Jaremy Creechley

Ok, thanks that's good to know the weights/biases seem correct. There's a good chance I am missing a part of the serialization or messing up the prediction. All of the...

> For the other I don't store the shape metadata in the layers (they are compile-time transformed away) Eventually that would be nice. I currently am just redefining the model...

Thanks, I tried reading the dsl.nim but got lost on where things were defined. Based on that those code snippets, the only place that I'm not sure is setup correctly...

I'm only saving the `model` after training. Are any of the above stored in the `ctx`? edit: Looking through this more, it doesn't appear so. I think the model is...

Would it be possible to include a lightweight pure C BLAS like https://github.com/michael-lehn/ulmBLAS as a fallback? I'm just learning about Nim, but it I under it should be possible to...

Ok, `ulmBLAS` doesn't seem like it'd be ideal. I tried compiling it on Linux and it breaks. What is the primary goal for this issue? Is it to write naive-Nim...

Just a quick response, that all sounds great! I'm impressed by how much detail you've already got. While I enjoy writing in Julia, the whole LLVM dependency is just a...

:+1: Running into this while trying to setup an R610 server for a test install.

Personally, the `#` symbol and the `{#label}` syntax would be easier to understand and use. In my mind citations and internal references follow very distinct "mental models". Many academic papers...

Just a note on usage, but I recently updated my embedded projects to use nimble 0.14 for the lock and sync mechanism. Mostly that was great, though theres a few...