@Marvin when using .deploy to setup schedules for deployed flows to run, how can i specify what parameters to run that flow with for each different schedule / deployment? If i'm using pydantic models as inputs to the flow

To specify parameters for each deployment and schedule when using Pydantic models as inputs to your flow in Prefect, you can follow these general steps: 1. Specify Parameters in Deployment YAML: You can define different sets of parameters in your

prefect.yaml

file for each deployment.

yaml
   deployments:
     - name: my_first_deployment
       schedule:
         cron: "0 0 * * *"
       parameters:
         pydantic_model_param:
           field1: value1
           field2: value2

     - name: my_second_deployment
       schedule:
         cron: "0 12 * * *"
       parameters:
         pydantic_model_param:
           field1: value3
           field2: value4

2. Using Pydantic Models in Your Flow:

python
   from prefect import flow
   from pydantic import BaseModel

   class MyParams(BaseModel):
       field1: str
       field2: str

   @flow(name="My Flow")
   def my_flow(params: MyParams):
       print(f"Running with parameters: {params}")

3. Build and Apply Deployments:

python
   from prefect.deployments import Deployment
   from prefect.server.schemas.schedules import CronSchedule

   def build_deployment():
       dep1 = Deployment.build_from_flow(
           flow=my_flow,
           name="my_first_deployment",
           schedule=(CronSchedule(cron="0 0 * * *")),
           parameters={"params": MyParams(field1="value1", field2="value2").dict()}
       )
       dep2 = Deployment.build_from_flow(
           flow=my_flow,
           name="my_second_deployment",
           schedule=(CronSchedule(cron="0 12 * * *")),
           parameters={"params": MyParams(field1="value3", field2="value4").dict()}
       )
       dep1.apply()
       dep2.apply()

This should allow you to create multiple deployments with different schedules and parameters. - For handling Pydantic models specifically, see the Prefect Docs on Parameters. - Issues related to serialization: Prefect uses

orjson

for speed, but

orjson

has limitations. If you face issues with complex Pydantic models, check GitHub Issue 9536. For detailed information, please refer to: - Prefect Schedules Documentation - Creating Deployments