Giuseppe Angelo Porcelli
Giuseppe Angelo Porcelli
[Stability.AI inference DLC] - TorchServe override entrypoint fails to extract .tar model cache file
The TorchServe override entry point script uses the following instruction to determine if the `-z `option has to be passed to the `tar` command run as a Python subprocess, which...
SM Distribution 1.6.0 includes `cron`, but the service is not running. ``` sagemaker-user@default:~$ service cron status * cron is not running ```
This would allow working with AWS CodeCommit with temporary IAM credentials available in JupyterLab and CodeEditor when using the SageMaker Distribution in SageMaker Studio. ``` git config --global credential.helper '!aws...
Include the `s3fs` package. Reason: `s3fs` is used by Pandas when reading from Amazon S3.
### 🚀 The feature, motivation and pitch Set the inference server port based on `SAGEMAKER_BIND_TO_PORT` environment variable, to support multi-container endpoints and inference pipelines. If the container needs to listen...
### 🚀 The feature, motivation and pitch Add `com.amazonaws.sagemaker.capabilities.accept-bind-to-port` Docker label to support multi-container and inference pipelines when deploying to SageMaker endpoints. Reference: https://docs.aws.amazon.com/sagemaker/latest/dg/multi-container-troubleshooting.html ### Use case description _No response_...