Julie Tibshirani

Hello @MaxTailt, grf's `quantile_forest` method does not actually implement Meinshausen's quantile regression forest algorithm. A major difference is that grf makes splits that are sensitive to quantiles, whereas Meinshausen's method...

We've already made the following improvements around memory usage: * **For each tree, switch to storing the samples which were drawn as opposed to out-of-bag (#182)**. This helps lower the...

@ferlocar thanks for your feedback, I tagged this issue as something we should look into.

Hello @xyztnull, thanks for the question. I understand how GPL can cause restrictions in an industry setting. We don't have current plans to change the license. However I'll leave this...

I think the increase in size is due to the fact that we now add a `_recovery_source` field to the document if `_source` is disabled, but soft deletes are enabled...

Here's an example of an error message containing the offending regexp: ``` handler panic: regexp: Compile((?i)

Initial eval results on CodeSearchNet: ``` Recall (files) 91/99 Recall (chunks) 70/99 Average chunk overlap 0.81 ``` Although we returned correct files for 91 queries, we only found relevant overlapping...

Experiment: I tried ordering the chunks by BM25 instead of the current ranking, which boosts matches on symbol definitions. It performed significantly worse on CodeSearchNet: ``` Recall (files) 91/99 Recall...

Exploring the alternative option: how hard would it be to support `content` filters correctly? As a user I'd be really confused by the limitation "the query contains a content: field...

@camdencheek could you give an example repro query? I've tried a `patterntype:regexp` query that matches the final line in a file, plus a newline. But I haven't repro'd the panic....