You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
LAMBKIN collects a lot of data for each benchmark run, which is then made available through lambkin.data APIs. This is fine and all, but it is overkill for many uses cases. Sometimes, you just need to look at the data on a spreadsheet. Having a CLI to quickly export benchmark data to common formats such as .csv would simplify some workflows.
Definition of Done
Update lambkin.data APIs to collect all data by default e.g.:
Feature description
LAMBKIN collects a lot of data for each benchmark run, which is then made available through
lambkin.data
APIs. This is fine and all, but it is overkill for many uses cases. Sometimes, you just need to look at the data on a spreadsheet. Having a CLI to quickly export benchmark data to common formats such as .csv would simplify some workflows.Definition of Done
lambkin.data
APIs to collect all data by default e.g.:lambkin.data.evo.stats()
should be optionalexport
verb to thelambkin
CLI to outputlambkin.data
frames in CSV e.g.:lambkin export [-f csv] [-i path/to/input] [-o output.csv] evo.stats
export
verb to take filtering predicates on benchmark iteration metadata e.g.:lambkin export -k "variation.parameters.some_parameter == 100" evo.stats
The text was updated successfully, but these errors were encountered: