You have until Sunday, May 2, 11:00pm EDT to submit problems 2, 3.2, and 3.3 on Matlab Grader.
1.1 Is the following implementation of instance method overlap for class Interval correct? Why or why not?
function Inter = overlap(self, other) % If self and the other Interval overlap, then Inter is the handle of the % overlapped Interval; otherwise Inter is an empty array of type Interval. left= max(self.left, other.left); right= min(self.right, other.right); if left >= right Inter= Interval.empty(); end end
1.2 Download the file Interval.m from the Exercises page. Complete/revise the three methods getWidth(), scale(), and spanWith(). Then in the Command Window write some code to try out the new class definition, e.g.,
a= Interval(3,7); w= a.getWidth() % w should be 4 a.scale(2) disp(a) % a should be (3,11) b= Interval(0,2); c= a.spanWith(b) % c should be (0,11)
Suppose we have an array of Intervals (.1,.5), (.2,.7), and (.8,.9). If we ask whether an Interval (.3,.4) “is in” the array of Intervals, we will answer “yes, yes, no” since
(.3,.4) is in (.1,.5),
(.3,.4) is in (.2,.7), but
(.3,.4) is not in (.8,.9).
Recall that in class Interval we have an isIn() method that would be useful here. Make effective use of the isIn() method in implementing the following function (in its own file). Solve this problem using the full Matlab environment first.
function tf = isInRange(inter, interArray) % inter: an Interval % interArray: 1-d array of Intervals; interArray is not empty % tf: a logical vector the same length as interArray where tf(k) is true % if inter "is in" interArray(k); otherwise tf(k) is false.
Test your function by typing the following in the Command Window (you will first need to download and read intervalArray.m):
a = intervalArray(4); % Create a length 4 array of Intervals using the % function implemented in lecture. b = Interval(.3,.5) v = isInRange(b,a) % Why isn’t the function call something.isInRange(...) ? % Answer: isInRange isn’t an instance method; it’s its own function. % Check the type of v; it should be logical (not double).
Download file LocalWeather.m and read it. Ask questions if there are parts from what we did in lecture (constructor, method showMonthData) that you do not understand. Two of the methods, getAnnualPrecip() and getMonthlyAveTemps(), are incomplete (contains “dummy code” that does no calculation and only assigns a value to the return parameter); you will complete them later using the full Matlab environment first.
3.1 Experiment! First, download the file ithacaWeather.txt which contains weather data for the City of Ithaca. In the Command Window, instantiate (create) a LocalWeather object using the data file:
ithaca= LocalWeather('ithacaWeather.txt')
You should see the properties of ithaca displayed. Note that one of the properties, temps, is an array of Interval objects. Type the following commands in the Command Window; make sure you understand the syntax for accessing values.
disp(ithaca.city) % display the value in the property city disp(ithaca.precip) % display the values in the property precip--a vector! disp(ithaca.precip(11)) % What is displayed? What is it? ____________________________ disp(ithaca.temps) % Matlab says it’s a 1-by-12 array of INTERVALs disp(ithaca.temps(11)) % Notice that the disp method in class Interval is % used to show the data using Interval notation. disp(ithaca.temps(11).left) % What is displayed? What is it? ________________________
3.2 Implement function getAnnualPrecip() which calculates and returns the total annual precipitation. If any month’s precipitation data is missing, the returned value should be NaN
, a value in Matlab of type double that indicates that a value is not-a-number.
Test your updated class. Save class LocalWeather, and type the following in the Command Window:
ithaca= LocalWeather('ithacaWeather.txt') % instantiate object % Which of the following two method calls is correct? Try them! a = getAnnualPrecip() b = ithaca.getAnnualPrecip()
Change some values in the data file ithacaWeather.txt and call your method again. Test your work thoroughly.
3.3 Implement function getMonthlyAveTemps() which returns the vector (length 12) of monthly average temperatures. Calculate a month’s average temperature as the average between the month’s high and low temperatures. See the function comment of getMonthlyAveTemps() for how any missing temperature (NaN
) should be handled. The built-in function isnan()
can be used to check whether a variable stores the value NaN
: isnan(x)
returns true
if x is NaN
and false
otherwise.
Again, save and test your updated class. Back in the Command Window, call the instance method getMonthlyAveTemps() to make sure that it works. Then change some data in the data file and test your method again.