our goal in the following sections is to use Python to

• create a simple decision tree based on a set of training instances
• classify (predict class labels for) for an instance using a simple decision tree
• evaluate the performance of the simple decision tree on classifying a set of test instances

First, we will explore some concepts and algorithms used in building and using decision trees.

### Entropy¶

When building a supervised classification model, the frequency distribution of attribute values is a potentially important factor in determining the relative importance of each attribute at various stages in the model building process.

In data modeling, we can use frequency distributions to compute entropy, a measure of disorder (impurity) in a set.

We compute the entropy of multiplying the proportion of instances with each class label by the log of that proportion, and then taking the negative sum of those terms.

More precisely, for a 2-class (binary) classification task:

$entropy(S) = - p_1 log_2 (p_1) - p_2 log_2 (p_2)$

where $p_i$ is proportion (relative frequency) of class i within the set S.

From the output above, we know that the proportion of clean_instances that are labeled 'e' (class edible) in the UCI dataset is $3488 \div 5644 = 0.618$, and the proportion labeled 'p' (class poisonous) is $2156 \div 5644 = 0.382$.

After importing the Python math module, we can use the math.log(x[, base]) function in computing the entropy of the clean_instances of the UCI mushroom data set as follows.

Note that you can use a backslash character (\) at the end of a line to continue the statement on the next line (this should generally be used sparingly).

In [75]:
import math

entropy = - (3488 / 5644.0) * math.log(3488 / 5644.0, 2) \
- (2156 / 5644.0) * math.log(2156 / 5644.0, 2)
print(entropy)

0.959441337353


### Exercise 6: define entropy()¶

Define a function, entropy(instances), that computes the entropy of instances. You may assume the class label is in position 0; we will later see how to specify default parameter values in function definitions.

[Note: the class label in many data files is the last rather than the first item on each line.]

In [76]:
# your function definition here

# delete 'simple_ml.' below to test your function
print(simple_ml.entropy(clean_instances))

0.959441337353


### Information Gain¶

Informally, a decision tree is constructed using a recursive algorithm that

• selects the best attribute
• splits the set into subsets based on the values of that attribute (each subset is composed of instances from the original set that have the same value for that attribute)
• repeats the process on each of these subsets until a stopping condition is met (e.g., a subset has no instances or has instances which all have the same class label)

Entropy is a metric that can be used in selecting the best attribute for each split: the best attribute is the one resulting in the largest decrease in entropy for a set of instances. [Note: other metrics can be used for determining the best attribute]

Information gain measures the decrease in entropy that results from splitting a set of instances based on an attribute.

$IG(S, a) = entropy(S) - [p(s_1) × entropy(s_1) + p(s_2) × entropy(s_2) ... + p(s_n) × entropy(s_n)]$

Where $n$ is the number of distinct values of attribute $a$, and $s_i$ is the subset of $S$ where all instances have the $i$th value of $a$.

In [77]:
print('Information gain for different attributes:', end='\n\n')
for i in range(1, len(attribute_names)):
print('{:5.3f}  {:2} {}'.format(
simple_ml.information_gain(clean_instances, i), i, attribute_names[i]))

Information gain for different attributes:

0.017   1 cap-shape
0.005   2 cap-surface
0.195   3 cap-color
0.140   4 bruises?
0.860   5 odor
0.004   6 gill-attachment
0.058   7 gill-spacing
0.032   8 gill-size
0.213   9 gill-color
0.275  10 stalk-shape
0.097  11 stalk-root
0.425  12 stalk-surface-above-ring
0.409  13 stalk-surface-below-ring
0.306  14 stalk-color-above-ring
0.279  15 stalk-color-below-ring
0.000  16 veil-type
0.002  17 veil-color
0.012  18 ring-number
0.463  19 ring-type
0.583  20 spore-print-color
0.110  21 population
0.101  22 habitat


We can sort the attributes based in decreasing order of information gain.

In [78]:
print('Information gain for different attributes:', end='\n\n')
sorted_information_gain_indexes = sorted([(simple_ml.information_gain(clean_instances, i), i) \
for i in range(1, len(attribute_names))],
reverse=True)
print(sorted_information_gain_indexes, end='\n\n')

for gain, i in sorted_information_gain_indexes:
print('{:5.3f}  {:2} {}'.format(gain, i, attribute_names[i]))

Information gain for different attributes:

[(0.8596704358849709, 5), (0.5828694793608379, 20), (0.46290566555455265, 19), (0.42456477093655975, 12), (0.40865780788318695, 13), (0.3062989793570199, 14), (0.27891994708759504, 15), (0.2750355212178639, 10), (0.2127971869976022, 9), (0.19495343617580085, 3), (0.1400386042032834, 4), (0.1097880400299237, 21), (0.10067585994181227, 22), (0.09733858997769329, 11), (0.05836192763098613, 7), (0.03242975884332899, 8), (0.01740692300090696, 1), (0.01205967443646827, 18), (0.004572013423856602, 2), (0.0044397141315495325, 6), (0.0019702590992403124, 17), (0.0, 16)]

0.860   5 odor
0.583  20 spore-print-color
0.463  19 ring-type
0.425  12 stalk-surface-above-ring
0.409  13 stalk-surface-below-ring
0.306  14 stalk-color-above-ring
0.279  15 stalk-color-below-ring
0.275  10 stalk-shape
0.213   9 gill-color
0.195   3 cap-color
0.140   4 bruises?
0.110  21 population
0.101  22 habitat
0.097  11 stalk-root
0.058   7 gill-spacing
0.032   8 gill-size
0.017   1 cap-shape
0.012  18 ring-number
0.005   2 cap-surface
0.004   6 gill-attachment
0.002  17 veil-color
0.000  16 veil-type


The following variation does not use a list comprehension:

In [79]:
print('Information gain for different attributes:', end='\n\n')

information_gain_values = []
for i in range(1, len(attribute_names)):
information_gain_values.append((simple_ml.information_gain(clean_instances, i), i))

sorted_information_gain_indexes = sorted(information_gain_values,
reverse=True)
print(sorted_information_gain_indexes, end='\n\n')

for gain, i in sorted_information_gain_indexes:
print('{:5.3f}  {:2} {}'.format(gain, i, attribute_names[i]))

Information gain for different attributes:

[(0.8596704358849709, 5), (0.5828694793608379, 20), (0.46290566555455265, 19), (0.42456477093655975, 12), (0.40865780788318695, 13), (0.3062989793570199, 14), (0.27891994708759504, 15), (0.2750355212178639, 10), (0.2127971869976022, 9), (0.19495343617580085, 3), (0.1400386042032834, 4), (0.1097880400299237, 21), (0.10067585994181227, 22), (0.09733858997769329, 11), (0.05836192763098613, 7), (0.03242975884332899, 8), (0.01740692300090696, 1), (0.01205967443646827, 18), (0.004572013423856602, 2), (0.0044397141315495325, 6), (0.0019702590992403124, 17), (0.0, 16)]

0.860   5 odor
0.583  20 spore-print-color
0.463  19 ring-type
0.425  12 stalk-surface-above-ring
0.409  13 stalk-surface-below-ring
0.306  14 stalk-color-above-ring
0.279  15 stalk-color-below-ring
0.275  10 stalk-shape
0.213   9 gill-color
0.195   3 cap-color
0.140   4 bruises?
0.110  21 population
0.101  22 habitat
0.097  11 stalk-root
0.058   7 gill-spacing
0.032   8 gill-size
0.017   1 cap-shape
0.012  18 ring-number
0.005   2 cap-surface
0.004   6 gill-attachment
0.002  17 veil-color
0.000  16 veil-type


### Exercise 7: define information_gain()¶

Define a function, information_gain(instances, i), that returns the information gain achieved by selecting the ith attribute to split instances. It should exhibit the same behavior as the simple_ml version of the function.

In [80]:
# your definition of information_gain(instances, i) here

# delete 'simple_ml.' below to test your function
sorted_information_gain_indexes = sorted([(simple_ml.information_gain(clean_instances, i), i) for i in range(1, len(attribute_names))],
reverse=True)

print('Information gain for different attributes:', end='\n\n')
for gain, i in sorted_information_gain_indexes:
print('{:5.3f}  {:2} {}'.format(gain, i, attribute_names[i]))

Information gain for different attributes:

0.860   5 odor
0.583  20 spore-print-color
0.463  19 ring-type
0.425  12 stalk-surface-above-ring
0.409  13 stalk-surface-below-ring
0.306  14 stalk-color-above-ring
0.279  15 stalk-color-below-ring
0.275  10 stalk-shape
0.213   9 gill-color
0.195   3 cap-color
0.140   4 bruises?
0.110  21 population
0.101  22 habitat
0.097  11 stalk-root
0.058   7 gill-spacing
0.032   8 gill-size
0.017   1 cap-shape
0.012  18 ring-number
0.005   2 cap-surface
0.004   6 gill-attachment
0.002  17 veil-color
0.000  16 veil-type


### Building a Simple Decision Tree¶

We will implement a modified version of the ID3 algorithm for building a simple decision tree.

ID3 (Examples, Target_Attribute, Attributes)
Create a root node for the tree
If all examples are positive, Return the single-node tree Root, with label = +.
If all examples are negative, Return the single-node tree Root, with label = -.
If number of predicting attributes is empty, then Return the single node tree Root,
with label = most common value of the target attribute in the examples.
Otherwise Begin
A ← The Attribute that best classifies examples.
Decision Tree attribute for Root = A.
For each possible value, v_i, of A,
Add a new tree branch below Root, corresponding to the test A = v_i.
Let Examples(v_i) be the subset of examples that have the value v_i for A
If Examples(v_i) is empty
Then below this new branch add a leaf node with label = most common target value in the examples
Else below this new branch add the subtree ID3 (Examples(v_i), Target_Attribute, Attributes – {A})
End
Return Root

In building a decision tree, we will need to split the instances based on the index of the best attribute, i.e., the attribute that offers the highest information gain. We will use separate utility functions to handle these subtasks. To simplify the functions, we will rely exclusively on attribute indexes rather than attribute names.

Note: the algorithm above is recursive, i.e., the there is a recursive call to ID3 within the definition of ID3. Covering recursion is beyond the scope of this primer, but there are a number of other resources on using recursion in Python. Familiarity with recursion will be important for understanding both the tree construction and classification functions below.

First, we will define a function, split_instances(instances, attribute_index), to split a set of instances based on any attribute. This function will return a dictionary where the key of each dictionary is a distinct value of the specified attribute_index, and the value of each dictionary is a list representing the subset of instances that have that attribute value.

In [81]:
def split_instances(instances, attribute_index):
'''Returns a list of dictionaries, splitting a list of instances according to their values of a specified attribute''

The key of each dictionary is a distinct value of attribute_index,
and the value of each dictionary is a list representing the subset of instances that have that value for the attribute'''
partitions = defaultdict(list)
for instance in instances:
partitions[instance[attribute_index]].append(instance)
return partitions

partitions = split_instances(clean_instances, 5)
print([(partition, len(partitions[partition])) for partition in partitions])

[('a', 400), ('c', 192), ('f', 1584), ('m', 36), ('l', 400), ('n', 2776), ('p', 256)]


Now that we can split instances based on a particular attribute, we would like to be able to choose the best attribute with which to split the instances, where best is defined as the attribute that provides the greatest information gain if instances were split based on that attribute. We will want to restrict the candidate attributes so that we don't bother trying to split on an attribute that was used higher up in the decision tree (or use the target attribute as a candidate).

### Exercise 8: define choose_best_attribute_index()¶

Define a function, choose_best_attribute_index(instances, candidate_attribute_indexes), that returns the index in the list of candidate_attribute_indexes that provides the highest information gain if instances are split based on that attribute index.

In [82]:
# your function here

# delete 'simple_ml.' below to test your function:
print('Best attribute index:',
simple_ml.choose_best_attribute_index(clean_instances, range(1, len(attribute_names))))

Best attribute index: 5


A leaf node in a decision tree represents the most frequently occurring - or majority - class value for that path through the tree. We will need a function that determines the majority value for the class index among a set of instances.

We earlier saw how the defaultdict container in the collections module can be used to simplify the construction of a dictionary containing the counts of all attribute values for all attributes, by automatically setting the count for any attribute value to zero when the attribute value is first added to the dictionary.

The collections module has another useful container, a Counter class, that can further simplify the construction of a specialized dictionary of counts. When a Counter object is instantiated with a list of items, it returns a dictionary-like container in which the keys are the unique items in the list, and the values are the counts of each unique item in that list.

This container has an additional method, most_common([n]), which returns a list of 2-element tuples representing the values and their associated counts for the most common n values; if n is omitted, the method returns all tuples.

The following is an example of how we can use a Counter to represent the frequency of different class labels, and how we can identify the most frequent value and its count.

In [83]:
from collections import Counter

class_counts = Counter([instance[0] for instance in clean_instances])
print('class_counts: {}; most_common(1): {}, most_common(1)[0][0]: {}'.format(
class_counts, # the Counter object
class_counts.most_common(1), # returns a list in which the 1st element is a tuple with the most common value and its count
class_counts.most_common(1)[0][0])) # the most common value (1st element in that tuple)

class_counts: Counter({'e': 3488, 'p': 2156}); most_common(1): [('e', 3488)], most_common(1)[0][0]: e


The following variation does not use a list comprehension:

In [84]:
class_values = []
for instance in clean_instances:
class_values.append(instance[0])

class_counts = Counter(class_values)
print ('class_counts: {}; most_common(1): {}, most_common(1)[0][0]: {}'.format(
class_counts, # the Counter object
class_counts.most_common(1), # returns a list in which the 1st element is a tuple with the most common value and its count
class_counts.most_common(1)[0][0])) # the most common value (1st element in that tuple)

class_counts: Counter({'e': 3488, 'p': 2156}); most_common(1): [('e', 3488)], most_common(1)[0][0]: e


Before putting all this together to define a decision tree construction function, it may be helpful to cover a few additional aspects of Python the function will utilize.

Python offers a very flexible mechanism for the testing of truth values: in an if condition, any null object, zero-valued numerical expression or empty container (string, list, dictionary or tuple) is interpreted as False (i.e., not True):

In [85]:
for x in [False, None, 0, 0.0, "", [], {}, ()]:
print('"{}" is'.format(x), end=' ')
if x:
print(True)
else:
print(False)

"False" is False
"None" is False
"0" is False
"0.0" is False
"" is False
"[]" is False
"{}" is False
"()" is False


Python also offers a conditional expression (ternary operator) that allows the functionality of an if/else statement that returns a value to be implemented as an expression. For example, the if/else statement in the code above could be implemented as a conditional expression as follows:

In [86]:
for x in [False, None, 0, 0.0, "", [], {}, ()]:
print('"{}" is {}'.format(x, True if x else False)) # using conditional expression as second argument to format()

"False" is False
"None" is False
"0" is False
"0.0" is False
"" is False
"[]" is False
"{}" is False
"()" is False


Python function definitions can specify default parameter values indicating the value those parameters will have if no argument is explicitly provided when the function is called. Arguments can also be passed using keyword parameters indicting which parameter will be assigned a specific argument value (which may or may not correspond to the order in which the parameters are defined).

The Python Tutorial page on default parameters includes the following warning:

Important warning: The default value is evaluated only once. This makes a difference when the default is a mutable object such as a list, dictionary, or instances of most classes.

Thus it is generally better to use the Python null object, None, rather than an empty list ([]), dict ({}) or other mutable data structure when specifying default parameter values for any of those data types.

In [87]:
def parameter_test(parameter1=None, parameter2=None):
'''Prints the values of parameter1 and parameter2'''
print('parameter1: {}; parameter2: {}'.format(parameter1, parameter2))

parameter_test() # no args are required
parameter_test(1) # if any args are provided, 1st arg gets assigned to parameter1
parameter_test(1, 2) # 2nd arg gets assigned to parameter2
parameter_test(2) # remember: if only 1 arg, 1st arg gets assigned to arg1
parameter_test(parameter2=2) # can use keyword to [only] provide an explicit value for parameter2
parameter_test(parameter2=2, parameter1=1) # can use keywords for either arg, in either order

parameter1: None; parameter2: None
parameter1: 1; parameter2: None
parameter1: 1; parameter2: 2
parameter1: 2; parameter2: None
parameter1: None; parameter2: 2
parameter1: 1; parameter2: 2


### Exercise 9: define majority_value()¶

Define a function, majority_value(instances, class_index), that returns the most frequently occurring value of class_index in instances. The class_index parameter should be optional, and have a default value of 0 (zero).

In [88]:
# your definition of majority_value(instances) here

# delete 'simple_ml.' below to test your function:
print('Majority value of index {}: {}'.format(
0, simple_ml.majority_value(clean_instances))) # note: relying on default parameter here
# although there is only one class_index for the dataset, we'll test it by providing non-default values
print('Majority value of index {}: {}'.format(
1, simple_ml.majority_value(clean_instances, 1))) # supplyling an optional 2nd argument
print('Majority value of index {}: {}'.format(
2, simple_ml.majority_value(clean_instances, class_index=2))) # supplying argument as a keyword

Majority value of index 0: e
Majority value of index 1: x
Majority value of index 2: y


The recursive create_decision_tree() function below uses an optional parameter, class_index, which defaults to 0. This is to accommodate other datasets in which the class label is the last element on each line (which would be most easily specified by using a -1 value). Most data files in the UCI Machine Learning Repository have the class labels as either the first element or the last element.

To show how the decision tree is being built, an optional trace parameter, when non-zero, will generate some trace information as the tree is constructed. The indentation level is incremented with each recursive call via the use of the conditional expression (ternary operator), trace + 1 if trace else 0.

In [89]:
def create_decision_tree(instances, candidate_attribute_indexes=None, class_index=0, default_class=None, trace=0):
'''Returns a new decision tree trained on a list of instances.

The tree is constructed by recursively selecting and splitting instances based on
the highest information_gain of the candidate_attribute_indexes.

The class label is found in position class_index.

The default_class is the majority value for the current node's parent in the tree.
A positive (int) trace value will generate trace information with increasing levels of indentation.

Derived from the simplified ID3 algorithm presented in Building Decision Trees in Python by Christopher Roach,
http://www.onlamp.com/pub/a/python/2006/02/09/ai_decision_trees.html?page=3'''

# if no candidate_attribute_indexes are provided, assume that we will use all but the target_attribute_index
if candidate_attribute_indexes is None:
candidate_attribute_indexes = [i for i in range(len(instances[0])) if i != class_index]
#candidate_attribute_indexes.remove(class_index)

class_labels_and_counts = Counter([instance[class_index] for instance in instances])

# If the dataset is empty or the candidate attributes list is empty, return the default value
if not instances or not candidate_attribute_indexes:
if trace:
print('{}Using default class {}'.format('< ' * trace, default_class))
return default_class

# If all the instances have the same class label, return that class label
elif len(class_labels_and_counts) == 1:
class_label = class_labels_and_counts.most_common(1)[0][0]
if trace:
print('{}All {} instances have label {}'.format(
'< ' * trace, len(instances), class_label))
return class_label
else:
default_class = simple_ml.majority_value(instances, class_index)

# Choose the next best attribute index to best classify the instances
best_index = simple_ml.choose_best_attribute_index(
instances, candidate_attribute_indexes, class_index)
if trace:
print('{}Creating tree node for attribute index {}'.format('> ' * trace, best_index))

# Create a new decision tree node with the best attribute index and an empty dictionary object (for now)
tree = {best_index:{}}

# Create a new decision tree sub-node (branch) for each of the values in the best attribute field
partitions = simple_ml.split_instances(instances, best_index)

# Remove that attribute from the set of candidates for further splits
remaining_candidate_attribute_indexes = [i for i in candidate_attribute_indexes if i != best_index]
for attribute_value in partitions:
if trace:
print('{}Creating subtree for value {} ({}, {}, {}, {})'.format(
'> ' * trace,
attribute_value,
len(partitions[attribute_value]),
len(remaining_candidate_attribute_indexes),
class_index,
default_class))

# Create a subtree for each value of the the best attribute
subtree = create_decision_tree(
partitions[attribute_value],
remaining_candidate_attribute_indexes,
class_index,
default_class,
trace + 1 if trace else 0)

# Add the new subtree to the empty dictionary object in the new tree/node we just created
tree[best_index][attribute_value] = subtree

return tree

# split instances into separate training and testing sets
training_instances = clean_instances[:-20]
testing_instances = clean_instances[-20:]
tree = create_decision_tree(training_instances, trace=1) # remove trace=1 to turn off tracing
print(tree)

> Creating tree node for attribute index 5
> Creating subtree for value a (400, 21, 0, e)
< < All 400 instances have label e
> Creating subtree for value c (192, 21, 0, e)
< < All 192 instances have label p
> Creating subtree for value f (1584, 21, 0, e)
< < All 1584 instances have label p
> Creating subtree for value m (28, 21, 0, e)
< < All 28 instances have label p
> Creating subtree for value l (400, 21, 0, e)
< < All 400 instances have label e
> Creating subtree for value n (2764, 21, 0, e)
> > Creating tree node for attribute index 20
> > Creating subtree for value k (1296, 20, 0, e)
< < < All 1296 instances have label e
> > Creating subtree for value r (72, 20, 0, e)
< < < All 72 instances have label p
> > Creating subtree for value w (100, 20, 0, e)
> > > Creating tree node for attribute index 21
> > > Creating subtree for value y (24, 19, 0, e)
< < < < All 24 instances have label e
> > > Creating subtree for value c (16, 19, 0, e)
< < < < All 16 instances have label p
> > > Creating subtree for value v (60, 19, 0, e)
< < < < All 60 instances have label e
> > Creating subtree for value n (1296, 20, 0, e)
< < < All 1296 instances have label e
> Creating subtree for value p (256, 21, 0, e)
< < All 256 instances have label p
{5: {'a': 'e', 'c': 'p', 'f': 'p', 'm': 'p', 'l': 'e', 'n': {20: {'k': 'e', 'r': 'p', 'w': {21: {'y': 'e', 'c': 'p', 'v': 'e'}}, 'n': 'e'}}, 'p': 'p'}}


The structure of the tree shown above is rather difficult to discern from the normal printed representation of a dictionary.

The Python pprint module has a number of useful methods for pretty-printing or formatting objects in a more human readable way.

The pprint.pprint(object, stream=None, indent=1, width=80, depth=None) method will print object to a stream (a default value of None will dictate the use of sys.stdout, the same destination as print function output), using indent spaces to differentiate nesting levels, using up to a maximum width columns and up to to a maximum nesting level depth (None indicating no maximum).

We will use the a variation on the import statement that imports one or more functions into the current namespace:

from pprint import pprint



This will to enable us to use pprint() rather than having to use dotted notation, i.e., pprint.pprint().

Note that if we wanted to define our own pprint() function, we would be best only using

import pprint



so that we can still access the pprint() function in the pprint module (since defining pprint() in the current namespace would otherwise override the imported definition of the function).

In [90]:
from pprint import pprint

pprint(tree)

{5: {'a': 'e',
'c': 'p',
'f': 'p',
'l': 'e',
'm': 'p',
'n': {20: {'k': 'e',
'n': 'e',
'r': 'p',
'w': {21: {'c': 'p', 'v': 'e', 'y': 'e'}}}},
'p': 'p'}}


### Classifying Instances with a Simple Decision Tree¶

Usually, when we construct a decision tree based on a set of training instances, we do so with the intent of using that tree to classify a set of one or more testing instances.

We will define a function, classify(tree, instance, default_class=None), to use a decision tree to classify a single instance, where an optional default_class can be specified as the return value if the instance represents a set of attribute values that don't have a representation in the decision tree.

We will use a design pattern in which we will use a series of if statements, each of which returns a value if the condition is true, rather than a nested series of if, elif and/or else clauses, as it helps constrain the levels of indentation in the function.

In [91]:
def classify(tree, instance, default_class=None):
'''Returns a classification label for instance, given a decision tree'''
if not tree:
return default_class
if not isinstance(tree, dict):
return tree
attribute_index = list(tree.keys())[0] # using list(dict.keys()) for Python 3 compatibility
attribute_values = list(tree.values())[0]
instance_attribute_value = instance[attribute_index]
if instance_attribute_value not in attribute_values:
return default_class
return classify(attribute_values[instance_attribute_value], instance, default_class)

for instance in testing_instances:
predicted_label = classify(tree, instance)
actual_label = instance[0]
print('predicted: {}; actual: {}'.format(predicted_label, actual_label))

predicted: p; actual: p
predicted: p; actual: p
predicted: p; actual: p
predicted: e; actual: e
predicted: e; actual: e
predicted: p; actual: p
predicted: e; actual: e
predicted: e; actual: e
predicted: e; actual: e
predicted: p; actual: p
predicted: e; actual: e
predicted: e; actual: e
predicted: e; actual: e
predicted: p; actual: p
predicted: e; actual: e
predicted: e; actual: e
predicted: e; actual: e
predicted: e; actual: e
predicted: p; actual: p
predicted: p; actual: p


### Evaluating the Accuracy of a Simple Decision Tree¶

It is often helpful to evaluate the performance of a model using a dataset not used in the training of that model. In the simple example shown above, we used all but the last 20 instances to train a simple decision tree, then classified those last 20 instances using the tree.

The advantage of this training/testing split is that visual inspection of the classifications (sometimes called predictions) is relatively straightforward, revealing that all 20 instances were correctly classified.

There are a variety of metrics that can be used to evaluate the performance of a model. Scikit Learn's Model Evaluation library provides an overview and implementation of several possible metrics. For now, we'll simply measure the accuracy of a model, i.e., the percentage of testing instances that are correctly classified (true positives and true negatives).

The accuracy of the model above, given the set of 20 testing instances, is 100% (20/20).

The function below calculates the classification accuracy of a tree over a set of testing_instances (with an optional class_index parameter indicating the position of the class label in each instance).

In [92]:
def classification_accuracy(tree, testing_instances, class_index=0, default_class=None):
'''Returns the accuracy of classifying testing_instances with tree, where the class label is in position class_index'''
num_correct = 0
for i in range(len(testing_instances)):
prediction = classify(tree, testing_instances[i], default_class)
actual_value = testing_instances[i][class_index]
if prediction == actual_value:
num_correct += 1
return num_correct / len(testing_instances)

print(classification_accuracy(tree, testing_instances))

1.0


The zip([iterable, ...]) function combines 2 or more sequences or iterables; the function returns a list of tuples, where the ith tuple contains the ith element from each of the argument sequences or iterables.

In [93]:
zip([0, 1, 2], ['a', 'b', 'c'])

Out[93]:
[(0, 'a'), (1, 'b'), (2, 'c')]

We can use list comprehensions, the Counter class and the zip() function to modify classification_accuracy() so that it returns a packed tuple with

• the number of correctly classified instances
• the number of incorrectly classified instances
• the percentage of instances correctly classified
In [94]:
def classification_accuracy(tree, instances, class_index=0, default_class=None):
'''Returns the accuracy of classifying testing_instances with tree, where the class label is in position class_index'''
predicted_labels = [classify(tree, instance, default_class) for instance in instances]
actual_labels = [x[class_index] for x in instances]
counts = Counter([x == y for x, y in zip(predicted_labels, actual_labels)])
return counts[True], counts[False], counts[True] / len(instances)

print(classification_accuracy(tree, testing_instances))

(20, 0, 1.0)


We sometimes want to partition the instances into subsets of equal sizes to measure performance. One metric this partitioning allows us to compute is a learning curve, i.e., assess how well the model performs based on the size of its training set. Another use of these partitions (aka folds) would be to conduct an n-fold cross validation) evaluation.

The following function, partition_instances(instances, num_partitions), partitions a set of instances into num_partitions relatively equally sized subsets.

We'll use this as yet another opportunity to demonstrate the power of using list comprehensions, this time, to condense the use of nested for loops.

In [95]:
def partition_instances(instances, num_partitions):
'''Returns a list of relatively equally sized disjoint sublists (partitions) of the list of instances'''
return [[instances[j] for j in range(i, len(instances), num_partitions)] \
for i in range(num_partitions)]


Before testing this function on the 5644 clean_instances from the UCI mushroom dataset, let's create a small number of simplified instances to verify that the function has the desired behavior.

In [96]:
instance_length = 3
num_instances = 5

simplified_instances = [[j for j in range(i, instance_length + i)] for i in range(num_instances)]

print('Instances:', simplified_instances)
partitions = partition_instances(simplified_instances, 2)
print('Partitions:', partitions)

Instances: [[0, 1, 2], [1, 2, 3], [2, 3, 4], [3, 4, 5], [4, 5, 6]]
Partitions: [[[0, 1, 2], [2, 3, 4], [4, 5, 6]], [[1, 2, 3], [3, 4, 5]]]


The following variations do not use list comprehensions.

In [97]:
def partition_instances(instances, num_partitions):
'''Returns a list of relatively equally sized disjoint sublists (partitions) of the list of instances'''
partitions = []
for i in range(num_partitions):
partition = []
# iterate over instances starting at position i in increments of num_paritions
for j in range(i, len(instances), num_partitions):
partition.append(instances[j])
partitions.append(partition)
return partitions

simplified_instances = []
for i in range(num_instances):
new_instance = []
for j in range(i, instance_length + i):
new_instance.append(j)
simplified_instances.append(new_instance)

print('Instances:', simplified_instances)
partitions = partition_instances(simplified_instances, 2)
print('Partitions:', partitions)

Instances: [[0, 1, 2], [1, 2, 3], [2, 3, 4], [3, 4, 5], [4, 5, 6]]
Partitions: [[[0, 1, 2], [2, 3, 4], [4, 5, 6]], [[1, 2, 3], [3, 4, 5]]]


The enumerate(sequence, start=0) function creates an iterator that successively returns the index and value of each element in a sequence, beginning at the start index.

In [98]:
for i, x in enumerate(['a', 'b', 'c']):
print(i, x)

0 a
1 b
2 c


We can use enumerate() to facilitate slightly more rigorous testing of our partition_instances function on our simplified_instances.

In [99]:
for i in range(5):
print('\n# partitions:', i)
for j, partition in enumerate(partition_instances(simplified_instances, i)):
print('partition {}: {}'.format(j, partition))

# partitions: 0

# partitions: 1
partition 0: [[0, 1, 2], [1, 2, 3], [2, 3, 4], [3, 4, 5], [4, 5, 6]]

# partitions: 2
partition 0: [[0, 1, 2], [2, 3, 4], [4, 5, 6]]
partition 1: [[1, 2, 3], [3, 4, 5]]

# partitions: 3
partition 0: [[0, 1, 2], [3, 4, 5]]
partition 1: [[1, 2, 3], [4, 5, 6]]
partition 2: [[2, 3, 4]]

# partitions: 4
partition 0: [[0, 1, 2], [4, 5, 6]]
partition 1: [[1, 2, 3]]
partition 2: [[2, 3, 4]]
partition 3: [[3, 4, 5]]


Returning our attention to the UCI mushroom dataset, the following will partition our clean_instances into 10 relatively equally sized disjoint subsets. We will use a list comprehension to print out the length of each partition

In [100]:
partitions = partition_instances(clean_instances, 10)
print([len(partition) for partition in partitions])

[565, 565, 565, 565, 564, 564, 564, 564, 564, 564]


The following variation does not use a list comprehension.

In [101]:
for partition in partitions:
print(len(partition), end=' ')  # note the comma at the end
print()

565 565 565 565 564 564 564 564 564 564


The following shows the different trees that are constructed based on partition 0 (first 10th) of clean_instances, partitions 0 and 1 (first 2/10ths) of clean_instances and all clean_instances.

In [102]:
tree0 = create_decision_tree(partitions[0])
print('Tree trained with {} instances:'.format(len(partitions[0])))
pprint(tree0)
print()

tree1 = create_decision_tree(partitions[0] + partitions[1])
print('Tree trained with {} instances:'.format(len(partitions[0] + partitions[1])))
pprint(tree1)
print()

tree = create_decision_tree(clean_instances)
print('Tree trained with {} instances:'.format(len(clean_instances)))
pprint(tree)

Tree trained with 565 instances:
{5: {'a': 'e',
'c': 'p',
'f': 'p',
'l': 'e',
'm': 'p',
'n': {20: {'k': 'e', 'n': 'e', 'r': 'p', 'w': 'e'}},
'p': 'p'}}

Tree trained with 1130 instances:
{5: {'a': 'e',
'c': 'p',
'f': 'p',
'l': 'e',
'm': 'p',
'n': {20: {'k': 'e',
'n': 'e',
'r': 'p',
'w': {21: {'c': 'p', 'v': 'e', 'y': 'e'}}}},
'p': 'p'}}

Tree trained with 5644 instances:
{5: {'a': 'e',
'c': 'p',
'f': 'p',
'l': 'e',
'm': 'p',
'n': {20: {'k': 'e',
'n': 'e',
'r': 'p',
'w': {21: {'c': 'p', 'v': 'e', 'y': 'e'}}}},
'p': 'p'}}


The only difference between the first two trees - tree0 and tree1 - is that in the first tree, instances with no odor (attribute index 5 is 'n') and a spore-print-color of white (attribute 20 = 'w') are classified as edible ('e'). With additional training data in the 2nd partition, an additional distinction is made such that instances with no odor, a white spore-print-color and a clustered population (attribute 21 = 'c') are classified as poisonous ('p'), while all other instances with no odor and a white spore-print-color (and any other value for the population attribute) are classified as edible ('e').

Note that there is no difference between tree1 and tree (the tree trained with all instances). This early convergence on an optimal model is uncommon on most datasets (outside the UCI repository).

Now that we can partition our instances into subsets, we can use these subsets to construct different-sized training sets in the process of computing a learning curve.

We will start off with an initial training set consisting only of the first partition, and then progressively extend that training set by adding a new partition during each iteration of computing the learning curve.

The list.extend(L) method enables us to extend list by appending all the items in another list, L, to the end of list.

In [103]:
x = [1, 2, 3]
x.extend([4, 5])
print(x)

[1, 2, 3, 4, 5]


We can now define the function, compute_learning_curve(instances, num_partitions=10), that will take a list of instances, partition it into num_partitions relatively equally sized disjoint partitions, and then iteratively evaluate the accuracy of models trained with an incrementally increasing combination of instances in the first num_partitions - 1 partitions then tested with instances in the last partition. That is, a model trained with the first partition will be constructed (and tested), then a model trained with the first 2 partitions will be constructed (and tested), and so on.

The function will return a list of num_partitions - 1 tuples representing the size of the training set and the accuracy of a tree trained with that set (and tested on the num_partitions - 1 set). This will provide some indication of the relative impact of the size of the training set on model performance.

In [104]:
def compute_learning_curve(instances, num_partitions=10):
'''Returns a list of training sizes and scores for incrementally increasing partitions.

The list contains 2-element tuples, each representing a training size and score.
The i-th training size is the number of instances in partitions 0 through num_partitions - 2.
The i-th score is the accuracy of a tree trained with instances
from partitions 0 through num_partitions - 2
and tested on instances from num_partitions - 1 (the last partition).'''

partitions = partition_instances(instances, num_partitions)
testing_instances = partitions[-1][:]
training_instances = []
accuracy_list = []
for i in range(0, num_partitions - 1):
# for each iteration, the training set is composed of partitions 0 through i - 1
training_instances.extend(partitions[i][:])
tree = create_decision_tree(training_instances)
partition_accuracy = classification_accuracy(tree, testing_instances)
accuracy_list.append((len(training_instances), partition_accuracy))
return accuracy_list

accuracy_list = compute_learning_curve(clean_instances)
print(accuracy_list)

[(565, (562, 2, 0.9964539007092199)), (1130, (564, 0, 1.0)), (1695, (564, 0, 1.0)), (2260, (564, 0, 1.0)), (2824, (564, 0, 1.0)), (3388, (564, 0, 1.0)), (3952, (564, 0, 1.0)), (4516, (564, 0, 1.0)), (5080, (564, 0, 1.0))]


Due to the quick convergence on an optimal decision tree for classifying the UCI mushroom dataset, we can use a larger number of smaller partitions to see a little more variation in acccuracy performance.

In [105]:
accuracy_list = compute_learning_curve(clean_instances, 100)
print(accuracy_list[:10])

[(57, (55, 1, 0.9821428571428571)), (114, (56, 0, 1.0)), (171, (55, 1, 0.9821428571428571)), (228, (56, 0, 1.0)), (285, (56, 0, 1.0)), (342, (56, 0, 1.0)), (399, (56, 0, 1.0)), (456, (56, 0, 1.0)), (513, (56, 0, 1.0)), (570, (56, 0, 1.0))]


### Object-Oriented Programming: Defining a Python Class to Encapsulate a Simple Decision Tree¶

The simple decision tree defined above uses a Python dictionary for its representation. One can imagine using other data structures, and/or extending the decision tree to support confidence estimates, numeric features and other capabilities that are often included in more fully functional implementations. To support future extensibility, and hide the details of the representation from the user, it would be helpful to have a user-defined class for simple decision trees.

Python is an object-oriented programming language, offering simple syntax and semantics for defining classes and instantiating objects of those classes. [It is assumed that the reader is already familiar with the concepts of object-oriented programming]

A Python class starts with the keyword class followed by a class name (identifier), a colon (':'), and then any number of statements, which typically take the form of assignment statements for class or instance variables and/or function definitions for class methods. All statements are indented to reflect their inclusion in the class definition.

The members - methods, class variables and instance variables - of a class are accessed by prepending self. to each reference. Class methods always include self as the first parameter.

All class members in Python are public (accessible outside the class). There is no mechanism for private class members, but identifiers with leading double underscores (__member_identifier) are 'mangled' (translated into _class_name__member_identifier), and thus not directly accessible outside their class, and can be used to approximate private members by Python programmers.

There is also no mechanism for protected identifiers - accessible only within a defining class and its subclasses - in the Python language, and so Python programmers have adopted the convention of using a single underscore (_identifier) at the start of any identifier that is intended to be protected (i.e., not to be accessed outside the class or its subclasses).

Some Python programmers only use the single underscore prefixes and avoid double underscore prefixes due to unintended consequences that can arise when names are mangled. The following warning about single and double underscore prefixes is issued in Code Like a Pythonista:

try to avoid the __private form. I never use it. Trust me. If you use it, you WILL regret it later

We will follow this advice and avoid using the double underscore prefix in user-defined member variables and methods.

Python has a number of pre-defined special method names, all of which are denoted by leading and trailing double underscores. For example, the object.__init__(self[, ...]) method is used to specify instructions that should be executed whenever a new object of a class is instantiated.

The code below defines a class, SimpleDecisionTree, with a single pseudo-protected member variable _tree and a pseudo-protected tree construction method _create(), two public methods - classify() and pprint() - and an initialization method that takes an optional list of training instances and a target_attribute_index.

The _create() method is identical to the create_decision_tree() function above, with the inclusion of the self parameter (as it is now a class method). The classify() method is a similarly modified version of the classify() and classification_accuracy() functions above, with references to tree converted to self._tree. The pprint() method prints the tree in a human-readable format.

Note that other machine learning libraries may use different terminology for the methods we've defined here. For example, in the sklearn.tree.DecisionTreeClassifier class (and in most sklearn classifier classes), the method for constructing a classifier is named fit() - since it "fits" the data to a model - and the method for classifying instances is named predict() - since it is predicting the class label for an instance.

Most comments and the use of the trace parameter have been removed to make the code more compact, but are included in the version found in SimpleDecisionTree.py.

In [106]:
class SimpleDecisionTree:

_tree = {} # this instance variable becomes accessible to class methods via self._tree

def __init__(self, instances=None, target_attribute_index=0): # note the use of self as the first parameter
if instances:
self._tree = self._create(instances, range(1, len(instances[0])), target_attribute_index)

def _create(self, instances, candidate_attribute_indexes, target_attribute_index=0, default_class=None):
class_labels_and_counts = Counter([instance[target_attribute_index] for instance in instances])
if not instances or not candidate_attribute_indexes:
return default_class
elif len(class_labels_and_counts) == 1:
class_label = class_labels_and_counts.most_common(1)[0][0]
return class_label
else:
default_class = simple_ml.majority_value(instances, target_attribute_index)
best_index = simple_ml.choose_best_attribute_index(instances, candidate_attribute_indexes, target_attribute_index)
tree = {best_index:{}}
partitions = simple_ml.split_instances(instances, best_index)
remaining_candidate_attribute_indexes = [i for i in candidate_attribute_indexes if i != best_index]
for attribute_value in partitions:
subtree = self._create(
partitions[attribute_value],
remaining_candidate_attribute_indexes,
target_attribute_index,
default_class)
tree[best_index][attribute_value] = subtree
return tree

# calls the internal "protected" method to classify the instance given the _tree
def classify(self, instance, default_class=None):
return self._classify(self._tree, instance, default_class)

# a method intended to be "protected" that can implement the recursive algorithm to classify an instance given a tree
def _classify(self, tree, instance, default_class=None):
if not tree:
return default_class
if not isinstance(tree, dict):
return tree
attribute_index = list(tree.keys())[0]  # using list(dict.keys()) for Python 3 compatibiity
attribute_values = list(tree.values())[0]
instance_attribute_value = instance[attribute_index]
if instance_attribute_value not in attribute_values:
return default_class
return self._classify(attribute_values[instance_attribute_value], instance, default_class)

def classification_accuracy(self, instances, default_class=None):
predicted_labels = [self.classify(instance, default_class) for instance in instances]
actual_labels = [x[0] for x in instances]
counts = Counter([x == y for x, y in zip(predicted_labels, actual_labels)])
return counts[True], counts[False], counts[True] / len(instances)

def pprint(self):
pprint(self._tree)


The following statements instantiate a SimpleDecisionTree, using all but the last 20 clean_instances, prints out the tree using its pprint() method, and then uses the classify() method to print the classification of the last 20 clean_instances.

In [107]:
simple_decision_tree = SimpleDecisionTree(training_instances)
simple_decision_tree.pprint()
print()
for instance in testing_instances:
predicted_label = simple_decision_tree.classify(instance)
actual_label = instance[0]
print('Model: {}; truth: {}'.format(predicted_label, actual_label))
print()
print('Classification accuracy:', simple_decision_tree.classification_accuracy(testing_instances))

{5: {'a': 'e',
'c': 'p',
'f': 'p',
'l': 'e',
'm': 'p',
'n': {20: {'k': 'e',
'n': 'e',
'r': 'p',
'w': {21: {'c': 'p', 'v': 'e', 'y': 'e'}}}},
'p': 'p'}}

Model: p; truth: p
Model: p; truth: p
Model: p; truth: p
Model: e; truth: e
Model: e; truth: e
Model: p; truth: p
Model: e; truth: e
Model: e; truth: e
Model: e; truth: e
Model: p; truth: p
Model: e; truth: e
Model: e; truth: e
Model: e; truth: e
Model: p; truth: p
Model: e; truth: e
Model: e; truth: e
Model: e; truth: e
Model: e; truth: e
Model: p; truth: p
Model: p; truth: p

Classification accuracy: (20, 0, 1.0)


## 5. Next steps¶

There are a variety of Python libraries - e.g., Scikit-Learn and xPatterns - for building more full-featured decision trees and other types of models based on a variety of machine learning algorithms. Hopefully, this primer will have prepared you for learning how to use those libraries effectively.

Many Python-based machine learning libraries use other external Python libraries such as NumPy, SciPy, Matplotlib and pandas. There are tutorials available for each of these libraries, including the following:

There are many machine learning or data science resources that may be useful to help you continue the journey. Here is a sampling:

Please feel free to contact the author (Joe McCarthy) to suggest additional resources.