根据条件从另一个数据帧的值向数据帧添加新列

Question

I have two different DataFrames of different sizes.

a = np.arange(0,100,10)
b = np.random.random_sample((10,))
df1 = pd.DataFrame({'a': a, 'b': b})
df1

    a   b
0   0   0.340319
1   10  0.821281
2   20  0.592853
3   30  0.589032
4   40  0.533543
5   50  0.628839
6   60  0.431976
7   70  0.306126
8   80  0.080576
9   90  0.533240

c = np.arange(4,14,2)
df2 = pd.DataFrame({'c': c})
df2
    c
0   4
1   6
2   8
3   10
4   12

By comparing the values in column 'c' in df2 to the values in column 'a' in df1, I'd like to add df2 a new column, which'll consist of the values in column 'b' of df1.

For example, the first three values of column c are 4, 6, and 8, which are all in the range defined by the first two rows of column a in df1 (0 to 10). That's why in the new column created, I'd like assign the b value of row a = 0 (0.340139) to all of them.

Similarly, for 10 and 12 in column c, they should get the b value of row a=10 (0.821281), since they are in between 10 and 20.

So at the end, I should end up with a DataFrame like this.

df2

    c   d
0   4   0.340319
1   6   0.340319
2   8   0.340319
3   10  0.340319
4   12  0.821281

So if you guys have any tips on that, it'd be greatly appreciated.

Answer 1

We can try merge_asof

out = pd.merge_asof(df2, df1, left_on='c',right_on='a', allow_exact_matches=False)
    c   a         b
0   4   0  0.340319
1   6   0  0.340319
2   8   0  0.340319
3  10   0  0.340319
4  12  10  0.821281