python - Pandas 用标签切割多列？

Question

我想剪切并标记多列中的值。基本上是这样的:

df[numericColumn] = pd.cut(df[numericColumn], 3, labels=["small", "medium", "big"])

我找到了这个代码(有效)但是没有做我想要的标签，当我用标签替换 bin 时它出错并说我需要包含一个 bin

df = pd.DataFrame(np.random.rand(10,4))
df.apply(pd.cut, bins=[0,0.5,1])

有没有办法在 pandas 中切割标签多列？

Answer 1

所以把它拆开一点来展示发生了什么。您会看到您可以设置分箱，然后逐行进行切割，然后将切割结果更改回您要使用的类别名称。

# set up a dataframe
df = pd.DataFrame(np.random.rand(10,4))
df.columns = ['col1','col2','col3','col4']

# set up the intervals
bins = pd.IntervalIndex.from_tuples([(0, 0.33), (0.33, 0.66), (0.66, 1)])

# do the cut on col1 for example
x = pd.cut(df["col1"].to_list(),bins)

# change the name of the catagories
x.categories = ['small','medium','large']
# put it back
df['col1'] = x

给予

    col1    col2        col3        col4
0   large   0.589432    0.545828    0.257144
1   medium  0.625025    0.087607    0.548300
2   small   0.538186    0.057027    0.023201
3   medium  0.686324    0.027694    0.819753
4   medium  0.850623    0.977317    0.782361
5   large   0.375888    0.209709    0.903763
6   medium  0.094957    0.583052    0.534926
7   large   0.101968    0.863916    0.929300
8   large   0.612073    0.005553    0.723863
9   large   0.814393    0.327281    0.463976

如果你想遍历整个数据框

# set up a dataframe
df = pd.DataFrame(np.random.rand(10,4))
df.columns = ['col1','col2','col3','col4']

# set up the intervals
bins = pd.IntervalIndex.from_tuples([(0, 0.33), (0.33, 0.66), (0.66, 1)])
names = ['small','medium','large']

for col in df.columns:
    x = pd.cut(df[col].to_list(),bins)
    x.categories = names
    df[col] = x