连接时出现 Pandas 错误“值长度与索引长度不匹配”-解网

问：

我弹出了这个“值的长度与索引的长度不匹配”错误并破坏了我的代码。

这里的代码是遍历字典，拉出参数和路径，发送到 api 以更新路径对象中的信息。

我的代码在第一次迭代后失败，我不知道为什么。

代码如下：

import dydx3 as dy
import pandas as pd
import time
import os

dir = os.getcwd()

client = dy.Client(
    host = 'https://api.dydx.exchange'
)

markets_1m = {'BTC-USD':f'{dir}\\markets\\BTC-USD_1MIN.csv', 
           'ETH-USD':f'{dir}\\markets\\ETH-USD_1MIN.csv', 
           'LINK-USD':f'{dir}\\markets\\LINK-USD_1MIN.csv',
           'SOL-USD':f'{dir}\\markets\\SOL-USD_1MIN.csv',
           'ATOM-USD':f'{dir}\\markets\\ATOM-USD_1MIN.csv',
           'XMR-USD':f'{dir}\\markets\\XMR-USD_1MIN.csv',
           'RUNE-USD':f'{dir}\\markets\\RUNE-USD_1MIN.csv',
           }

def get_price_data_1m():
  try:
    dataframe = pd.DataFrame()

    StartedAt = []
    UpdatedAt = []
    Market = []
    Resolution = []
    Low = []
    High = []
    Open = []
    Close = []
    BaseTokenVolume = []
    Trades = []
    UsdVolume = []
    StartingOpenInterest = []

    for ticker, path in markets_1m.items():

        candles = client.public.get_candles(
            market = ticker,
            resolution ='1MIN',
        )
        for i in candles.data['candles']:
            startedAt = i['startedAt']
            updatedAt = i['updatedAt']
            market = i['market']
            resolution = i['resolution']
            low = i['low']
            high = i['high']
            open = i['open']
            close = i['close']
            baseTokenVolume = i['baseTokenVolume']
            trades = i['trades']
            usdVolume = i['usdVolume']
            startingOpenInterest = i['startingOpenInterest']
            StartedAt.append(startedAt)
            UpdatedAt.append(updatedAt)
            Market.append(market)
            Resolution.append(resolution)
            Low.append(low)
            High.append(high)
            Open.append(open)
            Close.append(close)
            BaseTokenVolume.append(baseTokenVolume)
            Trades.append(trades)
            UsdVolume.append(usdVolume)
            StartingOpenInterest.append(startingOpenInterest)

        dataframe['startedAt'] = [value for value in StartedAt]
        dataframe['updatedAt'] = [value for value in UpdatedAt]
        dataframe['market'] = [value for value in Market]
        dataframe['resolution'] = [value for value in Resolution]
        dataframe['open'] = [value for value in Low]
        dataframe['high'] = [value for value in High]
        dataframe['low'] = [value for value in Open]
        dataframe['close'] = [value for value in Close]
        dataframe['baseTokenVolume'] = [value for value in BaseTokenVolume]
        dataframe['trades'] = [value for value in Trades]
        dataframe['usdVolume'] = [value for value in UsdVolume]
        dataframe['startingOpenInterest'] = [value for value in StartingOpenInterest]

        price_history = pd.read_csv(f'{path}') #input csv path

        combined = pd.concat([price_history.reset_index(drop=True), dataframe.reset_index(drop=True)], join='outer', ignore_index=True, axis=0)

        combined.drop_duplicates(subset=['startedAt'], keep='first', inplace=True)

        combined = combined.sort_values(by=['startedAt'])

        combined.to_csv(f'{dir}\\markets\\{ticker}_1MIN.csv', index=False)
    
        print("end of iteration")

  except Exception as e:
    print(f'{e}')

我试图重置两个数据帧中的索引长度并忽略 concat 函数中的索引，但没有奏效。我还尝试在主循环中使用 try/except 中的 continue 来抑制错误，这也产生了相同的错误。

我应该发生的事情是，该函数应该提取数据，将该数据附加到现有数据帧中，删除任何重复项，按“startedAt”列对数据帧进行排序，然后保存到相同的路径。

当我尝试这样做时，只会正确执行“markets_1m”字典中第一项的函数，然后在其余所有项上失败。

关于为什么它只对第一个有效或为什么会出现错误的任何想法？

谢谢

python-3.x pandas 数据帧循环串联

def get_api_data(ticker):
    """
    Call API and return select values in list of dicts.
    """
    candles = client.public.get_candles(
        market = ticker,
        resolution ='1MIN',
    )

    api_data = [
        {
            "ticker": ticker,
            "startedAt": i['startedAt'],
            "updatedAt": i['updatedAt'],
            "market": i['market'],
            "resolution": i['resolution'],
            "low": i['low'],
            "high": i['high'],
            "open": i['open'],
            "close": i['close'],
            "baseTokenVolume": i['baseTokenVolume'],
            "trades": i['trades'],
            "usdVolume": i['usdVolume'],
            "startingOpenInterest": i['startingOpenInterest']
        } 
        for i in candles.data['candles']
    ]

    return api_data


def build_price_data_1m_frames():
    """
    Build and return dict of data frames.

    Export each ticker historical and current data to CSVs.
    """
    try:
        dydx_dfs = {
            ticker: pd.concat(
                [
                    pd.read_csv(path).assign(ticker = ticker),
                    pd.DataFrame(get_api_data(ticker))                   
                ],
                ignore_index=True
            ).drop_duplicates(
               subset=['startedAt'], keep='first'
            ).sort_values(
               by=['startedAt']
            ).reset_index(drop=True)        
            for ticker, path in markets_1m.items()
        }

        for tkr, df in dydx.times():
            df.to_csv(os.path.join(dir, "markets", tkr, "_1MIN.csv"))

    except Exception as e:
        print(e)
        return None

    return dydx_dfs

上一个：MySQL CONCAT_WS无法使用外键

下一个：在 C 语言中定义无限类族“MatrixMxN”#

连接时出现 Pandas 错误“值长度与索引长度不匹配”

Pandas error "Length of values does not match length of index" when concatenating

评论