Tutorial 3: CSE101 Statistics

Exercise 1: Computing Averages

Create a new program file called stats.py. Make sure you get the filename right, or else the testing code won’t work.

Our first task is to define a function to compute the average of a list of numbers. Write a function average which takes as input a non-empty list of numbers and returns their average as a float.

We want floats to show (at most) 2 digits after the decimal point. We will use rounding for this; calling round(x, 2) on a float x returns x rounded to two decimal places.

Here is the start of your function:

def average(numlist):
    """Return average of list of numbers"""
    pass # remove this line and replace with your own code

Warning: Adding two numbers rounded to 2 digits does not necessarily yields a number rounded to two digits! Try for instance this:

In [1]: 0.99+1-0.90
Out[1]: 1.0899999999999999

Testing

Test your code in the console, after running stats.py. You should be able to replicate this example:

In [2]: average([7.5, 5.5, 7])
Out[2]: 6.67

Upload your solution via the form below for further automatic testing.

Exercise 2: Convert a string of floats to a list

We now define a function to convert a string of floats into a list of numbers. Write a new function string_floats_to_list which takes as input a string representing a series of numerical values, separated by single spaces, and returns the list of these numerical values as floats.

Here is the start of your function:

def string_floats_to_list(string_floats):
    """Return a list from the string of floats string_floats."""
    pass # remove this line and replace with your own code

Hints:

• Use float(s) to turn a string s into a float.
• If s is a string, then s.split() returns the space-separated pieces of s (it does not modify s).

Testing

Now test your code in the console (after re-running stats.py). Here’s what your output should look like:

In [3]: string_floats_to_list('4.3 2.1 -3.2')
Out[3]: [4.3, 2.1, -3.2]

Once this test is passing correctly you can upload your solution to the form below for further automatic testing.

Exercise 3: Student Data

Next we want a function which can read a line login_name s1 s2 ..., like the ones in our file and return the data in a usable format. Write a function student_data which takes as input a string as above and returns a tuple (login_name, results), where results is the list of scores s1, s2, … converted to floats.

Here is the start of your function:

def student_data(data_string):
    """Compute (name, results) tuple from the string data_string."""
    pass # remove this line and replace with your own code

Testing

Now test your code in the console, after running stats.py. Your output should look like this:

In [4]: student_data('gleb.pogudin 7.5 5.5 8')
Out[4]: ('gleb.pogudin', [7.5, 5.5, 8.0])

Upload your solution to the form below for further automatic testing.

Exercise 4: Convert a tuple to a string

Now we want the inverse of the previous exercise: that is, we want a function which converts the components of a (name, results) tuple into a string.

Write a function student_data_to_string which takes as input a string and a list of floats, and returns a string obtained by concatenation of the string and elements of the list, separated by single spaces.

Here is the start of your function:

def student_data_to_string(name, results):
    """Return string from (name, results) tuple"""
    pass # remove this line and replace with your own code

Hint:

Use ' '.join(l) to construct a space-separated string out of a list l of strings. Example: ' '.join(['a', 'b']) becomes 'a b'.

Testing

Test your code in the console, after running stats.py. You should be able to replicate the following example:

In [5]: student_data_to_string('gleb.pogudin', [9.2,8.1,10])
Out[5]: 'gleb.pogudin 9.2 8.1 10'

Upload your solution to the form below, it will be tested automatically.

Exercise 5: Reading Student Data

Now write a function read_student_data which reads a file with student data like in CSE101data.txt and returns a list containing student data tuples.

This is the start of your function:

def read_student_data(filename):
    """Return list of student data from file"""
    pass # remove this line and replace with your own code

Hint:

Use with open(filename,'r') as file: to open a file for reading. To read one line of text, use file.readline() and to read all the (remaining) lines, use file.readlines().

For instance, with

with open('sample.txt') as infile:
	list_lines = infile.readlines()

then list_lines is a list containing each line in sample.txt as a list item.

Testing

Test your code in the console (after re-running stats.py). You should be able to reproduce the following example:

In [6]: read_student_data('CSE101test.txt')
Out[6]: 
[('gleb.pogudin', [7.5, 5.5, 8.0, 3.6, 5.5, 8.0]),
 ('uli.fahrenberg', [2.0, 6.0, 2.5, 6.5, 3.0, 7.0]),
 ('stefan.mengel', [6.0, 6.0, 9.0, 5.0, 5.0, 8.0])]

Upload your solution to the form below, it will be tested automatically.

Exercise 6: Extracting Averages

We will do more complicated statistics on our student data below, but let’s start off with something simple. Write a function extract_averages which reads a student data file (like CSE101data.txt) and returns a list of tuples, each containing a student name and the average of that student’s scores. As before, round floats to (at most) two decimal places, using round( x, 2).

This is the start of your function:

def extract_averages(filename):
    """Return list of name and average for each line in file"""
    pass # remove this line and replace with your own code

Hint: Use the function read_student_data to read in the file.

Testing

Test your code in the console, after running stats.py. You should be able to replicate the following example exactly:

In [7]: extract_averages('CSE101test.txt')
Out[7]: [('gleb.pogudin', 6.35), ('uli.fahrenberg', 4.5), ('stefan.mengel', 6.5)]

Now, upload your solution using the form below for further automatic testing.

Exercise 7: Filtering Lists of Scores

To prepare for the next exercise, we need a function to filter lists of scores. Write a function discard_scores which takes a list numlist of (at least 5) numbers as input, and constructs a new list such that if numlist has \(n\) entries, then the result contains the top \(n-4\) of the last \(n-2\) entries in numlist.

In other words, moving from numlist to the new result list, we

• first, discard the first two numbers from numlist;
• then, discard the lowest two numbers from what remains;
• and finally, return the list of remaining numbers.

Warning:

• your function should not modify numlist!
• the order of remaining items should be the same: do not use sort()!

As an example, applied to the list [2.0, 6.0, 6.5, 9.0, 6.5, 2.5], discard_scores will:

• ignore the first two scores, 2.0 and 6.0,
• and ignore the worst two of the remaining scores: that is, from [6.5, 9.0, 6.5, 2.5], it will keep only 9.0 and (one of the) 6.5.
• the result is [9.0, 6.5].

In the second step, if the choice of the two lowest numbers is ambiguous, then we discard the first two of the lowest numbers.

Your function should start as follows:

def discard_scores(numlist):
    """Filter numlist: construct a new list from numlist with
    the first two, and then the lowest two, scores discarded.
    """
    pass # remove this line and replace with your own code

Testing

Now test your code in the console, after running stats.py. You should be able to reproduce the following example:

In [8]: discard_scores([2.0, 6.0, 6.5, 9.0, 6.5, 2.5])
Out[8]: [9.0, 6.5]

Can you think of some more interesting tests?

Once you have tried some different tests, you can upload your solution to the form below for automatic testing.

Exercise 8: Summaries per Student

We want to use the student data in CSE101data.txt to write a file with student summaries. For each student, we

• use discard_scores to discard the first two and lowest two scores;
• take the sum of the remaining scores;
• write a line to an output file containing
  • the student’s name, followed by a single space;
  • all remaining scores, separated by single spaces;
  • the string ' sum: ' followed by the computed sum.

Finally, the file should finish with a line containing the string 'total average: ' followed by the the average of all the summed scores. The file must end with a newline character \n. Also remember to round floats to (at most) two decimals.

Write a function summary_per_student which accomplishes this. Your function should start as follows:

def summary_per_student(infilename, outfilename):
    """Create summaries per student from the input file 
    and write the summaries to the output file.
    """
    data = read_student_data(infilename)
    pass # remove this line and replace with your own code

Hints:

• Use open(outfile, 'w') to open a file for writing.
• Use file.write('\n') to finish a line when writing to a file.

Testing

Test your code in the console, after re-running stats.py:

In [9]: summary_per_student('CSE101test.txt', 'out1.txt')

Now open the output file out1.txt (look in the list of files on the left of the main Spyder window). The generated file out1.txt should look like this:

gleb.pogudin 8.0 8.0 sum: 16.0
uli.fahrenberg 6.5 7.0 sum: 13.5
stefan.mengel 9.0 8.0 sum: 17.0
total average: 15.5

Now, a second test:

In [10]: summary_per_student('CSE101data.txt', 'out2.txt')

Have a look at out2.txt to see whether your code works on this more realistic data.

Finally, upload your solution using the form below; it will be tested automatically.

Exercise 9: Summaries per Tutorial

Now we want to create summaries for each tutorial. For each tutorial, we write to an output file a line containing the name of the tutorial together with its average, minimum, and maximum scores.

Since this is a French school, the tutorials are named TD1, TD2, TD3, etc (“TD” meaning travaux dirigés).

Note that no scores are to be removed from the information; we want the complete picture here. As usual, we use round on our floats before displaying them, to show (at most) 2 digits after the decimal point. The file must end with a newline character \n.

Write a function summary_per_tutorial which accomplishes this. Here’s a start:

def summary_per_tutorial(infilename, outfilename):
    """Create summaries per student from infile and write to outfile."""
    data = read_student_data(infilename)
    pass # remove this line and replace with your own code

Testing

Test your code in the console, after running stats.py, as follows:

In [11]: summary_per_tutorial('CSE101test.txt', 'out3.txt')

Here’s what the generated file out3.txt should look like:

TD1: average: 5.17 min: 2.0 max: 7.5
TD2: average: 5.83 min: 5.5 max: 6.0
TD3: average: 6.5 min: 2.5 max: 9.0
TD4: average: 5.03 min: 3.6 max: 6.5
TD5: average: 4.5 min: 3.0 max: 5.5
TD6: average: 7.67 min: 7.0 max: 8.0

Now try running your code on some more data:

In [12]: summary_per_tutorial('CSE101data.txt', 'out4.txt')

Have a look at out4.txt to see whether your code works on this data too.

Once this is done, you can upload your solution using the form below for automatic testing.