Trusted-AI / AIX360

Interpretability and explainability of data and machine learning models
https://aix360.res.ibm.com/
Apache License 2.0
1.63k stars 307 forks source link

Fixed TRXF Feature evaluation bug #155

Closed kmyusk closed 2 years ago

kmyusk commented 2 years ago

Problem: In the evaluation of the postfix representation, the stack was popped in the wrong order for the lhs and rhs operands. So the result was wrong for non-commutative operators.

Solution: Popped the stack in the correct order.

Testing: Corrected unit tests that failed to catch the bug. Signed-off-by: Yusik Kim kmyusk@gmail.com

vijay-arya commented 2 years ago

Thanks! @kmyusk