Python循环自动创建独立DataFrame并设置时间索引求助

Hey there! I see exactly where your loop is going off track. Let's break this down and fix it step by step.

问题根源 #

Your current code is storing each ticker's raw dictionary list directly as a column in a single data DataFrame. That's why you can't access data[t]['time'] properly—each column is just a list of dictionaries, not a structured DataFrame with columns like time, close, etc.

解决方案 #

Instead of shoving everything into one DataFrame's columns, we'll use a dictionary to store individual DataFrames for each ticker. This keeps each ticker's data isolated and lets you work with them properly.

修正后的代码 #

import requests
import pandas as pd

desired_width = 320
pd.set_option('display.width', desired_width)

# 用字典存储每个ticker的DataFrame，而不是单个DataFrame
crypto_data = {}
tickers = ['BTC', 'ETH', 'XRP']

for t in tickers:
    url = f'https://min-api.cryptocompare.com/data/histoday?fsym={t}&tsym=USD&limit=600000000000&aggregate=1&e=CCCAGG'
    response = requests.get(url)
    
    # 把JSON数据直接转换成DataFrame，而不是存在列里
    df = pd.DataFrame(response.json()['Data'])
    
    # 转换时间列并设置为索引
    df['time'] = pd.to_datetime(df['time'], unit='s')
    df.set_index('time', inplace=True)
    
    # 把处理好的DataFrame存入字典
    crypto_data[t] = df
    
    print(f"Downloading data for: {t}")
    print(f"Sample data for {t}:\n", df.head(5), "\n")

代码说明 #

• We use a dictionary crypto_data where each key is a ticker (like 'BTC') and the value is its dedicated DataFrame.
• Converting response.json()['Data'] directly to a DataFrame gives us proper columns (time, close, high, etc.) right away.
• pd.to_datetime(df['time'], unit='s') converts the Unix timestamp to a readable datetime format.
• df.set_index('time', inplace=True) sets the datetime column as the index, which is perfect for time-series data.

怎么访问单个ticker的数据 #

After running the loop, you can access each ticker's DataFrame like this:

# 获取BTC的数据
btc_df = crypto_data['BTC']
print(btc_df.describe())

# 获取ETH的收盘价
eth_close_prices = crypto_data['ETH']['close']

This way, each ticker's data is properly structured and ready for analysis!

内容的提问来源于stack exchange，提问作者Giladbi