James McDermott

"As this work is still in progress, these are preliminary results evaluated on grids up to 10×10." I guess this 80% is only on the grids of small sizes, which...

Sorry for being a slowpoke/off-topic -- could you please share the tool you used to make this tables?

* Keijzer https://link.springer.com/chapter/10.1007/3-540-36599-0_7 * Nicolau and Agapitos https://dl.acm.org/doi/pdf/10.1145/3205651.3205773?casa_token=ibAlv6UdJ70AAAAA:VP78XazPwLbSCvgFZanwgr76pKH0qeoeFWjGF6L028EJpLr3Hc3jkyMZNEAvNN_t_ICN7ZsKE_Kq * Ni et al https://dev.heuristiclab.com/trac.fcgi/raw-attachment/ticket/2915/06186815.pdf Short summary: they don't like discontinuities.

`closure_method` is slightly ambiguous, as there could be multiple smooth log operators, for example. It does create extra complexity, as the code would have to deal with the `closure_method` setting...

I would be in favour of putting the onus on the user here. If they specify two division operators, give it to them! Would require a lot of new code...

Very interesting discussion! I created a related issue (related to protected operators and closure, that is): https://github.com/trevorstephens/gplearn/issues/242.

Just an extra datapoint: on Mojave, with Py3.7 and requirements.txt items all installed using Anaconda, it works out-of-the-box. Only thing to note is that librosa is not mentioned in requirements.txt....

Agreed - just spotted your issue after making this fix in a pull request.

I think this is the same error I saw, due to an update in the pytorch API. The suggested change is to replace `.data[0]` with `.data.item()`. My pull request has...