You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I have been looking through the optimizer options and I can not figure out if there is an optimizer setting that will allow me to increase the speed of the optimizer at the cost of accuracy. I am wondering if there is something similar to learning rate or increasing the optimization steps size. I would really like this because my initial guesses for my optimization problem is very far from the correct solution and I would like to decrease the amount of time it takes for a rough optimization to be done.
The text was updated successfully, but these errors were encountered:
Difficult question to answer since there are many different optimiser algorithms in this package. If you are using one of the default DifferentialOptimizers there is no direct "step" per se but you can try to increase the size of the population, which should give you a broader initial sample, i.e. should lead to earlier finding some good "areas" of the search space, rather than focusing on some area and going deep.
I have been looking through the optimizer options and I can not figure out if there is an optimizer setting that will allow me to increase the speed of the optimizer at the cost of accuracy. I am wondering if there is something similar to learning rate or increasing the optimization steps size. I would really like this because my initial guesses for my optimization problem is very far from the correct solution and I would like to decrease the amount of time it takes for a rough optimization to be done.
The text was updated successfully, but these errors were encountered: