Pandas: how to operate with each pivottable value?

Question

This is the continuation of this . Let's say I have code that does this table I want

df.groupby('A')['B'].value_counts().unstack().stack(dropna=False
                ).reset_index(name="Count").set_index(['A', 'B'])

|----|----|-------|
| A  | B  | Count |
| a1 | b1 |  1    |
|    | b2 |  1    |
|    | b3 |  NaN  |
| a2 | b1 |  1    |
|    | b2 |  NaN  |
|    | b3 |  1    |

The problem is that there is the case where B column might have multivariate values, eg many unique values. So the groupping A column values are a bit far away :) Eventually, this all should be stored in some Excel format file df.to_excel() . The solution was proposed to generate such as Excel files per A values. Eg instead of groupped.xlsx where you have all this pivot table in once, to have A_a1.xlsx, A_a2.xlsx files.

Question: how do you do it?

I have some options in mind like to get the list of unique A values and just do something like df_loc = df.loc[df['A'] == 'a1'] , but maybe there is more cool way?

Answer 1

If I understand correctly you're looking for an individual excel file for each A value? If so, the following should work:

for i in df.a.unique(): df[df['a'] == i].to_excel(path+_i)

You can tweak the path for your needs but this is a pretty easy way to do what you're looking for.