Math domain error when using linalg.lstsq (numpy)

Question

So I'm trying to use linear regression to show a trendline when entering a set of data points. I'm using Tkinter to get an input for the data, and then converting them to float to put them in a list. I get this error code when I run the program though.

    Exception in Tkinter callback
    Traceback (most recent call last):
      File "C:\Python27\lib\lib-tk\Tkinter.py", line 1486, in __call__
        return self.func(*args)
      File     "C:/Users/NIBO9901/PycharmProjects/Matteuppgift/Trendlinje/Input.py", line 78,  in plot
        m, c = np.linalg.lstsq(a, y)[0]
      File "C:\Python27\lib\site-packages\numpy\linalg\linalg.py", line 1889, in lstsq
        nlvl = max( 0, int( math.log( float(min(m, n))/2. ) ) + 1 )
    ValueError: math domain error

And the code that it's refering to is here:

    xList = []
    yList = []

    x = np.array(xList)
    y = np.array(yList)

    if 0 < len(inpX0.get()):
        xList.append(float(inpX0.get()))

    if 0 < len(inpX1.get()):
        xList.append(float(inpX1.get()))

    if 0 < len(inpY0.get()):
        yList.append(float(inpY0.get()))

    if 0 < len(inpY1.get()):
        yList.append(float(inpY1.get()))

    a = np.vstack([x, np.ones(len(x))]).T
    m, c = np.linalg.lstsq(a, y)[0]

    plt.plot(x, y, 'o', label='Data', markersize=10)
    plt.plot(x, m*x + c, 'r', label='Trendlinje')
    plt.legend()
    plt.show()

inpX/Y are the Tkinter entries.

Answer 1

numpy arrays are static structures. You first have to populate your list and then convert it to a numpy array. After that, modifications to the list will not have an effect on the numpy array anymore.

So what you want to do is probably:

xList = []
yList = []

if 0 < len(inpX0.get()):
    xList.append(float(inpX0.get()))

if 0 < len(inpX1.get()):
    xList.append(float(inpX1.get()))

if 0 < len(inpY0.get()):
    yList.append(float(inpY0.get()))

if 0 < len(inpY1.get()):
    yList.append(float(inpY1.get()))

a = np.vstack([x, np.ones(len(x))]).T
m, c = np.linalg.lstsq(a, y)[0]

# the lists are complete, now convert them to numpy arrays
x = np.array(xList)
y = np.array(yList)

plt.plot(x, y, 'o', label='Data', markersize=10)
plt.plot(x, m*x + c, 'r', label='Trendlinje')
plt.legend()
plt.show()