Statistically, data science developers use 4.3-4.4 libraries for working with data. Does that mean t...

# datascience

[voddan]

Statistically, data science developers use 4.3-4.4 libraries for working with data. Does that mean that Kotlin needs 5-6 libraries for DS before it gets popular?

[thomasnield]

I think the question is more involved. 5-6 core libraries are needed, yes. But look at Scala which had those libraries, got lots of hype, and yet never really took off in the mainstream like people thought it would.

There are good linear algebra libs (e.g. NumPy -> ojAlgo, ND4J, Koma), data frame (Pandas -> TableSaw, Krangl), and deep learning (Smile, DL4J).

But I think there are more factors beyond libraries.

[stan0]

I wonder if the kotlin-jupyter effort would help as well. Is there any progress on it? I recently saw the last commits are from months ago.

[Thomas Legrand]

@stan0 otherwise there is BeakerX Notebook