How to use cond operator?

[iamaleksandra]

Hi,

I try to use 'cond' operator, but I keep receiving the error:
"The operator cond returned an output of type Int64, when it was given inputs -100.0 and -100.0 of type Float32. Please ensure that your operators return the same type as their inputs".

I'm using the following configuration in PySRRegressor:

model_pysr = PySRRegressor(
    niterations=50, 
    batching=True,
    batch_size=1000,
    binary_operators=["+", "*", "-", "/", 'cond'],
    unary_operators=['square', 'cube', 'exp', 'sqrt', 'log10', 'tan'],
    model_selection = 'accuracy', 
    elementwise_loss='L2DistLoss()',
    parsimony = 0.1
)

I've also tried to add 'cond' as a custom operator, but with the same outcome.
Any help resolving this problem is appreciated.

[MilesCranmer]

It could be an issue of your custom operator overwriting the internal one. Try restarting.

It’s defined here: https://github.com/MilesCranmer/SymbolicRegression.jl/blob/b0cf4efda8deef80976b757b233a3c64e5d5b592/src/Operators.jl#L113

[iamaleksandra]

Thank you! It is worked indeed.