Run operations
MLFlowClient.createrun
— Functioncreaterun(instance::MLFlow, experiment_id::String;
run_name::Union{String, Missing}=missing,
start_time::Union{Int64, Missing}=missing,
tags::Union{Dict{<:Any}, Array{<:Any}}=[])
Create a new Run
within an Experiment
. A Run
is usually a single execution of a machine learning or data ETL pipeline.
Arguments
instance
:MLFlow
configuration.experiment_id
: ID of the associatedExperiment
.run_name
: Name of theRun
.start_time
: Unix timestamp in milliseconds of when theRun
started.tags
: Additional metadata forRun
.
Returns
An instance of type Run
.
MLFlowClient.deleterun
— Functiondeleterun(instance::MLFlow, run_id::String)
deleterun(instance::MLFlow, run::Run)
Mark a Run
for deletion.
Arguments
Returns
true
if successful. Otherwise, raises exception.
MLFlowClient.restorerun
— Functionrestorerun(instance::MLFlow, run_id::String)
restorerun(instance::MLFlow, run::Run)
Restore a deleted Run
.
Arguments
Returns
true
if successful. Otherwise, raises exception.
MLFlowClient.getrun
— Functiongetrun(instance::MLFlow, run_id::String)
Get metadata, metrics, params, and tags for a Run
. In the case where multiple metrics with the same key are logged for a Run
, return only the value with the latest timestamp. If there are multiple values with the latest timestamp, return the maximum of these values.
Arguments
Returns
An instance of type Run
.
MLFlowClient.setruntag
— Functionsetruntag(instance::MLFlow, run_id::String, key::String, value::String)
setruntag(instance::MLFlow, run::Run, key::String, value::String)
setruntag(instance::MLFlow, run::Run, tag::Tag)
Arguments
instance
:MLFlow
configuration.run_id
: ID of theRun
under which to log theTag
.key
: Name of theTag
.value
: String value of theTag
being logged.
Returns
true
if successful. Otherwise, raises exception.
MLFlowClient.deleteruntag
— Functiondeleteruntag(instance::MLFlow, run_id::String, key::String)
deleteruntag(instance::MLFlow, run::Run, key::String)
deleteruntag(instance::MLFlow, run::Run, tag::Tag)
Arguments
instance
:MLFlow
configuration.run_id
: ID of theRun
that theTag
was logged under.key
: Name of theTag
.
Returns
true
if successful. Otherwise, raises exception.
MLFlowClient.searchruns
— Functionsearchruns(instance::MLFlow; experiment_ids::Array{String}=String[], filter::String="",
run_view_type::ViewType=ACTIVE_ONLY, max_results::Int=1000,
order_by::Array{String}=String[], page_token::String="")
Search for runs that satisfy expressions. Search expressions can use Metric
and Param
keys.
Arguments
instance
:MLFlow
configuration.experiment_ids
: List ofExperiment
IDs to search over.filter
: A filter expression over params, metrics, and tags, that allows returning a subset of runs. See MLFlow documentation.run_view_type
: Whether to display only active, only deleted, or all runs. Defaults to only active runs.max_results
: Maximum number of runs desired.order_by
: List of columns to be ordered by, including attributes, params, metrics, and tags with an optional “DESC” or “ASC” annotation, where “ASC” is the default.page_token
: Token indicating the page of runs to fetch.
Returns
- Vector of
Run
that were found in the specified experiments. - The next page token if there are more results.
MLFlowClient.updaterun
— Functionupdaterun(instance::MLFlow, run_id::String; status::Union{RunStatus, Missing}=missing,
end_time::Union{Int64, Missing}=missing, run_name::Union{String, Missing}=missing)
updaterun(instance::MLFlow, run::Run; status::Union{RunStatus, Missing}=missing,
end_time::Union{Int64, Missing}=missing, run_name::Union{String, Missing}=missing)
Update Run
metadata.
Arguments
instance
:MLFlow
configuration.run_id
: ID of theRun
to update.status
: Updated status of theRun
.end_time
: Unix timestamp in milliseconds of when theRun
ended.run_name
: Updated name of theRun
.
Returns
- An instance of type
RunInfo
with the updated metadata.