I have df which looks like:
df_table
Name exp1
bnb_q8a12 41
bnb_q8a8 36
bnb_q8a2 30
_xexp0 20
_xexp73 16
bnb_q8a7 10
_xexp216 10
bnb_q8a3 10
I would like the df to look like this, with all index which are named "Net: _xexp.." appended at the end of the df.
df_table
Name exp1
bnb_q8a12 41
bnb_q8a8 36
bnb_q8a2 30
bnb_q8a7 10
bnb_q8a3 10
Net: _xexp0 20
Net: _xexp73 16
Net: _xexp216 10
I am current appending the given index like this but Im sure there is a better way of doing this?
res = df_table.index.tolist()
nets = []
for i in df_table.index:
if 'Net: ' in i:
nets.append(res.pop(res.index(i)))
res.extend(nets)
df_table = df_table.reindex(res)
Use the vectorised str.contains method to mask the df for the rows that contain the string and prepend the str.
As it's currently your index, it's probably simpler to reset the index to restore it as a column, then filter the rows and prepend the text, use concat on the filtered rows to reorder them and then set the index back again:
In [64]:
df = df.reset_index()
df.loc[df.Name.str.contains('_xexp'), 'Name'] = 'Net: ' + df.Name
df
Out[64]:
Name exp1
0 bnb_q8a12 41
1 bnb_q8a8 36
2 bnb_q8a2 30
3 Net: _xexp0 20
4 Net: _xexp73 16
5 bnb_q8a7 10
6 Net: _xexp216 10
7 bnb_q8a3 10
In [65]:
df = pd.concat([df[~df.Name.str.contains('_xexp')], df[df.Name.str.contains('_xexp')]])
df
Out[65]:
Name exp1
0 bnb_q8a12 41
1 bnb_q8a8 36
2 bnb_q8a2 30
5 bnb_q8a7 10
7 bnb_q8a3 10
3 Net: _xexp0 20
4 Net: _xexp73 16
6 Net: _xexp216 10
In [66]:
df = df.set_index('Name')
df
Out[66]:
exp1
Name
bnb_q8a12 41
bnb_q8a8 36
bnb_q8a2 30
bnb_q8a7 10
bnb_q8a3 10
Net: _xexp0 20
Net: _xexp73 16
Net: _xexp216 10