Blended Prediction

Blend the predictions of several models with fixed weights. predictions has shape (n_models, n_samples) and weights has one non-negative entry per model; the weights are normalised to sum to 1 before use.

Implement solve(predictions,weights).

Returns. Return a NumPy array of length n_samples, the weighted average of the models' predictions.

Examples

Example 1

Input:

solve([[1.0, 2.0], [3.0, 4.0]], [1.0, 3.0])

Output:

[2.5, 3.5]

Example 2

Input:

solve([[10.0, 20.0], [0.0, 0.0]], [1.0, 1.0])

Output:

[5.0, 10.0]