Question

我正在尝试使用以下代码将489个csv文件列表读入pandas数据框

symbols = ['489 symbols']
def data(symbols):
    dates=pd.date_range(start_date,end_date)
    df=pd.DataFrame(index=dates)
    for symbol in symbols:
        df_temp=pd.read_csv('/home/furqan/Desktop/USSP/{}.csv'.format(str(symbol)),usecols=['Date','Price'],
                            parse_dates=True,index_col='Date',na_values=['nan'])
        df_temp = df_temp.rename(columns={'Price': symbol})
        df=df.join(df_temp)
        df=df.fillna(method='ffill')
        df=df.fillna(method='bfill')
    return df
start_date = '2006-01-01'
end_date = '2016-12-31'
data_frame = data(symbols)
test = data_frame.loc['2006-01-01':'2006-12-31',:]

我希望测试数据框的形状为（365,489），但形状结果为（9874,489），无法找出原因？

假设我没有使用489个元素，而是使用1个元素，换句话说，如果我读取一个csv文件而不是489个，我会得到（365,1）的正确形状。

Answer 1

以下代码工作

def data(symbols):
    dates=pd.date_range(start_date,end_date)
    df=pd.DataFrame(index=dates)
    for symbol in symbols:
        df_temp=pd.read_csv('/home/furqan/Desktop/USSP/{}.csv'.format(str(symbol)),usecols=['Date','Price'],
                            parse_dates=True,index_col='Date',na_values=['nan'])
        df_temp = df_temp.rename(columns={'Price': symbol})
        df_temp = df_temp.loc[~df_temp.index.duplicated(keep='first')]
        df=df.join(df_temp)
        df=df.fillna(method='ffill')
        df=df.fillna(method='bfill')
    return df

但仍然无法获得重复的内容？

Answer 2

我认为您需要在每个循环中追加到list of Dataframe并且最后concat：

symbols = ['489 symbols']
def data(symbols):
    dfs = []
    for symbol in symbols:
        df_temp=pd.read_csv('/home/furqan/Desktop/USSP/{}.csv'.format(str(symbol)),usecols=['Date','Price'],
                            parse_dates=True,index_col='Date',na_values=['nan'])
        df_temp = df_temp.rename(columns={'Price': symbol}).ffill().bfill()
        dfs.append(df_temp)

    return pd.concat(dfs, axis=1)

start_date = '2006-01-01'
end_date = '2016-12-31'
data_frame = data(symbols)
test = data_frame.loc['2006-01-01':'2006-12-31',:]

如有必要，还可以添加date_range添加reindex：

的值

symbols = ['489 symbols']
def data(symbols):
    dfs = []
    for symbol in symbols:
        df_temp=pd.read_csv('/home/furqan/Desktop/USSP/{}.csv'.format(str(symbol)),usecols=['Date','Price'],
                            parse_dates=True,index_col='Date',na_values=['nan'])
        df_temp = df_temp.rename(columns={'Price': symbol})
        dfs.append(df_temp)

    dates=pd.date_range(start_date,end_date)
    return pd.concat(dfs, axis=1).reindex(dates, method='ffill').bfill()

start_date = '2006-01-01'
end_date = '2016-12-31'
data_frame = data(symbols)
test = data_frame.loc['2006-01-01':'2006-12-31',:]

形状pandas DataFrame不正确

2 个答案: