+John Forbes Not usually. The program doesn't literally go through each possibility to identify the best: it looks at all the extreme vertices of the space defined by the constraints. As such, the more constraints there are, the more vertices need to be checked. E.g. when we have 2 decision variables. x and y, consider these three constraints - 2x + y >= 1 - x >= 0 - y >= 0 and we want to minimise f(x, y) = 3x + y^2 then we imagine the 3D-Space with the curved surface "z = 3x + y^2" drawn, and the three flat planes "2x + y = 1", "x >= 0", "y >= 0" drawn too. Then there is a space where the z-curve exists within the three constraints. Looking at the extremums of it (min, max, x infinity, y infinity) let's us find the solution quickly.
great video, she mentioned solving large scale linear programming problems in her company, do you use PULP ? does it scale well with size of the data ?
Great video with excellent explanation and implementation of python for solving real-world problems which can be modelled linearly👍👏
Great Video, taught me how to allocate quantities for a retail store, given quantities in the warehouse and with constraints
Wouldn't a more constrained problem be computed faster as there would be less possibilities that need to be considered? @13:15
+John Forbes Not usually. The program doesn't literally go through each possibility to identify the best: it looks at all the extreme vertices of the space defined by the constraints. As such, the more constraints there are, the more vertices need to be checked.
E.g. when we have 2 decision variables. x and y, consider these three constraints
- 2x + y >= 1
- x >= 0
- y >= 0
and we want to minimise f(x, y) = 3x + y^2
then we imagine the 3D-Space with the curved surface "z = 3x + y^2" drawn, and the three flat planes "2x + y = 1", "x >= 0", "y >= 0" drawn too. Then there is a space where the z-curve exists within the three constraints. Looking at the extremums of it (min, max, x infinity, y infinity) let's us find the solution quickly.
Thanks for the explanation.
it depends more on the topology of the feasible set (and the complexity of the constrains)
great video, she mentioned solving large scale linear programming problems in her company, do you use PULP ? does it scale well with size of the data ?
Thanks! This is what I'm looking for.
Excellent
I wish I had this alg. pgm to tell me to get my GED instead of hs diploma so I could've gotten my BS earlier.