Reorganize custom container folder (#1746)
* Delete existing examples
* Readd new examples
* Delete old workflows
* Re-add workflows
* Reformat minimal multimodel
* Add model version
* Add model version
* Edit base path
* Edit path
* Edit base path
* Error fixing
* Add half plus two integrated
* Add runner for half-plus-two-integrated
* Delete deploy r workflow
* Add r and test
* Add R readme/test
* Skip script validation
* Add textgen
* Add detached docker
* Revert "Add detached docker"
This reverts commit f2450516e5.
* Detached docker
* Detached docker
* Longer sleep
This commit is contained in:
Alex Wallace
GitHub
Родитель
027b82e58b
Коммит
37324eaff8
|
|
@ -1,4 +1,4 @@
|
|||
name: cli-scripts-deploy-torchserve
|
||||
name: cli-scripts-deploy-custom-container-minimal-multimodel
|
||||
on:
|
||||
workflow_dispatch:
|
||||
schedule:
|
||||
|
|
@ -8,8 +8,8 @@ on:
|
|||
- main
|
||||
- sdk-preview
|
||||
paths:
|
||||
- cli/deploy-torchserve.sh
|
||||
- .github/workflows/cli-scripts-deploy-torchserve.yml
|
||||
- cli/deploy-custom-container-minimal-multimodel.sh
|
||||
- .github/workflows/cli-scripts-deploy-custom-container-minimal-multimodel.yml
|
||||
- cli/setup.sh
|
||||
jobs:
|
||||
build:
|
||||
|
|
@ -26,5 +26,5 @@ jobs:
|
|||
working-directory: cli
|
||||
continue-on-error: true
|
||||
- name: test script script
|
||||
run: set -e; bash -x deploy-torchserve.sh
|
||||
run: set -e; bash -x deploy-custom-container-minimal-multimodel.sh
|
||||
working-directory: cli
|
||||
|
|
@ -1,4 +1,4 @@
|
|||
name: cli-scripts-deploy-r
|
||||
name: cli-scripts-deploy-custom-container-minimal-single-model
|
||||
on:
|
||||
workflow_dispatch:
|
||||
schedule:
|
||||
|
|
@ -8,8 +8,8 @@ on:
|
|||
- main
|
||||
- sdk-preview
|
||||
paths:
|
||||
- cli/deploy-r.sh
|
||||
- .github/workflows/cli-scripts-deploy-r.yml
|
||||
- cli/deploy-custom-container-minimal-single-model.sh
|
||||
- .github/workflows/cli-scripts-deploy-custom-container-minimal-single-model.yml
|
||||
- cli/setup.sh
|
||||
jobs:
|
||||
build:
|
||||
|
|
@ -26,5 +26,5 @@ jobs:
|
|||
working-directory: cli
|
||||
continue-on-error: true
|
||||
- name: test script script
|
||||
run: set -e; az ml model archive -n plumber -v 1 || true ; bash -x deploy-r.sh
|
||||
run: set -e; bash -x deploy-custom-container-minimal-single-model.sh
|
||||
working-directory: cli
|
||||
|
|
@ -1,4 +1,4 @@
|
|||
name: cli-scripts-deploy-custom-container-mlflow
|
||||
name: cli-scripts-deploy-custom-container-mlflow-multideployment-scikit
|
||||
on:
|
||||
workflow_dispatch:
|
||||
schedule:
|
||||
|
|
@ -6,9 +6,10 @@ on:
|
|||
pull_request:
|
||||
branches:
|
||||
- main
|
||||
- sdk-preview
|
||||
paths:
|
||||
- cli/deploy-custom-container-mlflow.sh
|
||||
- .github/workflows/cli-scripts-deploy-custom-container-mlflow.yml
|
||||
- cli/deploy-custom-container-mlflow-multideployment-scikit.sh
|
||||
- .github/workflows/cli-scripts-deploy-custom-container-mlflow-multideployment-scikit.yml
|
||||
- cli/setup.sh
|
||||
jobs:
|
||||
build:
|
||||
|
|
@ -24,8 +25,6 @@ jobs:
|
|||
run: bash setup.sh
|
||||
working-directory: cli
|
||||
continue-on-error: true
|
||||
- name: scripts installs
|
||||
run: sudo apt-get upgrade -y && sudo apt-get install uuid-runtime jq -y
|
||||
- name: test script script
|
||||
run: set -e; bash -x deploy-custom-container-mlflow.sh
|
||||
run: set -e; bash -x deploy-custom-container-mlflow-multideployment-scikit.sh
|
||||
working-directory: cli
|
||||
30
.github/workflows/cli-scripts-deploy-custom-container-r-multimodel-plumber.yml
поставляемый
Normal file
30
.github/workflows/cli-scripts-deploy-custom-container-r-multimodel-plumber.yml
поставляемый
Normal file
|
|
@ -0,0 +1,30 @@
|
|||
name: cli-scripts-deploy-custom-container-r-multimodel-plumber
|
||||
on:
|
||||
workflow_dispatch:
|
||||
schedule:
|
||||
- cron: "0 0/4 * * *"
|
||||
pull_request:
|
||||
branches:
|
||||
- main
|
||||
- sdk-preview
|
||||
paths:
|
||||
- cli/deploy-custom-container-r-multimodel-plumber.sh
|
||||
- .github/workflows/cli-scripts-deploy-custom-container-r-multimodel-plumber.yml
|
||||
- cli/setup.sh
|
||||
jobs:
|
||||
build:
|
||||
runs-on: ubuntu-latest
|
||||
steps:
|
||||
- name: check out repo
|
||||
uses: actions/checkout@v2
|
||||
- name: azure login
|
||||
uses: azure/login@v1
|
||||
with:
|
||||
creds: ${{secrets.AZ_CREDS}}
|
||||
- name: setup
|
||||
run: bash setup.sh
|
||||
working-directory: cli
|
||||
continue-on-error: true
|
||||
- name: test script script
|
||||
run: set -e; bash -x deploy-custom-container-r-multimodel-plumber.sh
|
||||
working-directory: cli
|
||||
30
.github/workflows/cli-scripts-deploy-custom-container-tfserving-half-plus-two-integrated.yml
поставляемый
Normal file
30
.github/workflows/cli-scripts-deploy-custom-container-tfserving-half-plus-two-integrated.yml
поставляемый
Normal file
|
|
@ -0,0 +1,30 @@
|
|||
name: cli-scripts-deploy-custom-container-tfserving-half-plus-two-integrated
|
||||
on:
|
||||
workflow_dispatch:
|
||||
schedule:
|
||||
- cron: "0 0/4 * * *"
|
||||
pull_request:
|
||||
branches:
|
||||
- main
|
||||
- sdk-preview
|
||||
paths:
|
||||
- cli/deploy-custom-container-tfserving-half-plus-two-integrated.sh
|
||||
- .github/workflows/cli-scripts-deploy-custom-container-tfserving-half-plus-two-integrated.yml
|
||||
- cli/setup.sh
|
||||
jobs:
|
||||
build:
|
||||
runs-on: ubuntu-latest
|
||||
steps:
|
||||
- name: check out repo
|
||||
uses: actions/checkout@v2
|
||||
- name: azure login
|
||||
uses: azure/login@v1
|
||||
with:
|
||||
creds: ${{secrets.AZ_CREDS}}
|
||||
- name: setup
|
||||
run: bash setup.sh
|
||||
working-directory: cli
|
||||
continue-on-error: true
|
||||
- name: test script script
|
||||
run: set -e; bash -x deploy-custom-container-tfserving-half-plus-two-integrated.sh
|
||||
working-directory: cli
|
||||
30
.github/workflows/cli-scripts-deploy-custom-container-tfserving-half-plus-two.yml
поставляемый
Normal file
30
.github/workflows/cli-scripts-deploy-custom-container-tfserving-half-plus-two.yml
поставляемый
Normal file
|
|
@ -0,0 +1,30 @@
|
|||
name: cli-scripts-deploy-custom-container-tfserving-half-plus-two
|
||||
on:
|
||||
workflow_dispatch:
|
||||
schedule:
|
||||
- cron: "0 0/4 * * *"
|
||||
pull_request:
|
||||
branches:
|
||||
- main
|
||||
- sdk-preview
|
||||
paths:
|
||||
- cli/deploy-custom-container-tfserving-half-plus-two.sh
|
||||
- .github/workflows/cli-scripts-deploy-custom-container-tfserving-half-plus-two.yml
|
||||
- cli/setup.sh
|
||||
jobs:
|
||||
build:
|
||||
runs-on: ubuntu-latest
|
||||
steps:
|
||||
- name: check out repo
|
||||
uses: actions/checkout@v2
|
||||
- name: azure login
|
||||
uses: azure/login@v1
|
||||
with:
|
||||
creds: ${{secrets.AZ_CREDS}}
|
||||
- name: setup
|
||||
run: bash setup.sh
|
||||
working-directory: cli
|
||||
continue-on-error: true
|
||||
- name: test script script
|
||||
run: set -e; bash -x deploy-custom-container-tfserving-half-plus-two.sh
|
||||
working-directory: cli
|
||||
|
|
@ -1,4 +1,4 @@
|
|||
name: cli-scripts-deploy-tfserving
|
||||
name: cli-scripts-deploy-custom-container-torchserve-densenet
|
||||
on:
|
||||
workflow_dispatch:
|
||||
schedule:
|
||||
|
|
@ -8,8 +8,8 @@ on:
|
|||
- main
|
||||
- sdk-preview
|
||||
paths:
|
||||
- cli/deploy-tfserving.sh
|
||||
- .github/workflows/cli-scripts-deploy-tfserving.yml
|
||||
- cli/deploy-custom-container-torchserve-densenet.sh
|
||||
- .github/workflows/cli-scripts-deploy-custom-container-torchserve-densenet.yml
|
||||
- cli/setup.sh
|
||||
jobs:
|
||||
build:
|
||||
|
|
@ -26,5 +26,5 @@ jobs:
|
|||
working-directory: cli
|
||||
continue-on-error: true
|
||||
- name: test script script
|
||||
run: set -e; bash -x deploy-tfserving.sh
|
||||
run: set -e; bash -x deploy-custom-container-torchserve-densenet.sh
|
||||
working-directory: cli
|
||||
30
.github/workflows/cli-scripts-deploy-custom-container-torchserve-huggingface-textgen.yml
поставляемый
Normal file
30
.github/workflows/cli-scripts-deploy-custom-container-torchserve-huggingface-textgen.yml
поставляемый
Normal file
|
|
@ -0,0 +1,30 @@
|
|||
name: cli-scripts-deploy-custom-container-torchserve-huggingface-textgen
|
||||
on:
|
||||
workflow_dispatch:
|
||||
schedule:
|
||||
- cron: "0 0/4 * * *"
|
||||
pull_request:
|
||||
branches:
|
||||
- main
|
||||
- sdk-preview
|
||||
paths:
|
||||
- cli/deploy-custom-container-torchserve-huggingface-textgen.sh
|
||||
- .github/workflows/cli-scripts-deploy-custom-container-torchserve-huggingface-textgen.yml
|
||||
- cli/setup.sh
|
||||
jobs:
|
||||
build:
|
||||
runs-on: ubuntu-latest
|
||||
steps:
|
||||
- name: check out repo
|
||||
uses: actions/checkout@v2
|
||||
- name: azure login
|
||||
uses: azure/login@v1
|
||||
with:
|
||||
creds: ${{secrets.AZ_CREDS}}
|
||||
- name: setup
|
||||
run: bash setup.sh
|
||||
working-directory: cli
|
||||
continue-on-error: true
|
||||
- name: test script script
|
||||
run: set -e; bash -x deploy-custom-container-torchserve-huggingface-textgen.sh
|
||||
working-directory: cli
|
||||
|
|
@ -1,4 +1,4 @@
|
|||
name: cli-scripts-deploy-custom-container-triton
|
||||
name: cli-scripts-deploy-custom-container-triton-single-model
|
||||
on:
|
||||
workflow_dispatch:
|
||||
schedule:
|
||||
|
|
@ -6,9 +6,10 @@ on:
|
|||
pull_request:
|
||||
branches:
|
||||
- main
|
||||
- sdk-preview
|
||||
paths:
|
||||
- cli/deploy-custom-container-triton.sh
|
||||
- .github/workflows/cli-scripts-deploy-custom-container-triton.yml
|
||||
- cli/deploy-custom-container-triton-single-model.sh
|
||||
- .github/workflows/cli-scripts-deploy-custom-container-triton-single-model.yml
|
||||
- cli/setup.sh
|
||||
jobs:
|
||||
build:
|
||||
|
|
@ -24,8 +25,6 @@ jobs:
|
|||
run: bash setup.sh
|
||||
working-directory: cli
|
||||
continue-on-error: true
|
||||
- name: scripts installs
|
||||
run: sudo apt-get upgrade -y && sudo apt-get install uuid-runtime jq -y
|
||||
- name: test script script
|
||||
run: set -e; bash -x deploy-custom-container-triton.sh
|
||||
run: set -e; bash -x deploy-custom-container-triton-single-model.sh
|
||||
working-directory: cli
|
||||
|
|
@ -1,31 +0,0 @@
|
|||
name: cli-scripts-deploy-minimal-inference-custom-container
|
||||
on:
|
||||
workflow_dispatch:
|
||||
schedule:
|
||||
- cron: "0 0/4 * * *"
|
||||
pull_request:
|
||||
branches:
|
||||
- main
|
||||
paths:
|
||||
- cli/deploy-minimal-inference-custom-container.sh
|
||||
- .github/workflows/cli-scripts-deploy-minimal-inference-custom-container.yml
|
||||
- cli/setup.sh
|
||||
jobs:
|
||||
build:
|
||||
runs-on: ubuntu-latest
|
||||
steps:
|
||||
- name: check out repo
|
||||
uses: actions/checkout@v2
|
||||
- name: azure login
|
||||
uses: azure/login@v1
|
||||
with:
|
||||
creds: ${{secrets.AZ_CREDS}}
|
||||
- name: setup
|
||||
run: bash setup.sh
|
||||
working-directory: cli
|
||||
continue-on-error: true
|
||||
- name: scripts installs
|
||||
run: sudo apt-get upgrade -y && sudo apt-get install uuid-runtime jq -y
|
||||
- name: test script script
|
||||
run: set -e; bash -x deploy-minimal-inference-custom-container.sh
|
||||
working-directory: cli
|
||||
|
|
@ -38,15 +38,20 @@ path|status|
|
|||
-|-
|
||||
[batch-score-rest.sh](batch-score-rest.sh)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-scripts-batch-score-rest.yml)
|
||||
[batch-score.sh](batch-score.sh)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-scripts-batch-score.yml)
|
||||
[deploy-custom-container-mlflow.sh](deploy-custom-container-mlflow.sh)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-scripts-deploy-custom-container-mlflow.yml)
|
||||
[deploy-custom-container-multimodel-minimal.sh](deploy-custom-container-multimodel-minimal.sh)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-scripts-deploy-custom-container-multimodel-minimal.yml)
|
||||
[deploy-custom-container-triton.sh](deploy-custom-container-triton.sh)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-scripts-deploy-custom-container-triton.yml)
|
||||
[deploy-custom-container-minimal-multimodel.sh](deploy-custom-container-minimal-multimodel.sh)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-scripts-deploy-custom-container-minimal-multimodel.yml)
|
||||
[deploy-custom-container-minimal-single-model.sh](deploy-custom-container-minimal-single-model.sh)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-scripts-deploy-custom-container-minimal-single-model.yml)
|
||||
[deploy-custom-container-mlflow-multideployment-scikit.sh](deploy-custom-container-mlflow-multideployment-scikit.sh)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-scripts-deploy-custom-container-mlflow-multideployment-scikit.yml)
|
||||
[deploy-custom-container-r-multimodel-plumber.sh](deploy-custom-container-r-multimodel-plumber.sh)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-scripts-deploy-custom-container-r-multimodel-plumber.yml)
|
||||
[deploy-custom-container-tfserving-half-plus-two-integrated.sh](deploy-custom-container-tfserving-half-plus-two-integrated.sh)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-scripts-deploy-custom-container-tfserving-half-plus-two-integrated.yml)
|
||||
[deploy-custom-container-tfserving-half-plus-two.sh](deploy-custom-container-tfserving-half-plus-two.sh)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-scripts-deploy-custom-container-tfserving-half-plus-two.yml)
|
||||
[deploy-custom-container-torchserve-densenet.sh](deploy-custom-container-torchserve-densenet.sh)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-scripts-deploy-custom-container-torchserve-densenet.yml)
|
||||
[deploy-custom-container-torchserve-huggingface-textgen.sh](deploy-custom-container-torchserve-huggingface-textgen.sh)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-scripts-deploy-custom-container-torchserve-huggingface-textgen.yml)
|
||||
[deploy-custom-container-triton-single-model.sh](deploy-custom-container-triton-single-model.sh)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-scripts-deploy-custom-container-triton-single-model.yml)
|
||||
[deploy-local-endpoint.sh](deploy-local-endpoint.sh)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-scripts-deploy-local-endpoint.yml)
|
||||
[deploy-managed-online-endpoint-access-resource-sai.sh](deploy-managed-online-endpoint-access-resource-sai.sh)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-scripts-deploy-managed-online-endpoint-access-resource-sai.yml)
|
||||
[deploy-managed-online-endpoint-access-resource-uai.sh](deploy-managed-online-endpoint-access-resource-uai.sh)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-scripts-deploy-managed-online-endpoint-access-resource-uai.yml)
|
||||
[deploy-managed-online-endpoint-mlflow.sh](deploy-managed-online-endpoint-mlflow.sh)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-scripts-deploy-managed-online-endpoint-mlflow.yml)
|
||||
[deploy-managed-online-endpoint.sh](deploy-managed-online-endpoint.sh)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-scripts-deploy-managed-online-endpoint.yml)
|
||||
[deploy-minimal-inference-custom-container.sh](deploy-minimal-inference-custom-container.sh)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-scripts-deploy-minimal-inference-custom-container.yml)
|
||||
[deploy-mlcompute-create-with-user-identity.sh](deploy-mlcompute-create-with-user-identity.sh)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-scripts-deploy-mlcompute-create-with-user-identity.yml)
|
||||
[deploy-mlcompute-create_with-system-identity.sh](deploy-mlcompute-create_with-system-identity.sh)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-scripts-deploy-mlcompute-create_with-system-identity.yml)
|
||||
[deploy-mlcompute-update-to-system-identity.sh](deploy-mlcompute-update-to-system-identity.sh)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-scripts-deploy-mlcompute-update-to-system-identity.yml)
|
||||
|
|
@ -54,12 +59,9 @@ path|status|
|
|||
[deploy-moe-autoscale.sh](deploy-moe-autoscale.sh)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-scripts-deploy-moe-autoscale.yml)
|
||||
[deploy-moe-vnet-mlflow.sh](deploy-moe-vnet-mlflow.sh)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-scripts-deploy-moe-vnet-mlflow.yml)
|
||||
[deploy-moe-vnet.sh](deploy-moe-vnet.sh)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-scripts-deploy-moe-vnet.yml)
|
||||
[deploy-r.sh](deploy-r.sh)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-scripts-deploy-r.yml)
|
||||
[deploy-rest.sh](deploy-rest.sh)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-scripts-deploy-rest.yml)
|
||||
[deploy-safe-rollout-kubernetes-online-endpoints.sh](deploy-safe-rollout-kubernetes-online-endpoints.sh)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-scripts-deploy-safe-rollout-kubernetes-online-endpoints.yml)
|
||||
[deploy-safe-rollout-online-endpoints.sh](deploy-safe-rollout-online-endpoints.sh)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-scripts-deploy-safe-rollout-online-endpoints.yml)
|
||||
[deploy-tfserving.sh](deploy-tfserving.sh)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-scripts-deploy-tfserving.yml)
|
||||
[deploy-torchserve.sh](deploy-torchserve.sh)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-scripts-deploy-torchserve.yml)
|
||||
[deploy-triton-managed-online-endpoint.sh](deploy-triton-managed-online-endpoint.sh)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-scripts-deploy-triton-managed-online-endpoint.yml)
|
||||
[misc.sh](misc.sh)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-scripts-misc.yml)
|
||||
[mlflow-uri.sh](mlflow-uri.sh)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-scripts-mlflow-uri.yml)
|
||||
|
|
@ -72,6 +74,7 @@ path|status|
|
|||
path|status|description
|
||||
-|-|-
|
||||
[jobs/basics/hello-automl/hello-automl-job-basic.yml](jobs/basics/hello-automl/hello-automl-job-basic.yml)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-jobs-basics-hello-automl-hello-automl-job-basic.yml)|A Classification job using bank marketing
|
||||
[jobs/pipelines-with-components/nyc_taxi_data_regression/single-job-pipeline.yml](jobs/pipelines-with-components/nyc_taxi_data_regression/single-job-pipeline.yml)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-jobs-pipelines-with-components-nyc_taxi_data_regression-single-job-pipeline.yml)|Single job pipeline to train regression model based on nyc taxi dataset
|
||||
[jobs/single-step/dask/nyctaxi/job.yml](jobs/single-step/dask/nyctaxi/job.yml)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-jobs-single-step-dask-nyctaxi-job.yml)|This sample shows how to run a distributed DASK job on AzureML. The 24GB NYC Taxi dataset is read in CSV format by a 4 node DASK cluster, processed and then written as job output in parquet format.
|
||||
[jobs/single-step/gpu_perf/gpu_perf_job.yml](jobs/single-step/gpu_perf/gpu_perf_job.yml)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-jobs-single-step-gpu_perf-gpu_perf_job.yml)|Runs NCCL-tests on gpu nodes.
|
||||
[jobs/single-step/julia/iris/job.yml](jobs/single-step/julia/iris/job.yml)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-jobs-single-step-julia-iris-job.yml)|Train a Flux model on the Iris dataset using the Julia programming language.
|
||||
|
|
@ -164,6 +167,7 @@ path|status|description
|
|||
[jobs/pipelines-with-components/basics/6c_r_iris/pipeline.yml](jobs/pipelines-with-components/basics/6c_r_iris/pipeline.yml)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-jobs-pipelines-with-components-basics-6c_r_iris-pipeline.yml)|Train an R model on the Iris dataset.
|
||||
[jobs/pipelines-with-components/image_classification_with_densenet/pipeline.yml](jobs/pipelines-with-components/image_classification_with_densenet/pipeline.yml)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-jobs-pipelines-with-components-image_classification_with_densenet-pipeline.yml)|Train densenet for image classification
|
||||
[jobs/pipelines-with-components/nyc_taxi_data_regression/pipeline.yml](jobs/pipelines-with-components/nyc_taxi_data_regression/pipeline.yml)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-jobs-pipelines-with-components-nyc_taxi_data_regression-pipeline.yml)|Train regression model based on nyc taxi dataset
|
||||
[jobs/pipelines-with-components/nyc_taxi_data_regression/single-job-pipeline.yml](jobs/pipelines-with-components/nyc_taxi_data_regression/single-job-pipeline.yml)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-jobs-pipelines-with-components-nyc_taxi_data_regression-single-job-pipeline.yml)|Single job pipeline to train regression model based on nyc taxi dataset
|
||||
[jobs/pipelines-with-components/pipeline_with_hyperparameter_sweep/pipeline.yml](jobs/pipelines-with-components/pipeline_with_hyperparameter_sweep/pipeline.yml)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-jobs-pipelines-with-components-pipeline_with_hyperparameter_sweep-pipeline.yml)|Tune hyperparameters using TF component
|
||||
[jobs/pipelines-with-components/rai_pipeline_adult_analyse/pipeline.yml](jobs/pipelines-with-components/rai_pipeline_adult_analyse/pipeline.yml)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-jobs-pipelines-with-components-rai_pipeline_adult_analyse-pipeline.yml)|Sample rai pipeline
|
||||
[jobs/pipelines-with-components/shakespear_sample_and_word_count_using_spark/pipeline.yml](jobs/pipelines-with-components/shakespear_sample_and_word_count_using_spark/pipeline.yml)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-jobs-pipelines-with-components-shakespear_sample_and_word_count_using_spark-pipeline.yml)|'submit a pipeline with spark job'
|
||||
|
|
@ -198,6 +202,8 @@ path|status|description
|
|||
[resources/compute/cluster-minimal.yml](resources/compute/cluster-minimal.yml)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-resources-compute-cluster-minimal.yml)|*no description*
|
||||
[resources/compute/cluster-ssh-password.yml](resources/compute/cluster-ssh-password.yml)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-resources-compute-cluster-ssh-password.yml)|*no description*
|
||||
[resources/compute/cluster-system-identity.yml](resources/compute/cluster-system-identity.yml)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-resources-compute-cluster-system-identity.yml)|*no description*
|
||||
[resources/registry/registry-storage-options.yml](resources/registry/registry-storage-options.yml)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-resources-registry-registry-storage-options.yml)|Registry with additional configuration for storage accounts
|
||||
[resources/registry/registry.yml](resources/registry/registry.yml)|[](https://github.com/Azure/azureml-examples/actions/workflows/cli-resources-registry-registry.yml)|Basic registry with one primary region and to additional regions
|
||||
|
||||
**Assets** ([assets](assets))
|
||||
|
||||
|
|
|
|||
|
|
@ -1,7 +1,5 @@
|
|||
set -e
|
||||
|
||||
pip install scikit-learn pandas joblib
|
||||
|
||||
# <set_variables>
|
||||
export ENDPOINT_NAME="<ENDPOINT_NAME>"
|
||||
export ACR_NAME="<CONTAINER_REGISTRY_NAME>"
|
||||
|
|
@ -10,30 +8,28 @@ export ACR_NAME="<CONTAINER_REGISTRY_NAME>"
|
|||
export ENDPOINT_NAME=endpt-moe-`echo $RANDOM`
|
||||
export ACR_NAME=$(az ml workspace show --query container_registry -o tsv | cut -d'/' -f9-)
|
||||
|
||||
# <setup_build_directory>
|
||||
export BASE_PATH=endpoints/online/custom-container/multimodel-minimal
|
||||
mkdir -p $BASE_PATH/{code,build,test-data}
|
||||
cp "$(dirname $BASE_PATH)/{multimodel-minimal*} $BASE_PATH/"
|
||||
mv $BASE_PATH/multmodel-minimal.dockerfile $BASE_PATH/build/
|
||||
mv $BASE_PATH/multmodel-minimal-score.py $BASE_PATH/code/
|
||||
# </setup_build_directory>
|
||||
export BASE_PATH=endpoints/online/custom-container/minimal/multimodel
|
||||
export ASSET_PATH=endpoints/online/custom-container/minimal/multimodel/models
|
||||
|
||||
cd $BASE_PATH
|
||||
|
||||
# <generate_models_and_test_data>
|
||||
python multimodel-minimal-train.py
|
||||
# </generate_models_and_test_data>
|
||||
|
||||
cd -
|
||||
|
||||
# <build_image>
|
||||
az acr build -t azureml-examples/minimal-multimodel:1 -r $ACR_NAME -f $BASE_PATH/build/multmodel-minimal.dockerfile $BASE_PATH/build
|
||||
# </build_image>
|
||||
# Helper function to change parameters in yaml files
|
||||
change_vars() {
|
||||
for FILE in "$@"; do
|
||||
TMP="${FILE}_"
|
||||
cp $FILE $TMP
|
||||
readarray -t VARS < <(cat $TMP | grep -oP '{{.*?}}' | sed -e 's/[}{]//g');
|
||||
for VAR in "${VARS[@]}"; do
|
||||
sed -i "s/{{${VAR}}}/${!VAR}/g" $TMP
|
||||
done
|
||||
done
|
||||
}
|
||||
|
||||
# <create_endpoint>
|
||||
az online-endpoint create -n $ENDPOINT_NAME -f $BASE_PATH/multimodel-minimal-endpoint.yml
|
||||
change_vars $BASE_PATH/minimal-multimodel-endpoint.yml
|
||||
az ml online-endpoint create -n $ENDPOINT_NAME -f $BASE_PATH/minimal-multimodel-endpoint.yml_
|
||||
# </create_endpoint>
|
||||
|
||||
rm $BASE_PATH/*.yml_
|
||||
|
||||
# Check if endpoint was successful
|
||||
endpoint_status=`az ml online-endpoint show --name $ENDPOINT_NAME --query "provisioning_state" -o tsv`
|
||||
echo $endpoint_status
|
||||
|
|
@ -45,12 +41,19 @@ else
|
|||
exit 1
|
||||
fi
|
||||
|
||||
# <build_image>
|
||||
az acr build -t azureml-examples/minimal-multimodel:1 -r $ACR_NAME -f $BASE_PATH/minimal-multimodel.dockerfile $BASE_PATH
|
||||
# </build_image>
|
||||
|
||||
# <create_deployment>
|
||||
az online-endpoint create -e $ENDPOINT_NAME -f $BASE_PATH/multimodel-minimal-deployment.yml
|
||||
change_vars $BASE_PATH/minimal-multimodel-deployment.yml
|
||||
az ml online-deployment create -e $ENDPOINT_NAME -f $BASE_PATH/minimal-multimodel-deployment.yml_ --all-traffic
|
||||
az ml online-deployment update -e $ENDPOINT_NAME -f $BASE_PATH/minimal-multimodel-deployment.yml_
|
||||
|
||||
# </create_deployment>
|
||||
|
||||
# Check if deployment was successful
|
||||
deploy_status=`az ml online-deployment show --name blue --endpoint $ENDPOINT_NAME --query "provisioning_state" -o tsv`
|
||||
deploy_status=`az ml online-deployment show --name minimal-multimodel --endpoint $ENDPOINT_NAME --query "provisioning_state" -o tsv`
|
||||
echo $deploy_status
|
||||
if [[ $deploy_status == "Succeeded" ]]
|
||||
then
|
||||
|
|
@ -70,8 +73,8 @@ SCORING_URL=$(az ml online-endpoint show -n $ENDPOINT_NAME --query scoring_uri -
|
|||
echo "Scoring url is $SCORING_URL"
|
||||
|
||||
# <test_online_endpoints>
|
||||
curl -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" -d @"$BASE_PATH/test-data/test-data-iris.json" $SCORING_URL
|
||||
curl -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" -d @"$BASE_PATH/test-data/test-data-diabetes.json" $SCORING_URL
|
||||
curl -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" -d @"$BASE_PATH/test-data/iris-test-data.json" $SCORING_URL
|
||||
curl -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" -d @"$BASE_PATH/test-data/diabetes-test-data.json" $SCORING_URL
|
||||
# </test_online_endpoints>
|
||||
|
||||
# <delete_online_endpoint>
|
||||
|
|
@ -0,0 +1,75 @@
|
|||
#!/bin/bash
|
||||
|
||||
set -e
|
||||
|
||||
# <set_variables>
|
||||
export ENDPOINT_NAME="<ENDPOINT_NAME>"
|
||||
export ACR_NAME="<CONTAINER_REGISTRY_NAME>"
|
||||
# </set_variables>
|
||||
|
||||
export ENDPOINT_NAME=endpt-moe-`echo $RANDOM`
|
||||
export ACR_NAME=$(az ml workspace show --query container_registry -o tsv | cut -d'/' -f9-)
|
||||
|
||||
export BASE_PATH=endpoints/online/custom-container/minimal/single-model
|
||||
export ASSET_PATH=endpoints/online/model-1
|
||||
|
||||
# Helper function to change parameters in yaml files
|
||||
change_vars() {
|
||||
for FILE in "$@"; do
|
||||
TMP="${FILE}_"
|
||||
cp $FILE $TMP
|
||||
readarray -t VARS < <(cat $TMP | grep -oP '{{.*?}}' | sed -e 's/[}{]//g');
|
||||
for VAR in "${VARS[@]}"; do
|
||||
sed -i "s/{{${VAR}}}/${!VAR}/g" $TMP
|
||||
done
|
||||
done
|
||||
}
|
||||
|
||||
# <create_endpoint>
|
||||
change_vars $BASE_PATH/minimal-single-model-endpoint.yml
|
||||
az ml online-endpoint create -f $BASE_PATH/minimal-single-model-endpoint.yml_
|
||||
# </create_endpoint>
|
||||
|
||||
rm $BASE_PATH/minimal-single-model-endpoint.yml_
|
||||
|
||||
# Get key and url
|
||||
echo "Getting access key and scoring URL..."
|
||||
KEY=$(az ml online-endpoint get-credentials -n $ENDPOINT_NAME --query primaryKey -o tsv)
|
||||
SCORING_URL=$(az ml online-endpoint show -n $ENDPOINT_NAME --query scoring_uri -o tsv)
|
||||
echo "Scoring url is $SCORING_URL"
|
||||
|
||||
# <build_image_pip_in_dockerfile>
|
||||
az acr build -f $BASE_PATH/pip-in-dockerfile/minimal-single-model-pip-in-dockerfile.dockerfile -t azureml-examples/minimal-single-model-pip-in-dockerfile:1 -r $ACR_NAME $ASSET_PATH
|
||||
# </build_with_pip_in_dockerfile>
|
||||
|
||||
# <create_deployment_pip_in_dockerfile>
|
||||
DEPLOYMENT_YML=$BASE_PATH/pip-in-dockerfile/minimal-single-model-pip-in-dockerfile-deployment.yml
|
||||
change_vars $DEPLOYMENT_YML
|
||||
az ml online-deployment create --endpoint-name $ENDPOINT_NAME -f "${DEPLOYMENT_YML}_" --all-traffic
|
||||
# </create_deployment_pip_in_dockerfile>
|
||||
|
||||
# <test_deployment_pip_in_dockerfile>
|
||||
curl -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" -d @$ASSET_PATH/sample-request.json $SCORING_URL
|
||||
# </test_deployment_pip_in_dockerfile>
|
||||
|
||||
rm $BASE_PATH/pip-in-dockerfile/*.yml_
|
||||
|
||||
# <build_image_conda_in_dockerfile>
|
||||
az acr build -f $BASE_PATH/conda-in-dockerfile/minimal-single-model-conda-in-dockerfile.dockerfile -t azureml-examples/minimal-single-model-conda-in-dockerfile:1 -r $ACR_NAME $ASSET_PATH
|
||||
# </build_with_conda_in_dockerfile>
|
||||
|
||||
# <create_deployment_conda_in_dockerfile>
|
||||
DEPLOYMENT_YML=$BASE_PATH/conda-in-dockerfile/minimal-single-model-conda-in-dockerfile-deployment.yml
|
||||
change_vars $DEPLOYMENT_YML
|
||||
az ml online-deployment create --endpoint-name $ENDPOINT_NAME -f "${DEPLOYMENT_YML}_" --all-traffic
|
||||
# </create_deployment_conda_in_dockerfile>
|
||||
|
||||
# <test_deployment_conda_in_dockerfile>
|
||||
curl -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" -d @$ASSET_PATH/sample-request.json $SCORING_URL
|
||||
# </test_deployment_conda_in_dockerfile>
|
||||
|
||||
rm $BASE_PATH/conda-in-dockerfile/*.yml_
|
||||
|
||||
# <delete_online_endpoint>
|
||||
az ml online-endpoint delete -y -n $ENDPOINT_NAME
|
||||
# </delete_online_endpoint>
|
||||
|
|
@ -12,7 +12,7 @@ export ACR_NAME=$(az ml workspace show --query container_registry -o tsv | cut -
|
|||
|
||||
# Create subdir "mlflow_context" and set BASE_PATH to it
|
||||
# <initialize_build_context>
|
||||
export PARENT_PATH=endpoints/online/custom-container
|
||||
export PARENT_PATH=endpoints/online/custom-container/mlflow/multideployment-scikit/
|
||||
export BASE_PATH="$PARENT_PATH/mlflow_context"
|
||||
export ASSET_PATH=endpoints/online/mlflow
|
||||
rm -rf $BASE_PATH && mkdir $BASE_PATH
|
||||
|
|
@ -101,6 +101,7 @@ az ml online-endpoint invoke -n $ENDPOINT_NAME --deployment-name sklearn-diabete
|
|||
az ml online-endpoint invoke -n $ENDPOINT_NAME --deployment-name lightgbm-iris --request-file "$BASE_PATH/sample-request-lightgbm.json"
|
||||
# </test_online_endpoints_with_invoke>
|
||||
|
||||
# <get_endpoint_details>
|
||||
# Get key
|
||||
echo "Getting access key..."
|
||||
KEY=$(az ml online-endpoint get-credentials -n $ENDPOINT_NAME --query primaryKey -o tsv)
|
||||
|
|
@ -109,6 +110,7 @@ KEY=$(az ml online-endpoint get-credentials -n $ENDPOINT_NAME --query primaryKey
|
|||
echo "Getting scoring url..."
|
||||
SCORING_URL=$(az ml online-endpoint show -n $ENDPOINT_NAME --query scoring_uri -o tsv)
|
||||
echo "Scoring url is $SCORING_URL"
|
||||
# </get_endpoint_details>
|
||||
|
||||
# <test_online_endpoints_with_curl>
|
||||
curl -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" -H "azureml-model-deployment: sklearn-diabetes" -d @"$BASE_PATH/sample-request-sklearn.json" $SCORING_URL
|
||||
|
|
@ -118,5 +120,3 @@ curl -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" -H "azu
|
|||
# <delete_online_endpoint>
|
||||
az ml online-endpoint delete -y -n $ENDPOINT_NAME
|
||||
# </delete_online_endpoint>
|
||||
|
||||
#rm -rf $BASE_PATH
|
||||
|
|
@ -1,4 +1,8 @@
|
|||
BASE_PATH=endpoints/online/custom-container/r
|
||||
#!/bin/bash
|
||||
|
||||
set -e
|
||||
|
||||
BASE_PATH=endpoints/online/custom-container/r/multimodel-plumber
|
||||
DEPLOYMENT_NAME=r-deployment
|
||||
|
||||
# <set_endpoint_name>
|
||||
|
|
@ -7,7 +11,7 @@ export ENDPOINT_NAME="<YOUR_ENDPOINT_NAME>"
|
|||
|
||||
export ENDPOINT_NAME=endpt-r-`echo $RANDOM`
|
||||
|
||||
# Get name of workspace ACR, build image
|
||||
# <get_workspace_details>
|
||||
WORKSPACE=$(az config get --query "defaults[?name == 'workspace'].value" -o tsv)
|
||||
ACR_NAME=$(az ml workspace show -n $WORKSPACE --query container_registry -o tsv | cut -d'/' -f9-)
|
||||
|
||||
|
|
@ -16,47 +20,48 @@ then
|
|||
echo "Getting ACR name failed, exiting"
|
||||
exit 1
|
||||
fi
|
||||
# </get_workspace_details>
|
||||
|
||||
# <build_image>
|
||||
az acr login -n $ACR_NAME
|
||||
IMAGE_TAG=${ACR_NAME}.azurecr.io/r_server
|
||||
az acr build $BASE_PATH -f $BASE_PATH/r.dockerfile -t $IMAGE_TAG -r $ACR_NAME
|
||||
# </build_image>
|
||||
|
||||
# Clean up utility
|
||||
cleanup(){
|
||||
sed -i 's/'$ACR_NAME'/{{acr_name}}/' $BASE_PATH/r-deployment.yml
|
||||
sed -i 's/'$ACR_NAME'/{{ACR_NAME}}/' $BASE_PATH/r-deployment.yml
|
||||
az ml online-endpoint delete -n $ENDPOINT_NAME -y
|
||||
az ml model archive -n plumber -v 1
|
||||
}
|
||||
|
||||
# Run image locally for testing
|
||||
# <run_locally>
|
||||
docker run --rm -d -p 8000:8000 -v "$PWD/$BASE_PATH/scripts:/var/azureml-app" \
|
||||
-v "$PWD/$BASE_PATH/models":"/var/azureml-models/models" \
|
||||
-e AZUREML_MODEL_DIR=/var/azureml-models \
|
||||
-e AML_APP_ROOT=/var/azureml-app \
|
||||
-e AZUREML_ENTRY_SCRIPT=plumber.R \
|
||||
--name="r_server" $IMAGE_TAG
|
||||
|
||||
sleep 10
|
||||
# </run_locally>
|
||||
|
||||
# <test_locally>
|
||||
# Check liveness, readiness, scoring locally
|
||||
curl "http://localhost:8000/live"
|
||||
curl "http://localhost:8000/ready"
|
||||
curl -d @$BASE_PATH/sample_request.json -H 'Content-Type: application/json' http://localhost:8000/score
|
||||
|
||||
docker stop r_server
|
||||
# </test_locally>
|
||||
|
||||
# Fill in placeholders in deployment YAML
|
||||
sed -i 's/{{acr_name}}/'$ACR_NAME'/' $BASE_PATH/r-deployment.yml
|
||||
sed -i 's/{{ACR_NAME}}/'$ACR_NAME'/' $BASE_PATH/r-deployment.yml
|
||||
|
||||
# <create_endpoint>
|
||||
az ml online-endpoint create --name $ENDPOINT_NAME -f endpoints/online/custom-container/r/r-endpoint.yml
|
||||
az ml online-endpoint create --name $ENDPOINT_NAME -f $BASE_PATH/r-endpoint.yml
|
||||
# </create_endpoint>
|
||||
|
||||
# <create_deployment>
|
||||
az ml online-deployment create --name r-deployment --endpoint $ENDPOINT_NAME -f endpoints/online/custom-container/r/r-deployment.yml --all-traffic
|
||||
# </create_deployment>
|
||||
|
||||
# check if create was successful
|
||||
# <check_endpoint_status>
|
||||
endpoint_status=`az ml online-endpoint show --name $ENDPOINT_NAME --query "provisioning_state" -o tsv`
|
||||
echo $endpoint_status
|
||||
if [[ $endpoint_status == "Succeeded" ]]
|
||||
|
|
@ -66,7 +71,13 @@ else
|
|||
echo "Endpoint creation failed"
|
||||
exit 1
|
||||
fi
|
||||
# </check_endpoint_status>
|
||||
|
||||
# <create_deployment>
|
||||
az ml online-deployment create --name r-deployment --endpoint $ENDPOINT_NAME -f $BASE_PATH/r-deployment.yml --all-traffic --skip-script-validation
|
||||
# </create_deployment>
|
||||
|
||||
# Check if deployment was successful
|
||||
deploy_status=`az ml online-deployment show --name r-deployment --endpoint $ENDPOINT_NAME --query "provisioning_state" -o tsv`
|
||||
echo $deploy_status
|
||||
if [[ $deploy_status == "Succeeded" ]]
|
||||
|
|
@ -78,13 +89,14 @@ else
|
|||
exit 1
|
||||
fi
|
||||
|
||||
# Test remotely
|
||||
# <test_endpoint>
|
||||
echo "Testing endpoint"
|
||||
for model in {a..c}
|
||||
do
|
||||
echo "Model $model tested successfully, response was $RESPONSE."
|
||||
RESPONSE=$(az ml online-endpoint invoke -n $ENDPOINT_NAME --request-file <(jq --arg m "$model" '.model=$m' $BASE_PATH/sample_request.json))
|
||||
echo "Model $model tested successfully, response was $RESPONSE."
|
||||
done
|
||||
# </test_endpoint>
|
||||
|
||||
# Clean up
|
||||
cleanup
|
||||
|
|
@ -0,0 +1,56 @@
|
|||
#!/bin/bash
|
||||
|
||||
set -e
|
||||
|
||||
BASE_PATH=endpoints/online/custom-container/tfserving/half-plus-two-integrated
|
||||
ENDPOINT_NAME=tfsintegrated-`echo $RANDOM`
|
||||
ACR_NAME=$(az ml workspace show --query container_registry -o tsv | cut -d'/' -f9-)
|
||||
|
||||
# Helper function to parameterize YAML
|
||||
change_vars() {
|
||||
for FILE in "$@"; do
|
||||
TMP="${FILE}_"
|
||||
cp $FILE $TMP
|
||||
readarray -t VARS < <(cat $TMP | grep -oP '{{.*?}}' | sed -e 's/[}{]//g');
|
||||
for VAR in "${VARS[@]}"; do
|
||||
sed -i "s/{{${VAR}}}/${!VAR}/g" $TMP
|
||||
done
|
||||
done
|
||||
}
|
||||
|
||||
# <build_image>
|
||||
az acr login -n $ACR_NAME
|
||||
az acr build -f $BASE_PATH/half-plus-two-integrated.Dockerfile -t azureml-examples/tfsintegrated:1 -r $ACR_NAME $BASE_PATH
|
||||
# </build_image>
|
||||
|
||||
# <test_locally>
|
||||
docker run -p 8501:8501 --rm -d --name tfsintegrated -t "$ACR_NAME.azurecr.io/azureml-examples/tfsintegrated:1"
|
||||
sleep 5
|
||||
curl -d '{"inputs": [[1,1]]}' -H "Content-Type: application/json" localhost:8501/v1/models/hpt:predict
|
||||
docker stop tfsintegrated
|
||||
# </test_locally>
|
||||
|
||||
# <create_endpoint>
|
||||
az ml online-endpoint create -n $ENDPOINT_NAME
|
||||
# </create_endpoint>
|
||||
|
||||
# <create_deployment>
|
||||
change_vars $BASE_PATH/half-plus-two-integrated-deployment.yml
|
||||
az ml online-deployment create -f $BASE_PATH/half-plus-two-integrated-deployment.yml_ --all-traffic
|
||||
rm $BASE_PATH/half-plus-two-integrated-deployment.yml_
|
||||
# </create_deployment>
|
||||
|
||||
# <get_endpoint_details>
|
||||
KEY=$(az ml online-endpoint get-credentials -n $ENDPOINT_NAME --query primaryKey -o tsv)
|
||||
SCORING_URL=$(az ml online-endpoint show -n $ENDPOINT_NAME --query scoring_uri -o tsv)
|
||||
echo "Scoring url is $SCORING_URL"
|
||||
echo "Key is $KEY"
|
||||
# </get_endpoint_details>
|
||||
|
||||
# <test_deployment>
|
||||
curl -d @$BASE_PATH/sample-data.json -H "Content-Type: application/json" -H "Authorization: Bearer $KEY" $SCORING_URL
|
||||
# </test_deployment>
|
||||
|
||||
# <delete_online_endpoint>
|
||||
az ml online-endpoint delete -y -n $ENDPOINT_NAME
|
||||
# </delete_online_endpoint>
|
||||
|
|
@ -3,7 +3,7 @@
|
|||
set -e
|
||||
|
||||
# <initialize_variables>
|
||||
BASE_PATH=endpoints/online/custom-container
|
||||
BASE_PATH=endpoints/online/custom-container/tfserving/half-plus-two
|
||||
AML_MODEL_NAME=tfserving-mounted
|
||||
MODEL_NAME=half_plus_two
|
||||
MODEL_BASE_PATH=/var/azureml-app/azureml-models/$AML_MODEL_NAME/1
|
||||
|
|
@ -49,11 +49,14 @@ export ENDPOINT_NAME="<YOUR_ENDPOINT_NAME>"
|
|||
export ENDPOINT_NAME=endpt-tfserving-`echo $RANDOM`
|
||||
|
||||
# <create_endpoint>
|
||||
az ml online-endpoint create --name $ENDPOINT_NAME -f endpoints/online/custom-container/tfserving-endpoint.yml
|
||||
az ml online-endpoint create --name $ENDPOINT_NAME -f $BASE_PATH/tfserving-endpoint.yml
|
||||
# </create_endpoint>
|
||||
|
||||
MODEL_VERSION=$RANDOM
|
||||
sed -e "s/{{MODEL_VERSION}}/$MODEL_VERSION/g" -i $BASE_PATH/tfserving-deployment.yml
|
||||
|
||||
# <create_deployment>
|
||||
az ml online-deployment create --name tfserving-deployment --endpoint $ENDPOINT_NAME -f endpoints/online/custom-container/tfserving-deployment.yml --all-traffic
|
||||
az ml online-deployment create --name tfserving-deployment --endpoint $ENDPOINT_NAME -f $BASE_PATH/tfserving-deployment.yml --all-traffic
|
||||
# </create_deployment>
|
||||
|
||||
# <get_status>
|
||||
|
|
@ -2,18 +2,8 @@
|
|||
|
||||
set -e
|
||||
|
||||
BASE_PATH=endpoints/online/custom-container
|
||||
AML_MODEL_NAME=torchserve-densenet161
|
||||
echo $AML_MODEL_NAME
|
||||
AZUREML_MODEL_DIR=azureml-models/$AML_MODEL_NAME/1
|
||||
MODEL_BASE_PATH=/var/azureml-app/$AZUREML_MODEL_DIR
|
||||
BASE_PATH=endpoints/online/custom-container/torchserve/densenet
|
||||
ENDPOINT_NAME=endpt-torchserve-`echo $RANDOM`
|
||||
DEPLOYMENT_NAME=torchserve-deployment
|
||||
|
||||
# Download model and config file
|
||||
echo "Downling model and config file..."
|
||||
mkdir $BASE_PATH/torchserve
|
||||
wget --progress=dot:mega https://aka.ms/torchserve-densenet161 -O $BASE_PATH/torchserve/densenet161.mar
|
||||
|
||||
# Get name of workspace ACR, build image
|
||||
WORKSPACE=$(az config get --query "defaults[?name == 'workspace'].value" -o tsv)
|
||||
|
|
@ -24,41 +14,51 @@ if [[ $ACR_NAME == "" ]]; then
|
|||
exit 1
|
||||
fi
|
||||
|
||||
az acr login -n $ACR_NAME
|
||||
IMAGE_TAG=${ACR_NAME}.azurecr.io/torchserve:8080
|
||||
az acr build $BASE_PATH/ -f $BASE_PATH/torchserve.dockerfile -t $IMAGE_TAG -r $ACR_NAME
|
||||
cleanTestingFiles() {
|
||||
rm -r $BASE_PATH/torchserve
|
||||
rm $BASE_PATH/kitten_small.jpg
|
||||
rm $BASE_PATH/torchserve-deployment.yml_
|
||||
}
|
||||
|
||||
# Run image locally for testing
|
||||
# <download_model>
|
||||
echo "Downling model and config file..."
|
||||
mkdir $BASE_PATH/torchserve
|
||||
wget --progress=dot:mega https://aka.ms/torchserve-densenet161 -O $BASE_PATH/torchserve/densenet161.mar
|
||||
# </download_model>
|
||||
|
||||
# <build_image>
|
||||
az acr login -n $ACR_NAME
|
||||
IMAGE_TAG=${ACR_NAME}.azurecr.io/torchserve:1
|
||||
az acr build -f $BASE_PATH/torchserve.dockerfile -t $IMAGE_TAG -r $ACR_NAME $BASE_PATH
|
||||
# <build_image>
|
||||
|
||||
# <run_image_locally>
|
||||
docker run --rm -d -p 8080:8080 --name torchserve-test \
|
||||
-e MODEL_BASE_PATH=$MODEL_BASE_PATH \
|
||||
-v $PWD/$BASE_PATH/torchserve:$MODEL_BASE_PATH/torchserve $IMAGE_TAG
|
||||
-e AZUREML_MODEL_DIR=/var/azureml-app/azureml-models/ \
|
||||
-e TORCHSERVE_MODELS="densenet161=densenet161.mar" \
|
||||
-v $PWD/$BASE_PATH/torchserve:/var/azureml-app/azureml-models/torchserve $IMAGE_TAG
|
||||
# </run_image_locally>
|
||||
|
||||
sleep 10
|
||||
|
||||
cleanTestingFiles() {
|
||||
rm -r $BASE_PATH/torchserve
|
||||
rm kitten_small.jpg
|
||||
}
|
||||
|
||||
# Check Torchserve health
|
||||
# <test_locally>
|
||||
echo "Checking Torchserve health..."
|
||||
curl http://localhost:8080/ping
|
||||
|
||||
# Download test image
|
||||
echo "Downloading test image..."
|
||||
wget https://aka.ms/torchserve-test-image -O kitten_small.jpg
|
||||
wget https://aka.ms/torchserve-test-image -O $BASE_PATH/kitten_small.jpg
|
||||
|
||||
# Check scoring locally
|
||||
echo "Uploading testing image, the scoring is..."
|
||||
curl http://localhost:8080/predictions/densenet161 -T kitten_small.jpg
|
||||
curl http://localhost:8080/predictions/densenet161 -T $BASE_PATH/kitten_small.jpg
|
||||
|
||||
docker stop torchserve-test
|
||||
# </test_locally>
|
||||
|
||||
# Deploy model to online endpoint
|
||||
echo "Deploying the model to a managed online endpoint..."
|
||||
sed -i 's/{{acr_name}}/'$ACR_NAME'/' $BASE_PATH/$DEPLOYMENT_NAME.yml
|
||||
# <create_endpoint>
|
||||
az ml online-endpoint create --name $ENDPOINT_NAME -f $BASE_PATH/torchserve-endpoint.yml
|
||||
# </create_endpoint>
|
||||
|
||||
# <check_endpoint_status>
|
||||
ENDPOINT_STATUS=$(az ml online-endpoint show --name $ENDPOINT_NAME --query "provisioning_state" -o tsv)
|
||||
echo "Endpoint status is $ENDPOINT_STATUS"
|
||||
|
||||
|
|
@ -69,12 +69,16 @@ else
|
|||
az ml online-endpoint delete --name $ENDPOINT_NAME
|
||||
exit 1
|
||||
fi
|
||||
# </check_endpoint_status>
|
||||
|
||||
# Create deployment
|
||||
echo "Creating deployment..."
|
||||
az ml online-deployment create --name $DEPLOYMENT_NAME --endpoint $ENDPOINT_NAME --file $BASE_PATH/$DEPLOYMENT_NAME.yml --all-traffic
|
||||
# <create_deployment>
|
||||
cp $BASE_PATH/torchserve-deployment.yml $BASE_PATH/torchserve-deployment.yml_
|
||||
sed -e "s/{{ACR_NAME}}/$ACR_NAME/g" -i $BASE_PATH/torchserve-deployment.yml_
|
||||
az ml online-deployment create -e $ENDPOINT_NAME -f $BASE_PATH/torchserve-deployment.yml_ --all-traffic
|
||||
# </create_deployment>
|
||||
|
||||
deploy_status=$(az ml online-deployment show --name $DEPLOYMENT_NAME --endpoint $ENDPOINT_NAME --query "provisioning_state" -o tsv)
|
||||
# <check_deployment_status>
|
||||
deploy_status=$(az ml online-deployment show --name torchserve-deployment --endpoint $ENDPOINT_NAME --query "provisioning_state" -o tsv)
|
||||
echo $deploy_status
|
||||
if [[ $deploy_status == "Succeeded" ]]; then
|
||||
echo "Deployment completed successfully"
|
||||
|
|
@ -85,28 +89,27 @@ else
|
|||
az ml model archive -n $AML_MODEL_NAME --version 1
|
||||
exit 1
|
||||
fi
|
||||
# </check_deployment_status>
|
||||
|
||||
# Get accessToken
|
||||
# <get_endpoint_details>
|
||||
echo "Getting access token..."
|
||||
TOKEN=$(az ml online-endpoint get-credentials -n $ENDPOINT_NAME --query accessToken -o tsv)
|
||||
|
||||
# Get scoring url
|
||||
echo "Getting scoring url..."
|
||||
SCORING_URL=$(az ml online-endpoint show -n $ENDPOINT_NAME --query scoring_uri -o tsv)
|
||||
echo "Scoring url is $SCORING_URL"
|
||||
# </get_endpoint_details>
|
||||
|
||||
# Check scoring
|
||||
# <test_endpoint>
|
||||
echo "Uploading testing image, the scoring is..."
|
||||
curl -H "Authorization: {Bearer $TOKEN}" -T kitten_small.jpg $SCORING_URL
|
||||
curl -H "Authorization: {Bearer $TOKEN}" -T $BASE_PATH/kitten_small.jpg $SCORING_URL
|
||||
# </test_endpoint>
|
||||
|
||||
echo "Tested successfully, cleaning up"
|
||||
|
||||
cleanTestingFiles
|
||||
|
||||
# Delete endpoint
|
||||
# <delete_endpoint>
|
||||
echo "Deleting endpoint..."
|
||||
az ml online-endpoint delete -n $ENDPOINT_NAME --yes
|
||||
# </delete_endpoint>
|
||||
|
||||
# Delete model
|
||||
echo "Deleting model..."
|
||||
az ml model archive -n $AML_MODEL_NAME --version 1
|
||||
|
|
@ -0,0 +1,123 @@
|
|||
#!/bin/bash
|
||||
|
||||
set -e
|
||||
|
||||
# <set_parameters>
|
||||
ENDPOINT_NAME=hf-tg-`echo $RANDOM`
|
||||
IMAGE_TAG=azureml-examples/huggingface-textgen:1
|
||||
|
||||
BASE_PATH=endpoints/online/custom-container/torchserve/huggingface-textgen
|
||||
SERVE_PATH=$BASE_PATH/serve/examples/Huggingface_Transformers
|
||||
ROOT_PATH=$PWD
|
||||
ACR_NAME=$(az ml workspace show --query container_registry -o tsv | cut -d'/' -f9-)
|
||||
# </set_parameters>
|
||||
|
||||
# <define_helper_functions>
|
||||
# Helper function to parameterize YAML
|
||||
change_vars() {
|
||||
for FILE in "$@"; do
|
||||
TMP="${FILE}_"
|
||||
cp $FILE $TMP
|
||||
readarray -t VARS < <(cat $TMP | grep -oP '{{.*?}}' | sed -e 's/[}{]//g');
|
||||
for VAR in "${VARS[@]}"; do
|
||||
sed -i "s#{{${VAR}}}#${!VAR}#g" $TMP
|
||||
done
|
||||
done
|
||||
}
|
||||
|
||||
cleanup () {
|
||||
sudo rm -rf $BASE_PATH/serve || true
|
||||
rm $BASE_PATH/ts-hf-tg-deployment.yml_
|
||||
az ml online-endpoint delete -y -n $ENDPOINT_NAME
|
||||
}
|
||||
# </define_helper_functions>
|
||||
|
||||
# <build_image>
|
||||
az acr login -n $ACR_NAME
|
||||
az acr build -t $IMAGE_TAG -f $BASE_PATH/ts-hf-tg.dockerfile -r $ACR_NAME $BASE_PATH
|
||||
# </build_image>
|
||||
|
||||
# <download_huggingface_assets>
|
||||
cd $BASE_PATH
|
||||
rm -rf serve
|
||||
git init serve
|
||||
cd serve
|
||||
git remote add -f origin https://github.com/pytorch/serve
|
||||
git config core.sparseCheckout true
|
||||
echo "examples/Huggingface_Transformers" >> .git/info/sparse-checkout
|
||||
git pull origin master
|
||||
cd $ROOT_PATH
|
||||
# </download_huggingface_assets>
|
||||
|
||||
# <generate_model>
|
||||
if [[ ! -f $SERVE_PATH/setup_config.json_ ]]; then
|
||||
cp $BASE_PATH/ts-hf-tg-setup_config.json $SERVE_PATH/setup_config.json_
|
||||
fi
|
||||
cp $BASE_PATH/ts-hf-tg-setup_config.json $SERVE_PATH/setup_config.json
|
||||
chmod -R o+w $SERVE_PATH
|
||||
cd $SERVE_PATH
|
||||
docker run --rm -v $PWD:/tmp/wd:z -w /tmp/wd -t "$ACR_NAME.azurecr.io/$IMAGE_TAG" "python Download_Transformer_models.py; \
|
||||
sed -i 's#\"max_length\": 50#\"max_length\": 300#g' ./Transformer_model/config.json; \
|
||||
torch-model-archiver --model-name Textgeneration --version 1.0 --serialized-file Transformer_model/pytorch_model.bin --handler ./Transformer_handler_generalized.py --extra-files Transformer_model/config.json,./setup_config.json; \
|
||||
exit;"
|
||||
cp setup_config.json_ setup_config.json
|
||||
rm setup_config.json_
|
||||
cd $ROOT_PATH
|
||||
# </generate_model>
|
||||
|
||||
# <test_locally>
|
||||
docker run --name huggingface-textgen --rm -d -p 8080:8080 -v "$PWD/$SERVE_PATH":/tmp/wd -e AZUREML_MODEL_DIR=/tmp/wd -e TORCHSERVE_MODELS="textgeneration=Textgeneration.mar" -t "$ACR_NAME.azurecr.io/$IMAGE_TAG"
|
||||
sleep 30
|
||||
curl -X POST http://127.0.0.1:8080/predictions/textgeneration -T "$SERVE_PATH/Text_gen_artifacts/sample_text.txt"
|
||||
docker stop huggingface-textgen
|
||||
# </test_locally>
|
||||
|
||||
# <create_endpoint>
|
||||
az ml online-endpoint create -n $ENDPOINT_NAME
|
||||
# </create_endpoint>
|
||||
|
||||
# <check_endpoint_status>
|
||||
endpoint_status=`az ml online-endpoint show --name $ENDPOINT_NAME --query "provisioning_state" -o tsv`
|
||||
echo $endpoint_status
|
||||
if [[ $endpoint_status == "Succeeded" ]]
|
||||
then
|
||||
echo "Endpoint created successfully"
|
||||
else
|
||||
echo "Endpoint creation failed"
|
||||
exit 1
|
||||
fi
|
||||
# </check_endpoint_status>
|
||||
|
||||
# <create_deployment>
|
||||
change_vars $BASE_PATH/ts-hf-tg-deployment.yml
|
||||
az ml online-deployment create -f $BASE_PATH/ts-hf-tg-deployment.yml_ --all-traffic
|
||||
# </create_deployment>
|
||||
|
||||
# <check_deployment_status>
|
||||
deploy_status=`az ml online-deployment show --endpoint-name $ENDPOINT_NAME --name textgeneration --query "provisioning_state" -o tsv`
|
||||
echo $deploy_status
|
||||
if [[ $deploy_status == "Succeeded" ]]
|
||||
then
|
||||
echo "Deployment completed successfully"
|
||||
else
|
||||
echo "Deployment failed"
|
||||
exit 1
|
||||
fi
|
||||
# </check_deployment_status>
|
||||
|
||||
# <get_endpoint_details>
|
||||
# Get key
|
||||
echo "Getting access key..."
|
||||
KEY=$(az ml online-endpoint get-credentials -n $ENDPOINT_NAME --query primaryKey -o tsv)
|
||||
|
||||
# Get scoring url
|
||||
echo "Getting scoring url..."
|
||||
SCORING_URL=$(az ml online-endpoint show -n $ENDPOINT_NAME --query scoring_uri -o tsv)
|
||||
echo "Scoring url is $SCORING_URL"
|
||||
# </get_endpoint_details>
|
||||
|
||||
# <test_endpoint>
|
||||
curl -X POST -H "Authorization: Bearer $KEY" -T "$SERVE_PATH/Text_gen_artifacts/sample_text.txt" $SCORING_URL
|
||||
# </test_endpoint>
|
||||
|
||||
cleanup
|
||||
|
|
@ -13,7 +13,7 @@ export ENDPOINT_NAME=endpt-moe-`echo $RANDOM`
|
|||
export ACR_NAME=$(az ml workspace show --query container_registry -o tsv | cut -d'/' -f9-)
|
||||
|
||||
# <set_base_path_and_copy_assets>
|
||||
export PARENT_PATH="endpoints/online/custom-container"
|
||||
export PARENT_PATH="endpoints/online/custom-container/triton/single-model"
|
||||
export ASSET_PATH="endpoints/online/triton/single-model"
|
||||
export BASE_PATH="$PARENT_PATH/triton_context"
|
||||
rm -rf $BASE_PATH && mkdir -p $BASE_PATH/models
|
||||
|
|
@ -68,5 +68,3 @@ python endpoints/online/triton/single-model/triton_densenet_scoring.py --base_ur
|
|||
# <delete_online_endpoint>
|
||||
az ml online-endpoint delete -y -n $ENDPOINT_NAME
|
||||
# </delete_online_endpoint>
|
||||
|
||||
#rm -rf $BASE_PATH
|
||||
|
|
@ -1,87 +0,0 @@
|
|||
#!/bin/bash
|
||||
|
||||
set -e
|
||||
|
||||
# <set_variables>
|
||||
export ENDPOINT_NAME="<ENDPOINT_NAME>"
|
||||
export ACR_NAME="<CONTAINER_REGISTRY_NAME>"
|
||||
# </set_variables>
|
||||
|
||||
export ENDPOINT_NAME=endpt-moe-`echo $RANDOM`
|
||||
export ACR_NAME=$(az ml workspace show --query container_registry -o tsv | cut -d'/' -f9-)
|
||||
|
||||
# <set_base_path_and_copy_model>
|
||||
export PARENT_PATH="endpoints/online/custom-container"
|
||||
export BASE_PATH="$PARENT_PATH/minimal_context"
|
||||
rm -rf $BASE_PATH && mkdir -p $BASE_PATH
|
||||
cp -r endpoints/online/model-1 $BASE_PATH
|
||||
cp $PARENT_PATH/minimal.dockerfile $BASE_PATH/Dockerfile
|
||||
cp $PARENT_PATH/minimal-deployment.yml $BASE_PATH/deployment.yaml
|
||||
cp $PARENT_PATH/minimal-endpoint.yml $BASE_PATH/endpoint.yaml
|
||||
sed -i "s/{{acr_name}}/$ACR_NAME/g;\
|
||||
s/{{endpoint_name}}/$ENDPOINT_NAME/g;" $BASE_PATH/deployment.yaml
|
||||
sed -i "s/{{endpoint_name}}/$ENDPOINT_NAME/g;" $BASE_PATH/endpoint.yaml
|
||||
# </set_base_path_and_copy_model>
|
||||
|
||||
# <build_image_locally>
|
||||
docker build -t azureml-examples/minimal-inf-cc $BASE_PATH
|
||||
# </build_image_locally>
|
||||
|
||||
# <run_image_locally>
|
||||
docker run -d -p 5001:5001 -v "$(pwd)/$BASE_PATH/model-1/onlinescoring":/var/azureml-app -v "$(pwd)/$BASE_PATH/model-1/model":/var/azureml-app/azureml-models/model -e AZUREML_APP_ROOT=/var/azureml-app -e AZUREML_ENTRY_SCRIPT=score.py -e AZUREML_MODEL_DIR=/var/azureml-app/azureml-models -t azureml-examples/minimal-inf-cc:latest
|
||||
# </run_image_locally>
|
||||
|
||||
sleep 10
|
||||
|
||||
# <test_local_image>
|
||||
curl -X POST -H "Content-Type: application/json" -d @$BASE_PATH/model-1/sample-request.json localhost:5001/score
|
||||
# </test_local_image>
|
||||
|
||||
# <login_to_acr>
|
||||
az acr login -n $ACR_NAME
|
||||
# </login_to_acr>
|
||||
|
||||
# We can either push the image to ACR
|
||||
# <tag_image>
|
||||
docker tag azureml-examples/minimal-inf-cc "$ACR_NAME.azurecr.io/azureml-examples/minimal-inf-cc:latest"
|
||||
# </tag_image>
|
||||
|
||||
# <push_to_acr>
|
||||
docker push "$ACR_NAME.azurecr.io/azureml-examples/minimal-inf-cc:latest"
|
||||
# </push_to_acr>
|
||||
|
||||
# Or build with ACR directly
|
||||
# <build_with_acr>
|
||||
az acr build -t azureml-examples/minimal-inf-cc:latest -r $ACR_NAME $BASE_PATH
|
||||
# </build_with_acr>
|
||||
|
||||
# <create_endpoint>
|
||||
az ml online-endpoint create -f $BASE_PATH/endpoint.yaml
|
||||
# </create_endpoint>
|
||||
|
||||
# <create_deployment>
|
||||
az ml online-deployment create --endpoint-name $ENDPOINT_NAME -f $BASE_PATH/deployment.yaml --all-traffic
|
||||
# </create_deployment>
|
||||
|
||||
# <test_online_endpoint_with_invoke>
|
||||
az ml online-endpoint invoke -n $ENDPOINT_NAME --request-file "$BASE_PATH/model-1/sample-request.json"
|
||||
# </test_online_endpoint_with_invoke>
|
||||
|
||||
# Get accessToken
|
||||
echo "Getting access token..."
|
||||
TOKEN=$(az ml online-endpoint get-credentials -n $ENDPOINT_NAME --query accessToken -o tsv)
|
||||
|
||||
# Get scoring url
|
||||
echo "Getting scoring url..."
|
||||
SCORING_URL=$(az ml online-endpoint show -n $ENDPOINT_NAME --query scoring_uri -o tsv)
|
||||
echo "Scoring url is $SCORING_URL"
|
||||
|
||||
# <test_online_endpoint_with_curl>
|
||||
curl -H "Authorization: {Bearer $TOKEN}" -H "Content-Type: application/json" -d @$BASE_PATH/model-1/sample-request.json $SCORING_URL
|
||||
# </test_online_endpoint_with_curl>
|
||||
|
||||
# <delete_online_endpoint>
|
||||
az ml online-endpoint delete -y -n $ENDPOINT_NAME
|
||||
# </delete_online_endpoint>
|
||||
|
||||
#rm -rf $BASE_PATH
|
||||
|
|
@ -1 +0,0 @@
|
|||
asset-file-contents
|
||||
|
|
@ -1,201 +0,0 @@
|
|||
Apache License
|
||||
Version 2.0, January 2004
|
||||
http://www.apache.org/licenses/
|
||||
|
||||
TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
|
||||
|
||||
1. Definitions.
|
||||
|
||||
"License" shall mean the terms and conditions for use, reproduction,
|
||||
and distribution as defined by Sections 1 through 9 of this document.
|
||||
|
||||
"Licensor" shall mean the copyright owner or entity authorized by
|
||||
the copyright owner that is granting the License.
|
||||
|
||||
"Legal Entity" shall mean the union of the acting entity and all
|
||||
other entities that control, are controlled by, or are under common
|
||||
control with that entity. For the purposes of this definition,
|
||||
"control" means (i) the power, direct or indirect, to cause the
|
||||
direction or management of such entity, whether by contract or
|
||||
otherwise, or (ii) ownership of fifty percent (50%) or more of the
|
||||
outstanding shares, or (iii) beneficial ownership of such entity.
|
||||
|
||||
"You" (or "Your") shall mean an individual or Legal Entity
|
||||
exercising permissions granted by this License.
|
||||
|
||||
"Source" form shall mean the preferred form for making modifications,
|
||||
including but not limited to software source code, documentation
|
||||
source, and configuration files.
|
||||
|
||||
"Object" form shall mean any form resulting from mechanical
|
||||
transformation or translation of a Source form, including but
|
||||
not limited to compiled object code, generated documentation,
|
||||
and conversions to other media types.
|
||||
|
||||
"Work" shall mean the work of authorship, whether in Source or
|
||||
Object form, made available under the License, as indicated by a
|
||||
copyright notice that is included in or attached to the work
|
||||
(an example is provided in the Appendix below).
|
||||
|
||||
"Derivative Works" shall mean any work, whether in Source or Object
|
||||
form, that is based on (or derived from) the Work and for which the
|
||||
editorial revisions, annotations, elaborations, or other modifications
|
||||
represent, as a whole, an original work of authorship. For the purposes
|
||||
of this License, Derivative Works shall not include works that remain
|
||||
separable from, or merely link (or bind by name) to the interfaces of,
|
||||
the Work and Derivative Works thereof.
|
||||
|
||||
"Contribution" shall mean any work of authorship, including
|
||||
the original version of the Work and any modifications or additions
|
||||
to that Work or Derivative Works thereof, that is intentionally
|
||||
submitted to Licensor for inclusion in the Work by the copyright owner
|
||||
or by an individual or Legal Entity authorized to submit on behalf of
|
||||
the copyright owner. For the purposes of this definition, "submitted"
|
||||
means any form of electronic, verbal, or written communication sent
|
||||
to the Licensor or its representatives, including but not limited to
|
||||
communication on electronic mailing lists, source code control systems,
|
||||
and issue tracking systems that are managed by, or on behalf of, the
|
||||
Licensor for the purpose of discussing and improving the Work, but
|
||||
excluding communication that is conspicuously marked or otherwise
|
||||
designated in writing by the copyright owner as "Not a Contribution."
|
||||
|
||||
"Contributor" shall mean Licensor and any individual or Legal Entity
|
||||
on behalf of whom a Contribution has been received by Licensor and
|
||||
subsequently incorporated within the Work.
|
||||
|
||||
2. Grant of Copyright License. Subject to the terms and conditions of
|
||||
this License, each Contributor hereby grants to You a perpetual,
|
||||
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
|
||||
copyright license to reproduce, prepare Derivative Works of,
|
||||
publicly display, publicly perform, sublicense, and distribute the
|
||||
Work and such Derivative Works in Source or Object form.
|
||||
|
||||
3. Grant of Patent License. Subject to the terms and conditions of
|
||||
this License, each Contributor hereby grants to You a perpetual,
|
||||
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
|
||||
(except as stated in this section) patent license to make, have made,
|
||||
use, offer to sell, sell, import, and otherwise transfer the Work,
|
||||
where such license applies only to those patent claims licensable
|
||||
by such Contributor that are necessarily infringed by their
|
||||
Contribution(s) alone or by combination of their Contribution(s)
|
||||
with the Work to which such Contribution(s) was submitted. If You
|
||||
institute patent litigation against any entity (including a
|
||||
cross-claim or counterclaim in a lawsuit) alleging that the Work
|
||||
or a Contribution incorporated within the Work constitutes direct
|
||||
or contributory patent infringement, then any patent licenses
|
||||
granted to You under this License for that Work shall terminate
|
||||
as of the date such litigation is filed.
|
||||
|
||||
4. Redistribution. You may reproduce and distribute copies of the
|
||||
Work or Derivative Works thereof in any medium, with or without
|
||||
modifications, and in Source or Object form, provided that You
|
||||
meet the following conditions:
|
||||
|
||||
(a) You must give any other recipients of the Work or
|
||||
Derivative Works a copy of this License; and
|
||||
|
||||
(b) You must cause any modified files to carry prominent notices
|
||||
stating that You changed the files; and
|
||||
|
||||
(c) You must retain, in the Source form of any Derivative Works
|
||||
that You distribute, all copyright, patent, trademark, and
|
||||
attribution notices from the Source form of the Work,
|
||||
excluding those notices that do not pertain to any part of
|
||||
the Derivative Works; and
|
||||
|
||||
(d) If the Work includes a "NOTICE" text file as part of its
|
||||
distribution, then any Derivative Works that You distribute must
|
||||
include a readable copy of the attribution notices contained
|
||||
within such NOTICE file, excluding those notices that do not
|
||||
pertain to any part of the Derivative Works, in at least one
|
||||
of the following places: within a NOTICE text file distributed
|
||||
as part of the Derivative Works; within the Source form or
|
||||
documentation, if provided along with the Derivative Works; or,
|
||||
within a display generated by the Derivative Works, if and
|
||||
wherever such third-party notices normally appear. The contents
|
||||
of the NOTICE file are for informational purposes only and
|
||||
do not modify the License. You may add Your own attribution
|
||||
notices within Derivative Works that You distribute, alongside
|
||||
or as an addendum to the NOTICE text from the Work, provided
|
||||
that such additional attribution notices cannot be construed
|
||||
as modifying the License.
|
||||
|
||||
You may add Your own copyright statement to Your modifications and
|
||||
may provide additional or different license terms and conditions
|
||||
for use, reproduction, or distribution of Your modifications, or
|
||||
for any such Derivative Works as a whole, provided Your use,
|
||||
reproduction, and distribution of the Work otherwise complies with
|
||||
the conditions stated in this License.
|
||||
|
||||
5. Submission of Contributions. Unless You explicitly state otherwise,
|
||||
any Contribution intentionally submitted for inclusion in the Work
|
||||
by You to the Licensor shall be under the terms and conditions of
|
||||
this License, without any additional terms or conditions.
|
||||
Notwithstanding the above, nothing herein shall supersede or modify
|
||||
the terms of any separate license agreement you may have executed
|
||||
with Licensor regarding such Contributions.
|
||||
|
||||
6. Trademarks. This License does not grant permission to use the trade
|
||||
names, trademarks, service marks, or product names of the Licensor,
|
||||
except as required for reasonable and customary use in describing the
|
||||
origin of the Work and reproducing the content of the NOTICE file.
|
||||
|
||||
7. Disclaimer of Warranty. Unless required by applicable law or
|
||||
agreed to in writing, Licensor provides the Work (and each
|
||||
Contributor provides its Contributions) on an "AS IS" BASIS,
|
||||
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
|
||||
implied, including, without limitation, any warranties or conditions
|
||||
of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
|
||||
PARTICULAR PURPOSE. You are solely responsible for determining the
|
||||
appropriateness of using or redistributing the Work and assume any
|
||||
risks associated with Your exercise of permissions under this License.
|
||||
|
||||
8. Limitation of Liability. In no event and under no legal theory,
|
||||
whether in tort (including negligence), contract, or otherwise,
|
||||
unless required by applicable law (such as deliberate and grossly
|
||||
negligent acts) or agreed to in writing, shall any Contributor be
|
||||
liable to You for damages, including any direct, indirect, special,
|
||||
incidental, or consequential damages of any character arising as a
|
||||
result of this License or out of the use or inability to use the
|
||||
Work (including but not limited to damages for loss of goodwill,
|
||||
work stoppage, computer failure or malfunction, or any and all
|
||||
other commercial damages or losses), even if such Contributor
|
||||
has been advised of the possibility of such damages.
|
||||
|
||||
9. Accepting Warranty or Additional Liability. While redistributing
|
||||
the Work or Derivative Works thereof, You may choose to offer,
|
||||
and charge a fee for, acceptance of support, warranty, indemnity,
|
||||
or other liability obligations and/or rights consistent with this
|
||||
License. However, in accepting such obligations, You may act only
|
||||
on Your own behalf and on Your sole responsibility, not on behalf
|
||||
of any other Contributor, and only if You agree to indemnify,
|
||||
defend, and hold each Contributor harmless for any liability
|
||||
incurred by, or claims asserted against, such Contributor by reason
|
||||
of your accepting any such warranty or additional liability.
|
||||
|
||||
END OF TERMS AND CONDITIONS
|
||||
|
||||
APPENDIX: How to apply the Apache License to your work.
|
||||
|
||||
To apply the Apache License to your work, attach the following
|
||||
boilerplate notice, with the fields enclosed by brackets "[]"
|
||||
replaced with your own identifying information. (Don't include
|
||||
the brackets!) The text should be enclosed in the appropriate
|
||||
comment syntax for the file format. We also recommend that a
|
||||
file or class name and description of purpose be included on the
|
||||
same "printed page" as the copyright notice for easier
|
||||
identification within third-party archives.
|
||||
|
||||
Copyright 2016, The TensorFlow Serving Authors.
|
||||
|
||||
Licensed under the Apache License, Version 2.0 (the "License");
|
||||
you may not use this file except in compliance with the License.
|
||||
You may obtain a copy of the License at
|
||||
|
||||
http://www.apache.org/licenses/LICENSE-2.0
|
||||
|
||||
Unless required by applicable law or agreed to in writing, software
|
||||
distributed under the License is distributed on an "AS IS" BASIS,
|
||||
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||
See the License for the specific language governing permissions and
|
||||
limitations under the License.
|
||||
Двоичный файл не отображается.
Двоичный файл не отображается.
Двоичный файл не отображается.
|
|
@ -0,0 +1,24 @@
|
|||
import joblib
|
||||
import os
|
||||
import pandas as pd
|
||||
from pathlib import Path
|
||||
import json
|
||||
|
||||
models = None
|
||||
|
||||
|
||||
def init():
|
||||
global models
|
||||
model_dir = Path(os.getenv("AZUREML_MODEL_DIR")) / "models"
|
||||
models = {m[:-4]: joblib.load(model_dir / m) for m in os.listdir(model_dir)}
|
||||
|
||||
|
||||
def run(data):
|
||||
data = json.loads(data)
|
||||
model = models[data["model"]]
|
||||
payload = pd.DataFrame(data["data"])
|
||||
try:
|
||||
ret = model.predict(payload)
|
||||
return pd.DataFrame(ret).to_json()
|
||||
except KeyError:
|
||||
raise KeyError("No such model")
|
||||
|
|
@ -0,0 +1,31 @@
|
|||
# Deploy multiple models to one deployment
|
||||
|
||||
This example shows how to deploy multiple models to a single deployment by extending the AzureML Inference Minimal CPU image.
|
||||
|
||||
It can be run end-to-end by executing the `deploy-custom-container-minimal-multimodel.sh` script in the CLI directory.
|
||||
|
||||
## Build the image
|
||||
|
||||
The image used for this example is defined in the `minimal-multimodel.dockerfile` and uses the pip method of adding additional requirements needed to score the model. To add requirements to an AzureML Inference base image using conda, see the `deploy-custom-container-minimal-single-model` example, which demonstrates both methods.
|
||||
|
||||
## Models
|
||||
|
||||
The two models used in this example are included in the `models` subdirectory. They are both sklearn models and need the same requirements. The two models are registered as a single "registered model" on Azure by including them in the same directory and passing that directory as the `path` of the model. In this example, the model is defined inline in the deployment.
|
||||
|
||||
## Score script
|
||||
|
||||
The score script for this example is located in the `code` directory.
|
||||
|
||||
At deployment time, both models are loaded into memory using the standard method of using the `AZUREML_MODEL_DIR` environment variable. Even though this is a custom container deployment, the `AZUREML_MODEL_DIR` is still present in the container. This happens any time a model is mounted to an endpoint container, even if the custom container does not use an Azure image as a base image.
|
||||
|
||||
The models are loaded into a dictionary keyed on their names. When a request is received, the desired model is retrieved from the JSON payload, and the relevant model is used to score the payload.
|
||||
|
||||
## Environment
|
||||
|
||||
The environment is defined in the deployment yaml and references the ACR url of the image. The ACR must be associated with the workspace (or have a user-assigned managed identity that enables ACRPull) in order to successfully deploy. Both models share the same environment.
|
||||
|
||||
The environment also contains an `inference_config` block that defines the `liveness`, `readiness`, and `scoring` routes by path and port. Because the images used in this examples are based on the AzureML Inference Minimal images, these values are the same as those in a non-BYOC deployment, however they must be included since we are now using a custom image.
|
||||
|
||||
### Python requirements
|
||||
|
||||
Both models are loaded in the same process by the same Python interpreter. Models loaded in a multi-model deployment using this technique must share the same Python version and - even if they do not have the same dependencies in a strict sense (i.e. a scikit model loaded alongside a torch model) - their dependiences must not conflict with each other and be able to be simultaneously imported.
|
||||
|
|
@ -0,0 +1,14 @@
|
|||
$schema: https://azuremlschemas.azureedge.net/latest/managedOnlineDeployment.schema.json
|
||||
name: minimal-multimodel
|
||||
endpoint_name: {{ENDPOINT_NAME}}
|
||||
model:
|
||||
name: minimal-multimodel
|
||||
path: models
|
||||
code_configuration:
|
||||
code: code
|
||||
scoring_script: minimal-multimodel-score.py
|
||||
environment:
|
||||
name: minimal-multimodel
|
||||
image: {{ACR_NAME}}.azurecr.io/azureml-examples/minimal-multimodel:1
|
||||
instance_type: Standard_DS2_v2
|
||||
instance_count: 1
|
||||
|
|
@ -1,3 +1,3 @@
|
|||
$schema: https://azuremlschemas.azureedge.net/latest/managedOnlineEndpoint.schema.json
|
||||
name: multimodel-minimal
|
||||
name: {{ENDPOINT_NAME}}
|
||||
auth_mode: key
|
||||
Двоичные данные
cli/endpoints/online/custom-container/minimal/multimodel/models/diabetes.sav
Normal file
Двоичные данные
cli/endpoints/online/custom-container/minimal/multimodel/models/diabetes.sav
Normal file
Двоичный файл не отображается.
Двоичные данные
cli/endpoints/online/custom-container/minimal/multimodel/models/iris.sav
Normal file
Двоичные данные
cli/endpoints/online/custom-container/minimal/multimodel/models/iris.sav
Normal file
Двоичный файл не отображается.
|
|
@ -0,0 +1 @@
|
|||
{"model": "diabetes", "data": {"age": {"0": -0.0018820165277906047, "1": 0.03081082953138418, "2": -0.0018820165277906047, "3": -0.027309785684926546, "4": -0.005514554978810025}, "sex": {"0": 0.05068011873981862, "1": 0.05068011873981862, "2": -0.044641636506989144, "3": -0.044641636506989144, "4": 0.05068011873981862}, "bmi": {"0": -0.02452875939178067, "1": 0.0563071461492793, "2": -0.06656343027313188, "3": 0.08864150836570328, "4": -0.015906262800734303}, "bp": {"0": 0.052858044296680055, "1": 0.07695800112762488, "2": 0.0012152796589411327, "3": -0.02519138732582271, "4": -0.06764173985804409}, "s1": {"0": 0.027326050202012293, "1": 0.04934129593323023, "2": -0.0029449126784123676, "3": 0.02182223876920781, "4": 0.04934129593323023}, "s2": {"0": 0.03000096875273476, "1": -0.012274073588852453, "2": 0.003070201038834776, "3": 0.04252690722431616, "4": 0.07916527725369175}, "s3": {"0": 0.030231910429713918, "1": -0.036037570043851025, "2": 0.01182372140927921, "3": -0.03235593223976409, "4": -0.028674294435677143}, "s4": {"0": -0.002592261998183278, "1": 0.07120997975363674, "2": -0.002592261998183278, "3": 0.03430885887772673, "4": 0.03430885887772673}, "s5": {"0": -0.021395309255276825, "1": 0.12005149644350945, "2": -0.020292321339471356, "3": 0.002861309289833047, "4": -0.018113692315690322}, "s6": {"0": 0.036201264733044136, "1": 0.09004865462589207, "2": -0.025930338989472702, "3": 0.07762233388138869, "4": 0.044485478562713045}}}
|
||||
|
|
@ -0,0 +1 @@
|
|||
{"model": "iris", "data": {"sepal length (cm)": {"0": 5.8, "1": 6.1, "2": 6.8, "3": 7.1, "4": 4.5}, "sepal width (cm)": {"0": 2.7, "1": 3.0, "2": 3.0, "3": 3.0, "4": 2.3}, "petal length (cm)": {"0": 5.1, "1": 4.9, "2": 5.5, "3": 5.9, "4": 1.3}, "petal width (cm)": {"0": 1.9, "1": 1.8, "2": 2.1, "3": 2.1, "4": 0.3}}}
|
||||
|
|
@ -0,0 +1,24 @@
|
|||
$schema: https://azuremlschemas.azureedge.net/latest/managedOnlineDeployment.schema.json
|
||||
name: conda-in-dockerfile
|
||||
endpoint_name: {{ENDPOINT_NAME}}
|
||||
model:
|
||||
name: minimal-cc-model
|
||||
path: ../../../../model-1/model
|
||||
code_configuration:
|
||||
code: ../../../../model-1/onlinescoring
|
||||
scoring_script: score.py
|
||||
environment:
|
||||
name: conda-in-dockerfile
|
||||
image: {{ACR_NAME}}.azurecr.io/azureml-examples/minimal-single-model-conda-in-dockerfile:1
|
||||
inference_config:
|
||||
liveness_route:
|
||||
path: /
|
||||
port: 5001
|
||||
readiness_route:
|
||||
path: /
|
||||
port: 5001
|
||||
scoring_route:
|
||||
path: /score
|
||||
port: 5001
|
||||
instance_type: Standard_DS2_v2
|
||||
instance_count: 1
|
||||
|
|
@ -1,7 +1,5 @@
|
|||
FROM mcr.microsoft.com/azureml/minimal-ubuntu20.04-py38-cpu-inference:latest
|
||||
|
||||
ARG MODEL_NAME=model-1
|
||||
|
||||
USER dockeruser
|
||||
|
||||
# Conda is already installed
|
||||
|
|
@ -12,7 +10,7 @@ ENV CONDA_ENV_DIR=/opt/miniconda/envs
|
|||
# or can be added alongside the image in the environment yaml definition for Azure to build
|
||||
|
||||
# Create a new conda environment and install the same version of the server
|
||||
COPY $MODEL_NAME/environment/conda.yml /tmp/conda.yaml
|
||||
COPY ./environment/conda.yml /tmp/conda.yaml
|
||||
RUN conda env create -n userenv -f /tmp/conda.yaml && \
|
||||
export SERVER_VERSION=$(pip show azureml-inference-server-http | grep Version | sed -e 's/.*: //') && \
|
||||
$CONDA_ENV_DIR/userenv/bin/pip install azureml-inference-server-http==$SERVER_VERSION
|
||||
|
|
@ -0,0 +1,31 @@
|
|||
# Deploy a minimal custom container
|
||||
This example shows how to deploy a simple custom container endpoint by extending the AzureML Inference Server minimal cpu image in two different ways - via a conda file or via pip requirements.
|
||||
|
||||
It can be run end-to-end by executing the `deploy-minimal-inference-custom-container.sh` script in the CLI directory.
|
||||
|
||||
## Build the image(s)
|
||||
The subdirectories `conda-in-dockerfile` and `pip-in-dockerfile` correspond to two different techniques for adding custom Python requirements at build time to the AzureML Inference base image. Requirements can also be added at deployment time via pip or conda through environment variables. Alternatively, environments can be defined on Azure as a combination of an inference base image and a conda file. In this case, Azure will manage the actual build.
|
||||
|
||||
### Using pip
|
||||
Adding pip requirements to a Dockerfile can be accomplished through a simple `pip install` statement as in `minimal-single-model-pip-in-dockerfile.dockerfile` or via a `requirements.txt` file copied in during the build process. No additional environment variables are needed.
|
||||
|
||||
### Conda
|
||||
|
||||
The file `minimal-single-model-conda-in-dockerfile.dockerfile` in the `conda-in-dockerfile` directory shows how to add a new conda environment to the minimal inferencence base image. The inference image uses a conda environment called `amlenv`, however in this example we create a new conda environment called `userenv`, reinstall the `azureml-inference-server-http` package and then modify three environment variables `AZUREML_CONDA_ENVIRONMENT`, `PATH` and `LD_LIBRARY_PATH` to make our conda environment take precedence. The conda file used for this image is located at `endpoints/online/model-1/environment`.
|
||||
|
||||
## Model
|
||||
|
||||
Both deployments uses the `model-1` sklearn model located at `endpoints/online/model-1/model` as well as the sample request located in the parent directory. It is defined inline in the deployment yaml and mounted at deployment time.
|
||||
|
||||
## Score script
|
||||
|
||||
The scoring script for this model is located at `endpoints/online/model-1/onlinescoring` and is identitcal to the scoring script for this model that would be used in a non-custom container scenario.
|
||||
|
||||
Even though this deployment uses a custom container, the `AZUREML_MODEL_DIR` environment variable is still present in the container, and so the model init process that uses it is the same as well.
|
||||
|
||||
## Environment
|
||||
|
||||
The environment is defined inline in the deployment yaml and references the ACR url of the image. The ACR must be associated with the workspace (or have a user-assigned managed identity that enables ACRPull) in order to successfully deploy.
|
||||
|
||||
The environment also contains an `inference_config` block that defines the `liveness`, `readiness`, and `scoring` routes by path and port. Because the images used in this examples are based on the AzureML Inference Minimal images, these values are the same as those in a non-BYOC deployment, however they must be included since we are now using a custom image.
|
||||
|
||||
|
|
@ -0,0 +1,3 @@
|
|||
$schema: https://azuremlschemas.azureedge.net/latest/managedOnlineEndpoint.schema.json
|
||||
name: {{ENDPOINT_NAME}}
|
||||
auth_mode: key
|
||||
|
|
@ -0,0 +1,16 @@
|
|||
# Deploy a minimal custom container
|
||||
This example shows how to deploy a simple custom container endpoint by extending the AzureML Inference Server minimal cpu image.
|
||||
|
||||
This example can be run end-to-end by executing the deploy-minimal-inference-custom-container.sh script in the CLI directory.
|
||||
|
||||
### Build the image
|
||||
Minimal.dockerfile in this directory shows how to add a new conda environment to the minimal inferencence base image. The inference image uses a conda environment called `amlenv`, however in this example we create a new conda environment called `userenv`, reinstall the `azureml-inference-server-http` package and then modify three environment variables `AZUREML_CONDA_ENVIRONMENT`, `PATH` and `LD_LIBRARY_PATH` to make our conda environment take precedence.
|
||||
|
||||
Dependencies can also be added to the image through pip using a `requirements.txt` file without the need to modify environment variables.
|
||||
|
||||
### Run the image locally
|
||||
Containers using the AzureML inference images can be run locally by mounting the code and model directories to their expected locations. The inference images find the directories via the environment variables `AML_APP_ROOT` for the code directory and `AZUREML_MODEL_DIR` for the model directory. The default locations for thse are /var/azureml-app and /var/azureml-app/azureml-model.
|
||||
|
||||
### Add the image to ACR
|
||||
The locally built image can either be added to ACR through pushing the locally-built image or delegating the build to ACR.
|
||||
|
||||
|
|
@ -1,15 +1,15 @@
|
|||
$schema: https://azuremlschemas.azureedge.net/latest/managedOnlineDeployment.schema.json
|
||||
name: blue
|
||||
endpoint_name: {{endpoint_name}}
|
||||
name: pip-in-dockerfile
|
||||
endpoint_name: {{ENDPOINT_NAME}}
|
||||
model:
|
||||
name: minimal-cc-model
|
||||
path: ./model-1/model
|
||||
path: ../../../../model-1/model
|
||||
code_configuration:
|
||||
code: ./model-1/onlinescoring
|
||||
code: ../../../../model-1/onlinescoring
|
||||
scoring_script: score.py
|
||||
environment:
|
||||
name: minimal-inf-cc-env
|
||||
image: {{acr_name}}.azurecr.io/azureml-examples/minimal-inf-cc:latest
|
||||
name: pip-in-dockerfile
|
||||
image: {{ACR_NAME}}.azurecr.io/azureml-examples/minimal-single-model-pip-in-dockerfile:1
|
||||
inference_config:
|
||||
liveness_route:
|
||||
path: /
|
||||
|
|
@ -0,0 +1,3 @@
|
|||
FROM mcr.microsoft.com/azureml/minimal-ubuntu20.04-py38-cpu-inference:latest
|
||||
|
||||
RUN pip install numpy pip scikit-learn scipy joblib
|
||||
|
|
@ -0,0 +1,24 @@
|
|||
# Deploy a custom container using MLFlow
|
||||
This example shows how to deploy two MLFlow models that each have different conda environments using a custom container. In this example, we deploy the MLFlow models to two separate deployments (as opposed to one single deployment) to accommodate their different requirements needs. For a non-MLFlow single-deployment multimodel example, see the `deploy-custom-containerminimal-multimodel` example.
|
||||
|
||||
This example can be run end-to-end by executing the deploy-custom-container-mlflow.sh script in the CLI directory.
|
||||
|
||||
## Build the images
|
||||
The custom container is defined in the `mlflow.dockerfile` in this directory. The container uses the AzureML minimal inference image as its base image and creates a new conda environment from a conda yaml. For more details on adding conda environments to the AzureML minimal inference images, see the `deploy-custom-container-minimal` example.
|
||||
|
||||
Each of the two models has different conda requirements, so we build the image twice. A build argument `MLFLOW_MODEL_NAME` enables us to choose which conda yaml we want to use for our newly-created `userenv` environment.
|
||||
|
||||
Images can be built with ACR using the `az acr build` command.
|
||||
|
||||
## Environment
|
||||
|
||||
The environment is defined inline in the deployment yaml and references the ACR url of the image. The ACR must be associated with the workspace (or have a user-assigned managed identity that enables ACRPull) in order to successfully deploy.
|
||||
|
||||
The environment also contains an `inference_config` block that defines the `liveness`, `readiness`, and `scoring` routes by path and port. Because the images used in this examples are based on the AzureML Inference Minimal images, these values are the same as those in a non-BYOC deployment, however they must be included since we are now using a custom image.
|
||||
|
||||
|
||||
## Deployment
|
||||
We create a separate deployment for each model behind the same endpoint using the two containers. Each model has a deployment yaml that registers a separate model and environment.
|
||||
|
||||
## Testing the model
|
||||
Traffic can be routed to a specific deployment using the header `azureml-model-deployment` or the `--deployment-name` flag of the `az ml online-endpoint invoke` CLI command.
|
||||
|
|
@ -1,14 +0,0 @@
|
|||
$schema: https://azuremlschemas.azureedge.net/latest/managedOnlineDeployment.schema.json
|
||||
name: multimodel-minimal
|
||||
endpoint_name: multimodel-minimal
|
||||
model:
|
||||
name: multimodel-minimal
|
||||
path: models
|
||||
code_configuration:
|
||||
code: code
|
||||
scoring_script: multimodel-minimal-score.py
|
||||
environment:
|
||||
name: multimodel-minimal
|
||||
image: ACR_NAME.azurecr.io/azureml-examples/minimal-multimodel:1
|
||||
instance_type: Standard_DS2_v2
|
||||
instance_count: 1
|
||||
|
|
@ -1,24 +0,0 @@
|
|||
import joblib
|
||||
import os
|
||||
import pandas as pd
|
||||
from pathlib import Path
|
||||
|
||||
|
||||
def init():
|
||||
global model_iris
|
||||
global model_diabetes
|
||||
|
||||
model_dir = Path(os.getenv("AZUREML_MODEL_DIR"))
|
||||
model_iris = joblib.load(model_dir / "iris.sav")
|
||||
model_diabetes = joblib.load(model_dir / "diabetes.sav")
|
||||
|
||||
|
||||
def run(data):
|
||||
model = data["model"]
|
||||
data = pd.DataFrame(data["data"])
|
||||
if model == "iris":
|
||||
return model_iris.predict(data)
|
||||
elif model == "diabetes":
|
||||
return model_diabetes.predict(data)
|
||||
else:
|
||||
return "No such model"
|
||||
|
|
@ -1,36 +0,0 @@
|
|||
import sklearn.datasets
|
||||
from sklearn import linear_model
|
||||
from sklearn.model_selection import train_test_split
|
||||
import joblib
|
||||
from pathlib import Path
|
||||
import json
|
||||
|
||||
test_size = 0.3
|
||||
model_dir = Path("models")
|
||||
test_data_dir = Path("test-data")
|
||||
|
||||
datasets = {
|
||||
"diabetes": sklearn.datasets.load_diabetes(as_frame=True).frame,
|
||||
"iris": sklearn.datasets.load_iris(as_frame=True).frame,
|
||||
}
|
||||
|
||||
|
||||
def train_model(name, dataset, test_size=0.3):
|
||||
ycol = "target"
|
||||
Xcol = [c for c in dataset.columns if c != ycol]
|
||||
X = dataset[Xcol]
|
||||
y = dataset[ycol]
|
||||
Xtrain, Xtest, ytrain, ytest = train_test_split(X, y, test_size=test_size)
|
||||
Xtest = Xtest.reset_index(drop=True)
|
||||
model = linear_model.Lasso(alpha=0.1)
|
||||
model.fit(Xtrain, ytrain)
|
||||
joblib.dump(model, filename=model_dir / f"{name}.sav")
|
||||
with open(test_data_dir / f"{name}-test-data.json", "w") as f:
|
||||
test_data = {}
|
||||
test_data["model"] = name
|
||||
test_data["data"] = Xtest.loc[0:4, :].to_dict()
|
||||
f.write(json.dumps(test_data))
|
||||
|
||||
|
||||
for name, dataset in datasets.items():
|
||||
train_model(name, dataset)
|
||||
|
|
@ -11,7 +11,7 @@ model:
|
|||
environment:
|
||||
#name: r-environment
|
||||
#version: 1
|
||||
image: {{acr_name}}.azurecr.io/r_server
|
||||
image: {{ACR_NAME}}.azurecr.io/r_server
|
||||
inference_config:
|
||||
liveness_route:
|
||||
port: 8000
|
||||
|
|
@ -0,0 +1,34 @@
|
|||
# Deploy an integrated inference container
|
||||
This example shows how to deploy a fully-integrated inference BYOC image using TFServing with the model integrated into the container itself and both `model` and `code` absent from the deployment yaml.
|
||||
|
||||
## How to deploy
|
||||
To deploy this example execute the deploy-custom-container-half-plus-two-integrated.sh script in the CLI directory.
|
||||
|
||||
## Testing
|
||||
The endpoint can be tested using the code at the end of the deployment script:
|
||||
```bash
|
||||
curl -d @sample-data.json -H "Content-Type: application/json" -H "Authorization: Bearer $KEY" $SCORING_URL
|
||||
```
|
||||
|
||||
The inputs are a list of tensors of dimension 2.
|
||||
```json
|
||||
{
|
||||
"inputs" : [
|
||||
[[1,2,3,4]],
|
||||
[[0,1,1,1]]
|
||||
]
|
||||
}
|
||||
```
|
||||
|
||||
## Model
|
||||
This model is a simple [Half Plus Two](https://www.tensorflow.org/tfx/serving/docker) model which returns `2X+0.5` for input tensors of dimension 2.
|
||||
|
||||
```python
|
||||
class HPT(tf.Module):
|
||||
@tf.function(input_signature=[tf.TensorSpec(shape=[None,None], dtype=tf.float32)])
|
||||
def __call__(self, x):
|
||||
return tf.math.add(tf.math.multiply(x, 0.5),2)
|
||||
```
|
||||
|
||||
## Image
|
||||
This is a BYOC TFServing image with the model integrated into the container itself.
|
||||
|
|
@ -0,0 +1,17 @@
|
|||
$schema: https://azuremlschemas.azureedge.net/latest/managedOnlineDeployment.schema.json
|
||||
name: integratedbyoc
|
||||
endpoint_name: {{ENDPOINT_NAME}}
|
||||
environment:
|
||||
image: {{ACR_NAME}}.azurecr.io/azureml-examples/tfsintegrated:1
|
||||
inference_config:
|
||||
liveness_route:
|
||||
port: 8501
|
||||
path: /v1/models/hpt
|
||||
readiness_route:
|
||||
port: 8501
|
||||
path: /v1/models/hpt
|
||||
scoring_route:
|
||||
port: 8501
|
||||
path: /v1/models/hpt:predict
|
||||
instance_type: Standard_DS2_v2
|
||||
instance_count: 1
|
||||
|
|
@ -0,0 +1,5 @@
|
|||
FROM docker.io/tensorflow/serving:latest
|
||||
|
||||
ENV MODEL_NAME=hpt
|
||||
|
||||
COPY models /models
|
||||
Двоичные данные
cli/endpoints/online/custom-container/tfserving/half-plus-two-integrated/models/hpt/1/saved_model.pb
Normal file
Двоичные данные
cli/endpoints/online/custom-container/tfserving/half-plus-two-integrated/models/hpt/1/saved_model.pb
Normal file
Двоичный файл не отображается.
Двоичный файл не отображается.
Двоичные данные
cli/endpoints/online/custom-container/tfserving/half-plus-two-integrated/models/hpt/1/variables/variables.index
Normal file
Двоичные данные
cli/endpoints/online/custom-container/tfserving/half-plus-two-integrated/models/hpt/1/variables/variables.index
Normal file
Двоичный файл не отображается.
|
|
@ -0,0 +1,6 @@
|
|||
{
|
||||
"inputs" : [
|
||||
[[1,2,3,4]],
|
||||
[[0,1,1,1]]
|
||||
]
|
||||
}
|
||||
|
|
@ -0,0 +1,17 @@
|
|||
# Deploy the Half Plus Two model using TFServing
|
||||
In this example, we deploy a single model (half-plus-two) using a TFServing custom container.
|
||||
|
||||
This example can be run end-to-end by executing the `deploy-custom-container-tfserving-half-plus-two.sh` script in the `CLI` directory.
|
||||
|
||||
## Model
|
||||
This example uses the `half-plus-two` model, which is downloaded in the script. In the deployment yaml it is registered as a model and mounted at runtime at the `AZUREML_MODEL_DIR` environment variable as in standard deployments. The default location for model mounting is `/var/azureml-app/azureml-models/<MODEL_NAME>/<MODEL_VERSION>` unless overridden by the `model_mount_path` field in the deployment yaml.
|
||||
|
||||
This path is passed to TFServing as an environment variable in the deployment YAML.
|
||||
|
||||
## Build the image
|
||||
This example uses the `tensorflow/serving` image with no modifications as defined in the `tfserving.dockerfile`. Although this example demonstrates the usual workflow of building the image on an ACR instance, this deployment could bypass the ACR build step and include the `docker.io` path of the image as the image URL in the deployment YAML.
|
||||
|
||||
## Environment
|
||||
The environment is defined inline in the deployment yaml and references the ACR url of the image. The ACR must be associated with the workspace (or have a user-assigned managed identity that enables ACRPull) in order to successfully deploy.
|
||||
|
||||
The environment also contains an `inference_config` block that defines the `liveness`, `readiness`, and `scoring` routes by path and port. Because the images used in this examples are based on the AzureML Inference Minimal images, these values are the same as those in a non-BYOC deployment, however they must be included since we are now using a custom image.
|
||||
|
|
@ -3,10 +3,10 @@ name: tfserving-deployment
|
|||
endpoint_name: tfserving-endpoint
|
||||
model:
|
||||
name: tfserving-mounted
|
||||
version: 1
|
||||
version: {{MODEL_VERSION}}
|
||||
path: ./half_plus_two
|
||||
environment_variables:
|
||||
MODEL_BASE_PATH: /var/azureml-app/azureml-models/tfserving-mounted/1
|
||||
MODEL_BASE_PATH: /var/azureml-app/azureml-models/tfserving-mounted/{{MODEL_VERSION}}
|
||||
MODEL_NAME: half_plus_two
|
||||
environment:
|
||||
#name: tfserving
|
||||
|
|
@ -1,3 +0,0 @@
|
|||
FROM pytorch/torchserve:latest-cpu
|
||||
|
||||
CMD ["torchserve","--start","--model-store","$MODEL_BASE_PATH/torchserve","--models","densenet161=densenet161.mar"]
|
||||
|
|
@ -0,0 +1,16 @@
|
|||
# Deploy a Torchserve custom container using the Densenet model
|
||||
|
||||
In this example, we deploy a single model (Densenet) using a Torchserve custom container.
|
||||
|
||||
This example can be run end-to-end by executing the `deploy-custom-container-torchserve-densenet.sh` script in the `CLI` directory.
|
||||
|
||||
## Model
|
||||
|
||||
This example uses the `densenet161` model. It is registered as a model ahead-of-time using a separate call to `az ml model create`. Upon instantiation, the `AZUREML_MODEL_DIR` environment variable as in standard deployments. The default location for model mounting is `/var/azureml-app/azureml-models/<MODEL_NAME>/<MODEL_VERSION>` unless overridden by the `model_mount_path` field in the deployment yaml.
|
||||
|
||||
This path is passed to TFServing as an environment variable in the deployment YAML.
|
||||
|
||||
## Environment
|
||||
The environment is defined inline in the deployment yaml and references the ACR url of the image. The ACR must be associated with the workspace (or have a user-assigned managed identity that enables ACRPull) in order to successfully deploy.
|
||||
|
||||
The environment also contains an `inference_config` block that defines the `liveness`, `readiness`, and `scoring` routes by path and port. Because the images used in this examples are based on the AzureML Inference Minimal images, these values are the same as those in a non-BYOC deployment, however they must be included since we are now using a custom image.
|
||||
|
|
@ -1,13 +1,12 @@
|
|||
$schema: https://azuremlschemas.azureedge.net/latest/managedOnlineDeployment.schema.json
|
||||
name: torchserve-deployment
|
||||
endpoint_name: torchserve-endpoint
|
||||
model: azureml:torchserve-densenet161:1
|
||||
model:
|
||||
path: torchserve
|
||||
environment_variables:
|
||||
MODEL_BASE_PATH: /var/azureml-app/azureml-models/torchserve-densenet161/1
|
||||
TORCHSERVE_MODELS: "densenet161=densenet161.mar"
|
||||
environment:
|
||||
#name: torchserve-environment
|
||||
#version: 1
|
||||
image: {{acr_name}}.azurecr.io/torchserve:8080
|
||||
image: {{ACR_NAME}}.azurecr.io/torchserve:1
|
||||
inference_config:
|
||||
liveness_route:
|
||||
port: 8080
|
||||
|
|
@ -0,0 +1,3 @@
|
|||
FROM pytorch/torchserve:latest-cpu
|
||||
|
||||
CMD ["torchserve","--start","--model-store","$AZUREML_MODEL_DIR/torchserve","--models","$TORCHSERVE_MODELS"]
|
||||
|
|
@ -0,0 +1,24 @@
|
|||
$schema: https://azuremlschemas.azureedge.net/latest/managedOnlineDeployment.schema.json
|
||||
name: textgeneration
|
||||
endpoint_name: {{ENDPOINT_NAME}}
|
||||
model:
|
||||
name: hf-tg
|
||||
path: serve/examples/Huggingface_Transformers/Textgeneration.mar
|
||||
environment_variables:
|
||||
TORCHSERVE_MODELS: "textgeneration=Textgeneration.mar"
|
||||
environment:
|
||||
name: hf-tg
|
||||
image: {{ACR_NAME}}.azurecr.io/{{IMAGE_TAG}}
|
||||
inference_config:
|
||||
liveness_route:
|
||||
port: 8080
|
||||
path: /ping
|
||||
readiness_route:
|
||||
port: 8080
|
||||
path: /ping
|
||||
scoring_route:
|
||||
port: 8080
|
||||
path: /predictions/textgeneration
|
||||
instance_type: Standard_DS2_v2
|
||||
instance_count: 1
|
||||
|
||||
|
|
@ -0,0 +1,12 @@
|
|||
{
|
||||
"model_name":"gpt2",
|
||||
"mode":"text_generation",
|
||||
"do_lower_case":true,
|
||||
"num_labels":"0",
|
||||
"save_mode":"pretrained",
|
||||
"max_length":"20",
|
||||
"captum_explanation":false,
|
||||
"FasterTransformer":false,
|
||||
"embedding_name": "gpt2",
|
||||
"model_parallel":false
|
||||
}
|
||||
|
|
@ -0,0 +1,5 @@
|
|||
FROM pytorch/torchserve:latest-cpu
|
||||
|
||||
RUN pip install transformers==4.6.0
|
||||
|
||||
CMD ["torchserve","--start","--model-store","$AZUREML_MODEL_DIR","--models","$TORCHSERVE_MODELS","--ncs"]
|
||||
|
|
@ -0,0 +1,16 @@
|
|||
# Deploy a Triton model using a custom container.
|
||||
|
||||
In this example, we deploy a single model (Densenet) using a Torchserve custom container.
|
||||
|
||||
This example can be run end-to-end by executing the `deploy-custom-container-triton-single-model.sh` script in the `CLI` directory.
|
||||
|
||||
## Model
|
||||
|
||||
This example uses the `densenet161` model. The default location for model mounting is `/var/azureml-app/azureml-models/<MODEL_NAME>/<MODEL_VERSION>` unless overridden by the `model_mount_path` field in the deployment yaml. a
|
||||
|
||||
This path is passed to Triton as an environment variable in the deployment YAML and passed to Triton using the customized CMD command from the Dockerfile.
|
||||
|
||||
## Environment
|
||||
The environment is defined inline in the deployment yaml and references the ACR url of the image. The ACR must be associated with the workspace (or have a user-assigned managed identity that enables ACRPull) in order to successfully deploy.
|
||||
|
||||
The environment also contains an `inference_config` block that defines the `liveness`, `readiness`, and `scoring` routes by path and port. Because the images used in this examples are based on the AzureML Inference Minimal images, these values are the same as those in a non-BYOC deployment, however they must be included since we are now using a custom image.
|
||||
|
|
@ -1,7 +1,7 @@
|
|||
$schema: https://azuremlschemas.azureedge.net/latest/managedOnlineDeployment.schema.json
|
||||
name: triton-cc-deployment
|
||||
endpoint_name: {{endpoint_name}}
|
||||
model:
|
||||
model:
|
||||
path: ./models/model_1
|
||||
model_mount_path: /models
|
||||
environment:
|
||||
Загрузка…
Ссылка в новой задаче