HyperparameterTuner object doesn't provide a method of "create_model()" with best params for deployment?

[sam-zen-dev]

Please fill out the form below.

System Information

• Framework (e.g. TensorFlow) / Algorithm (e.g. KMeans): XGBoost, HyperparameterTuner
• Framework Version:
• Python Version: 3.6
• CPU or GPU:
• Python SDK Version:
• Are you using a custom image: No. using xgboost container.

Describe the problem

I am using the xgboost(sagemaker) to tune the model and it provided the best tuning job when all jobs are done. It seems sagemaker only provides the console button to create_model and deploy it.

what if i just want to use HyperparameterTuner object with best params tuned to create a model object first and combine it with other models into a pipeline model.

the only way i found out is to copy the best params(tuned) and re-fit the model again then use create_model() method to init the xgboost model for pipelinemodel.... which is very inconvenient...

I double checked the source code and didn't find related methods....any idea?

Minimal repro / logs

Please provide any logs and a bare minimum reproducible test case, as this will be helpful to diagnose the problem. If including tracebacks, please include the full traceback. Large logs and files should be attached.

• Exact command to reproduce:

[sam-zen-dev]

Oh sorry, i think i found the method:

1: get the best training job.
2: attach this training job in the xgb estimator object.

Done....