Convert dataframe to a numpy matrix

Question

I have dataframe of the form

user_id  item_id  rating
1          abc       5
1          abcd      3
2          abc       3
2          fgh       5

I want to convert it into a numpy matrix such as

# abc  abcd  fgh
[[5,    3,    0]  # user_id 1
[3,    0,    5]] # user_id 2

Can anyone help?

jezrael · Accepted Answer · 2016-12-01 19:09:11Z

3

You can use pivot with fillna, cast to int and last convert to array by values:

arr = df.pivot('user_id', 'item_id', 'rating').fillna(0).astype(int).values
print (arr)
[[5 3 0]
 [3 0 5]]

Another solution with set_index, unstack and values:

arr = df.set_index(['user_id','item_id']).unstack(fill_value=0).values
print (arr)
[[5 3 0]
 [3 0 5]]

answered Dec 1, 2016 at 19:03

jezrael

868k103 gold badges1.4k silver badges1.3k bronze badges

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

1 Answer 1