通过迭代对数据框的所有行应用函数 - Python

Question

我正在处理这个数据框：

Detection_Location	Blast Hole	East Coordinate	North Coordinate	Tag Detector ID	Detection Start Time	Detection end time	Tags
CV22	105,100,99	16764.83,16752.74,16743.1	107347.67,107360.32,107362.96	385742468,385112050,385087366	2018-09-06 20:02:46	2018-09-06 20:49:21	3
CV23	63,64,61	16755.07,16745.42,16773.48	107387.68,107390.32,107382.6	385262370,385656531,385760755	2018-09-08 14:12:42	2018-09-08 14:24:19	3
CV22	5,35,19	16757.27,16747.75,16770.89	107452.4,107417.68,107420.83	385662254,385453358,385826979	2018-09-23 05:01:12	2018-09-23 05:52:54	3

我正在尝试从“东坐标”列中提取 X 坐标，从“北坐标”列中提取 y 坐标。我编写了一个函数来计算每行中 3 个点的质心，然后计算到每个点的距离并进行验证。它适用于一排。

我想为每一行应用该函数，然后返回答案。我也尝试过 df.iterrows 和 df.apply 但两者都对所有行给出相同的答案，所以很明显它不起作用。

这是代码：（只有最后一部分是如何在每一行上应用该函数并获得结果作为原始数据集中的附加列将是一个挑战，其余代码仅用于理解。）

def Calculate_dispersion(row):
    #Picking up the columns with x coordinates.
    df2 = df['East Coordinate'].tolist()
    #Picking up the columns with y coordinates.
    df3 = df['North Coordinate'].tolist()


    #Splitting the list into seperate x coordinates.
    df4 = pd.DataFrame([sub.split(",") for sub in df2])
    #Splitting the list into seperate y coordinates.
    df5 = pd.DataFrame([sub.split(",") for sub in df3])


    #Creating a tuple with x coordinates
    x1 = df4.iloc[0]
    x2 = x1.tolist()
    x3 = tuple(float(x) for x in x2)
    #Creating a tuple with y coordinates
    y1 = df5.iloc[0]
    y2 = y1.tolist()
    y3 = tuple(float(x) for x in y2)


    #Creating the Coordinate tuple for centroid calculation.
    c = (x3,y3)

    #Calculating centroid.
    centroid = (sum(c[0])/len(c[0]),sum(c[1])/len(c[1]))
    Centroid1 = (round(centroid[0],2), round(centroid[1],2))

    #Converting tuple in (x,y) form.
    a = (x3[0], y3[0])
    b = (x3[1], y3[1])
    c = (x3[2], y3[2])

    #Function for calculating distance from centroid.
    def get_distance(x1, x2, y1, y2):
        dist = math.sqrt((x2-x1)**2 +(y2-y1)**2)
        return dist


    #Assinging first coordinate points value.
    x1 = a[0]
    x2 = Centroid1[0]
    y1 = a[1]
    y2 = Centroid1[1]
    #Calculating distance for second coordinate point and centroid.
    distance_a_centroid = get_distance(x1,x2,y1,y2)
    print(distance_a_centroid)

    #Assinging second coordinate points value.
    x_1 = b[0]
    y_1 = b[1]
    #Calculating distance for second coordinate point and centroid.
    distance_b_centroid = get_distance(x_1, x2, y_1, y2)
    print(distance_b_centroid)


    #Assinging second coordinate points value.
    x_2 = c[0]
    y_2 = c[1]
    #Calculating distance for third coordinate point and centroid.
    distance_c_centroid = get_distance(x_2,x2,y_2,y2)
    print(distance_c_centroid)

    #calculate average dispersion
    Average_dispersion = (distance_a_centroid+distance_b_centroid+distance_c_centroid)/3
    print(Average_dispersion)

    #Validation statement
    if distance_a_centroid <= 16.00 and distance_b_centroid <= 16.00 and distance_c_centroid <=16.00 :
        print ("True")
    else:
        print("False")





for index, row in df.iterrows():
    Final_PHD = Calculate_dispersion(row)
print(Final_PHD)

提前致谢

Answer 1

很简单，如果你在函数内部看不到任何东西，取决于你执行函数时使用的row，这意味着函数将始终返回相同的值。请注意，函数的变量是局部变量，因此在每次函数调用时都会重新启动。

试试这个：

def Calculate_dispersion(row):
    #Picking up the columns with x coordinates.
    df2 = df.loc[row,'East Coordinate'].tolist()
    #Picking up the columns with y coordinates.
    df3 = df.loc[row,'North Coordinate'].tolist()


    #Splitting the list into seperate x coordinates.
    df4 = pd.DataFrame([sub.split(",") for sub in df2])
    #Splitting the list into seperate y coordinates.
    df5 = pd.DataFrame([sub.split(",") for sub in df3])


    #Creating a tuple with x coordinates
    x1 = df4.iloc[0]
    x2 = x1.tolist()
    x3 = tuple(float(x) for x in x2)
    #Creating a tuple with y coordinates
    y1 = df5.iloc[0]
    y2 = y1.tolist()
    y3 = tuple(float(x) for x in y2)


    #Creating the Coordinate tuple for centroid calculation.
    c = (x3,y3)

    #Calculating centroid.
    centroid = (sum(c[0])/len(c[0]),sum(c[1])/len(c[1]))
    Centroid1 = (round(centroid[0],2), round(centroid[1],2))

    #Converting tuple in (x,y) form.
    a = (x3[0], y3[0])
    b = (x3[1], y3[1])
    c = (x3[2], y3[2])

    #Function for calculating distance from centroid.
    def get_distance(x1, x2, y1, y2):
        dist = math.sqrt((x2-x1)**2 +(y2-y1)**2)
        return dist


    #Assinging first coordinate points value.
    x1 = a[0]
    x2 = Centroid1[0]
    y1 = a[1]
    y2 = Centroid1[1]
    #Calculating distance for second coordinate point and centroid.
    distance_a_centroid = get_distance(x1,x2,y1,y2)
    print(distance_a_centroid)

    #Assinging second coordinate points value.
    x_1 = b[0]
    y_1 = b[1]
    #Calculating distance for second coordinate point and centroid.
    distance_b_centroid = get_distance(x_1, x2, y_1, y2)
    print(distance_b_centroid)


    #Assinging second coordinate points value.
    x_2 = c[0]
    y_2 = c[1]
    #Calculating distance for third coordinate point and centroid.
    distance_c_centroid = get_distance(x_2,x2,y_2,y2)
    print(distance_c_centroid)

    #calculate average dispersion
    Average_dispersion = (distance_a_centroid+distance_b_centroid+distance_c_centroid)/3
    print(Average_dispersion)

    #Validation statement
    if distance_a_centroid <= 16.00 and distance_b_centroid <= 16.00 and distance_c_centroid <=16.00 :
        print ("True")
    else:
        print("False")





row=0
while row<len(df.index)
    Final_PHD = Calculate_dispersion(row)
    row+=1
    print(Final_PHD)