python - Pandas str.count()

Question

我有一个包含 2 列的数据框，我正在尝试创建第三列来计算第一列在第二列中出现的次数。

样本_df =

Object  Text
Banana  Banana Banana Banana
Banana  Apple Apple Apple
Apple   Banana Apple

现在我正在尝试以下代码:

sample_df['Mentions'] = sample_df['Text'].count(sample_df['Object'])

这会产生以下错误:

AttributeErrorTraceback (most recent call last)
<ipython-input-65-c9ae4ce28088> in <module>()
----> 1 sample_df['Mentions'] = sample_df['Text'].count(sample_df['Object'])

/usr/local/lib/python2.7/dist-packages/pandas/core/series.pyc in count(self, 
level)
1177             level = self.index._get_level_number(level)
1178 
-> 1179         lev = self.index.levels[level]
1180         lab = np.array(self.index.labels[level], subok=False, copy=True)
1181 

AttributeError: 'RangeIndex' object has no attribute 'levels'

Answer 1

如果您阅读 pd.Series.count 的文档，你会发现它并没有像你想象的那样做:

Series.count(level=None)

Return number of non-NA/null observations in the Series

您提供了 pandas Series 作为 level 参数，该参数无效，这就是您收到错误的原因。为了您的使用，请尝试以下操作:

df['counter'] = df.apply(lambda x: x.Text.count(x.Object), axis=1)

   Object                  Text  counter
0  Banana  Banana Banana Banana        3
1  Banana     Apple Apple Apple        0
2   Apple          Banana Apple        1

如果您关心性能，您还可以在此处使用简单的列表理解:

df['counter'] = [i.count(j) for i, j in zip(df.Text, df.Object)]

时间(使用列表理解:D)

df = pd.concat([df]*10000)

%timeit df.apply(lambda x: x.Text.count(x.Object), axis=1)
1.14 s ± 14.3 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)

%timeit [i.count(j) for i, j in zip(df.Text, df.Object)]
6.71 ms ± 25 µs per loop (mean ± std. dev. of 7 runs, 100 loops each)