magma-nvim VS jupyter-vim

You use magma-nvim for interactive jupyter experience (or my fork which also includes three last PRs)

FYSA: https://github.com/dccsillag/magma-nvim & https://gitlab.com/usmcamp0811/nvim-julia-autotest

Check out Magma.. it works with Julia.. see my open PR for examples... does require IJulia.

Just throwing this out there... cause its pretty cool!! https://github.com/dccsillag/magma-nvim

I use them all the time and there are many features that differentiates it from vim, which is why I asked for clarification. In any case this plugin is pretty sweet if you want to evaluate cells within vim: https://github.com/dccsillag/magma-nvim

Have you taken a look at https://github.com/dccsillag/magma-nvim?

My current workflow is a combined version of Maxwellrules' solution and magma.nvim. I am mostly satisfied. I could maybe use vim-jukit's cell division, but otherwise I am good for now. Though again: I am definitely open for better solutions.

You can check out https://github.com/dccsillag/magma-nvim A nice, integrated solution in my opinion.

For vim users, vim-jupyter [0] coupled with a jupyter QtConsole is hard to beat. The short video [1] is maybe self-explanatory, but in short:

- vim on the left half of the screen, a jupyter QtConsole on the right, showing any plots possibly interactive.

- the kernel on the jupyter QtConsole can be running on a powerful remote host, e.g., GPU, but the plots are displayed locally

- Focused window is always vim. From vim editing a .py and without ever leaving vim or touching the mouse, one connects to the jupyter kernel of the QtConsole and one can send a selection of lines, or vim text objects, to be evaluated in the QtConsole.

One gets the full power of both vim and jupyter kernels with native plots. No more browser based notebooks or other editors with half-baked vim bindings.

[0]: https://github.com/jupyter-vim/jupyter-vim

[1]: https://www.youtube.com/watch?v=h59cbg4HqpY

In this config, under the Plugins folder there are various plugins that aren't found in the Astrocommunity repo (aka "Custom" plugins). A great example that I found interesting and included into my own config was mehalters repl.lua file that installs the jupyter-vim plugin. From my limited understanding to install packages into astrovim, we need to make a file in the Plugins folder, make the necessary configurations, and then it should load.

Is there any way to communicate with jupyter like jupyter-vim? I am aware of python-repl in emacs, but it does not support inline plots/figures. Moreover, it is nice to nice to show outputs in different windows (especially using multiple monitors).

Now here comes the endboss: Jupyter. For the first time, I feel like I'm missing out on stuff when using vim. I've started a job in datascience, which is actually awesome. However, I work a lot with image data. I also do a lot of analysis on results, meaning I do a lot of fancy plots that hopefully show the weaknesses of our prediction models. I recently wrote an augmentation algorithm where I had to see the output in form of an image after every step to make sure it's correct. This is not a possible workflow in vim right now. I know of many solutions that I already tried, like for example jupyter-vim or the jupyter vim mode. I'd like to work inside my terminal though. I'm not this kind of purist who needs to have a terminal that is compatible with VT100 or whatever people came up with in the 80ies. I also don't care if my terminal in based on an ascii like grid or actually rendered in HTML. I just want (Neo)vim, with the functionality of jupyter (inline plotting) even if this means vim has to be rendered inside an electron app or whatever people use these days for fancy GUIs. Imagine an electron based editor like Oni which not only runs the "real" neovim in the background, but is also able to do inline figures, images, plots and even interactive stuff. It seems to me like I can't be the only one who wants this. So after all this, here's the question: Is there anything you now of that allows for this kind of stuff? Is there any other workflow that I'm not aware of? Or do people just not use those features when working with vim? Pls help a vimmer stay at vim.

For Matlab-like "cell-mode", have a look at https://github.com/jupyter-vim/jupyter-vim/.

To work in a py file (or py codelines selection from whatever format): https://github.com/jupyter-vim/jupyter-vim