LinearRegression.py

# A dependent column must have continuous data.  If it is continuous you should use a linear regression model.  
    
# we create a dictionary 
eruptions = {
        "eruptions": [3.6, 1.8, 3.333, 2.283, 4.533, 2.883],
        "waiting":   [79,  54,  74,    62,    85,    55]
    }

# we convert the dictionary into a dataframe
edf = pd.DataFrame(eruptions)

# we plot the dataframe
edfPlot = plt.plot(edf["eruptions"], edf["waiting"])

# we confirm there is a correlation
if edf.corr()["eruptions"][1] > 0.5:
    
    # We store each line into different variables
    x = edf[["eruptions"]]
    y = edf["waiting"]
    
    # create a linear regression object
    m = LinearRegression()
    
    # Trains out model.  y = mx + c a very simple model for calculating slope...
    # chooses the line with the lowest error lines sum. 
    m.fit(x, y)
    
    # provides the slope. 
    slope = m.coef_
    
    # intercept
    intercept = m.intercept_
    
    # 100 is the x value 
    yvalue = slope * 100 + intercept
    print(yvalue)
    
    # predict for one single value
    single = m.predict([[100]])
    print(single)
    
    # predict for multiple values create a dataframe.
    pr = pd.DataFrame({"eruptions": [30, 400, 990, 2349]})
    waiting = m.predict(pr)
    print(waiting)
    prd = m.predict(x)
    
    plt.plot(edf["eruptions"], edf["waiting"])