Worksheet 2#

This worksheet is due Tuesday of Week 2, by 11:59pm.

You are encouraged to work in a group of up to 3 total students (working alone is also okay). Everyone should make a submission on Canvas, but it’s fine for all group members to submit the same thing.

See Worksheet 1 for instructions on how to make a Education workspace (which will provide higher-priority access to Deepnote resources).

Loading a NumPy array from a pickle file#

  • Import numpy using the usual abbreviation np, and also import the Python pickle module (no abbreviation).

  • Notice that there is an attached file wkst2-starter.pickle. Load the data from that file and store it in the variable arr using the following code.

with open("wkst2-starter.pickle", "rb") as f:
    arr = pickle.load(f)

  • Check the type of this object using the function type and check its dimensions by computing the shape attribute (no parentheses after shape). The variable arr should represent an 8000 by 7 NumPy array.

Practice with Boolean indexing#

  • Define the variable x to be the column at index 4.

  • Check that x is a length-8000 NumPy array.

  • How many values in x are equal to 10?

  • Define arrsub to be the sub-array of arr containing all the rows for which the column at index 4 is equal to the integer 10. Use Boolean indexing with arr and x.

  • What is the shape of arrsub? Use the shape attribute again. (It should be 7 columns, but how many rows?)

  • Define a new variable, a, to be equal to the number of rows of arrsub.

Counting rows satisfying a condition using NumPy#

  • How many rows in arr contain the number 40? You can use arr == 40 to create a Boolean array. From the resulting Boolean array, you can use the sum method, in particular, .sum(axis=1), to count for each row, how many times 40 appears. We want to find all the rows where 40 occurs at least once. Once you have a length-8000 array of True and False values (with True if 40 occurs in the row and with False otherwise). You want to count the number of times True occurs. Save the result with the variable name b. (Reality check: the answer should be between 500 and 700.)

Submission#

  • Store these answers, as well as the names of everyone in your group, as a Python dictionary

ansdict = {"names": [name1, name2, name3], "a": a, "b": b}

The "names" value must be a list, with length the number of group members, and with entries the full names of your group members. The names of your group members should be strings. For example, ["Chris Davis", "Jinghao Chen"].

  • Save the resulting dictionary in a pickle file using the following code.

with open("wkst2-ans.pickle", "wb") as f:
    pickle.dump(ansdict, f)

  • Submit this pickle file on Canvas.

Created in deepnote.com Created in Deepnote