Run#

`Run`	Run representation
`RunStage`	Run stage
`RunStatus`	Run status

`RunInfo`	Run information representation
`RunViewType`	Run view type

`RunData`	Run data representation
`Param`	Run parameter
`Metric`	Run metric representation
`RunTag`	Run tag

class mlflow_rest_client.run.Run(**data)#

Run representation

Parameters:

infodict or RunInfo: Run info
datadict or RunData, optional: Run data

Examples

run_info = RunInfo(id="some_id")
run_data = RunData(params=..., metrics=..., tags=...)

run = Run(run_info, run_data)

Attributes:

infoRunInfo: Run info
dataRunData: Run data

Create a new model by parsing and validating input data from keyword arguments.

Raises ValidationError if the input data cannot be parsed to form a valid model.

class mlflow_rest_client.run.RunStage(value, names=None, *, module=None, qualname=None, type=None, start=1, boundary=None)#

Run stage

ACTIVE = 'active'#: Run is active

DELETED = 'deleted'#: Run was deleted

class mlflow_rest_client.run.RunStatus(value, names=None, *, module=None, qualname=None, type=None, start=1, boundary=None)#

Run status

FAILED = 'FAILED'#: Run is failed

FINISHED = 'FINISHED'#: Run was finished successfully

KILLED = 'KILLED'#: Run was killed

SCHEDULED = 'SCHEDULED'#: Run is scheduled for run

STARTED = 'RUNNING'#: Run is running or created

class mlflow_rest_client.run.RunInfo(**data)#

Run information representation

Parameters:

idstr: Run ID
experiment_idint, optional: Experiment ID
statusstr or RunStatus, optional: Run status
stagestr or RunStage, optional: Run stage
start_timeint (UNIX timestamp) or datetime.datetime, optional: Run start time
end_timeint (UNIX timestamp) or datetime.datetime, optional: Run end time
artifact_uristr, optional: Artifact URL

Examples

run_info = RunInfo("some_id")

Attributes:

id:str: Run ID
experiment_idint: Experiment ID
statusRunStatus: Run status
stageRunStage: Run stage
start_timedatetime.datetime: Run start time
end_timedatetime.datetime: Run end time
artifact_uristr: Artifact URL

Create a new model by parsing and validating input data from keyword arguments.

Raises ValidationError if the input data cannot be parsed to form a valid model.

class mlflow_rest_client.run.RunViewType(value, names=None, *, module=None, qualname=None, type=None, start=1, boundary=None)#

Run view type

ACTIVE = 'ACTIVE_ONLY'#: Show only active runs

ALL = 'ALL'#: Show all runs

DELETED = 'DELETED_ONLY'#: Show only deleted runs

class mlflow_rest_client.run.RunData(**data)#

Run data representation

Parameters:

paramsdict or list of dict, optional: Params list
metricsdict or list of dict, optional: Metrics list
tagsdict or list of dict, optional: Run tags list

Examples

param = Param(key="some.param", value="some_value")
metric = Metric(name="some.metric", value=1.23)
tag = RunTag(key="some.tag", value="some.val")

run_data = RunData(params=[param], metrics=[metric], tags=[tag])

Attributes:

paramsParamList: Params list
metricsMetricList: Metrics list
tagsRunTagList: Run tags list

Create a new model by parsing and validating input data from keyword arguments.

Raises ValidationError if the input data cannot be parsed to form a valid model.

class mlflow_rest_client.run.Param(**data)#

Bases: Tag

Run parameter

Parameters:

keystr: Param name
valuestr: Param value

Attributes:

keystr: Param name
valuestr: Param value

Create a new model by parsing and validating input data from keyword arguments.

Raises ValidationError if the input data cannot be parsed to form a valid model.

class mlflow_rest_client.run.Metric(**data)#

Run metric representation

Parameters:

namestr: Metric name
valuefloat, optional: Metric value
stepint, optional: Metric step
timestampint (UNIX timestamp) or datetime.datetime, optional: Metric timestamp

Examples

metric = Metric(name="some.metric")
metric = Metric(name="some.metric", value=1.23)
metric = Metric(name="some.metric", value=1.23, step=2)
metric = Metric(
    name="some.metric", value=1.23, step=2, timestamp=datetime.datetime.now()
)

Attributes:

namestr: Metric name
valuefloat: Metric value
stepint: Metric step
timestampdatetime.datetime: Metric timestamp

Create a new model by parsing and validating input data from keyword arguments.

Raises ValidationError if the input data cannot be parsed to form a valid model.

class mlflow_rest_client.run.RunTag(**data)#

Bases: Tag

Run tag

Parameters:

keystr: Tag name
valuestr: Tag value

Examples

tag = RunTag(key="some.tag", value="some.val")

Attributes:

keystr: Tag name
valuestr: Tag value

Create a new model by parsing and validating input data from keyword arguments.

Raises ValidationError if the input data cannot be parsed to form a valid model.