Mycchaka Kleinbort
03/07/2023, 12:50 PM@asset
def training_data():
...
@asset
def ml_model(training_data):
...
@asset
def model_report(ml_model):
...
This works well, but it overwrites the earlier models and model reports.
Is there a way to version the assets without creating copies? Something to replace this:
@asset
def model_report_jan(ml_model_jan):
...
@asset
def model_report_feb(ml_model_feb):
...
@asset
def model_report_mar(ml_model_mar):
...
Mycchaka Kleinbort
03/07/2023, 12:52 PMMycchaka Kleinbort
03/07/2023, 12:54 PMAndras Somi
03/07/2023, 12:59 PMThis works well, but it overwrites the earlier models and model reports.Static (or even dynamic) partitions might help with this. You could define your own io manager to handle partitions to avoid overwriting.
Mycchaka Kleinbort
03/07/2023, 1:01 PMMycchaka Kleinbort
03/07/2023, 1:01 PMMycchaka Kleinbort
03/07/2023, 1:03 PMjan_model = SourceAsset(...) # Hard coded?
@asset
def model_report_jan(jan_model):
...
Vinnie
03/07/2023, 1:14 PMMycchaka Kleinbort
03/07/2023, 1:19 PMVinnie
03/07/2023, 1:20 PMMycchaka Kleinbort
03/07/2023, 1:22 PMRahul Dave
03/07/2023, 2:55 PMRahul Dave
03/07/2023, 2:56 PMRahul Dave
03/07/2023, 2:57 PMMycchaka Kleinbort
03/07/2023, 3:00 PMHmm... the new model is predicting strange things on new data, I want to apply the old model to the new data to compare
I have an asset
@asset
def y_pred(model, X):
...
and I wanted to run it with a previous materialization of the model (which I have saved somewhere)Andras Somi
03/07/2023, 4:00 PMRahul Dave
03/07/2023, 4:53 PMMycchaka Kleinbort
03/07/2023, 4:56 PMRahul Dave
03/07/2023, 5:06 PMMycchaka Kleinbort
03/07/2023, 5:16 PMchris
03/07/2023, 9:13 PMChris Histe
03/10/2023, 2:56 PM