The documentation says that word weights can be real-valued. Real numbers can be negative. This is convenient because outputs of models one could use for scoring often output log-likelihoods, which are negative numbers.
However, using negative weights results in a negative cost value, which, if minimized, maximizes the error:
If this is a mistake in the documentation, could we update the documentation to say "non-negative numbers"?
If not, could we update the documentation with the steps to be taken to avoid divergence with negative weights?
The documentation says that word weights can be real-valued. Real numbers can be negative. This is convenient because outputs of models one could use for scoring often output log-likelihoods, which are negative numbers.
However, using negative weights results in a negative cost value, which, if minimized, maximizes the error:
![image](https://user-images.githubusercontent.com/3306933/221141280-75d8a4a0-4bfb-4738-919d-b024b5af74f1.png)
Cheers, Toms