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
Following the script that was created under the Repeat benchmark, there needs to be a way to run the script that compiles the JSON object that the script reads into a table that is more readable.
There is a PR in tachometer where you can turn on 'collate' and it would filter the table to separate tables for better readability of the results, so to achieve this with my script, I currently have to symlink to that PR/branch
The next task around this work is to figure out a 'base suite' of benchmarks that would be able to test Repeat holistically.
-----------Update-----------------------------
Currently, each 'configuration' is generated to a unique url that is passed into the benchmarks array (for tachometer config), and the entryName under measurement all point to the same operation
This work is moved back to backlog, as this works locally, but to merge into the main branch, would have to wait until the collate PR gets merged to tachometer.
Following the script that was created under the Repeat benchmark, there needs to be a way to run the script that compiles the JSON object that the script reads into a table that is more readable.
Tracking work here https://github.com/microsoft/fast/tree/users/wendy/custom-script
-----------Update-----------------------------
Currently, each 'configuration' is generated to a unique url that is passed into the benchmarks array (for tachometer config), and the entryName under measurement all point to the same operation
another possible approach is to create unique entryNames under the measurement array (for tachometer config):
The text was updated successfully, but these errors were encountered: