TypeError：'type'对象在pandas DataFrame中没有属性'__getitem__'

Question

I was trying to analyse the play-by-play data of a basketball team What I did was to read a csv file into a DataFrame object.

I want to preserve the functionality of the DataFrame object while adding in new attributes to the existing object. Thus I wrote a class called Basketball:

from data_math import *
import pandas as pd
class Basketball(pd.DataFrame):
    def __init__(self,*args,**kargs):
        pd.DataFrame.__init__(self,*args,**kargs)
        self.FGM = calculate_FGM(pd.DataFrame)
        self.FGA = calculate_FGA(pd.DateFrame)
        self.FGP = self.FGM / self.FGA
        self.M3  = calculate_3M(pd.DataFrame)
        self.A3  = calcualte_3A(pd.DataFrame)
        self.P3  = self.M3 / self.A3
        self.FTM = calcualte_FTM(pd.DataFrame)
        self.FTA = calculate_FTA(pd.DataFrame)
        self.FTP = self.FTM / self.FTA
    # self.P = score_calculate(pd.DataFrame)

I wrote another data_math.py file to help calculate the different attributes I wanted to include into the Basketball class.

from pandas import DataFrame

def score_calculate(df):
    df_pt_scored = df[((df['etype']=='shot') & (df['result']=='made'))]
    df_ft_scored = df[((df['etype']=='free throw') & (df['result']=='made'))]
    return df_pt_scored['points'].sum()+len(df_ft_scored.index)

def calculate_FGM(df):
    cond_pt = (df['etype']=='shots') & (df['results']=='made')
    cond_ft = (df['etype']=='freethrow') & (df['results']=='made')
    return len(df[cond_pt].index)+len(df[cond_ft].index)

def calculate_FGA(df):
    shot_cond= df['etype']=='shot'
    free_throw_cond = df['etype']=='free throw'
    return len(df[shot_cond].index)+len(df[free_throw_cond].index)

def calculate_3M(df):
    cond_3M= (df['etype']=='shot')&(df['type']=='3pt')&(df['result']=='made')
    return len(df[cond_3M].index)

def calcualte_3A(df):
    cond_3A = (df['etype']=='shot')&(df['type']=='3pt')
    return len(df[cond_3A].index)

def calculate_FTM(df):
    cond_FTM =(df['etype']=='free throw') & (df['result']=='made')
    return len(df[cond_FTM].index)

def calcualte_FTA(df):
    cond_FTA =(df['etype']=='free throw')
    return len(df[cond_FTA].index)

In the end I start my program from main.py which I hope would give me the correct output. However while executing on this line:

team1= Basketball(tm1)

I received the following Traceback

 Traceback (most recent call last):
  File "/Users/luoyicheng/Developer/STAR-Research/data_analysis/source code/main.py", line 20, in <module>
    team1= Basketball(tm1)
  File "/Users/luoyicheng/Developer/STAR-Research/data_analysis/source code/Basketball.py", line 6, in __init__
    self.FGM = calculate_FGM(pd.DataFrame)
  File "/Users/luoyicheng/Developer/STAR-Research/data_analysis/source code/data_math.py", line 9, in calculate_FGM
    cond_pt = (df['etype']=='shots') & (df['results']=='made')
TypeError: 'type' object has no attribute '__getitem__'

I am new to python programming and could not figure out why this error has occurred. To my understanding, this error means I am unable to use indexing feature of the DataFrame. However, if I try to code in my main function similar things I am able to get the output I want. I am also not clear of how to extend the existing DataFrame class so that I can still access the methods in the DataFrame class while extending the team1 object to have attributes such as FGM, FGA, etc.

The idea of extending this class is to allow me to pass any DataFrame object in the Basketball() so that I can have an object with extending attributes and methods. I think I also lack an understanding of the use of init and self.

Please don't blame for not describing the problem clearly as I am not familiar with all the terminology in OOP.

Thank you so much!

Answer 1

You're passing each function pd.DataFrame which is of type type:

In [11]: type(pd.DataFrame)
Out[11]: type

Hence the exception message.

You mean to be passing self (which is of type DataFrame):

self.FGM = calculate_FGM(pd.DataFrame)
...

should read:

self.FGM = calculate_FGM(self)
...