https://github.com/stitchfix/hamilton logo
Join Slack
Powered by
This message was deleted.
# hamilton-help
s
This message was deleted.
p
Currently I think my best option is to make 4 nodes, one with 
n
parameterized inputs and the other 3 with 
m
parameterized inputs, which should produce the set of outputs I need, but wanted to check in case it was possible to make it happen in a single node!
e
Hey! Awesome question. So currently you can't actually do a cartesian product, but there are a few good options: 1. You can do what you're doing (somewhat verbose) 2. You can use 
@parameterized
to put it all in one and generate the `m`x`n` outside of it, in a nested loop. 3. You can use the new 
@parameterize_frame
to make this a little more concise -- its experimental, but we're looking for beta testers! https://hamilton-opensource.slack.com/archives/C03AJNGDGQL/p1669601959199269
My guess would be that dbt-fal isn't installed but that's just a wild guess... Whoops wrong thread :/
p
Ah interesting, thanks! I'll try to give the 
@parameterize_frame
on the beta version!
e
Awesome! Appreciate you taking it for a spin — definitely looking for feedback before we release it to the world
👍 1
Hey -- realized I misspoke slightly. The name was called 
@parameterize_frame
while I was developing and now its called 
@parameterize_extract_columns
and you can create it from a dataframe by doing: 
@parameterize_extract_columns.from_df(df)
. We welcome feedback on naming 🙂
👍 1
Hey @Peter Robinson! Following up — have you had success with the latest decorator/rc? We’re planning to publish it shortly…
p
Hi Elijah - sorry for the delay, didnt spot this at the time! Our use case changed since and I found just the standard parameterised decorator was enough - sorry! I'll give the extract_columns one a try when a find a way to use it!
e
Awesome, sounds good! We’re planning to release shortly — will be announcing on slack.
Open in Slack
PreviousNext
Powered by