user
08/25/2021, 1:28 PMAnish Shah @ WANDB
09/02/2021, 6:29 PMkedro-mlflow
?
I have trained a custom mock model in which I define operations to save internal data sources to an mlflow run and load them into the desired custom mlflow.pyfunc.PythonModel
.
I am able to serve the model using --no-conda
. Without this flag I get an error raised in which ModuleNotFoundError: No module named '<kedro_package_name>'
. In my MLModel file I set loader_module: <kedro_package_name>.extras.mlflow.loader_cosine_model
. I can also provide any other additional details if anyone can help!Arnaldo
09/02/2021, 6:39 PMuser
09/03/2021, 8:28 PMuser
09/09/2021, 7:16 PMdatajoely
09/09/2021, 8:05 PMuser
09/09/2021, 11:20 PMuser
09/10/2021, 3:28 AMdatajoely
09/10/2021, 6:21 AMuser
09/10/2021, 9:07 AMuser
09/10/2021, 1:07 PMIsaac89
09/13/2021, 6:36 AMYetunde
09/14/2021, 10:33 AMSolarer
09/14/2021, 12:37 PMkedro run --env test
and still have my settings in 'test' overwritten by my settings in 'local'? I think it only works when NOT specifying an environmentSolarer
09/14/2021, 12:37 PMdatajoely
09/14/2021, 3:01 PMSolarer
09/14/2021, 3:11 PMdatajoely
09/14/2021, 3:14 PMIsaac89
09/15/2021, 12:31 PMWaldrill
09/20/2021, 4:39 PMlocal
, production_vm
, production_cloud
, etc.
That said … how can I keep the pipeline reusable with substantial changes in catalog and parameters?
Thanks for reading! 👍Waldrill
09/20/2021, 4:49 PMuser
09/20/2021, 5:28 PMconf/
, you can maintain the following structure:
conf/
|__ app1/
|__ local
|__ vm
|__ app2/
And run the pipeline with kedro run --env="app1/local"
Waldrill
09/20/2021, 6:03 PMconf/
├── app1/
│ ├── parameters.yml
│ ├── local/
│ │ └── catalog.yml
│ └── vm/
│ ├── catalog.yml
└── app2
will it read the conf/app1/parameters.yml
if passed --env="app1/local"
?datajoely
09/20/2021, 6:27 PMWaldrill
09/20/2021, 6:32 PMdatajoely
09/20/2021, 6:33 PMdatajoely
09/20/2021, 6:33 PMdatajoely
09/20/2021, 6:33 PMdatajoely
09/20/2021, 6:34 PM