在熊猫数据框的每一行上应用功能

Question

How can I apply the function point_rotation function on Coordinates1 , Coordinates2 columns in the below dataframe:

def point_rotation(point):
    """
    Changing axis
    """
    xcor, ycor = (700, 0)
    xcor1, ycor1 = (0, 0)
    xpoint, ypoint = point

    xnew = xpoint - xcor
    ynew = ypoint - ycor    
    xnew = xcor1 + math.cos(math.radians(270)) * (xnew - xcor1) - math.sin(math.radians(90)) * (ynew - ycor1)
    ynew = ycor1 + math.sin(math.radians(270)) * (xnew - xcor1) + math.cos(math.radians(90)) * (ynew - ycor1)
    return round(xnew, 0), round(ynew, 0)

Here's my current dataframe:

df = pd.DataFrame({'X_1': [-34.58, -15.78, -33.45, 4.60, 10.48],
                   'Y_1': [-58.66, -47.91, -70.66, -74.08, -66.86],
                   'X_2': [-3.58, -1.8, -3.5, 4.0, 1.48],
                   'Y_2': [-5.66, -4.1, -7.6, -7.8, -6.86]})

df['Coordinates1'] = list(zip(df.X_1, df.Y_1))
df['Coordinates2'] = list(zip(df.X_2, df.Y_2))

Output wanted: should have columns Coordinates3 and Coordinates4 which are basically derived from the point_rotation function by passing columns Coordinates1 and Coordinates2 .

I tried using the apply function, but it throws me an error: too many values to unpack (expected 2) .

Thanks for the help!

Answer 1

If you use list comprehension. Your new point will be a tuple (similar to how you have defined Coordinates1 and Coordinates2 )

df['Coordinates3'] = [point_rotation(point) for point in df['Coordinates1']]

apply also seems to work :)

df['Coordinates4'] = df['Coordinates2'].apply(point_rotation)

Answer 2

Just do

df['Coordinates3'] = df['Coordinates1'].apply(point_rotation)
df['Coordinates4'] = df['Coordinates2'].apply(point_rotation)