Page - 48 - in Programming for Computations – Python - A Gentle Introduction to Numerical Simulations with Python 3.6, Volume Second Edition

48 2 AFewMoreSteps In [4]: type(x) # check type of array as a whole Out[4]: numpy.ndarray In [5]: type(x[0]) # check type of array element Out[5]: numpy.float64 The line x = linspace(0, 2, 3) makes Python reserve, or allocate, space in memory for the array produced by linspace. With the assignment, x becomes a reference to the array object from linspace, i.e. x becomes the “name of the array”.Thearraywillhavethreeelementswithnamesx[0],x[1]andx[2],where thebracketednumbersare referred toas indices (note thatwhenreading,wesay“x ofzero” tox[0], “xofone” tox[1], andsoon). Observethat the indexingstartswith0,so thatanarraywithnelementswillhave n-1 as the last index. We say that Python has zero based indexing, which differs fromone based indexingwherearray indexingstarts with 1 (as, e.g., in Matlab). In x, the value ofx[0] is 0.0, the value ofx[1] is 1.0 and, finally, the value ofx[2] is2.0.Thesevaluesaregivenby theprintoutarray([ 0., 1., 2.])above. With thecommandtype(x),we confirmthat thearrayobjectnamedxhas type numpy.ndarray.Note that, at thesame time, the individualarrayelements refer to objects with another type. We see this from the very last command,type(x[0]), which makes Python respond withnumpy.float64 (being just a certain float data type in NumPy9). If we continue the previous dialogue with a few lines, we can also demonstrate thatuseof individualarrayelements is straight forward: In [4]: sum_elements = x[0] + x[1] + x[2] In [5]: sum_elements Out[5]: 3.0 In [6]: product_2_elements = x[1]*x[2] In [7]: product_2_elements Out[7]: 2.0 In [8]: x[0] = 5.0 # overwrite previous value In [9]: x Out[9]: array([ 5., 1., 2.]) The Zeros Function There are other common ways to generate arrays too. One way is to use anothernumpy function namedzeros, which (as the name suggests) may be used to produce an array with zeros. These zeros can be either floating point numbers or integers, depending on the arguments provided when zeros is called.10 Often, the zerosare overwritten in a secondstep to arriveat an array with thenumbersactuallywanted. 9 You may check out the many numerical data types in NumPy at https://docs.scipy.org/ doc/numpy-1.13.0/user/basics.types.html. 10 https://docs.scipy.org/doc/numpy-1.13.0/reference/generated/numpy.zeros.html.