rhayes777 / PyAutoFit

PyAutoFit: Classy Probabilistic Programming
https://pyautofit.readthedocs.io/
MIT License
59 stars 11 forks source link

Feature/mle #1029

Closed Jammy2211 closed 1 month ago

Jammy2211 commented 1 month ago

Methods which found the maximum likelihood model were previously called optimize, which this PR renames to MLE (maximum likelihood estimator).

This PR improves the MLE methods with functionality including:

The BFGS and LBFGS MLE searches have been improved in this PR, including visualization.

The main use case is fits where a good starting point is known and simple gradient descent can find the maximum likelihood solution.

The intent is to use this for strong lens sensitivity mapping.

Jammy2211 commented 1 month ago

The following two issues need to be complete before merge:

https://github.com/rhayes777/PyAutoFit/issues/1031 https://github.com/rhayes777/PyAutoFit/issues/1030