aws / aws/sagemaker-python-sdk

MultiRecord Batch Transform is not working for a Multi container Model (Inference Pipeline)

Open
#4,042 0 comments 0 reactions 1 assignee Claimed by @nargokul View on GitHub
component: Inference APIs and Interfaces type: bug type: logging/error reporting
Dominant language
Python
Stars
2.3k
Forks
1.3k
Avg merge
1d 22h
Merged PRs (30d)
35

Description

**Describe the bug**
I'm trying to run a batch transform using a model with two containers, one is a scikit-learn model that has all the feature engineering steps an other that has the actual estimator using an xgboost model. The thing is that when I set the BatchStrategy param as 'SingleRecord' everything works just fine, but the problem comes when I set the parameter to 'MultiRecord', the batch transform failed and the logs are not clear about the error. I am using a csv file as input, with no headers of course.

**To reproduce**
Try to run a batch transform with the conditions explained above.

**Expected behavior**
The batch transform should work just fine without any errors.

Contributor guide

Open the contributing guide

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.