Eduardo Salinas comments

Results 30 comments of


                                            Eduardo Salinas

Stable math mode for cross-platform testing

can https://github.com/VowpalWabbit/vowpal_wabbit/pull/3999 be leveraged here?

[Roadmap] Logging and Observability

I've been exploring with this notebook to consume the sqlite trace - sort of relevant to this issue: https://github.com/microsoft/autogen/pull/2526

prototype: pytorch base learner

update https://github.com/VowpalWabbit/vowpal_wabbit/blob/master/ThirdPartyNotices.txt if you are planning on merging

Segfault on ccb_explore_adf -cb_type dr

@NegatioN how are you consuming VW? python package? or something else

is vowpal wabbit can work with NER?

The closest thing would be this POS tagger https://github.com/hal3/vowpal_wabbit/blob/master/python/examples/Learning_to_Search.ipynb Not exactly a NER but you can tag nouns.

Expose commit hash from pyvw metadata

https://github.com/VowpalWabbit/vowpal_wabbit/pull/3446

Expose commit hash from pyvw metadata

Re-opening since that PR is just a breadcrumb from a previous approach we did.

Inconsistent and unclear loss calculation for contextual bandits

we also have these other estimator impls (not part of cb_adf): https://github.com/VowpalWabbit/vowpal_wabbit/tree/master/vowpalwabbit/core/include/vw/core/estimators

[Feature Request]: Observability of agents' actions taken, tools used, and data accessed

https://github.com/microsoft/autogen/pull/2526

How can I correspond the database data in runtime_logging with actual runtime outputs?

https://github.com/microsoft/autogen/pull/2526