Look. A web service inside the notebook: https://gist.github.com/altavir/c959ca422288924b493fb7fc4c68220a. I saw a task for C++ recruitment in twitter and decided to try to do it in the notebook. It works!

Hi, Is version 0.11.0-45 of the jupyter gradle plugin (org.jetbrains.kotlin.jupyter.api) meant as replacement of 0.10.0-131-1 ? Why is the latest version on maven-central https://mvnrepository.com/artifact/org.jetbrains.kotlin.jupyter.api/org.jetbrains.kotlin.jupyter.api.gradle.plugin still 0.10.0-53? I'm a bit lost here.

By the way, I am finalizing some things in VisionForge. Now I have fully working forms and widgets with bi-directional support as well as plots and 3d rendering. I am thinking about adding interactive tables and we are going to do a production release and a paper.

Double.MAX_VALUE-1< Double.MAX_VALUE

is wrong on the JVM (because of limited precision) although it is conceptually true. But what's the largest

Double

for which the assumption holds also on the JVM?

londogard-nlp-toolkit 1.1.0 release! 🥳 One of the, if not the, best NLP toolkit on the JVM. GitHub - londogard/londogard-nlp-toolkit Some of the updates includes optimizations, classifiers, vectorizers & unsupervised keyword extraction. See 🧵 for more details

Is it just me, or is the new Duration API still struggling in Jupyter (or vice versa):

Added bi-directional integration with DataFrame to Tables.kt: https://github.com/mipt-npm/tables-kt/blob/master/tables-kt-dataframe/src/test/kotlin/space/kscience/dataforge/dataframe/DataFrameTableTest.kt. Working with DataFrame outside of notebook is hard

Could you also ping https://nteract.io/kernels already/again regarding the addition of the kotlin kernel?

Anyone used

@tensorflow/tfjs

from KotlinJS?

It feels like dark magic: When using the kotlin-notebook IDE integration plugin, it "eats " away commas and opening round brackets. That is: I type them, but they do not appear irrespective of the cell type. Is this a known problem?

In the past, I could use

Class.forName("jupyter.kotlin.KotlinContext")

to test if my code is running within the jupyter kernel. But with the improved integration via

kotlin-jupyter-api-0.11.0-45.jar

this test does no longer work, because the jar ships with the

KotlinContext

. How could I now perform such a test?

message has been deleted

Is there some tutorial or documentation teaching how to build an interactive vis in a jupyter notebook? I was just reading https://mesa.readthedocs.io/en/latest/ and the engine bits are already part of kalasim, but the fun bit about mesa is its integrated interactive viewer, which I guess could also live in a jupyter notebook.

Anyone that built predictors on top of either Kmath or Multik? Like SKLearn.

Does anybody know how hard it is to configure a notebook to run both Python and Kotlin? Been having a good time writing Kotlin notebooks for a customer but occasionally I want to show Python sidebars too.

What's the state of Kotlin notebooks and 3D plotting? As well as 3D animation? Had anybody had success animating physics simulations in 3D? Or just plotting 3D charts…

The thing I did try was adding fake datapoints at the beginning of the plot, one for each label. By controlling the order, I hoped to control the colors. But it didn’t seem to have the effect I wanted.

Did a small writeup this weekend; https://blog.londogard.com/blog/kotlinjs-onnx-deep-learning-browser It’s on how you can use ONNX-webruntime from KotlinJS to run inference inside the browser (which works the same on all OS:s)! A .gif of the DL inference can be found here

Hey, do you know a jvm/Kotlin library for signal correlation, similar to scipy? https://docs.scipy.org/doc/scipy/reference/generated/scipy.signal.correlate.html and https://docs.scipy.org/doc/scipy/reference/generated/scipy.signal.correlation_lags.html#scipy.signal.correlation_lags

sorry for the off-topic, but if you are a scientist, postdoc, phd or master based in Ukraine, Casus may help you relocating into Germany (feel free to PM me)

I’m trying to import the kotlinx datetime library into a jupyter notebook. I thought I understand the syntax, but apparently not.

@file:DependsOn("org.jetbrains.kotlinx:kotlinx-datetime:0.3.2")
%useLatestDescriptors
%use lets-plot

import java.util.Random
import kotlinx.datetime.Instant

Running that yields the error

Line_214.jupyter-kts (6:25 - 32) Unresolved reference: Instant

What am I missing?

I’m using LetsPlot to graph a line (electrocardiogram) plus a scatterplot of detected R peaks. This helps visualize whether the R peak detection algorithm is working correctly. The ECG line looks great. The problem is that the R peak scatterplot is behaving strangely. About half the points appear on top of the line graph as expected, but then some points cluster at the end of the graph in the top right corner.

As a quick followup: When I use the same data with JFreeChart, the graph looks as expected. So I’m suspecting something specific to LetsPlot.

Hello, is MultiK still maintained by JetBrains? Looks like all development suddenly stopped... https://github.com/Kotlin/multik

@Pavel Gorgulov Hello, could you provide an update on this issue ?https://github.com/Kotlin/multik/issues/92 We are seeing very high memory usage spikes from 750MB to 9GB of RAM after preprocessing where we create a tensor of size 500, 3, 512, 612 using

mk.ndarray(it, batchSize, 3, targetHeight, targetWidth)

I tried chunking it to batches of 4 but I still hit 5GB RAM usage from this line

Hi I'm using spark mllib with kotlin . but get an error

fun main() {
    withSpark {
        dsOf(c(1, 2, 0.4f), c(2, 2, 0.5f)).map {
            ALS.Rating(it._1, it._2, it._3)
        }.show()
    }
}

Exception in thread "main" java.lang.ClassCastException: kotlin.reflect.jvm.internal.KTypeImpl cannot be cast to kotlin.jvm.internal.TypeReference

Is there a way to run pre-release versions in Kotlin notebooks? Have been playing with context receivers, but when enabling that in 1.6.20 it creates pre-prerelease versions of my libs that cannot be loaded in Jupyter kernels.

Used

JupyterReplTestCase

since a long time to automate some regression testing of notebooks and it always worked fine. But now I’m trying slightly different setup with a lib.json file containing some imports (before this was handled by

JupyterIntegration

) and I cannot get it to work. In my notebook:

%use roboquant@url[<https://roboquant.org/roboquant.json>]

In my unit test:

class NotebookTester() : JupyterReplTestCase(ReplProvider.forLibrariesTesting(listOf("roboquant"))) { ... }

But if I try this, the imports that are defined in the roboquant.json are ignored. Anyone know what I’m doing wrong or how I can get the imports in the roboquant.json files to be used for my unit test (other then specifying them explicitly in my notebooks) ?