from metaflow import NBDeployer, FlowSpec, step

class HelloFlow(FlowSpec):
    @step
    def start(self):
        self.x = 1
        self.next(self.end)

    @step
    def end(self):
        self.x += 1
        print("Hello world! The value of x is", self.x)

deployer = NBDeployer(HelloFlow)

then, in the next cell, you could have...

deployed_flow = deployer.argo_workflows().create()
df_name = deployed_flow.name # for later use...
triggered_run = deployed_flow.trigger()

then, you can access the

Run

object with:

run_obj = triggered_run.run

more details about Run object here: https://docs.metaflow.org/api/client also, since you have already deployed the flow, you can trigger it again as follows:

from metaflow import DeployedFlow

df = DeployedFlow.from_deployment(identifier=df_name)
triggered_run = df.trigger()

this is for the use-case when you have already deployed to argo workflows and don't want to call

.create()

again but fetch the already deployed flow and just trigger it again... (eg: from a separate notebook or terminal where you don't have access to the original

deployed_flow

object)

more details about

Deployer

here: https://docs.metaflow.org/api/deployer

also https://docs.metaflow.org/metaflow/managing-flows/deployer