I m a little confused about the `upstream tasks` arg on `Tas

Question

I m a little confused about the `upstream tasks` arg on `Task map` a list of upstream dependencies to map over What would mapping over the upstream dependencies do

Luke Orland · Answer

assign different upstream dependencies to each input being mapped over

nicholas · Answer

Hi < Luke Orland> The `upstream tasks` arg on `Task map` will be resolved as the dependencies for all the mapped tasks An example ```root = Root node1 1 = Node name= Node 1 1 map upstream tasks= root node1 2 = Node name= Node 1 2 map upstream tasks= root ``` This would mean that `root` task would be the upstream dependency of all the tasks generated from `node1 1` and `node1 2` to generate a schematic like this

nicholas · Answer

And to further clarify using the `upstream tasks` argument creates state dependencies between tasks but does NOT create data dependencies For mapping we DO generate multiple task instances based on the length of the upstream result but we don t pass the data In this example `hi` would print 3 times ```from prefect import task Flow task def print me print hi task def return list return 1 2 3 with Flow demo as flow print me map upstream tasks= return list flow run ```

Luke Orland · Answer

I think what I experienced was ```downstream task map twelve data objects upstream tasks= task outputting a list with six items ``` generating 6 mapped tasks instead of 12 I fixed it by using `upstream tasks= unmapped task outputting a list with six items `