A data scientist deploys a custom pyfunc model to Mosaic AI Model Serving. During deployment, the endpoint shows a dependency error for a private internal package `data_transforms`. Which `log_model()` parameter should be used to bundle this private package with the model?

Question

Accepted Answer

C. `code_path=["path/to/data_transforms"]`. To bundle a private Python module or package that is not pip-installable with the model, `code_path` accepts a list of local file or directory paths. These are bundled into the MLflow model artifact and are available for import at serving time. `pip_requirements` and `extra_pip_requirements` are for pip-installable packages only.

A data scientist deploys a custom pyfunc model to Mosaic AI Model Serving. During deployment, the endpoint shows a dependency error for a private internal package `data_transforms`. Which `log_model()` parameter should be used to bundle this private package with the model?

Related Questions

Discussion

A data scientist deploys a custom pyfunc model to Mosaic AI Model Serving. During deployment, the endpoint shows a dependency error for a private internal package data_transforms. Which log_model() parameter should be used to bundle this private package with the model?

Related Questions

Discussion

A data scientist deploys a custom pyfunc model to Mosaic AI Model Serving. During deployment, the endpoint shows a dependency error for a private internal package `data_transforms`. Which `log_model()` parameter should be used to bundle this private package with the model?