feat(workshop): clean up for workshop (#26)

* feat(workshop): clean up for workshop * feat(workshop): clean up for workshop * feat(workshop): clean up for workshop * feat(workshop): clean up for workshop * feat(workshop): clean up for workshop * feat(workshop): clean up for workshop * feat(workshop): clean up for workshop * feat(workshop): clean up for workshop * feat(workshop): clean up for workshop * Update main.yml * Update main.yml * Update main.yml * Update main.yml
guillaume-chervet · Dec 17, 2023 · eea9c48 · eea9c48
1 parent 66e00cf
commit eea9c48
Show file tree

Hide file tree

Showing 9 changed files with 240 additions and 83 deletions.
diff --git a/.github/workflows/main.yml b/.github/workflows/main.yml
@@ -14,8 +14,8 @@ env:
   AZURE_LOCATION: "northeurope"
   AZURE_ML_WORKSPACE_NAME: "cats-dogs-yola"
   AZURE_WEBAPP_NAME: "cats-dogs-yola"
-  DELETE_WEBAPP: "true"
-  DOCKER_IMAGE_NAME: "guillaumechervet/mlopspython"
+  DELETE_WEBAPP: "false"
+  DOCKER_IMAGE_NAME: "guillaumechervetlunique/mlopspython"
 
 permissions:
   id-token: write
@@ -196,7 +196,10 @@ jobs:
           poetry run python azureml_run_pipeline.py \
             --subscription_id ${{ secrets.AZURE_SUBSCRIPTION_ID }}  \
             --resource_group_name ${{ env.AZURE_RESOURCE_GROUP_NAME }} \
-            --workspace_name ${{ env.AZURE_ML_WORKSPACE_NAME }} > train_output.txt
+            --workspace_name ${{ env.AZURE_ML_WORKSPACE_NAME }} \
+            --location ${{ env.AZURE_LOCATION }} \
+            --tags "{\"triggering_actor\":\"${{github.triggering_actor}}\"}" \
+            > train_output.txt
           cat train_output.txt
       working-directory: train
     - name: download model
@@ -272,6 +275,7 @@ jobs:
       AZURE_TENANT_ID: ${{ secrets.AZURE_TENANT_ID }}
       AZURE_SUBSCRIPTION_ID: ${{ secrets.AZURE_SUBSCRIPTION_ID }}
   deploy:
+    environment: MLOpsPython
     runs-on: ubuntu-latest
     needs: [tags, build_docker]
     steps:
@@ -286,6 +290,7 @@ jobs:
         # https://learn.microsoft.com/en-us/cli/azure/container?view=azure-cli-latest#az-container-create()
         az container create --resource-group ${{ env.AZURE_RESOURCE_GROUP_NAME}} --name ${{ env.AZURE_WEBAPP_NAME }} --dns-name-label ${{ env.AZURE_WEBAPP_NAME }} --image docker.io/${{ env.DOCKER_IMAGE_NAME }}:${{ needs.tags.outputs.new_version }} --ports 5000
   integration_tests:
+    environment: MLOpsPython
     runs-on: ubuntu-latest
     needs: [deploy, train]
     steps:
@@ -312,7 +317,7 @@ jobs:
         poetry install
         az extension add -n ml
         poetry run python azureml_run_pipeline.py \
-            --subscription_id ${{ env.AZURE_SUBSCRIPTION_ID }}  \
+            --subscription_id ${{ secrets.AZURE_SUBSCRIPTION_ID }}  \
             --resource_group_name ${{ env.AZURE_RESOURCE_GROUP_NAME }} \
             --workspace_name ${{ env.AZURE_ML_WORKSPACE_NAME }} \
             --experiment_id ${{ needs.train.outputs.EXPERIMENT_ID }} \

diff --git a/bin/init_repository.ps1 b/bin/init_repository.ps1
@@ -12,7 +12,7 @@ az login
 gh auth login
 
 # Fork MLOPsPython repository
-gh repo fork https://github.com/guillaume-chervet/MLOpsPythonDemo1 --default-branch-only --fork-name $repositoryName --clone
+gh repo fork https://github.com/guillaume-chervet/MLOpsPython --default-branch-only --fork-name $repositoryName --clone
 cd $repositoryName
 git remote remove upstream
 git push --set-upstream origin main
@@ -63,9 +63,9 @@ gh secret set DOCKER_USENAME --body "dckr_pat_e2lZ9YgpMt8APE-Qxzn89u6mt28" --env
 
 # Run workflow
 gh workflow enable main.yml
-gh workflow run main.yml
-$runId=$(gh run list --workflow=main.yml --json databaseId -q ".[0].databaseId")
-gh run watch $runId
+#gh workflow run main.yml
+#$runId=$(gh run list --workflow=main.yml --json databaseId -q ".[0].databaseId")
+#gh run watch $runId
 
-# Open the repostory in the browser
+# Open the repository in the browser
 gh repo view -w
diff --git a/bin/init_repository.sh b/bin/init_repository.sh
@@ -62,13 +62,13 @@ gh secret set DOCKER_USENAME --body "dckr_pat_e2lZ9YgpMt8APE-Qxzn89u6mt28" --env
 
 # Run the GitHub workflow
 gh workflow enable main.yml
-gh workflow run main.yml
+#gh workflow run main.yml
 
 # Get the run ID
-runId=$(gh run list --workflow=main.yml --json databaseId -q ".[0].databaseId")
+#runId=$(gh run list --workflow=main.yml --json databaseId -q ".[0].databaseId")
 
 # Watch the run
-gh run watch "$runId"
+#gh run watch "$runId"
 
 # Open the repository in the browser
 gh repo view -w
diff --git a/documentation/workshop.md b/documentation/workshop.md
@@ -1,4 +1,4 @@
-# Workshop Cloud Nord
+# Workshop MLOps 
 
 Welcome aboard the MLOpsPython team !
 
@@ -34,9 +34,9 @@ You can also check health check route with HTTP GET:
 
 http://cats-dogs-yolo.northeurope.azurecontainer.io:5000/health
 
-## 2. Fork the project MLOpsPython
+## 3. Initialize your own project
 
-### 2.1. Create your Azure Account
+### 3.1. Create your Azure Account
 
 Important point:
 - About azure coupon:
@@ -53,25 +53,28 @@ https://www.microsoftazurepass.com/?WT.mc_id=DOP-MVP-5003370
 Documentation:
 https://www.microsoftazurepass.com/Home/HowTo?WT.mc_id=DOP-MVP-5003370 
 
-### 2.3. Fork the GitHub project MLOpsPython
-
-Prerequisite to install on your laptop:
-
-1. Fork https://github.com/guillaume-chervet/MLOpsPython repository
-2. Go to GitHub Action Tab and activate it !
-3. Clone your new own **MLOpsPython** repository
-
-Je vous mets à disposition les fichiers de script ci-dessous :
-Copier un de ces deux fichiers disponibles ici https://github.com/guillaume-chervet/... chez vous et exécuter le Powershell (.ps1) si vous êtes sous Windows et le bash (.sh) si vous êtes sous linux ou mac.
-
-Il se base sur ce super Article réalisé par Alexandre Nédélec que je remercie : https://www.techwatching.dev/posts/sc...
-
+### 3.2. Fork the GitHub project MLOpsPython
 
+**On Windows**: 
+1. Download https://github.com/guillaume-chervet/MLOpsPython/blob/main/bin/init_repository.ps1 PowerShell Script
+2. Open a PowerShell Terminal then run 
+````ps
+./init_repository.ps1
+````
+3. Go to GitHub Action Tab and activate it !
+
+**On Linux or Mac**:
+1. Download https://github.com/guillaume-chervet/MLOpsPython/blob/main/bin/init_repository.sh Bash Script
+2. Open a Bash Terminal then run
+````bach
+chmod +x ./init_repository.sh
+./init_repository.ps1
+````
 
 
-### 2.6. Adapt GitHub Action Parameters
+### 3.3. Adapt GitHub Action Parameters
 
-Inside "./.github/workflows/ci.yml" file
+Inside "./.github/workflows/main.yml" file
 
 ````bash
 
@@ -83,7 +86,7 @@ env:
 
 ````
 
-### 2.7. Run GitHub Action
+### 3.4. Run GitHub Action
 
 Commit and push your code
 
@@ -97,7 +100,7 @@ git push origin main
 
 This will trigger the GitHub Action.
 
-## 3. Getting Started on your laptop
+## 4. Getting Started on your laptop
 
 Follow "Get Started" section to run the project on your laptop :
 
@@ -107,7 +110,7 @@ https://github.com/your-github-login/MLOpsPython
 - Drift dataset: https://github.com/guillaume-chervet/dataset-cats-dogs-others-drift
 - You can test with your own files :)
 
-## 4. First Contribution : Images Labelling
+## 5. First Contribution : Images Labelling
 
 We need you to annotate 200 images of classification of :
 - cat
@@ -123,7 +126,7 @@ https://axaguildev-ecotag.azurewebsites.net/projects/2329f843-fa3d-45df-bec5-08d
 For you culture, Ecotag is an awesome Open Source tool available here :
 https://github.com/AxaGuilDEv/ecotag
 
-## 5. Second Contribution : CleanCode & PullRequest
+## 6. Second Contribution : CleanCode & PullRequest
 
 Our team Kanban:
 
@@ -146,7 +149,7 @@ git commit -m "refactor(myfonctionnality): commit message"
 git push
 ```
 
-## 6. How to check your API result  
+## 7. How to check your API result  
 
 1. Follow "ml-cli" readme and download "ml-cli" version v0.54.2 for your OS https://github.com/AxaFrance/ecotag/blob/master/README-ML-CLI.md
 2. Download https://github.com/guillaume-chervet/dataset-cats-dogs-others-mlcli as a zip and unzip content in "demo" folder
@@ -172,7 +175,7 @@ try {
 ```
 
 
-## 7. More AzureML
+## 8. More AzureML
 
 1. Create a AzureML Compute Instance
 2. Open JupyterLab with Python 3.10 and SDK v2