Â
Â
Â
Â
Â
Â
Project 5 (Markov Model)
Clarifications and Hints
1 / 11
Prologue
Project goal: use a Markov chain to create a statistical model of a piece of English text
and use the model to generate stylized pseudo-random text and decode noisy messages
The zip file (
http://www.swamiiyer.net/cs110/markov_model.zip
) for the project contains
•
project specification (
markov_model.pdf
)
•
starter files
•
markov_model.py
•
text_generator.py
•
fix_corrupted.py
•
test script (
run_tests.py
)
•
test data (
data/
)
•
report template (
report.txt
)
This checklist will help only if you have read the writeup for the project and have
a good understanding of the problems involved. So, please read the project
writeup
m
before you continue with this checklist.
2 / 11
Â
Prologue
Understanding how dictionaries work is crucial for this project, so make sure you
understand the following example which illustrates how you create a dictionary of
dictionaries and how you manipulate them
>>> M = {} # create an empty dictionary M
>>> M.setdefault(’ba ’, {}) # add key/value pair ’ba ’/{} to M
{} # since ’ba ’ didn ’t exist in M,
# {} (the value just added) is returned
>>> M # check M
{’ba ’: {}}
>>> M[’ba ’]. setdefault(’n’, 0) # add key/value pair ’n’/0 to the
# dictionary M[’ba ’]
0 # since ’n’ didn ’t exist in M[’ba ’],
# 0 (the value just added) is returned
>>> M # check M
{’ba ’: {’n’: 0}}
>>> M[’ba ’][’n’] += 1 # increment the value corresponding to the
# key ’n’ in the dictionary M[’ba ’] by 1
>>> M # check M
{’ba ’: {’n’: 1}}
>>> M[’ba ’]. setdefault(’n’, 42) # add key/value pair ’n’/42 to the
# dictionary M[’ba ’]
1 # since ’n’ exists in M[’ba ’],
# setdefault () simply returns (without
# changing) the corresponding value , 1
>>> M # check M
{’ba ’: {’n’: 1}}
>>> M.setdefault(’an ’, {}) # add key/value pair ’an ’/{} to M
{}
>>> M # check M
{’ba ’: {’n’: 1}, ’an ’: {}}
3 / 11
