Reputation: 4697
I have a list like:
[array([ 0.6, 0.6, -0.6, -0.6, -0.6, -0.6, -0.6, -0.6, -0.6, -0.6]), array([ 0.35, 0.35, 0.35, 0.35, 0.35, 0.35, 0.35, 0.35, -0.35,
-0.35])]
and I can guarantee that each sublist is the same size. I need to sum every element at each position (producing one list of the same size as the sublists).
This is generated like this:
sums = []
for key in est_dict.keys():
model, alpha = est_dict[key]
fX = model.predict(X)
y_fx = np.where(fX==0, -1, fX)
sums.append(y_fx * alpha)
if there's a way to build the sum immediately, without knowing the size of the sublists in advance, that would be even better.
Upvotes: 1
Views: 63
Reputation: 1052
np.sum(sums, axis=0)
You even don't need to create a new 2D numpy array.
Upvotes: 1