如何在多索引数组中隔离特征

Question

我想在多索引数组中隔离特征。 这就是我的阵法

我试过寻找间谍

date,symbol,open,high,low,close,volume
2012-01-03,AAPL,58.49,58.93,58.43,58.8,75555200.0
2012-01-03,CSCO,18.55,18.86,18.48,18.63,41200000.0
2012-01-03,INTC,24.62,24.89,24.39,24.54,44800000.0
2012-01-03,MSFT,26.55,26.96,26.39,26.77,64700000.0
2012-01-03,SPY,127.76,128.38,127.43,127.5,194000000.0
2012-01-04,AAPL,58.57,59.24,58.47,59.1,65005500.0
2012-01-04,CSCO,18.44,19.0,18.35,18.99,52900000.0
2012-01-04,INTC,24.57,25.13,24.53,25.11,47500000.0
2012-01-04,MSFT,26.82,27.47,26.78,27.4,80500000.0
2012-01-04,SPY,127.2,127.81,126.71,127.7,127000000.0

• 阅读中的DataFrame，使用：

df = pd.read_clipboard(sep=',')
df.set_index(['date', 'symbol'], inplace=True)

如何在上面隔离间谍符号

Answer 1

给定以下DataFrame：

---

不重置索引：

df_spy = df.loc[df.index.get_level_values('symbol') == 'SPY']

---

重置索引：

df.reset_index(inplace=True)
df_spy = df[df.symbol == 'SPY']