Priorities surveys

The tidymodels team has periodically fielded short community surveys to gather feedback on development priorities and possible next steps. The reports below summarize the results of each survey.

2024 report

Announced on the tidyverse blog. Almost 340 people responded. The priority given the most weight, across most groups, was causal inference; chattr, cost-sensitive learning, and sparse tibbles were among the most likely to be given zero weight.

2022 report

Announced on the tidyverse blog. Over 600 people responded. The priorities given the most weight included supervised feature selection, model fairness metrics, and probability calibrations; H2O and spatial analysis were among the most likely to be given zero weight.

2020 report

The first priorities survey. Over 300 people responded. The priorities given the most weight included model stacking and a system for model monitoring, updating, and organization; priorities involving the inner workings of tidymodels (skipping recipe steps, sparse data structures, etc.) were among the most likely to be given zero weight.