Description
In today’s lab, you will:
1. Utilize cProfile to profile code that takes a while to run. You will identify which parts of
the code are inefficient and change them.
2. Use comprehensions, built-in functions, data structures and other tricks to optimize code.
3. Get first hand experience working with not-so-good code. This is going to happen a lot
once you work. The standard isn’t very high out there.
Grading
This lab and all future labs will be marked out of 10
For full marks this week, you must:
1. (5 points) Improve the given code using profiling to make it run faster . (Part 1)
2. (5 point) Further optimize the improved code after you have reduced the run-time. (Part
2)
Requirements
Make sure you read both parts before starting on Part 1.
Part 1: Profiling
Download the lab files from D2L. The zip file containes a book_analyzer.py module and a
House of Usher.txt text file. Place these files in your lab07 folder and run the code in
book_analyzer.
The BookAnalyzer class is responsible for reading a text file and providing a method to extract
all the words that appear only once. This code is not written very well and takes a few seconds to
run. This will be different for each machine. On my machine it took approximately 10 seconds.
When faced with a situation like this, profilers are a super convenient tool to help us identify
what is taking so long to run. We can then edit and change only those parts of the code that
contribute to the run time.
1. Duplicate this module, or copy the code into a new module called
book_analyzer_profiled.py.
2. In this new module, you will repeat the following steps until you have improved the code
as much as you can without re-designing the code and changing the data types. :
1. Profile the code and identify which parts of the code take the longest to run/ are
inefficient.
2. Interpret the results and identify the parts of the code that are contributing to this
long run time. Change only those parts so that the code runs faster. (You can
restructure and change the code, but you cannot re-design the code or change the
data structures. Not yet anyway.)
By the end of this part, My code was running at approximately 3-4 seconds. This is a lot better
but still isn’t the best. Remember the goal here is to improve the performance of the program, not
to re-design our code. Once this done move on to Part 2!
Part 2: Optimizing and Re-designing
1. Duplicate your code from book_analyzer_profiled.py into a new module called
book_analyzer_optimized.py .
2. Go through the Lab 07 Optimization Slides on D2L to learn how to optimize python
code.
3. Optimize and re-design how the Book Analyzer class reads and processes data. Also rewrite the code in find_unique_words(self) to run in O(N) time instead of O(N^2).
o You can save a lot of time and energy by creating a new attribute called
word_count when reading data from the file. What data structure would you use
for this?).
o Remember to re-profile and check how your modifications effect the results.
o What built-in methods can you use to improve the code? Can generators and/or
comprehensions be used? Can you reduce the number of times you iterate over
the data?
4. By the end of this part your code should run in less than a second.
After you submit your Lab to GitHub, send the marker a private message on Discord telling them
you finished, along with your student number, gitName, and collaboration url.
• ie: “I uploaded my work to gitHub. My student number is A00XXXXXX and my
gitName is YYYYY, my git collaboration url is: ZZZZZZZZZZZZZZZ”
That’s it. Good luck, and have fun!
