Thomas Parnell

From Page 107 of "Advances in Financial Machine Learning" > For those cases where purging is not able to prevent all leakage, we can impose an embargo on training observations...

Yes, this is looks like what I would expect given the definition of PurgedKFold

Thanks for the quick and helpful response. It's clear now - thanks!

@simon-mo sure, will try to get to that later today

@YuWang916 I had actually implemented support in the engine for this at some point, but wasn't sure if there was any interest in this feature beyond the support in the...

CI errors do not look related to these changes

@diego898 @simon-mo I made the `truncation_side` configurable via a new option in the `ModelConfig`. Please take a look and let me know what you think.

@simon-mo Do you agree with @njhill that it makes sense to hard-code `truncation_side` to `left`? If so, I will revert this branch to the earlier commit.

@njhill I've reverted back to the version with truncation side fixed to left, and resolves some minor conflicts with changes on main branch. I think it is ready to merge.

IDK why ruff is failing - the formatting checks are passing for me locally: ``` $ bash format.sh vLLM yapf: Done vLLM codespell: Done vLLM ruff: vLLM isort: Done ```