Richard Stotz

I believe that the issue you're seeing is the following: You are combining an outdated Tensorflow version (2.8.0) with an earlier TF-DF version (1.2.0). This is not possible since TF-DF...

Closing this as obsolete

Closing this issue since it has been stale for >1 year now. However, I want to point out that TF-DF has since added a parameter [maximum_model_size_in_memory_in_bytes](https://ydf.readthedocs.io/en/latest/hyper_parameters.html#maximum-model-size-in-memory-in-bytes) to limit the size...

Hi everyone, just wanted to share some quick tangentially related info: While there is still no GPU-implementation for TF-DF, TF-DF models can now run on even faster FPGAs for really...

Since this bug was opened, distributed training has been rewritten and is now stable. See https://github.com/tensorflow/decision-forests/blob/main/examples/distributed_training.py for an example. If there are still feature requests or issues with TF-DF distributed...

Closing as obsolete, TF-DF, TF and bazel have changed significantly since. Feel free to open a new issue if you still have this problem.

Years later, this can finally be closed :) - Mean Average Error has been a built-in loss in TF-DF for a while now - You can train models with custom...

I believe that there could be a fundamental misundestanding. CART trees are not generated the same way neural networks are; in fact, the two algorithms are fundamentally different. For neural...

Thank you for the clarification! Probability and number of examples are used by the decision tree to output confidence estimates along with the predictions, as well as e.g. tree pruning...

Unfortunately, TF-DF does not offer a specific API for this :( You're probably able to bootstrap this with a bit of code: Say you have `n` leaves in your hand-made...