根据条件将数据框的值移动到列表

Question

I am trying to move the data from a single column in a dataframe to a list.

Account Name    Renewal % Change
Client 1        0%
Client 2        0%
Client 3        2%
Client 4        0%
Client 5        1%
Client 6        1%

So If a client has 0% It should get added to the list Cost0 and it it has 1% it should get added to Cost1.

I tried using a pd.series.tolist() but it kept giving me an error.

if brl['Renewal % Change'] == '0%':
    a = pd.Series(brl['Account Name'])
    a.tolist()

ValueError: The truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all()

Please advise

Answer 1

Try this

Cost0 = brl[blr['Renewal % Change']=='0%']['Account Name'].tolist()
Cost1 = brl[blr['Renewal % Change']=='1%']['Account Name'].tolist()

Update

Using .loc

Cost0 = brl.loc[blr['Renewal % Change']=='0%','Account Name'].tolist()
Cost1 = brl.loc[blr['Renewal % Change']=='1%','Account Name'].tolist()

Answer 2

You can use a bit changed boolean indexing with selecting column by loc :

mask = brl['Renewal % Change'] == '0%'
print (mask)
0     True
1     True
2    False
3     True
4    False
5    False
Name: Renewal % Change, dtype: bool

print (brl.loc[mask, 'Account Name'].tolist())
['Client 1', 'Client 2', 'Client 4']

All together:

Cost0 = brl.loc[brl['Renewal % Change'] == '0%', 'Account Name'].tolist()
print (Cost0)
['Client 1', 'Client 2', 'Client 4']

Cost1 = brl.loc[brl['Renewal % Change'] == '1%', 'Account Name'].tolist()
print (Cost1)
['Client 5', 'Client 6']

---

You get error, because comparing return boolean Series - array, not scalar value, see docs :

print (brl['Renewal % Change'] == '0%')
0     True
1     True
2    False
3     True
4    False
5    False
Name: Renewal % Change, dtype: bool

Comparing solutions - with loc is faster:

In [137]: %timeit brl.loc[brl['Renewal % Change'] == '0%', 'Account Name'].tolist()
1000 loops, best of 3: 536 µs per loop

In [138]: %timeit brl[brl['Renewal % Change']=='0%']['Account Name'].tolist()
1000 loops, best of 3: 657 µs per loop