Aurélien Geron

Ah, actually I just realised that some test cases included floats without the `f64` qualifier, so that's the problem, sorry for the noise. However, I do get the issue in...

Thanks for filing this issue @Anton-4 . Here's the corresponding Roc code. It's about 30 lines of code, but it's probably possible to reduce that a bit. ```roc module [saddlePoints]...

This issue also affects records. I've opened a separate issue #7144, but feel free to close it if you think it's the same underlying cause.

As @bhansconnect suggested, I can confirm that `isEq` does not seem to be called. Here's a little module defining a `CustomSet` opaque type: ```roc # CustomSet.roc module [create, isEq] CustomSet...

I think the compiler panics when there's more than one `?` in an expression.

I got the same bug with the following code: ```roc app [main] { pf: platform "https://github.com/roc-lang/basic-cli/releases/download/0.15.0/SlwdbJ-3GR7uBWQo6zlmYWNYOxnvo8r6YABXD-45UOw.tar.br", } import pf.Stdout table = Dict.fromList [ ("AUG", Methionine), ("UAA", Stop), ("UUC", Phenylalanine), ]...

Hi @jab2727 , Thanks for your question. You are correct: we are indeed pretending that whatever action the model chose was the correct one, and we're saving the corresponding gradients....

I believe this that saying "A is outperformed by B" means that A is worse than B. If I had written "A outperforms B" then it would mean the opposite....

Thanks for your question @tc-git-1 , and thanks to @cf-chmod for the best possible answer! That's *exactly* right.

Thanks for your question @siavashr99. As @LittleCottage said, the dot product of two vectors is commutative: **a** · **b** = **b** · **a**. By the way, in Machine Learning, the...