How do you edit your kotlin jupiter files? I’ve tried editing the files in the browser, but I miss all the shortcuts I am used to in Intellj. The editing experience in the browser just does not feel right to me. I’ve had the same issue with python, but there I was able to use pycharm to edit my files directly which felt much better.

Hello all. I’m new to this channel, and somewhat new to data science. Do any of you have any recommended learning resources? Also any preferred deep learning frameworks in Kotlin? Forgive me if this has been asked a bunch.

What would be required to publish the Kotlin Kernel jars to a package repository like maven so I could simply specify it as a dependency in a gradle file to install it for embedding purposes?

I assume that managing a maven package requires at least some effort, and I am aware that until recently there has been absolutely no point to it, because the kernel wasn't designed for embedding into other JVM projects before. My ideal user experience would basically be that I can specify something like

compile "jetbrains:kotlin-jupyter-kernel:0.8.2"

in my gradle file and this automatically gets the jar from an official source and resolves the dependencies to the relevant kotlin lib/runtime/reflect jars

There's also a #deeplearning channel here now, everyone interested in the topic is welcome to join :)

Is anyone using kotlin for data science in production?

Please RT this tweet to get more people interested in Kotlin and Data Science 🙂 https://twitter.com/prpatel/status/1337424798461419520

Hi folks! We are happy to share with you a very first preview of Multik  — a library of multidimensional arrays we are working on. It provides Kotlin-idiomatic, type- and dimension-safe API for mathematical operations on arrays. API has two pluggable implementations in Kotlin/JVM and Native (via JNI), and they can be switched on the fly to get the best performance. Right now we are in the first alpha stage and we are dying for your feedback. Read more: https://github.com/Kotlin/multik

@Pavel Gorgulov I've asked already the same question in #deeplearning room, are there some plans to use multik with other JB's data science libs (like DL) ?

Another question to understand. Why do you need 2 different implementations? Because of memory model on native?

When defining a type renderer in kotlin-jupyter for library such as https://github.com/Kotlin/kotlin-jupyter/blob/master/libraries/krangl.json#L21 , is it possible to add another renderer for a package-internal class? I'd love to add support for

krangl.GroupedDataFrame

but this class is package-internal, and I'm somehow reluctant to make it part of the public library API just to get a more pretty rendering in jupyter. For corresponding sources see https://github.com/holgerbrandl/krangl/blob/master/src/main/kotlin/krangl/GroupedDataFrame.kt#L15

Hi folks, I've worked with the kotlin-jupyter kernel last weekend and I was amazed by its stability and its wonderful module-definition feature. See yourself (just in case you like spaghetti and philosophy, but who doesn't...) https://github.com/holgerbrandl/kalasim/blob/master/simulations/notebooks/dining.ipynb . Or if you prefer a more gentle/verbose introduction see the corresponding manual page https://www.kalasim.org/examples/dining_philosophers/ FYI @Maria Khalusova data science with kotlin!

Hi, check out new CorrPlot builder https://nbviewer.jupyter.org/github/JetBrains/lets-plot-kotlin/blob/master/docs/examples/jupyter-notebooks/correlation_plot.ipynb in the last week's Lets-Plot Kotlin release: https://github.com/JetBrains/lets-plot-kotlin/releases/tag/v1.2.0

Hi are there any good tutorials on Kotlin ML pls

Does KotlinDL support reinforcement learning? E.g to support kotlin implementations such as MuZero?

When building datascience APIs with kotlin, do you refer to missing values as NA nor null? Data-scientists are likely to prefer NA, and most kotlin programmer will prefer null. I'm asking because I want to add https://www.rdocumentation.org/packages/tidyr/versions/0.8.3/topics/drop_na to krangl (which is a data-frame implementation for kotlin) but the correct kotlin name seems to be

filterNotNull

.

how did you integrate with the JVM?

Wrote a Markov chain sampler. It seems to work well, but it's kinda slow, any ideas how to speed it up? https://github.com/breandan/markovian/blob/master/src/main/kotlin/edu/mcgill/markovian/mcmc/MarkovChain.kt

house advertising warning! Just in case you guys ever feel the urge to discuss how to create a discrete event simulation with kotlin --> #kalasim

How about renaming the channel into KotlinData or KotlinForData? We seem to discuss a lot of things like kotlin jupyter kernell not directly related to data science. @Maria Khalusova?

Maybe I missed something big but... is there any support for import project local Kotlin Classes from other Files into a jupyter notebook? Found nothing online regarding this.

Are you also experiencing issues with ipynb preview of kotlin-kernel rendered/computed notebook in gitlab? It shows text output but strips all images.

For those of you not subscribed for #science channel, a little trick for using Jupyter with locally built libraries and for distribution non-published buncles. The video is . The short recepie is the following: 1. Create a project in the idea and do what you want, exposing APIs that you want to use from the jupyter. 2. Use shadow plugin to create fat-jar distribution 3. Copy resulting jar from

build/libs

to the location of your notebook or send it to the consumer alongside the notebook 4. Load it using

@file:dependsOn("yourJar.jar")

5. OR create a module json file which references local jar file in the same way. In a module you can also define default imports and startup scripts if you want. 6. The resulting notebook + jar + [optional] module file form a stand-alone bundle that will work for anybody you send it to without the requirement to install environment or build anything.

Aaaand big news: https://github.com/Kotlin/kotlin-jupyter/pull/99

@Ilya Muradyan As far as I understand, the annotation processor is used only for library definition discovery. Are there any reasons not to use Java SPI? Also why definition is not an interface it needs to access something in JupyterIntegration?

Hi all! I'm new here and I'm assessing the capabilities of Kotlin when working in the Data Science field. Any suggestions? Thanks!