PySR

PySR copied to clipboard

[Feature] Support for minimum and maximum result value

4

**Is your feature request related to a problem? Please describe.** I have a complex equation with 9 independent variables and I'm evaluating a small viewport of it. I only care...

sirisian

How to exclude translational constants from the learning process?

2

Hi Miles Cranmer, Thank you again for developing this useful tool. I have a question to ask about a parameter setting. Is there a way for me to exclude transnational...

jstan15

PySR compared to heuristic lab

1

hey, I've been using your Library for one of my research purposes and it's been very insightful, I switched from gplearn and could reasonably tell the difference. Recently I came...

Shreynger

Predefined function form

2

Hi Miles, In the regression process, can we pre-define a function form first, and let the regression start from this function? for example, If our objective function is x0**2 +...

nice-mon

Solver convergence

2

Hi Miles, Thanks for sharing your code. I was trying the example case with some different solver configurations. The calculation is terminated after few 2 iterations. Is this because the...

Rdfing

Partial derivative loss

4

Hi,Miles I learned that Regression losses work on the distance between targets and predictions in PySR. However, in the regression case for having a complex formula, the regression results maybe...

nice-mon

Benchmark / GPlearn

6

Can you do a basic comparison between this and gplearn with regards to speed and flexibility?

eladmw

Symbolic deep learning

11

Trying to recreate the examples from this [paper](https://arxiv.org/abs/2006.11287) PySR is always predicting scalars as a low complexity solution, which doesn't make much sense, can you please elaborate on that? And...

abdalazizrashid

Not sure if this is possible in the current setup, but would be nice to have a progress bar display inside the jupyter notebooks. To implement the multi-line progress bar,...

MilesCranmer

Recursive feature selection

3

I think something that may give a large algorithmic performance is recursive feature selection. Here is the idea: 1. Hold "next-mutation" feature importances for every equation separately. 2. Start these...

MilesCranmer