https://flyte.org
Join Slack
How are you supposed to launch a workflow/task with a `StructuredDataset` or dataframe input? I get ...
# flyte-support
m
How are you supposed to launch a workflow/task with a 
StructuredDataset
or dataframe input? I get the following error when running `flytectl get task ...`:
Copy code
Error: failed to convert to a known Literal. Input Type [structured_dataset_type:<columns:<name:"prediction" literal_type:<simple:FLOAT > > columns:<name:"target" literal_type:<simple:FLOAT > > > ] not supported
Is the only option here to wrap it in a workflow?
t
How are you launching your task/workflow? Also, you're seeing that error when running 
flytectl get task
?
m
Trying to launch the task with 
flytectl executions create
and yes. sorry typo above
t
I think you can provide a URI that contains the dataframe. @glamorous-carpet-83516, does this work? And is there a better way to send structured dataset input?
m
@tall-lock-23197 I can do this with a 
FlyteFile
but not 
StructuredDataset
which is strange.
(given that a 
StructuredDataset
is backed by a 
FlyteFile
this force me to make a dummy node which wraps a file as a StructuredDataset and wrap the task in a workflow.)
t
Yeah. I'm hoping that should unblock you for now. But we should definitely simplify the UX.
f
Structured dataset is infact not a file. I think flytectl has missing converter for this type?
h
what version of flytectl are you running, @melodic-magician-71351?
m
Copy code
{
  "App": "flytectl",
  "Build": "5aefa07",
  "Version": "0.6.30",
  "BuildTime": "2023-02-28 14:20:51.381452824 +0100 CET m=+0.026807396"
}
{
  "App": "controlPlane",
  "Build": "1ccd59c",
  "Version": "v1.1.70",
  "BuildTime": "2023-02-21 14:20:38.834313886 +0000 UTC m=+0.113812103"
}
149 Views
Open in Slack
PreviousNext
Powered by

Flyte

Flyte enables production-grade orchestration for machine learning workflows and data processing created to accelerate local workflows to production.

Powered by