Erik Sverdrup

Hi @hanneleer, you could calculate that using the function `get_tree` that gives you details on the split variable and level for every tree. You can also fit a new forest...

Hi @hanneleer, something like the heatmaps that visualize covariate levels across HTE predictions in the above tutorial link is typically what we'd recommend over focusing on on every single split...

Yes, it's a good idea to do sample splitting where you fit CATEs on one subset and evaluate them on another ([here](https://github.com/grf-labs/maq/blob/c812ef4d678564a926d43c96c2fe11986aa09d31/experiments/swaw/voting.R#L27) is an example of how you could draw...

If you want an abstract algorithm description, `instrumental_forest` is Algorithm 1 in https://arxiv.org/abs/1610.01271 with the estimating equation $\psi$ given by the moment functions described in section 7.1 of the same...

Hi @bcjaeger, thank you! Oblique forests look very interesting! It's a good question, causal_survival_forests has many complicated nuisance components, and for simplicity we opted out of user-specified estimates. I think...

Very cool! Yes the Y.grid shouldn't be too large, we emit a warning with suggestions if that is the case, but it could be nice further down the line to...

Hi @bcjaeger, I'm just curious if you ended up discovering any interesting differences when forming nuisance estimates with oblique random forests over plain random forests? Further down the line it...

Hi, I did not have anything in particular in mind, I thought just in case you'd already tried this out! I could also imagine your oblique RSF is faster than...

Thank you, that's nice to know!

**UPDATE:** We've finally decided to change the default GRF behavior to **not** have the random seed interact with the number of threads used to train the forest (#1447). This will...