roofAI 🏛️ meets ⚒️ SageMaker + more- 3

Next, validate Amazon_JumpStart_Semantic_Segmentation.ipynb

Using model ‘mxnet-semseg-fcn-resnet50-ade’ with … ‘*’. … pin to … ‘2.0.0’ for more stable results. … models may have different input/output signatures after a major version upgrade…

fixed.

ClientError: An error occurred (404) when calling the HeadObject operation: Not Found

trying mxnet-semseg-fcn-resnet101-ade, same error.

▶️ https://github.com/aws/amazon-sagemaker-examples/issues/4478

continuing,

check deploy_image_uri: 763104351884.dkr.ecr.ca-central-1.amazonaws.com/mxnet-inference:1.9.0-gpu-py38

aws ecr get-login-password \
  --region ca-central-1 | docker login \
  --username AWS \
  --password-stdin 763104351884.dkr.ecr.ca-central-1.amazonaws.com

docker pull 763104351884.dkr.ecr.ca-central-1.amazonaws.com/mxnet-inference:1.9.0-gpu-py38

✅

check deploy_source_uri: s3://jumpstart-cache-prod-ca-central-1/source-directory-tarballs/mxnet/inference/semseg/v1.2.1/sourcedir.tar.gz

@select
aws s3 cp \
  s3://jumpstart-cache-prod-ca-central-1/source-directory-tarballs/mxnet/inference/semseg/v1.2.1/sourcedir.tar.gz \
  .
tar -xvf sourcedir.tar.gz

✅

base_model_uri: s3://jumpstart-cache-prod-ca-central-1/mxnet-semseg/mxnet-semseg-fcn-resnet101-coco/artifacts/inference-prepack/v1.0.0/

aws s3 ls \
  s3://jumpstart-cache-prod-ca-central-1/mxnet-semseg/mxnet-semseg-fcn-resnet101-coco/artifacts/inference-prepack/v1.0.0/

✅

Repacking model artifact (s3://jumpstart-cache-prod-ca-central-1/mxnet-semseg/mxnet-semseg-fcn-resnet101-coco/artifacts/inference-prepack/v1.0.0/), script artifact (s3://jumpstart-cache-prod-ca-central-1/source-directory-tarballs/mxnet/inference/semseg/v1.2.1/sourcedir.tar.gz), and dependencies ([]) into single tar.gz file located at s3://sagemaker-ca-central-1–120429650996/jumpstart-example-infer-mxnet-semseg-fc-2024–01–28–00–02–10–748/model.tar.gz. This may take some time depending on model size…

✅

https://sagemaker-examples.readthedocs.io/en/latest/introduction_to_amazon_algorithms/semantic_segmentation_pascalvoc/semantic_segmentation_pascalvoc.html#Training

https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/semantic_segmentation_pascalvoc/semantic_segmentation_pascalvoc.ipynb

continues.