Extracting the max, min or std from a DF for a par

问题:

I have a df with columns X1, Y1, Z3. df.describe shows the stats for each column

I would like to extract the min, max and std for say column Z3. df[df.z3].idxmax() doesn't seem to work

回答1:

Demo:

In [7]: df = pd.DataFrame(np.random.rand(10, 3), columns=['X1','Y1','Z3'])

In [8]: df
Out[8]:
         X1        Y1        Z3
0  0.258116  0.667943  0.954830
1  0.584975  0.546284  0.045745
2  0.698974  0.409223  0.307409
3  0.073166  0.356393  0.722233
4  0.339093  0.146043  0.614686
5  0.624361  0.062805  0.574546
6  0.886631  0.217291  0.258432
7  0.403345  0.035377  0.096881
8  0.663185  0.376171  0.593964
9  0.789347  0.489057  0.564688

In [9]: df.describe()
Out[9]:
              X1         Y1         Z3
count  10.000000  10.000000  10.000000
mean    0.532119   0.330659   0.473341
std     0.255544   0.210851   0.287745
min     0.073166   0.035377   0.045745
25%     0.355156   0.163855   0.270676
50%     0.604668   0.366282   0.569617
75%     0.690027   0.469099   0.609506
max     0.886631   0.667943   0.954830

In [10]: df.describe().loc[['min','max','std'], 'Z3']
Out[10]:
min    0.045745
max    0.954830
std    0.287745
Name: Z3, dtype: float64

回答2:

New in pandas 0.20.0, pd.DataFrame.agg

df.agg(['min', 'max', 'std'])

           X1        Y1        Z3
min  0.074689  0.032109  0.074912
max  0.833935  0.981217  0.840831
std  0.252231  0.315192  0.294872