gonb VS gomlx

I've had these strong feelings and the OP describes it really well. Despite being a polyglot programmer, I really struggle with Python, both in expression and performance (unless it's just config for GPUs).

Some of this frustration was recently an "Unpopular Opinion" on the Go Time Podcast regarding Python being great for "data exploration" but not for "data engineering": https://changelog.com/gotime/304#t=3196

I've been yearning for better interactive tooling and ML-related libraries bridge this gap and started using some even in just the last week:

* GoNB (Golang-support for Jupyter notebooks, also from a Googler) https://github.com/janpfeifer/gonb

* That uses Go-Plotly for graphs/UI: https://github.com/MetalBlueberry/go-plotly

* GoMLX (GoNB author is also on that project, many thanks Jan!) https://github.com/gomlx/gomlx

* Hidden at the end of OP is LangChainGo for LLMs, which I haven't used yet: https://github.com/tmc/langchaingo

Pick those up and let's make the Go community stronger together!

When I created a new Jupyter file in Go, I faced a challenge trying to replicate the development process I usually follow with Python. In Python and Jupyter Notebook I can conveniently run code in separate parts, saving previous values in memory and using cells to organize code. This flexibility was missing in Go, and it took me some time to figure out a solution. However, I came across a helpful tutorial that explained how to use caching with the Go Kernel, making the process smoother with gonb.

I needed one more thing to make myself feel at home, something I usually use with Python. When working with data, I often turned to the Jupyter VSCode extension for its convenience. To my relief, I discovered that a Go kernel existed, tailored perfectly for my needs.

Training library, with some pretty-printing. Including plots for GoNB Jupyter notebook.

Tutorial (and demo) here. Source code in github.com/janpfeifer/gonb.

Well, since you offered, and you have Python experience, if you are looking for a giant itch to scratch -- actually what started me in this project -- check github.com/gomlx/gomlx.

On that topic, let me advertise an experimental alternative in Go for the ML market: GoMLX -- it uses XLA, same engine that powers TensorFlow and Jax (so presumably same speed/accelerator support). It has a tutorial, examples and already has a working diffusion model -- LLMs coming soon.

A bit experimental still, but moving fast and many examples (even a [Diffusion example](https://github.com/gomlx/gomlx/blob/main/examples/oxfordflowers102/OxfordFlowers102_Diffusion.ipynb)).

GoMLX (github.com/gomlx/gomlx) is a fast and (relatively) easy-to-use set of ML libraries built on top of OpenXLA, a just-in-time compiler of numeric computations for CPU, TPU and GPUs.