Intelligence / ATH documentation

ath grade

Record the outcome and compare it with a prediction.

Usage

$ath grade fran --actual 4:32

Arguments

<benchmark>: Required benchmark ID for the attempt being graded.

What it does

Previews the result, then saves it and grades an open prediction. Without an open prediction it records the result. An existing matching result can be graded without adding a duplicate.

Use a clock time for timed efforts, a number for repetitions, or a load such as 100kg. --date selects an earlier attempt. --again is for a genuine second attempt that day. An agent can attach its explanation of a miss with --analysis after reviewing the returned evidence.

$ath grade fran --actual 4:32 --dry-run

Options

--actual <score>: what happened: 4:32 for a time, 245 for reps, 100kg for a load
--date <date>: the day of the attempt, as YYYY-MM-DD. Leave it off for today
--scaling <rx|scaled>: whether the workout was done as written, or scaled
--analysis <json>: the agent's write-up of a miss, after it has read the dossier
--file <path>: change a file other than the one in this folder
-y, --yes: write it without asking — for scripts, and when you are sure
--again: record a second attempt on a day that already has a result
--dry-run: show what would be written, and write nothing
--json: structured output, for an agent rather than a person

Related commands

Up nextath backtest →