I have a dataframe called df_cv with column that contains this string _log. I would like to delete _log from the name of each column in a dataframe So i didlike this :
modify_cols=[]
for c in df_cv.columns:
if c.find("_log") != -1:
modify_cols.append(c)
for c in modify_cols:
c = c.replace('_log',' ')
But it didn't work , there is no error message but the comluns name doesn't change.
Any idea please to help me?
Thanks
Use str.replace:
df_cv = pd.DataFrame(columns=['col1','col2_log','1_log'])
df_cv.columns=df_cv.columns.str.replace('_log', '')
print (df_cv)
Empty DataFrame
Columns: [col1, col2, 1]
Index: []
Use list-comprehension:
df_cv.columns = [i.replace("_log","") for i in df_cv.columns]
You should collect results in a collection, or at least process them
modified_cols=[]
for c in modify_cols:
modified_cols.append(c.replace('_log',' '))
Or shorter version, using list comprehensions:
my_list = [c.replace('_log',' ') for c in df_cv.columns if c.find('_log')!=-1]
Otherwise you will lost each c variable after each loop block.
= makes the reference to c becomes a new string, but the string referenced by the modify_cols is still the original. and because of string is immutable， so you can not do like this：
s = [['a_log'], ['b_log']]
for i in s:
# i reference to the mutable list
i[0] = i[0].replace('_log', '')
# now s is [['a'], ['b']]
you can use list comprehension:
modify_cols = [c.replace('_log',' ') for c in modify_cols]
df_cv.columns=df_cv.columns.str.strip('_log')??_, l, o, g