akka optimize training workloads
Manage training workloads
Synopsis
A workload defines a training target, specifying the dataset and grader used to evaluate candidate models, along with reference pointers to model entries.
Runs and models reference workloads by immutable ID rather than name, so renaming a workload preserves all associations. Workload commands accept either workload name or ID.
Workloads support three standard reference pointers: "base" points to the baseline model, while "candidate" and "champion" track models under evaluation or in active use.
Examples
# Create a workload with an evaluation dataset akka-optimize training workloads create commit-router --dataset eval-set # Set the base model reference pointer akka-optimize training workloads refs set commit-router base 4f3a91c07b2e5d68 # Compare candidate model scores against the baseline akka-optimize training workloads scores commit-router
Options inherited from parent commands
--cache-file string location of cache file (default "~/.akka/cache.yaml")
--config string location of config file (default "~/.akka/config.yaml")
--context string configuration context to use
--disable-prompt Disable all interactive prompts when running akka commands. If input is required, defaults will be used, or an error will be raised.
This is equivalent to setting the environment variable AKKA_DISABLE_PROMPTS to true.
--force Skip the confirmation prompt
--jq string Filter the JSON output with a jq expression
--json string Output JSON with only these fields; --json= lists the fields a command offers
--no-color Print without color, as NO_COLOR does
-o, --output string set output format to one of [text,json,json-compact,go-template=] (default "text")
--page-mode string the mode for paging, either paged, buffered or auto. (default "auto")
-q, --quiet set quiet output (helpful when used as part of a script)
--region string Region to use if the project has more than one region
--timeout duration client command timeout (default 10s)
--trainer-url string Trainer service URL (default http://localhost:9001, or TRAINER_URL)
--url string Optimize service URL (default http://localhost:9010, or OPTIMIZE_URL)
--use-grpc-web use grpc-web when talking to Akka APIs. This is useful when behind corporate firewalls that decrypt traffic but don't support HTTP/2.
--verbose set verbose output
SEE ALSO
-
akka optimize training - Base models, workloads, runs, snapshots, comparisons, pipelines, evaluations, endpoints and their data
-
akka optimize training workloads annotate - Update the description for a training workload
-
akka optimize training workloads create - Create a new training workload
-
akka optimize training workloads get - Display details for a training workload
-
akka optimize training workloads history - Display change history for a training workload
-
akka optimize training workloads list - List training workloads
-
akka optimize training workloads refs - Manage workload reference pointers
-
akka optimize training workloads rename - Rename a training workload
-
akka optimize training workloads scores - Compare candidate model scores against baselines for a workload
-
akka optimize training workloads set-evaluation - Set evaluation dataset and grader for a workload