Search code examples
pythonpython-3.xloopsmean-square-error

How do I create a loop repeating the following but each time deducting 0.1 from a?


How do I create a loop repeating the following but each time deducting 0.1 more from a=10? It should repeat a 100 times and then stop. Thanks!

for i in x:
    yt = (a - 0.1)* i
    MSE = np.square(np.subtract(y,yt)).mean()

Solution

  • You could use a while loop instead, like the following:

    a = 10
    while a > 0:
        yt = (a-0.1)
        MSE = np.square(np.subtract(y,yt)).mean()
        a -= 0.1
    

    That way, if a == 0 the loop stops and yt won't become 0. If that is what you are asking for. Due to accurancy problems, often a repeated -0.1 will result in rounding errors and could create unwanted results. Thus I recommand using something like this:

    a = 100
    while a > 0:
        yt = (a/10-0.1)
        MSE = np.square(np.subtract(y,yt)).mean()
        a -= 1
    

    Alternatively after the newest comment: Using temporary values which get compared iteratively every loop index:

    import numpy as np
    a = 10
    y = 5.423           #example value
    tmp_MSE = np.infty  #the first calculated MSE is always smaller then infty
    tmp_a = a           #if no modified a results in a better MSE, a itself is closest 
    for i in range(100):
        yt = a-0.1*(i+1)
        MSE = np.square(np.subtract(y,yt)).mean()
        if MSE < tmp_MSE: #new MSE comparison
            tmp_MSE = MSE #tmp files are updated
            tmp_a = yt
    
    print("nearest value of a and MSE:",tmp_a,tmp_MSE)