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
Description
There is a table result which is a bit of a misnomer. It describes some parameters related to a run of an algorithm container, such as its input, started and finished times, and ...its results!
The idea is that for the next major version release, we rename this table to something like algorithm_run
The text was updated successfully, but these errors were encountered:
Also rename the result endpoint to run but only include ref to result
Rename the task.run_id field to avoid confusion
Keep client.result.get() but that should only return result
Check if there are other places where this needs to be replaced to be consistent
Have 2 endpoints: /result and /run
The endpoint /result should give ONLY the result, not the other fields
The /run endpoint should give the other fields but with only a link to obtain the result field Create endpoint to patch result (?). Note that doing so would require changes in the nodes
Take into account in all other places that result is no longer a field in the /run endpoint -> update client and node (and CLI?)
Rename / change return values of get_results(), patch_result(), proxy_task_result() and proxy_result()
Check output for /task?include=result -> should only include result. Also add include=run?
Adapt UI to the changes
The endpoint /task/<id>/run is used by nodes to collect algorithm results and other run data. Should we refactor into two endpoints?
Description
There is a table
result
which is a bit of a misnomer. It describes some parameters related to a run of an algorithm container, such as its input, started and finished times, and ...its results!The idea is that for the next major version release, we rename this table to something like
algorithm_run
The text was updated successfully, but these errors were encountered: