Make Numpy Symmetric Matrix from a list of pairs

Question

I have a correlation matrix, but specified as pairs, like:

cm = pd.DataFrame({'name1': ['A', 'A', 'B'], 
                   'name2': ['B', 'C', 'C'], 
                   'corr': [0.1, 0.2, 0.3]})
cm
    name1   name2   corr
0   A       B       0.1
1   A       C       0.2
2   B       C       0.3

What is the simplest way to make this into a numpy 2d array correlation matrix?

    A   B   C
A 1.0 0.1 0.2
B 0.1 1.0 0.3
C 0.2 0.3 1.0

Answer 1

Not sure about pure numpy since you are dealing with a pandas dataframe. Here's a pure pandas solution:

s = cm.pivot(*cm)

ret = s.add(s.T, fill_value=0).fillna(1)

Output:

     A    B    C
A  1.0  0.1  0.2
B  0.1  1.0  0.3
C  0.2  0.3  1.0

---

Extra: for reverse (ret is as above)

(ret.where(np.triu(np.ones(ret.shape, dtype=bool),1))
    .stack()
    .reset_index(name='corr')
)

Output:

  level_0 level_1  corr
0       A       B   0.1
1       A       C   0.2
2       B       C   0.3