You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository was archived by the owner on Jun 2, 2025. It is now read-only.
Copy file name to clipboardExpand all lines: README.md
+18-18Lines changed: 18 additions & 18 deletions
Original file line number
Diff line number
Diff line change
@@ -223,20 +223,20 @@ Calling `CreateModel` is required for creating a `Model` in SageMaker with this
223
223
SageMaker works with Docker images stored in [Amazon ECR](https://aws.amazon.com/ecr/). SageMaker team has prepared and uploaded the Docker images for SageMaker SparkML Serving Container in all regions where SageMaker operates.
224
224
Region to ECR container URL mapping can be found below. For a mapping from Region to Region Name, please see [here](https://docs.aws.amazon.com/AmazonRDS/latest/UserGuide/Concepts.RegionsAndAvailabilityZones.html).
@@ -263,7 +263,7 @@ First you need to ensure that have installed [Docker](https://www.docker.com/) o
263
263
In order to build the Docker image, you need to run a single Docker command:
264
264
265
265
```
266
-
docker build -t sagemaker-sparkml-serving:2.2 .
266
+
docker build -t sagemaker-sparkml-serving:2.4 .
267
267
```
268
268
269
269
#### Running the image locally
@@ -272,7 +272,7 @@ In order to run the Docker image, you need to run the following command. Please
272
272
The command will start the server on port 8080 and will also pass the schema as an environment variable to the Docker container. Alternatively, you can edit the `Dockerfile` to add `ENV SAGEMAKER_SPARKML_SCHEMA=schema` as well before building the Docker image.
273
273
274
274
```
275
-
docker run -p 8080:8080 -e SAGEMAKER_SPARKML_SCHEMA=schema -v /tmp/model:/opt/ml/model sagemaker-sparkml-serving:2.2 serve
275
+
docker run -p 8080:8080 -e SAGEMAKER_SPARKML_SCHEMA=schema -v /tmp/model:/opt/ml/model sagemaker-sparkml-serving:2.4 serve
The `Dockerfile` can be found at the root directory of the package. SageMaker SparkML Serving Container tags the Docker images using the Spark major version it is compatible with. Right now, it only supports Spark 2.2 and as a result, the Docker image is tagged with 2.2.
290
+
The `Dockerfile` can be found at the root directory of the package. SageMaker SparkML Serving Container tags the Docker images using the Spark major version it is compatible with. Right now, it only supports Spark 2.4 and as a result, the Docker image is tagged with 2.4.
291
291
292
292
In order to save the effort of building the Docker image everytime you are making a code change, you can also install [Maven](http://maven.apache.org/) and run `mvn clean package` at your project root to verify if the code is compiling fine and unit tests are running without any issue.
0 commit comments