CODEWITHSUNDEEP
pythonpython-3.xpandasdataframe
sharathchandramandadi
sharathchandramandadi

Reputation: 540

How to compare two columns both with list of strings and create a new column with unique items?

I have two columns both with list of strings. Basically one column df['products'] which are in all capitals. The other column is product description df['desc'].

I want to check what all items in df['products'] are present in df['desc'] and make a new column out of it.

I tried the following code:

df['uniq'] = df.apply(lambda x : [i for i in x['products'] if i.lower() in x['desc']])

I checked the other similar questions and built the above code, but it's not working.

The data looks something like this:

enter image description here

Upvotes: 0

Views: 2299

Answers (2)

jezrael
jezrael

Reputation: 863721

It seems you need add axis=1 if need check per rows:

df = pd.DataFrame({'products':[['A','B'],['D','C']],
                   'desc':[['a', 'c'],['c', 'e']]})

df['uniq'] = df.apply(lambda x: [i for i in x['products'] if i.lower() in x['desc']], axis=1)
print (df)
     desc products uniq
0  [a, c]   [A, B]  [A]
1  [c, e]   [D, C]  [C]

Upvotes: 2

John Zwinck
John Zwinck

Reputation: 249642

Don't use apply() when you don't absolutely need to. It's slow.

Instead, do it the vectorized way:

desc_upper = df.desc.str.upper()
matches = df.products.isin(desc_upper)
result = df.products[matches]

Upvotes: 3

Related Questions