pulp.lpDot update

[nariaki3551]

to address issue #483

With this change, the following lpDot operation behaves as expected.

import pandas as pd
import numpy as np
import pulp as pl

x = pd.Series(pl.LpVariable.dicts("x", range(3)), index=range(3))  # Series of LpVariable
y = 2 * x  # Series of LpAffineExpression [2*x_0, 2*x_1, 2*x_2]
a = np.arange(3) + 1  # [1, 2, 3]

print(pl.lpDot(y, a))
>>> 2*x_0 + 4*x_1 + 6*x_2   # but, current version is x_0 + 2*x_1 + 3*x_2

[CLAassistant]

All committers have signed the CLA.

[CLAassistant]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}

[pchtsp]

could you add a unit test to show what did not work and how it works now? thanks.

[nariaki3551]

Thanks for your comment! I added a test for this change. With this change, lpDot outputs the correct LpAffineExpression even when an LpAffineExpression is given to lpDot.

current version

x = LpVariable("x")
y = LpVariable("y")
z = LpVariable("z")
a = [1, 2, 3]

dict(lpDot([2*x, 2*y, 2*z], a))
>>> {x: 1, y: 2, z: 3}

with this change

x = LpVariable("x")
y = LpVariable("y")
z = LpVariable("z")
a = [1, 2, 3]

dict(lpDot([2*x, 2*y, 2*z], a))
>>> {x: 2, y: 4, z: 6}

[pchtsp]

thanks and sorry for the late reply. Could you take out PSCIP_CMD from this PR? so I can add only the lpDot use? thanks. You can add a new PR with the new solver.

[nariaki3551]

I have completed the requested changes. The PSCIP_CMD addition has been removed from this pull request, focusing solely on the lpDot updating as you suggested.

I apologize for my delayed response. Thanks.