creating multiple dataframes which contain subsets of an orginal pandas dataframe

Question

I have a dataframe (called df) and want to split it into multiple dataframes based on the values in one of the columns.

I think the syntax would be:

month = df.month.unique().tolist()
for item in month:
    [item] = df[df[month]==[item]]

Answer 1

The DataFrameGroupBy object is an iterator that returns sub-DataFrames:

for month, subdf in df.groupby(['month']):
    ...

---

If instead you want random access to the sub-DataFrames, specified by month, you could change the month column into an index:

df = df.set_index(['month'])

and then you could select rows by month with:

df.loc[month]

---

For example,

In [4]: df = pd.DataFrame({'month': ['Jan','Jan','Feb'], 'val':[1,2,3]})        
In [6]: df = df.set_index(['month'])    

Given this DataFrame:

In [7]: df
Out[7]: 
       val
month     
Jan      1
Jan      2
Feb      3

This selects the rows where the month (index) is 'Jan':

In [8]: df.loc['Jan']
Out[8]: 
       val
month     
Jan      1
Jan      2