How to get the Endpoint for a model built in my machine

Sukalpo Banerjee 0
Viewed 34

I have built a model in my machine and want to deploy the model (catboost classifier) in the amazon sagemaker. Not able to get the endpoint of the model. Looking for a code/ process to get the end point. I tried with deploy function, but it is not giving the endpoint, the message is deploy function don't exist for catboostclassifier()

endpoint catboost amazon-sagemaker deployment python
1 Answers

You are looking for “bring your own model” in Sagemaker.

Here is an example with Xgboost : https://github.com/aws/amazon-sagemaker-examples/blob/main/advanced_functionality/xgboost_bring_your_own_model/xgboost_bring_your_own_model.ipynb

Check the section after “upload pertained model to s3”.

Another blogpost with similar information - again with Xgboost though :( https://medium.com/geekculture/84af8989d065

  1. Upload the trained model object in tar.gz format.
  2. Use image_uris.retrieve with catboost-classification-model to create a container for catboost
  3. create the end point. This notebook has the detailed steps.

Also this post discusses similar issue : cannot load locally pretrained model that I deployed via sagemaker notebook

Gaurav Chawla0
Related