count values by name in pandas dataframe

Question

How to count number of values by their name? I have a pandas data frame:

word token 
physical I-Problem 
design I-Problem 
for I-Problem 
XML I-Problem 
databases I-Problem 
which o 
is o 
the o 
process o

I need to count how many I-Problem in the token. I tried df['token'].count('I-Problem') but it doesn't work.

expected output :

I-Problem : 5

or

Could've used [str.count](https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.Series.str.count.html) instead of just count `df['token'].str.count('I-Problem').sum()` though apparently numpy sum is fastest `(df['token'].values == 'I-Problem').sum()` things may have changed over time though [timings](https://stackoverflow.com/a/35277776/15497888) — Henry Ecker, Sep 21 '21 at 02:09

score 2 · Answer 1 · answered Sep 21 '21 at 01:35

2

Use value_counts:

>>> df['token'].value_counts()['I-Problem']
5
>>>

Or try tolist:

>>> df['token'].tolist().count('I-Problem')
5
>>>

answered Sep 21 '21 at 01:35

U13-Forward

1 Answers1