python類：逐步運行函數並保存它們

Question

我有一個讀取數據幀的類，然后是另一個處理該數據幀的類。 處理類中的函數應逐步應用於同一數據幀，以形成最終數據幀，然后將其保存為 csv 文件。

from pydantic import BaseModel
from config import DATA_REPO
import pandas as pd
import os

class PandaDataFrame(BaseModel):
data: pd.DataFrame

      class Config:
          arbitrary_types_allowed = True

class Directory(BaseModel):
    data_directory: str

class DataToPandaReader(object):

    def csv_file_reader(self, directory: Directory):
        directory = directory.data_directory
        for file in os.listdir(directory):
            if file.endswith('.csv'):
               return pd.read_csv(os.path.join(directory, file))

class DataProcessor(object):

    def remove_punctuation(self, my_: PandaDataFrame):
        my_data_to_process = my_.data
        for col in my_data_to_process:
            if any(word in col for word in ['example', 'text', 'Answer']):
                my_data_to_process = my_data_to_process[col].str.replace('[^\w\s]', '', regex=True)
                return add_number_column(my_data_to_process)

    def add_number_column(self, my_: PandaDataFrame):
        my_data_to_process = my_.data
        my_data_to_process['sentence_number'] = range(len(my_data_to_process))
        return save_final_dataframe(my_data_to_process)

    def save_final_dataframe(self, my_:PandaDataFrame):
        my_data_to_process = my_.data
        return my_data_to_process.to_csv('final_data.csv')

def parse_data_process(directory_to_csv_file):
    toprocess = DataProcessor()
    toprocess.save_final_dataframe(directory_to_csv_file)
    toprocess.remove_punctuation(directory_to_csv_file)
    toprocess.add_number_column(directory_to_csv_file)
    return toprocess

if __name__ == '__main__':
    parse_data_process(PandaDataFrame(data= DataToPandaReader().csv_file_reader(Directory(data_directory = os.path.join(DATA_REPO, 'input_data')))))

現在，例如要實例化 DataProcessor 類中的第一個函數，我將執行以下操作

DataProcessor().remove_punctuation(PandaDataFrame(data= DataToPandaReader().csv_file_reader(Directory(data_directory = os.path.join(DATA_REPO, 'input_data')))))

但我的目的是在 DataProcessor 類中逐步運行所有這些函數，因此 save_final_dataset 函數將保存已刪除標點符號並且還有一個數字列的數據幀。

更新：

按照給出的答案，我進行了這些更改，但得到函數未知的錯誤。

def parse_data_process(directory_to_csv_file):
    toprocess = DataProcessor()
    toprocess.save_final_dataframe(directory_to_csv_file)
    toprocess.remove_punctuation(directory_to_csv_file)
    toprocess.add_number_column(directory_to_csv_file)
    return toprocess

if __name__ == '__main__':
    parse_data_process(PandaDataFrame(data= DataToPandaReader().csv_file_reader(Directory(data_directory = os.path.join(DATA_REPO, 'input_data')))))

Answer 1

除非我誤解了您的用例，否則您需要做的就是替換

return my_data_to_process

...在 remove_punctuation 函數中

return add_number_column(my_data_to_process)

...然后更換

return my_data_to_process

...在 add_number_column 函數中

return save_final_dataframe(my_data_to_process)