Richard Stotz

You're correct that YDF models can vary across platforms due to differences in the underlying randomness primitives. YDF uses C++'s and occasionally Abseil's random library, neither of which guarantees identical...

Hi, apologies for the late reply. I will try to reproduce this - it very much sounds like a bug somewhere. If you can provide any more details on the...

Hi, I cannot reproduce this yet in Colab, so let's try to dig deeper here. I'm training and predicting 10K rows and 5000 features, no validation data / no early...

Tl;Dr: It's likely possible by [first converting to JAX and then to Tensorflow](https://ydf.readthedocs.io/en/latest/tutorial/compose_with_jax/), but our team has not yet tried it. There are two paths for serving from Tensorflow: ##...

That's super interesting, thank you for letting us know!

Great suggestion. I know it's possible (we've experimented with it in the past), including an example in the docs is a good idea. If anyone wants to write something up,...

Great, thank you for reporting, closing this

I think you can patch the YDF standalone example at https://github.com/google/yggdrasil-decision-forests/tree/main/examples/standalone and replace the Bazel file with ``` package( default_visibility = ["//visibility:public"], licenses = ["notice"], ) cc_library( name = "ydf_tutorial_model",...

I agree that this is a nice feature to have, we'll consider it for our roadmap! Anyone seeing this issue - Please comment or react with emoji to express support...

In the current implementation, YDF does not support unpacking pandas dataframes to multi-dimensional features, but I'll take this as a feature request to add this functionality. In the first case...