How to create keys from lists and get key-value pairs from tuples

Question

I have tuples in a dataframe that look like this:

	items
0	(strawberry, ice cream)
1	(apple, banana, biscuits)

I have also two separate lists of "fruit" and "snack".

fruit = ["strawberry", "apple", "banana"]
snack = ["ice cream", "biscuits"]

How do I change them to key:value pairs that look like this:

	items
0	{"fruit": "strawberry" , "snack": "ice cream"}
1	{"fruit": ["apple", "banana"] , "snack": "biscuits"}

Answer 1

You haven't provided that much information. Based on your sample data :

df1 =pd.DataFrame()
df1['items'] = [('strawberry', 'ice cream'), ('apple', 'banana', 'biscuits')]

A solution would be to make a simple function associating each item with his category.

def findCategory(x):
    fruit = ["strawberry", "apple", "banana"]
    snack = ["ice cream", "biscuits"]

    result = {
        'fruit': [],
        'snack': []
    }

    for item in x['items']:
        if item in fruit:
            result['fruit'].append(item)
        elif item in snack:
            result['snack'].append(item)

    return result

Then you can apply that function on each row of your DataFrame with apply() :

df1['items'] = df1.apply(findCategory, axis=1)
# Output
                                               items
0  {'fruit': ['strawberry'], 'snack': ['ice cream']}
1  {'fruit': ['apple', 'banana'], 'snack': ['bisc...

The findCategory() function created above is working specifically for your example, but can be generalised with more categories, and with dictionnary comprehension if you are confortable with it.