Pablo Duboue
Pablo Duboue
Hi, it is correct that if you are evaluating a single unknown feature the system will always pick the same class for it. In general it is the majority class,...
Why don't you write a PR for that? But if you care about performance, some changes might affect speed. You might want to test both options (with a variable vs....
What you describe seems more in line with application code and it is beyond what a classifier is or does. But working out confidence levels on the predictions is a...
Free Pascal default Video is cp437. It was ported to unicode recently (see https://www.mail-archive.com/[email protected]/msg14512.html). To run in in cp437 the terminal should be set to cp437 and it should work,...
moving the recursive call to `tail_recursion` doesn't work either: ```php function fac($n, $acc = 1) { if ($n == 1) { echo (new Exception())->getTraceAsString()."\n"; return $acc; } return tail_recursion('fac')($n -...
Hmm, looking at https://github.com/functional-php/trampoline, a full solution might need two functions.
> > In particular, supporting _only_ PHP 8 would save a good bit of conditional code -- not sure if that's fine for @DrDub's purposes or not. For the RubixML...
Sure thing, I put it separately as I wasn't sure if you'd accept the PR. Great project, by the way. Fun times.
Let me know how it looks now. I think I got all the changes in.
I'd love it if we can support models exported in the ONNX format: https://onnx.ai/ But we'll need much more contributors to achieve that.