diff --git a/docs/modelserving/v1beta1/transformer/torchserve_image_transformer/README.md b/docs/modelserving/v1beta1/transformer/torchserve_image_transformer/README.md index 292efa74e..0162c3bee 100644 --- a/docs/modelserving/v1beta1/transformer/torchserve_image_transformer/README.md +++ b/docs/modelserving/v1beta1/transformer/torchserve_image_transformer/README.md @@ -70,7 +70,7 @@ class ImageTransformer(kserve.Model): return infer_request ``` -Please see the code example [here](https://github.com/kserve/kserve/tree/release-0.11/python/custom_transformer). +Please see the code example [here](https://github.com/kserve/kserve/tree/release-0.14/python/custom_transformer). ### Transformer Server Entrypoint For single model you just create a transformer object and register that to the model server. @@ -100,10 +100,10 @@ Kserve allows users to override the default logger configuration of serving runt You can follow the [logger configuration documentation](../../custom/custom_model/README.md#configuring-logger-for-serving-runtime) to configure the logger. ### Build Transformer docker image -Under `kserve/python` directory, build the transformer docker image using [Dockerfile](https://github.com/kserve/kserve/blob/release-0.11/python/custom_transformer.Dockerfile) +Under `kserve/python` directory, build the transformer docker image using [Dockerfile](https://github.com/kserve/kserve/blob/release-0.14/python/custom_transformer.Dockerfile) ```bash cd python -docker build -t $DOCKER_USER/image-transformer:latest -f transformer.Dockerfile . +docker build -t $DOCKER_USER/image-transformer:latest -f custom_transformer.Dockerfile . docker push {username}/image-transformer:latest ```