Introduction
Last updated on 2023-04-25 | Edit this page
Overview
Questions
- How can I make my results easier to reproduce?
Objectives
- Explain what Make is for.
- Explain why Make differs from shell scripts.
- Name other popular build tools.
Let’s imagine that we’re interested in testing Zipf’s Law in some of our favorite books.
Zipf’s Law
The most frequently-occurring word occurs approximately twice as often as the second most frequent word. This is Zipf’s Law.
We’ve compiled our raw data i.e. the books we want to analyze and have prepared several Python scripts that together make up our analysis pipeline.
Let’s take quick look at one of the books using the command
head books/isles.txt
.
Our directory has the Python scripts and data files we will be working with:
OUTPUT
|- books
| |- abyss.txt
| |- isles.txt
| |- last.txt
| |- LICENSE_TEXTS.md
| |- sierra.txt
|- plotcounts.py
|- countwords.py
|- testzipf.py
The first step is to count the frequency of each word in a book. For
this purpose we will use a python script countwords.py
which takes two command line arguments. The first argument is the input
file (books/isles.txt
) and the second is the output file
that is generated (here isles.dat
) by processing the
input.
Let’s take a quick peek at the result.
This shows us the top 5 lines in the output file:
OUTPUT
the 3822 6.7371760973
of 2460 4.33632998414
and 1723 3.03719372466
to 1479 2.60708619778
a 1308 2.30565838181
We can see that the file consists of one row per word. Each row shows the word itself, the number of occurrences of that word, and the number of occurrences as a percentage of the total number of words in the text file.
We can do the same thing for a different book:
OUTPUT
the 4044 6.35449402891
and 2807 4.41074795726
of 1907 2.99654305468
a 1594 2.50471401634
to 1515 2.38057825267
Let’s visualize the results. The script plotcounts.py
reads in a data file and plots the 10 most frequently occurring words as
a text-based bar plot:
OUTPUT
the ########################################################################
of ##############################################
and ################################
to ############################
a #########################
in ###################
is #################
that ############
by ###########
it ###########
plotcounts.py
can also show the plot graphically:
Close the window to exit the plot.
plotcounts.py
can also create the plot as an image file
(e.g. a PNG file):
Finally, let’s test Zipf’s law for these books:
OUTPUT
Book First Second Ratio
abyss 4044 2807 1.44
isles 3822 2460 1.55
So we’re not too far off from Zipf’s law.
Together these scripts implement a common workflow:
- Read a data file.
- Perform an analysis on this data file.
- Write the analysis results to a new file.
- Plot a graph of the analysis results.
- Save the graph as an image, so we can put it in a paper.
- Make a summary table of the analyses
Running countwords.py
and plotcounts.py
at
the shell prompt, as we have been doing, is fine for one or two files.
If, however, we had 5 or 10 or 20 text files, or if the number of steps
in the pipeline were to expand, this could turn into a lot of work.
Plus, no one wants to sit and wait for a command to finish, even just
for 30 seconds.
The most common solution to the tedium of data processing is to write a shell script that runs the whole pipeline from start to finish.
So to reproduce the tasks that we have just done we create a new file
named run_pipeline.sh
in which we place the commands one by
one. Using a text editor of your choice (e.g. for nano use the command
nano run_pipeline.sh
) copy and paste the following text and
save it.
BASH
# USAGE: bash run_pipeline.sh
# to produce plots for isles and abyss
# and the summary table for the Zipf's law tests
python countwords.py books/isles.txt isles.dat
python countwords.py books/abyss.txt abyss.dat
python plotcounts.py isles.dat isles.png
python plotcounts.py abyss.dat abyss.png
# Generate summary table
python testzipf.py abyss.dat isles.dat > results.txt
Run the script and check that the output is the same as before:
This shell script solves several problems in computational reproducibility:
- It explicitly documents our pipeline, making communication with colleagues (and our future selves) more efficient.
- It allows us to type a single command,
bash run_pipeline.sh
, to reproduce the full analysis. - It prevents us from repeating typos or mistakes. You might not get it right the first time, but once you fix something it’ll stay fixed.
Despite these benefits it has a few shortcomings.
Let’s adjust the width of the bars in our plot produced by
plotcounts.py
.
Edit plotcounts.py
so that the bars are 0.8 units wide
instead of 1 unit. (Hint: replace width = 1.0
with
width = 0.8
in the definition of
plot_word_counts
.)
Now we want to recreate our figures. We could just
bash run_pipeline.sh
again. That would work, but it could
also be a big pain if counting words takes more than a few seconds. The
word counting routine hasn’t changed; we shouldn’t need to recreate
those files.
Alternatively, we could manually rerun the plotting for each word-count file. (Experienced shell scripters can make this easier on themselves using a for-loop.)
With this approach, however, we don’t get many of the benefits of having a shell script in the first place.
Another popular option is to comment out a subset of the lines in
run_pipeline.sh
:
BASH
# USAGE: bash run_pipeline.sh
# to produce plots for isles and abyss
# and the summary table for the Zipf's law tests.
# These lines are commented out because they don't need to be rerun.
#python countwords.py books/isles.txt isles.dat
#python countwords.py books/abyss.txt abyss.dat
python plotcounts.py isles.dat isles.png
python plotcounts.py abyss.dat abyss.png
# Generate summary table
# This line is also commented out because it doesn't need to be rerun.
#python testzipf.py abyss.dat isles.dat > results.txt
Then, we would run our modified shell script using
bash run_pipeline.sh
.
But commenting out these lines, and subsequently uncommenting them, can be a hassle and source of errors in complicated pipelines.
What we really want is an executable description of our pipeline that allows software to do the tricky part for us: figuring out what steps need to be rerun.
For our pipeline Make can execute the commands needed to run our analysis and plot our results. Like shell scripts it allows us to execute complex sequences of commands via a single shell command. Unlike shell scripts it explicitly records the dependencies between files - what files are needed to create what other files - and so can determine when to recreate our data files or image files, if our text files change. Make can be used for any commands that follow the general pattern of processing files to create new files, for example:
- Run analysis scripts on raw data files to get data files that summarize the raw data (e.g. creating files with word counts from book text).
- Run visualization scripts on data files to produce plots (e.g. creating images of word counts).
- Parse and combine text files and plots to create papers.
- Compile source code into executable programs or libraries.
There are now many build tools available, for example Apache ANT, doit, and nmake for Windows. Which is best for you depends on your requirements, intended usage, and operating system. However, they all share the same fundamental concepts as Make.
Also, you might come across build generation scripts e.g. GNU Autoconf and CMake. Those tools do not run the pipelines directly, but rather generate files for use with the build tools.
Why Use Make if it is Almost 40 Years Old?
Make development was started by Stuart Feldman in 1977 as a Bell Labs summer intern. Since then it has been undergoing an active development and several implementations are available. Since it solves a common issue of workflow management, it remains in widespread use even today.
Researchers working with legacy codes in C or FORTRAN, which are very common in high-performance computing, will, very likely encounter Make.
Researchers can use Make for implementing reproducible research workflows, automating data analysis and visualisation (using Python or R) and combining tables and plots with text to produce reports and papers for publication.
Make’s fundamental concepts are common across build tools.
GNU Make is a free-libre, fast, well-documented, and very popular Make implementation. From now on, we will focus on it, and when we say Make, we mean GNU Make.
Key Points
- Make allows us to specify what depends on what and how to update things that are out of date.