I tried to append a new column to my dataframe. It doesn't work... well... it does work, but it always throws SettingWithCopyWarnings.
I used all the methods proposed here, and searched also other Stackoverflow threads for a solution, but still I don't arrive.
I have a dynamic list of words that shall be added to my DataFrame as new Columns. In each column there shall be the count of words as found in a line of text.
import re
import pandas as pd
df = pd.read_csv('manyLinesOfText.csv')
word_lst = ['foo', 'bar', 'spam']
def element_word_count(text, **kwds):
return float(len(re.findall(kwds['kwds']['countword'], text)))
for word in word_lst:
df[word] = df['textcolumn'].apply(element_word_count, kdws = {'countword':word})
This works, but throws an annoying warning.
I tried assign but it doesn't work at all, because the syntax of assign interprets the "words" as the title of the column.
I tried df.insert, but it throws a NoneType exception, which is weird because my df definitely is not NoneType.
I tried df.loc[:,word] but it does solve nothing. It works, and the same exception is thrown.
Maybe I have just a knot in my brain after the entire day worth of work, but I cant't find the solution.
I also think my use of df.apply is a bit strange - so if someone finds an easier way to count a given but dynamic list of words, I am open to anything.
Thanks a lot!
