How to assign value to an empty dataframe with multiindexes columns with Pandas?

Question

I would like to create an empty dataframe with multiindexes columns with Pandas, and then fill values later on. How could I do that ?

In this example I can create the dataframe but can't assign value to it with .loc.

segment = 'segment_1'
tuples = [(segment, 'market', 'symbol'),
          (segment, 'market', 'base'),
          (segment, 'market', 'quote'),
          (segment, 'market', 'wallet'),
          (segment, 'market', 'type'),
          (segment, 'market', 'margined'),
          (segment, 'market', 'derivative'),
          (segment, 'funds', 'wallet'),
          (segment, 'funds', 'code'),
          (segment, 'type', 'action'),
          (segment, 'type', 'transfer'),
          (segment, 'type', 'priority'),
         ]
index = pd.MultiIndex.from_tuples(tuples, names=["level_1", "level_2", 'level_3'])
df = pd.DataFrame(columns=index)
df.loc[:, ('segment_1', 'market', 'symbol')] = 1
df

level_1 segment_1
level_2 market  funds   type
level_3 symbol  base    quote   wallet  type    margined    derivative  wallet  code    action  transfer    priority

As you can it's not possible to set value to this dataframe. What is the reason for this ? do I need to create an index first ?

jezrael · Accepted Answer · 2021-06-17 09:36:55Z

1

There is used : in loc for select all rows, but because rows not exist is not set value. So solution is set index label like 0 here:

df.loc[0, ('segment_1', 'market', 'symbol')] = 1

print (df)
level_1 segment_1                                                         \
level_2    market                                             funds        
level_3    symbol base quote wallet type margined derivative wallet code   
0               1  NaN   NaN    NaN  NaN      NaN        NaN    NaN  NaN   

level_1                           
level_2   type                    
level_3 action transfer priority  
0          NaN      NaN      NaN

answered Jun 17, 2021 at 9:36

jezrael

868k103 gold badges1.4k silver badges1.3k bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

Florent Over a year ago

So there is no need to create an index, thank you it exactly what I was looking for.

jezrael Over a year ago

@Florent - Before not, index is created in loc.

Collectives™ on Stack Overflow

How to assign value to an empty dataframe with multiindexes columns with Pandas?

1 Answer 1

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related