kotlindl VS jackson-module-kotlin

This is a really nice sample app (works out of the box) that Jetbrains have made to showcase the capabilities of this Kotlin Deep Learning library https://github.com/Kotlin/kotlindl

There are Java bindings for TensorFlow, but that's quite low level. I tried to see if I can get some Keras API for Scala, but I'm no expert and haven't had enough time to invest in this, so it's stuck in alpha. Maybe I develop it slow burning over the next year. A bit envious that Kotlin has a Keras-like library.

KotlinDL 0.3 is available now on Maven Central with a variety of new features — check out all the changes that are coming to the new release!

Could you please add an issue with your use-case and proposed solution? Or write here more details. Do you need just serialization to objects in memory? or just in bytes?

While KotlinDL seems to be a good solution by Jetbrains, I would personally stick to Java frameworks like DL4J for a better community support and likely more features.

For deep learning this is your best bet: https://github.com/jetbrains/kotlindl

So, the KotlinDL library could solve classification and regression tasks on tabular data and image data.

Thank you for your feedback! Kotlin∇ is a research prototype for building differentiable programs from basic primitives. If you’re looking for a higher-level API, KotlinDL [1] or KInference [2] might be a better fit for building end-user applications such as image, speech or language processing.

[1]: https://github.com/JetBrains/KotlinDL

[2]: https://github.com/JetBrains-Research/kinference/

strictNullCheck option in the Kotlin module doesn't cover non-object-field situations at all. https://github.com/FasterXML/jackson-module-kotlin/issues/479

But, it sucked. I was constantly finding bugs related to nullability, around generics, inline/value classes, etc. And it wasn't the typical trivial ones, either. Figuring out that you're calling a Java library and encountering a "platform type" is the least problematic. My biggest struggle came from using libraries that were written in Java first, but then added a Kotlin API on top. So, my code didn't know I was dealing with a platform type and neither did I. Yet they were still broken. Some of them are broken in an unfixable way, such as JacksonXML's Kotlin module: https://github.com/FasterXML/jackson-module-kotlin/issues/479. So many Java libraries rely on reflection even when it's not obvious that they might, so you pretty much can't ever use value classes in Kotlin, including the unsigned number types.

You should add Jackson Module Kotlin in your project. After that, you can't specify a type of object explicitly.

Not really, you just need to handle some standard types and have a good extension mechanism to let the user implement factories. That library having built in support for UUID and datetime is probably a bit much, I'd prefer a plugin like https://github.com/FasterXML/jackson-module-kotlin

There is a bug at this very moment that makes it literally impossible to deserialize [1, 2, 3, null] into a List with Jackson's Kotlin module. Yes- even if you set KotlinModule(strictNullChecks = true). It's just broken. (Here's the link: https://github.com/FasterXML/jackson-module-kotlin/issues/479)

Got you, thanks. I didn’t mean to sound offensive - a better wording would be that they don’t seem to prioritize this part, at least based on https://github.com/FasterXML/jackson-module-kotlin/issues/199, which has one abandoned branch from a 3rd party and a couple of comments from from maintainers.

If you use Jackson, don't forget to include the Kotlin extensions https://github.com/FasterXML/jackson-module-kotlin

Did you include jackson-module-kotlin?