Lesson 7 - Miscellany

Learning Objectives

Concepts - After completing this lesson, students will be able to:

• Compare and contrast plain text and binary file formats

• Select the best data structure for different types of data

Skills - After completing this lesson, students will be able to:

• Download files from a url from the command line

• Read files line-by-line and perform actions on each line

Tasks - This lesson is complete when students have:

• Read Chapter: 9 from Think Julia

• Read Chapter: 14 from Think Julia

• Completed Assignment06

• Run all code examples from Lesson 7 on their own computers

Beyond Base - Statistics

All of the functionality we've used thus far has been found in the main julia library, called Base. But one of the great things about programming languages, especially open-source ones like julia, is that many people have written a lot of other functionality and shared it with the world in "packages."^[1]

This functionality is not available out-of-the-box, but you can easily bring this world of additional functions and types into your own code. Some commonly used packages are installed along with julia (these are called the "standard library" or "stdlib"), and others can be installed using the "package manager."

For now, we'll just deal with the packages in stdlib.

Bringing in other functions

Julia was designed for numerical computing, but there are a many functions commonly used in statistics that are not available when you first start up julia. For example, it is often useful to calculate the mean, median, and standard deviation of numbers in a vector.

julia> v = rand(100); # create vector of random numbers

julia> mean(v)
ERROR: UndefVarError: mean not defined

julia> median(v)
ERROR: UndefVarError: median not defined

Defining my_mean() isn't so hard, but does your function also work on Tuples? What about a matrix (a 2-dimensional array) like m = [1 2; 1 2; 4 5]? And what about my_median() or my_standard_deviation()? You could probably figure out how to define those functions given what you've learned so far, but lots of other people have probably needed this functionality before, and the code is already written and tested.

In this case, the functionality is part of the stdlib in the Statistics package. To load the functionality into your own code, we use the keyword using:

julia> using Statistics

julia> mean(v)
0.48948120851162236

julia> median(v)
0.47657406120014545

julia> std(v)
0.2725349745172383

Beyond Base - other libraries

In the stdlib, there are also packages for working with dates, generating random numbers, and for doing linear algebra, among others, but a huge amount of additional functionality can be found in the broader package ecosystem.

Installing these packages takes just a little bit of extra effort, and we'll learn about that in future lessons, but it's important to realize that these packages are just like the code you've been writing this whole time. In fact - the lessons you've already completed are packages!

Practice, practice, practice

Both chapters from Think Julia this week are practical exercises, but contain a lot of concepts that will be very helpful as we start to more earnestly deal with biological data. Several of the assignment questions are based on the examples in the book. Try to solve the problems in the book before looking at the answers, as struggling through these practice problems will help when it comes time to do the assignment.

Good luck!

Key Terms