converting tuple into pandas dataframe

Question

Assume that we have the following tuple

    S=({'country':[('India')],'state':[('Telangana')],'city':[('Hyderabad'),('Vizag')]}, {'date':[{'year': 2021, 'month':10}]})
    
    <class 'tuple'>

Can we convert this into a dataframe and get the date part into a variable

    df = key      value
         country  India
         state    Telangana
         city     Hyderabad
         city     Vizag
         
    Date = {'date':[{'year': 2021, 'month':10}]}

I have tried pd.DataFrame(json.loads(json.dumps(s))) but its not clean

I don't understand the need for a dataframe here...why not just S[1]? — rahlf23
– rahlf23, Commented Jun 30, 2021 at 20:27

SeaBean · Accepted Answer · 2021-06-30 20:47:40Z

1

For the first part, you can try:

pd.Series(S[0]).explode()

Result:

country    India    
state      Telangana
city       Hyderabad
city       Vizag    
dtype: object

If you need a dataframe, you can use:

df = pd.DataFrame(pd.Series(S[0]).explode().reset_index().values, columns=['key', 'value'])

Result:

print(df)

       key      value
0  country      India
1    state  Telangana
2     city  Hyderabad
3     city      Vizag

For the second part on `Date`, you can use:

Date = S[1]

Result:

print(Date)

{'date': [{'year': 2021, 'month': 10}]}

edited Jun 30, 2021 at 20:47

answered Jun 30, 2021 at 20:34

SeaBean

23.4k3 gold badges16 silver badges28 bronze badges

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

5 Comments

andy Over a year ago

looks good but the result dataframe has braces and commas in the value column key value 0 country ( India) 1 state (Telangana) 2 city (Hyderabad,) 3 city ( Vizag,) how do I remove braces and commas for the value column

SeaBean Over a year ago

@andy That's weird. I tested with your tuple S and arrived at the dataframe I shown in the result. No braces and comma there. Did you run the code with the tuple S given here, or run on another tuple ?

SeaBean Over a year ago

@andy Just by dumping S[0] on my system gives {'country': ['India'], 'state': ['Telangana'], 'city': ['Hyderabad', 'Vizag']} which already have no braces there. So, how come the braces come out afterwards ? Is the tuple that you run contains list of strings instead of list of tuple ?

SeaBean Over a year ago

@andy You can also test out using the code: df = pd.Series(S[0]).explode().to_frame(name='value').rename_axis(index='key').reset_index() and see how it goes.

SeaBean Over a year ago

@andy There is absolutely no reason to get comma in the value column together with the text. This could happen only when inside your tuple you have something like 'city':['(Hyderabad,)', ' (Vizag,)'] Note the placement of single quote character ' is different from that in S that you provided. There probably the braces are inside single quotes as a string. Please double check with your data.

0sharp · Accepted Answer · 2021-06-30 21:16:34Z

0

For the Date part, you can try

Date = S[1]
print(Date)

{'date': [{'year': 2021, 'month': 10}]}

As for the first part of your question, you may find the following code less than perfect, but it can easily be edited to match the desired output perfectly.

df = pd.DataFrame(
    {k:pd.Series(v) for k, v in dict(S[0]).items()}
).transpose()
print(df)

               0      1
country      India    NaN
state    Telangana    NaN
city     Hyderabad  Vizag

For more information, refer to this post

answered Jun 30, 2021 at 21:16

0sharp

431 silver badge7 bronze badges

Collectives™ on Stack Overflow

converting tuple into pandas dataframe

2 Answers 2

For the first part, you can try:

For the second part on `Date`, you can use:

5 Comments

Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

For the first part, you can try:

For the second part on Date, you can use:

5 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related

For the second part on `Date`, you can use: