Katib official release v0.18.0

* [feature] move manifest image references to ghcr (#2529)

* move to ghcr

Signed-off-by: mahdikhashan <mahdikhashan1@gmail.com>

* move images to ghcr

Signed-off-by: mahdikhashan <mahdikhashan1@gmail.com>

* manifests

Signed-off-by: mahdikhashan <mahdikhashan1@gmail.com>

* change registry in all path

Signed-off-by: mahdikhashan <mahdikhashan1@gmail.com>

* update script

Signed-off-by: mahdikhashan <mahdikhashan1@gmail.com>

* fix

Signed-off-by: mahdikhashan <mahdikhashan1@gmail.com>

* fix

Signed-off-by: mahdikhashan <mahdikhashan1@gmail.com>

* slight fix

Signed-off-by: mahdikhashan <mahdikhashan1@gmail.com>

---------

Signed-off-by: mahdikhashan <mahdikhashan1@gmail.com>
Signed-off-by: Mahdi Khashan <58775404+mahdikhashan@users.noreply.github.com>

* revert changes for ipynb

Signed-off-by: sailesh duddupudi <saileshradar@gmail.com>

---------

Signed-off-by: mahdikhashan <mahdikhashan1@gmail.com>
Signed-off-by: Mahdi Khashan <58775404+mahdikhashan@users.noreply.github.com>
Signed-off-by: sailesh duddupudi <saileshradar@gmail.com>
Co-authored-by: Mahdi Khashan <58775404+mahdikhashan@users.noreply.github.com>

.github/workflows/build-and-publish-images.yaml

@@ -1,5 +1,5 @@
 # Reusable workflows for publishing Katib images.
-name: Build And Publish Images
+name: Build and Publish Images
 
 on:
   workflow_call:
@@ -21,31 +21,50 @@ on:
 
 jobs:
   build-and-publish:
-    name: Publish Image
+    name: Build and Publish Images
     runs-on: ubuntu-22.04
+
     steps:
       - name: Checkout
         uses: actions/checkout@v4
 
-      - name: Docker Login
-        # Trigger workflow only for kubeflow/katib repository with specific branch (master, release-.*) or tag (v.*).
-        if: >-
-          github.repository == 'kubeflow/katib' &&
-          (github.ref == 'refs/heads/master' || startsWith(github.ref, 'refs/heads/release-') || startsWith(github.ref, 'refs/tags/v'))
+      - name: Set Publish Condition
+        id: publish-condition
+        shell: bash
+        run: |
+          if [[ "${{ github.repository }}" == 'kubeflow/katib' && \
+                ( "${{ github.ref }}" == 'refs/heads/master' || \
+                  "${{ github.ref }}" =~ ^refs/heads/release- || \
+                  "${{ github.ref }}" =~ ^refs/tags/v ) ]]; then
+            echo "should_publish=true" >> $GITHUB_OUTPUT
+          else
+            echo "should_publish=false" >> $GITHUB_OUTPUT
+          fi
+
+      - name: GHCR Login
+        if: steps.publish-condition.outputs.should_publish == 'true'
         uses: docker/login-action@v3
         with:
+          registry: ghcr.io
+          username: ${{ github.actor }}
+          password: ${{ secrets.GITHUB_TOKEN }}
+
+      - name: DockerHub Login
+        if: steps.publish-condition.outputs.should_publish == 'true'
+        uses: docker/login-action@v3
+        with:
+          registry: docker.io
           username: ${{ secrets.DOCKERHUB_USERNAME }}
           password: ${{ secrets.DOCKERHUB_TOKEN }}
 
       - name: Publish Component ${{ inputs.component-name }}
-        # Trigger workflow only for kubeflow/katib repository with specific branch (master, release-.*) or tag (v.*).
-        if: >-
-          github.repository == 'kubeflow/katib' &&
-          (github.ref == 'refs/heads/master' || startsWith(github.ref, 'refs/heads/release-') || startsWith(github.ref, 'refs/tags/v'))
+        if: steps.publish-condition.outputs.should_publish == 'true'
         id: publish
         uses: ./.github/workflows/template-publish-image
         with:
-          image: docker.io/kubeflowkatib/${{ inputs.component-name }}
+          image: |
+            ghcr.io/kubeflow/katib/${{ inputs.component-name }}
+            docker.io/kubeflowkatib/${{ inputs.component-name }}
           dockerfile: ${{ inputs.dockerfile }}
           platforms: ${{ inputs.platforms }}
           push: true
@@ -54,7 +73,9 @@ jobs:
         if: steps.publish.outcome == 'skipped'
         uses: ./.github/workflows/template-publish-image
         with:
-          image: docker.io/kubeflowkatib/${{ inputs.component-name }}
+          image: |
+            ghcr.io/kubeflow/katib/${{ inputs.component-name }}
+            docker.io/kubeflowkatib/${{ inputs.component-name }}
           dockerfile: ${{ inputs.dockerfile }}
           platforms: ${{ inputs.platforms }}
           push: false

Makefile

@@ -5,7 +5,7 @@ HAS_SETUP_ENVTEST := $(shell command -v setup-envtest;)
 HAS_MOCKGEN := $(shell command -v mockgen;)
 
 COMMIT := v1beta1-$(shell git rev-parse --short=7 HEAD)
-KATIB_REGISTRY := docker.io/kubeflowkatib
+KATIB_REGISTRY := ghcr.io/kubeflow/katib
 CPU_ARCH ?= linux/amd64,linux/arm64
 ENVTEST_K8S_VERSION ?= 1.31
 MOCKGEN_VERSION ?= $(shell grep 'go.uber.org/mock' go.mod | cut -d ' ' -f 2)

docs/images-location.md

@@ -22,7 +22,7 @@ The following table shows images for the
     </tr>
     <tr align="center">
       <td>
-        <code>docker.io/kubeflowkatib/katib-controller</code>
+        <code>ghcr.io/kubeflow/katib/katib-controller</code>
       </td>
       <td>
         Katib Controller
@@ -33,7 +33,7 @@ The following table shows images for the
     </tr>
     <tr align="center">
       <td>
-        <code>docker.io/kubeflowkatib/katib-ui</code>
+        <code>ghcr.io/kubeflow/katib/katib-ui</code>
       </td>
       <td>
         Katib User Interface
@@ -44,7 +44,7 @@ The following table shows images for the
     </tr>
     <tr align="center">
       <td>
-        <code>docker.io/kubeflowkatib/katib-db-manager</code>
+        <code>ghcr.io/kubeflow/katib/katib-db-manager</code>
       </td>
       <td>
         Katib DB Manager
@@ -87,7 +87,7 @@ The following table shows images for the
     </tr>
     <tr align="center">
       <td>
-        <code>docker.io/kubeflowkatib/file-metrics-collector</code>
+        <code>ghcr.io/kubeflow/katib/file-metrics-collector</code>
       </td>
       <td>
         File Metrics Collector
@@ -98,7 +98,7 @@ The following table shows images for the
     </tr>
     <tr align="center">
       <td>
-        <code>docker.io/kubeflowkatib/tfevent-metrics-collector</code>
+        <code>ghcr.io/kubeflow/katib/tfevent-metrics-collector</code>
       </td>
       <td>
         Tensorflow Event Metrics Collector
@@ -131,7 +131,7 @@ and the [Katib Early Stopping algorithms](https://www.kubeflow.org/docs/componen
     </tr>
     <tr align="center">
       <td>
-        <code>docker.io/kubeflowkatib/suggestion-hyperopt</code>
+        <code>ghcr.io/kubeflow/katib/suggestion-hyperopt</code>
       </td>
       <td>
         <a href="https://github.com/hyperopt/hyperopt">Hyperopt</a> Suggestion
@@ -142,7 +142,7 @@ and the [Katib Early Stopping algorithms](https://www.kubeflow.org/docs/componen
     </tr>
     <tr align="center">
       <td>
-        <code>docker.io/kubeflowkatib/suggestion-skopt</code>
+        <code>ghcr.io/kubeflow/katib/suggestion-skopt</code>
       </td>
       <td>
         <a href="https://github.com/scikit-optimize/scikit-optimize">Skopt</a> Suggestion
@@ -153,7 +153,7 @@ and the [Katib Early Stopping algorithms](https://www.kubeflow.org/docs/componen
     </tr>
     <tr align="center">
       <td>
-        <code>docker.io/kubeflowkatib/suggestion-optuna</code>
+        <code>ghcr.io/kubeflow/katib/suggestion-optuna</code>
       </td>
       <td>
         <a href="https://github.com/optuna/optuna">Optuna</a> Suggestion
@@ -164,7 +164,7 @@ and the [Katib Early Stopping algorithms](https://www.kubeflow.org/docs/componen
     </tr>
     <tr align="center">
       <td>
-        <code>docker.io/kubeflowkatib/suggestion-goptuna</code>
+        <code>ghcr.io/kubeflow/katib/suggestion-goptuna</code>
       </td>
       <td>
         <a href="https://github.com/c-bata/goptuna">Goptuna</a> Suggestion
@@ -175,7 +175,7 @@ and the [Katib Early Stopping algorithms](https://www.kubeflow.org/docs/componen
     </tr>
     <tr align="center">
       <td>
-        <code>docker.io/kubeflowkatib/suggestion-hyperband</code>
+        <code>ghcr.io/kubeflow/katib/suggestion-hyperband</code>
       </td>
       <td>
         <a href="https://www.kubeflow.org/docs/components/katib/experiment/#hyperband">Hyperband</a> Suggestion
@@ -186,7 +186,7 @@ and the [Katib Early Stopping algorithms](https://www.kubeflow.org/docs/componen
     </tr>
     <tr align="center">
       <td>
-        <code>docker.io/kubeflowkatib/suggestion-enas</code>
+        <code>ghcr.io/kubeflow/katib/suggestion-enas</code>
       </td>
       <td>
         <a href="https://www.kubeflow.org/docs/components/katib/experiment/#enas">ENAS</a> Suggestion
@@ -197,7 +197,7 @@ and the [Katib Early Stopping algorithms](https://www.kubeflow.org/docs/componen
     </tr>
     <tr align="center">
       <td>
-        <code>docker.io/kubeflowkatib/suggestion-darts</code>
+        <code>ghcr.io/kubeflow/katib/suggestion-darts</code>
       </td>
       <td>
         <a href="https://www.kubeflow.org/docs/components/katib/experiment/#differentiable-architecture-search-darts">DARTS</a> Suggestion
@@ -208,7 +208,7 @@ and the [Katib Early Stopping algorithms](https://www.kubeflow.org/docs/componen
     </tr>
     <tr align="center">
       <td>
-        <code>docker.io/kubeflowkatib/earlystopping-medianstop</code>
+        <code>ghcr.io/kubeflow/katib/earlystopping-medianstop</code>
       </td>
       <td>
         <a href="https://www.kubeflow.org/docs/components/katib/early-stopping/#median-stopping-rule">Median Stopping Rule</a>
@@ -240,7 +240,7 @@ The following table shows images for training containers which are used in the
     </tr>
     <tr align="center">
       <td>
-        <code>docker.io/kubeflowkatib/pytorch-mnist-cpu</code>
+        <code>ghcr.io/kubeflow/katib/pytorch-mnist-cpu</code>
       </td>
       <td>
         PyTorch MNIST example with printing metrics to the file or StdOut with CPU support
@@ -251,7 +251,7 @@ The following table shows images for training containers which are used in the
     </tr>
     <tr align="center">
       <td>
-        <code>docker.io/kubeflowkatib/pytorch-mnist-gpu</code>
+        <code>ghcr.io/kubeflow/katib/pytorch-mnist-gpu</code>
       </td>
       <td>
         PyTorch MNIST example with printing metrics to the file or StdOut with GPU support
@@ -262,7 +262,7 @@ The following table shows images for training containers which are used in the
     </tr>
     <tr align="center">
       <td>
-        <code>docker.io/kubeflowkatib/tf-mnist-with-summaries</code>
+        <code>ghcr.io/kubeflow/katib/tf-mnist-with-summaries</code>
       </td>
       <td>
         Tensorflow MNIST example with saving metrics in the summaries
@@ -273,7 +273,7 @@ The following table shows images for training containers which are used in the
     </tr>
     <tr align="center">
       <td>
-        <code>docker.io/kubeflowkatib/xgboost-lightgbm</code>
+        <code>ghcr.io/kubeflow/katib/xgboost-lightgbm</code>
       </td>
       <td>
         Distributed LightGBM example for XGBoostJob
@@ -306,7 +306,7 @@ The following table shows images for training containers which are used in the
     </tr>
     <tr align="center">
       <td>
-        <code>docker.io/kubeflowkatib/enas-cnn-cifar10-gpu</code>
+        <code>ghcr.io/kubeflow/katib/enas-cnn-cifar10-gpu</code>
       </td>
       <td>
         Keras CIFAR-10 CNN example for ENAS with GPU support
@@ -317,7 +317,7 @@ The following table shows images for training containers which are used in the
     </tr>
     <tr align="center">
       <td>
-        <code>docker.io/kubeflowkatib/enas-cnn-cifar10-cpu</code>
+        <code>ghcr.io/kubeflow/katib/enas-cnn-cifar10-cpu</code>
       </td>
       <td>
         Keras CIFAR-10 CNN example for ENAS with CPU support
@@ -328,7 +328,7 @@ The following table shows images for training containers which are used in the
     </tr>
     <tr align="center">
       <td>
-        <code>docker.io/kubeflowkatib/darts-cnn-cifar10-gpu</code>
+        <code>ghcr.io/kubeflow/katib/darts-cnn-cifar10-gpu</code>
       </td>
       <td>
         PyTorch CIFAR-10 CNN example for DARTS with GPU support
@@ -339,7 +339,7 @@ The following table shows images for training containers which are used in the
     </tr>
     <tr align="center">
       <td>
-        <code>docker.io/kubeflowkatib/darts-cnn-cifar10-cpu</code>
+        <code>ghcr.io/kubeflow/katib/darts-cnn-cifar10-cpu</code>
       </td>
       <td>
         PyTorch CIFAR-10 CNN example for DARTS with CPU support

examples/v1beta1/argo/argo-workflow.yaml

@@ -74,7 +74,7 @@ spec:
                 - name: epochs
             container:
               name: model-training
-              image: docker.io/kubeflowkatib/pytorch-mnist-cpu:latest
+              image: ghcr.io/kubeflow/katib/pytorch-mnist-cpu:v0.18.0
               command:
                 - "python3"
                 - "/opt/pytorch-mnist/mnist.py"

examples/v1beta1/early-stopping/median-stop-with-json-format.yaml

@@ -62,7 +62,7 @@ spec:
           spec:
             containers:
               - name: training-container
-                image: docker.io/kubeflowkatib/pytorch-mnist-cpu:latest
+                image: ghcr.io/kubeflow/katib/pytorch-mnist-cpu:v0.18.0
                 command:
                   - "python3"
                   - "/opt/pytorch-mnist/mnist.py"

examples/v1beta1/early-stopping/median-stop.yaml

@@ -52,7 +52,7 @@ spec:
           spec:
             containers:
               - name: training-container
-                image: docker.io/kubeflowkatib/pytorch-mnist-cpu:latest
+                image: ghcr.io/kubeflow/katib/pytorch-mnist-cpu:v0.18.0
                 command:
                   - "python3"
                   - "/opt/pytorch-mnist/mnist.py"

examples/v1beta1/hp-tuning/bayesian-optimization.yaml

@@ -45,7 +45,7 @@ spec:
           spec:
             containers:
               - name: training-container
-                image: docker.io/kubeflowkatib/pytorch-mnist-cpu:latest
+                image: ghcr.io/kubeflow/katib/pytorch-mnist-cpu:v0.18.0
                 command:
                   - "python3"
                   - "/opt/pytorch-mnist/mnist.py"

examples/v1beta1/hp-tuning/cma-es.yaml

@@ -45,7 +45,7 @@ spec:
           spec:
             containers:
               - name: training-container
-                image: docker.io/kubeflowkatib/pytorch-mnist-cpu:latest
+                image: ghcr.io/kubeflow/katib/pytorch-mnist-cpu:v0.18.0
                 command:
                   - "python3"
                   - "/opt/pytorch-mnist/mnist.py"

examples/v1beta1/hp-tuning/grid.yaml

@@ -44,7 +44,7 @@ spec:
           spec:
             containers:
               - name: training-container
-                image: docker.io/kubeflowkatib/pytorch-mnist-cpu:latest
+                image: ghcr.io/kubeflow/katib/pytorch-mnist-cpu:v0.18.0
                 command:
                   - "python3"
                   - "/opt/pytorch-mnist/mnist.py"

examples/v1beta1/hp-tuning/hyperband.yaml

@@ -57,7 +57,7 @@ spec:
           spec:
             containers:
               - name: training-container
-                image: docker.io/kubeflowkatib/pytorch-mnist-cpu:latest
+                image: ghcr.io/kubeflow/katib/pytorch-mnist-cpu:v0.18.0
                 command:
                   - "python3"
                   - "/opt/pytorch-mnist/mnist.py"

examples/v1beta1/hp-tuning/hyperopt-distribution.yaml

@@ -63,7 +63,7 @@ spec:
           spec:
             containers:
               - name: training-container
-                image: docker.io/kubeflowkatib/pytorch-mnist-cpu:latest
+                image: ghcr.io/kubeflow/katib/pytorch-mnist-cpu:v0.18.0
                 command:
                   - "python3"
                   - "/opt/pytorch-mnist/mnist.py"

examples/v1beta1/hp-tuning/multivariate-tpe.yaml

@@ -42,7 +42,7 @@ spec:
           spec:
             containers:
               - name: training-container
-                image: docker.io/kubeflowkatib/pytorch-mnist-cpu:latest
+                image: ghcr.io/kubeflow/katib/pytorch-mnist-cpu:v0.18.0
                 command:
                   - "python3"
                   - "/opt/pytorch-mnist/mnist.py"

examples/v1beta1/hp-tuning/optuna-distribution.yaml

@@ -63,7 +63,7 @@ spec:
           spec:
             containers:
               - name: training-container
-                image: docker.io/kubeflowkatib/pytorch-mnist-cpu:latest
+                image: ghcr.io/kubeflow/katib/pytorch-mnist-cpu:v0.18.0
                 command:
                   - "python3"
                   - "/opt/pytorch-mnist/mnist.py"

examples/v1beta1/hp-tuning/random.yaml

@@ -42,7 +42,7 @@ spec:
           spec:
             containers:
               - name: training-container
-                image: docker.io/kubeflowkatib/pytorch-mnist-cpu:latest
+                image: ghcr.io/kubeflow/katib/pytorch-mnist-cpu:v0.18.0
                 command:
                   - "python3"
                   - "/opt/pytorch-mnist/mnist.py"

examples/v1beta1/hp-tuning/simple-pbt.yaml

@@ -43,7 +43,7 @@ spec:
           spec:
             containers:
               - name: training-container
-                image: docker.io/kubeflowkatib/simple-pbt:latest
+                image: ghcr.io/kubeflow/katib/simple-pbt:v0.18.0
                 command:
                   - "python3"
                   - "/opt/pbt/pbt_test.py"

examples/v1beta1/hp-tuning/sobol.yaml

@@ -42,7 +42,7 @@ spec:
           spec:
             containers:
               - name: training-container
-                image: docker.io/kubeflowkatib/pytorch-mnist-cpu:latest
+                image: ghcr.io/kubeflow/katib/pytorch-mnist-cpu:v0.18.0
                 command:
                   - "python3"
                   - "/opt/pytorch-mnist/mnist.py"

examples/v1beta1/hp-tuning/tpe.yaml

@@ -42,7 +42,7 @@ spec:
           spec:
             containers:
               - name: training-container
-                image: docker.io/kubeflowkatib/pytorch-mnist-cpu:latest
+                image: ghcr.io/kubeflow/katib/pytorch-mnist-cpu:v0.18.0
                 command:
                   - "python3"
                   - "/opt/pytorch-mnist/mnist.py"

examples/v1beta1/kubeflow-training-operator/pytorchjob-mnist.yaml

@@ -46,7 +46,7 @@ spec:
               spec:
                 containers:
                   - name: pytorch
-                    image: docker.io/kubeflowkatib/pytorch-mnist-cpu:latest
+                    image: ghcr.io/kubeflow/katib/pytorch-mnist-cpu:v0.18.0
                     command:
                       - "python3"
                       - "/opt/pytorch-mnist/mnist.py"
@@ -61,7 +61,7 @@ spec:
               spec:
                 containers:
                   - name: pytorch
-                    image: docker.io/kubeflowkatib/pytorch-mnist-cpu:latest
+                    image: ghcr.io/kubeflow/katib/pytorch-mnist-cpu:v0.18.0
                     command:
                       - "python3"
                       - "/opt/pytorch-mnist/mnist.py"

examples/v1beta1/kubeflow-training-operator/tfjob-mnist-with-summaries.yaml

@@ -56,7 +56,7 @@ spec:
               spec:
                 containers:
                   - name: tensorflow
-                    image: docker.io/kubeflowkatib/tf-mnist-with-summaries:latest
+                    image: ghcr.io/kubeflow/katib/tf-mnist-with-summaries:v0.18.0
                     command:
                       - "python"
                       - "/opt/tf-mnist-with-summaries/mnist.py"

examples/v1beta1/kubeflow-training-operator/xgboostjob-lightgbm.yaml

@@ -56,7 +56,7 @@ spec:
               spec:
                 containers:
                   - name: xgboost
-                    image: docker.io/kubeflowkatib/xgboost-lightgbm:1.0
+                    image: ghcr.io/kubeflow/katib/xgboost-lightgbm:1.0
                     ports:
                       - containerPort: 9991
                         name: xgboostjob-port
@@ -90,7 +90,7 @@ spec:
               spec:
                 containers:
                   - name: xgboost
-                    image: docker.io/kubeflowkatib/xgboost-lightgbm:1.0
+                    image: ghcr.io/kubeflow/katib/xgboost-lightgbm:1.0
                     ports:
                       - containerPort: 9991
                         name: xgboostjob-port

examples/v1beta1/metrics-collector/custom-metrics-collector.yaml

@@ -26,7 +26,7 @@ spec:
           - katib-db-manager.kubeflow:6789
           - -path
           - /katib/mnist.log
-        image: kubeflowkatib/custom-metrics-collector:latest
+        image: ghcr.io/kubeflow/katib/custom-metrics-collector:latest
         imagePullPolicy: Always
         name: custom-metrics-logger-and-collector
         env:
@@ -67,7 +67,7 @@ spec:
           spec:
             containers:
               - name: training-container
-                image: docker.io/kubeflowkatib/pytorch-mnist-cpu:latest
+                image: ghcr.io/kubeflow/katib/pytorch-mnist-cpu:v0.18.0
                 command:
                   - "python3"
                   - "/opt/pytorch-mnist/mnist.py"

examples/v1beta1/metrics-collector/file-metrics-collector-with-json-format.yaml

@@ -52,7 +52,7 @@ spec:
           spec:
             containers:
               - name: training-container
-                image: docker.io/kubeflowkatib/pytorch-mnist-cpu:latest
+                image: ghcr.io/kubeflow/katib/pytorch-mnist-cpu:v0.18.0
                 command:
                   - "python3"
                   - "/opt/pytorch-mnist/mnist.py"

examples/v1beta1/metrics-collector/file-metrics-collector.yaml

@@ -54,7 +54,7 @@ spec:
           spec:
             containers:
               - name: training-container
-                image: docker.io/kubeflowkatib/pytorch-mnist-cpu:latest
+                image: ghcr.io/kubeflow/katib/pytorch-mnist-cpu:v0.18.0
                 command:
                   - "python3"
                   - "/opt/pytorch-mnist/mnist.py"

examples/v1beta1/metrics-collector/metrics-collection-strategy.yaml

@@ -49,7 +49,7 @@ spec:
           spec:
             containers:
               - name: training-container
-                image: docker.io/kubeflowkatib/pytorch-mnist-cpu:latest
+                image: ghcr.io/kubeflow/katib/pytorch-mnist-cpu:v0.18.0
                 command:
                   - "python3"
                   - "/opt/pytorch-mnist/mnist.py"

examples/v1beta1/nas/darts-cpu.yaml

@@ -60,7 +60,7 @@ spec:
           spec:
             containers:
               - name: training-container
-                image: docker.io/kubeflowkatib/darts-cnn-cifar10-cpu:latest
+                image: ghcr.io/kubeflow/katib/darts-cnn-cifar10-cpu:v0.18.0
                 command:
                   - python3
                   - run_trial.py

examples/v1beta1/nas/darts-gpu.yaml

@@ -77,7 +77,7 @@ spec:
           spec:
             containers:
               - name: training-container
-                image: docker.io/kubeflowkatib/darts-cnn-cifar10-gpu:latest
+                image: ghcr.io/kubeflow/katib/darts-cnn-cifar10-gpu:v0.18.0
                 command:
                   - python3
                   - run_trial.py

examples/v1beta1/nas/enas-cpu.yaml

@@ -139,7 +139,7 @@ spec:
           spec:
             containers:
               - name: training-container
-                image: docker.io/kubeflowkatib/enas-cnn-cifar10-cpu:latest
+                image: ghcr.io/kubeflow/katib/enas-cnn-cifar10-cpu:v0.18.0
                 command:
                   - python3
                   - -u

examples/v1beta1/nas/enas-gpu.yaml

@@ -136,7 +136,7 @@ spec:
           spec:
             containers:
               - name: training-container
-                image: docker.io/kubeflowkatib/enas-cnn-cifar10-gpu:latest
+                image: ghcr.io/kubeflow/katib/enas-cnn-cifar10-gpu:v0.18.0
                 command:
                   - python3
                   - -u

examples/v1beta1/resume-experiment/from-volume-resume.yaml

@@ -43,7 +43,7 @@ spec:
           spec:
             containers:
               - name: training-container
-                image: docker.io/kubeflowkatib/pytorch-mnist-cpu:latest
+                image: ghcr.io/kubeflow/katib/pytorch-mnist-cpu:v0.18.0
                 command:
                   - "python3"
                   - "/opt/pytorch-mnist/mnist.py"

examples/v1beta1/resume-experiment/long-running-resume.yaml

@@ -43,7 +43,7 @@ spec:
           spec:
             containers:
               - name: training-container
-                image: docker.io/kubeflowkatib/pytorch-mnist-cpu:latest
+                image: ghcr.io/kubeflow/katib/pytorch-mnist-cpu:v0.18.0
                 command:
                   - "python3"
                   - "/opt/pytorch-mnist/mnist.py"

examples/v1beta1/tekton/README.md

@@ -30,13 +30,13 @@ set this `nop` image to Metrics Collector image.
 
 For example, if you are using
 [StdOut](https://www.kubeflow.org/docs/components/katib/experiment/#metrics-collector) Metrics Collector,
-`nop` image must be equal to `docker.io/kubeflowkatib/file-metrics-collector`.
+`nop` image must be equal to `ghcr.io/kubeflow/katib/file-metrics-collector`.
 
 Run the following command to modify the `nop` image:
 
 ```bash
 kubectl patch deploy tekton-pipelines-controller -n tekton-pipelines --type='json' \
-  -p='[{"op": "replace", "path": "/spec/template/spec/containers/0/args/9", "value": "docker.io/kubeflowkatib/file-metrics-collector"}]'
+  -p='[{"op": "replace", "path": "/spec/template/spec/containers/0/args/9", "value": "ghcr.io/kubeflow/katib/file-metrics-collector"}]'
 ```
 
 Check that Tekton Pipelines Controller's pod was restarted:
@@ -54,7 +54,7 @@ Verify that `nop` image was modified:
 ```bash
 $ kubectl get $(kubectl get pods -o name -n tekton-pipelines | grep tekton-pipelines-controller) -n tekton-pipelines -o yaml | grep katib
 
-   - docker.io/kubeflowkatib/file-metrics-collector
+   - ghcr.io/kubeflow/katib/file-metrics-collector
 ```
 
 ### Katib Controller

examples/v1beta1/tekton/pipeline-run.yaml

@@ -88,7 +88,7 @@ spec:
                     description: Number of epochs
                 steps:
                   - name: model-training
-                    image: docker.io/kubeflowkatib/pytorch-mnist-cpu:latest
+                    image: ghcr.io/kubeflow/katib/pytorch-mnist-cpu:v0.18.0
                     command:
                       - "python3"
                       - "/opt/pytorch-mnist/mnist.py"

examples/v1beta1/trial-template/trial-metadata-substitution.yaml

@@ -66,7 +66,7 @@ spec:
           spec:
             containers:
               - name: training-container
-                image: docker.io/kubeflowkatib/pytorch-mnist-cpu:latest
+                image: ghcr.io/kubeflow/katib/pytorch-mnist-cpu:v0.18.0
                 command:
                   - "python3"
                   - "/opt/pytorch-mnist/mnist.py"

manifests/v1beta1/components/controller/controller.yaml

@@ -23,7 +23,7 @@ spec:
       serviceAccountName: katib-controller
       containers:
         - name: katib-controller
-          image: docker.io/kubeflowkatib/katib-controller
+          image: ghcr.io/kubeflow/katib/katib-controller
           command: ["./katib-controller"]
           args:
             - --katib-config=/katib-config.yaml

manifests/v1beta1/components/controller/trial-templates.yaml

@@ -15,7 +15,7 @@ data:
         spec:
           containers:
             - name: training-container
-              image: docker.io/kubeflowkatib/pytorch-mnist-cpu:latest
+              image: ghcr.io/kubeflow/katib/pytorch-mnist-cpu:v0.18.0
               command:
                 - "python3"
                 - "/opt/pytorch-mnist/mnist.py"
@@ -33,7 +33,7 @@ data:
         spec:
           containers:
             - name: training-container
-              image: docker.io/kubeflowkatib/enas-cnn-cifar10-cpu:latest
+              image: ghcr.io/kubeflow/katib/enas-cnn-cifar10-cpu:v0.18.0
               command:
                 - python3
                 - -u
@@ -54,7 +54,7 @@ data:
             spec:
               containers:
                 - name: pytorch
-                  image: docker.io/kubeflowkatib/pytorch-mnist-cpu:latest
+                  image: ghcr.io/kubeflow/katib/pytorch-mnist-cpu:v0.18.0
                   command:
                     - "python3"
                     - "/opt/pytorch-mnist/mnist.py"
@@ -68,7 +68,7 @@ data:
             spec:
               containers:
                 - name: pytorch
-                  image: docker.io/kubeflowkatib/pytorch-mnist-cpu:latest
+                  image: ghcr.io/kubeflow/katib/pytorch-mnist-cpu:v0.18.0
                   command:
                     - "python3"
                     - "/opt/pytorch-mnist/mnist.py"

manifests/v1beta1/components/db-manager/db-manager.yaml

@@ -20,7 +20,7 @@ spec:
     spec:
       containers:
         - name: katib-db-manager
-          image: docker.io/kubeflowkatib/katib-db-manager
+          image: ghcr.io/kubeflow/katib/katib-db-manager
           env:
             - name: DB_NAME
               value: "mysql"

manifests/v1beta1/components/ui/ui.yaml

@@ -20,7 +20,7 @@ spec:
     spec:
       containers:
         - name: katib-ui
-          image: docker.io/kubeflowkatib/katib-ui
+          image: ghcr.io/kubeflow/katib/katib-ui
           command:
             - "./katib-ui"
           args:

manifests/v1beta1/installs/katib-cert-manager/katib-config.yaml

@@ -13,40 +13,40 @@ init:
 runtime:
   metricsCollectors:
     - kind: StdOut
-      image: docker.io/kubeflowkatib/file-metrics-collector:latest
+      image: ghcr.io/kubeflow/katib/file-metrics-collector:v0.18.0
     - kind: File
-      image: docker.io/kubeflowkatib/file-metrics-collector:latest
+      image: ghcr.io/kubeflow/katib/file-metrics-collector:v0.18.0
     - kind: TensorFlowEvent
-      image: docker.io/kubeflowkatib/tfevent-metrics-collector:latest
+      image: ghcr.io/kubeflow/katib/tfevent-metrics-collector:v0.18.0
       resources:
         limits:
           memory: 1Gi
   suggestions:
     - algorithmName: random
-      image: docker.io/kubeflowkatib/suggestion-hyperopt:latest
+      image: ghcr.io/kubeflow/katib/suggestion-hyperopt:v0.18.0
     - algorithmName: tpe
-      image: docker.io/kubeflowkatib/suggestion-hyperopt:latest
+      image: ghcr.io/kubeflow/katib/suggestion-hyperopt:v0.18.0
     - algorithmName: grid
-      image: docker.io/kubeflowkatib/suggestion-optuna:latest
+      image: ghcr.io/kubeflow/katib/suggestion-optuna:v0.18.0
     - algorithmName: hyperband
-      image: docker.io/kubeflowkatib/suggestion-hyperband:latest
+      image: ghcr.io/kubeflow/katib/suggestion-hyperband:v0.18.0
     - algorithmName: bayesianoptimization
-      image: docker.io/kubeflowkatib/suggestion-skopt:latest
+      image: ghcr.io/kubeflow/katib/suggestion-skopt:v0.18.0
     - algorithmName: cmaes
-      image: docker.io/kubeflowkatib/suggestion-goptuna:latest
+      image: ghcr.io/kubeflow/katib/suggestion-goptuna:v0.18.0
     - algorithmName: sobol
-      image: docker.io/kubeflowkatib/suggestion-goptuna:latest
+      image: ghcr.io/kubeflow/katib/suggestion-goptuna:v0.18.0
     - algorithmName: multivariate-tpe
-      image: docker.io/kubeflowkatib/suggestion-optuna:latest
+      image: ghcr.io/kubeflow/katib/suggestion-optuna:v0.18.0
     - algorithmName: enas
-      image: docker.io/kubeflowkatib/suggestion-enas:latest
+      image: ghcr.io/kubeflow/katib/suggestion-enas:v0.18.0
       resources:
         limits:
           memory: 400Mi
     - algorithmName: darts
-      image: docker.io/kubeflowkatib/suggestion-darts:latest
+      image: ghcr.io/kubeflow/katib/suggestion-darts:v0.18.0
     - algorithmName: pbt
-      image: docker.io/kubeflowkatib/suggestion-pbt:latest
+      image: ghcr.io/kubeflow/katib/suggestion-pbt:v0.18.0
       persistentVolumeClaimSpec:
         accessModes:
           - ReadWriteMany
@@ -55,4 +55,4 @@ runtime:
             storage: 5Gi
   earlyStoppings:
     - algorithmName: medianstop
-      image: docker.io/kubeflowkatib/earlystopping-medianstop:latest
+      image: ghcr.io/kubeflow/katib/earlystopping-medianstop:v0.18.0

manifests/v1beta1/installs/katib-cert-manager/kustomization.yaml

@@ -20,15 +20,15 @@ resources:
   # Cert-manager certificate for webhooks
   - certificate.yaml
 images:
-  - name: docker.io/kubeflowkatib/katib-controller
-    newName: docker.io/kubeflowkatib/katib-controller
-    newTag: latest
-  - name: docker.io/kubeflowkatib/katib-db-manager
-    newName: docker.io/kubeflowkatib/katib-db-manager
-    newTag: latest
-  - name: docker.io/kubeflowkatib/katib-ui
-    newName: docker.io/kubeflowkatib/katib-ui
-    newTag: latest
+  - name: ghcr.io/kubeflow/katib/katib-controller
+    newName: ghcr.io/kubeflow/katib/katib-controller
+    newTag: v0.18.0
+  - name: ghcr.io/kubeflow/katib/katib-db-manager
+    newName: ghcr.io/kubeflow/katib/katib-db-manager
+    newTag: v0.18.0
+  - name: ghcr.io/kubeflow/katib/katib-ui
+    newName: ghcr.io/kubeflow/katib/katib-ui
+    newTag: v0.18.0
 
 patchesStrategicMerge:
   - patches/katib-cert-injection.yaml

manifests/v1beta1/installs/katib-external-db/katib-config.yaml

@@ -15,40 +15,40 @@ init:
 runtime:
   metricsCollectors:
     - kind: StdOut
-      image: docker.io/kubeflowkatib/file-metrics-collector:latest
+      image: ghcr.io/kubeflow/katib/file-metrics-collector:v0.18.0
     - kind: File
-      image: docker.io/kubeflowkatib/file-metrics-collector:latest
+      image: ghcr.io/kubeflow/katib/file-metrics-collector:v0.18.0
     - kind: TensorFlowEvent
-      image: docker.io/kubeflowkatib/tfevent-metrics-collector:latest
+      image: ghcr.io/kubeflow/katib/tfevent-metrics-collector:v0.18.0
       resources:
         limits:
           memory: 1Gi
   suggestions:
     - algorithmName: random
-      image: docker.io/kubeflowkatib/suggestion-hyperopt:latest
+      image: ghcr.io/kubeflow/katib/suggestion-hyperopt:v0.18.0
     - algorithmName: tpe
-      image: docker.io/kubeflowkatib/suggestion-hyperopt:latest
+      image: ghcr.io/kubeflow/katib/suggestion-hyperopt:v0.18.0
     - algorithmName: grid
-      image: docker.io/kubeflowkatib/suggestion-optuna:latest
+      image: ghcr.io/kubeflow/katib/suggestion-optuna:v0.18.0
     - algorithmName: hyperband
-      image: docker.io/kubeflowkatib/suggestion-hyperband:latest
+      image: ghcr.io/kubeflow/katib/suggestion-hyperband:v0.18.0
     - algorithmName: bayesianoptimization
-      image: docker.io/kubeflowkatib/suggestion-skopt:latest
+      image: ghcr.io/kubeflow/katib/suggestion-skopt:v0.18.0
     - algorithmName: cmaes
-      image: docker.io/kubeflowkatib/suggestion-goptuna:latest
+      image: ghcr.io/kubeflow/katib/suggestion-goptuna:v0.18.0
     - algorithmName: sobol
-      image: docker.io/kubeflowkatib/suggestion-goptuna:latest
+      image: ghcr.io/kubeflow/katib/suggestion-goptuna:v0.18.0
     - algorithmName: multivariate-tpe
-      image: docker.io/kubeflowkatib/suggestion-optuna:latest
+      image: ghcr.io/kubeflow/katib/suggestion-optuna:v0.18.0
     - algorithmName: enas
-      image: docker.io/kubeflowkatib/suggestion-enas:latest
+      image: ghcr.io/kubeflow/katib/suggestion-enas:v0.18.0
       resources:
         limits:
           memory: 400Mi
     - algorithmName: darts
-      image: docker.io/kubeflowkatib/suggestion-darts:latest
+      image: ghcr.io/kubeflow/katib/suggestion-darts:v0.18.0
     - algorithmName: pbt
-      image: docker.io/kubeflowkatib/suggestion-pbt:latest
+      image: ghcr.io/kubeflow/katib/suggestion-pbt:v0.18.0
       persistentVolumeClaimSpec:
         accessModes:
           - ReadWriteMany
@@ -57,4 +57,4 @@ runtime:
             storage: 5Gi
   earlyStoppings:
     - algorithmName: medianstop
-      image: docker.io/kubeflowkatib/earlystopping-medianstop:latest
+      image: ghcr.io/kubeflow/katib/earlystopping-medianstop:v0.18.0

manifests/v1beta1/installs/katib-external-db/kustomization.yaml

@@ -16,15 +16,15 @@ resources:
   # Katib webhooks.
   - ../../components/webhook/
 images:
-  - name: docker.io/kubeflowkatib/katib-controller
-    newName: docker.io/kubeflowkatib/katib-controller
-    newTag: latest
-  - name: docker.io/kubeflowkatib/katib-db-manager
-    newName: docker.io/kubeflowkatib/katib-db-manager
-    newTag: latest
-  - name: docker.io/kubeflowkatib/katib-ui
-    newName: docker.io/kubeflowkatib/katib-ui
-    newTag: latest
+  - name: ghcr.io/kubeflow/katib/katib-controller
+    newName: ghcr.io/kubeflow/katib/katib-controller
+    newTag: v0.18.0
+  - name: ghcr.io/kubeflow/katib/katib-db-manager
+    newName: ghcr.io/kubeflow/katib/katib-db-manager
+    newTag: v0.18.0
+  - name: ghcr.io/kubeflow/katib/katib-ui
+    newName: ghcr.io/kubeflow/katib/katib-ui
+    newTag: v0.18.0
 patchesStrategicMerge:
   - patches/db-manager.yaml
 # Modify katib-mysql-secrets with parameters for the DB.

manifests/v1beta1/installs/katib-leader-election/katib-config.yaml

@@ -16,40 +16,40 @@ init:
 runtime:
   metricsCollectors:
     - kind: StdOut
-      image: docker.io/kubeflowkatib/file-metrics-collector:latest
+      image: ghcr.io/kubeflow/katib/file-metrics-collector:v0.18.0
     - kind: File
-      image: docker.io/kubeflowkatib/file-metrics-collector:latest
+      image: ghcr.io/kubeflow/katib/file-metrics-collector:v0.18.0
     - kind: TensorFlowEvent
-      image: docker.io/kubeflowkatib/tfevent-metrics-collector:latest
+      image: ghcr.io/kubeflow/katib/tfevent-metrics-collector:v0.18.0
       resources:
         limits:
           memory: 1Gi
   suggestions:
     - algorithmName: random
-      image: docker.io/kubeflowkatib/suggestion-hyperopt:latest
+      image: ghcr.io/kubeflow/katib/suggestion-hyperopt:v0.18.0
     - algorithmName: tpe
-      image: docker.io/kubeflowkatib/suggestion-hyperopt:latest
+      image: ghcr.io/kubeflow/katib/suggestion-hyperopt:v0.18.0
     - algorithmName: grid
-      image: docker.io/kubeflowkatib/suggestion-optuna:latest
+      image: ghcr.io/kubeflow/katib/suggestion-optuna:v0.18.0
     - algorithmName: hyperband
-      image: docker.io/kubeflowkatib/suggestion-hyperband:latest
+      image: ghcr.io/kubeflow/katib/suggestion-hyperband:v0.18.0
     - algorithmName: bayesianoptimization
-      image: docker.io/kubeflowkatib/suggestion-skopt:latest
+      image: ghcr.io/kubeflow/katib/suggestion-skopt:v0.18.0
     - algorithmName: cmaes
-      image: docker.io/kubeflowkatib/suggestion-goptuna:latest
+      image: ghcr.io/kubeflow/katib/suggestion-goptuna:v0.18.0
     - algorithmName: sobol
-      image: docker.io/kubeflowkatib/suggestion-goptuna:latest
+      image: ghcr.io/kubeflow/katib/suggestion-goptuna:v0.18.0
     - algorithmName: multivariate-tpe
-      image: docker.io/kubeflowkatib/suggestion-optuna:latest
+      image: ghcr.io/kubeflow/katib/suggestion-optuna:v0.18.0
     - algorithmName: enas
-      image: docker.io/kubeflowkatib/suggestion-enas:latest
+      image: ghcr.io/kubeflow/katib/suggestion-enas:v0.18.0
       resources:
         limits:
           memory: 400Mi
     - algorithmName: darts
-      image: docker.io/kubeflowkatib/suggestion-darts:latest
+      image: ghcr.io/kubeflow/katib/suggestion-darts:v0.18.0
     - algorithmName: pbt
-      image: docker.io/kubeflowkatib/suggestion-pbt:latest
+      image: ghcr.io/kubeflow/katib/suggestion-pbt:v0.18.0
       persistentVolumeClaimSpec:
         accessModes:
           - ReadWriteMany
@@ -58,4 +58,4 @@ runtime:
             storage: 5Gi
   earlyStoppings:
     - algorithmName: medianstop
-      image: docker.io/kubeflowkatib/earlystopping-medianstop:latest
+      image: ghcr.io/kubeflow/katib/earlystopping-medianstop:v0.18.0

manifests/v1beta1/installs/katib-openshift/katib-config.yaml

@@ -13,40 +13,40 @@ init:
 runtime:
   metricsCollectors:
     - kind: StdOut
-      image: docker.io/kubeflowkatib/file-metrics-collector:latest
+      image: ghcr.io/kubeflow/katib/file-metrics-collector:v0.18.0
     - kind: File
-      image: docker.io/kubeflowkatib/file-metrics-collector:latest
+      image: ghcr.io/kubeflow/katib/file-metrics-collector:v0.18.0
     - kind: TensorFlowEvent
-      image: docker.io/kubeflowkatib/tfevent-metrics-collector:latest
+      image: ghcr.io/kubeflow/katib/tfevent-metrics-collector:v0.18.0
       resources:
         limits:
           memory: 1Gi
   suggestions:
     - algorithmName: random
-      image: docker.io/kubeflowkatib/suggestion-hyperopt:latest
+      image: ghcr.io/kubeflow/katib/suggestion-hyperopt:v0.18.0
     - algorithmName: tpe
-      image: docker.io/kubeflowkatib/suggestion-hyperopt:latest
+      image: ghcr.io/kubeflow/katib/suggestion-hyperopt:v0.18.0
     - algorithmName: grid
-      image: docker.io/kubeflowkatib/suggestion-optuna:latest
+      image: ghcr.io/kubeflow/katib/suggestion-optuna:v0.18.0
     - algorithmName: hyperband
-      image: docker.io/kubeflowkatib/suggestion-hyperband:latest
+      image: ghcr.io/kubeflow/katib/suggestion-hyperband:v0.18.0
     - algorithmName: bayesianoptimization
-      image: docker.io/kubeflowkatib/suggestion-skopt:latest
+      image: ghcr.io/kubeflow/katib/suggestion-skopt:v0.18.0
     - algorithmName: cmaes
-      image: docker.io/kubeflowkatib/suggestion-goptuna:latest
+      image: ghcr.io/kubeflow/katib/suggestion-goptuna:v0.18.0
     - algorithmName: sobol
-      image: docker.io/kubeflowkatib/suggestion-goptuna:latest
+      image: ghcr.io/kubeflow/katib/suggestion-goptuna:v0.18.0
     - algorithmName: multivariate-tpe
-      image: docker.io/kubeflowkatib/suggestion-optuna:latest
+      image: ghcr.io/kubeflow/katib/suggestion-optuna:v0.18.0
     - algorithmName: enas
-      image: docker.io/kubeflowkatib/suggestion-enas:latest
+      image: ghcr.io/kubeflow/katib/suggestion-enas:v0.18.0
       resources:
         limits:
           memory: 400Mi
     - algorithmName: darts
-      image: docker.io/kubeflowkatib/suggestion-darts:latest
+      image: ghcr.io/kubeflow/katib/suggestion-darts:v0.18.0
     - algorithmName: pbt
-      image: docker.io/kubeflowkatib/suggestion-pbt:latest
+      image: ghcr.io/kubeflow/katib/suggestion-pbt:v0.18.0
       persistentVolumeClaimSpec:
         accessModes:
           - ReadWriteMany
@@ -55,4 +55,4 @@ runtime:
             storage: 5Gi
   earlyStoppings:
     - algorithmName: medianstop
-      image: docker.io/kubeflowkatib/earlystopping-medianstop:latest
+      image: ghcr.io/kubeflow/katib/earlystopping-medianstop:v0.18.0

manifests/v1beta1/installs/katib-openshift/kustomization.yaml

@@ -28,15 +28,15 @@ resources:
   # Katib webhooks.
   - ../../components/webhook/
 images:
-  - name: docker.io/kubeflowkatib/katib-controller
-    newName: docker.io/kubeflowkatib/katib-controller
-    newTag: latest
-  - name: docker.io/kubeflowkatib/katib-db-manager
-    newName: docker.io/kubeflowkatib/katib-db-manager
-    newTag: latest
-  - name: docker.io/kubeflowkatib/katib-ui
-    newName: docker.io/kubeflowkatib/katib-ui
-    newTag: latest
+  - name: ghcr.io/kubeflow/katib/katib-controller
+    newName: ghcr.io/kubeflow/katib/katib-controller
+    newTag: v0.18.0
+  - name: ghcr.io/kubeflow/katib/katib-db-manager
+    newName: ghcr.io/kubeflow/katib/katib-db-manager
+    newTag: v0.18.0
+  - name: ghcr.io/kubeflow/katib/katib-ui
+    newName: ghcr.io/kubeflow/katib/katib-ui
+    newTag: v0.18.0
 
 patchesJson6902:
   # Annotate Service to delegate TLS-secret generation to OpenShift service controller

manifests/v1beta1/installs/katib-standalone-postgres/katib-config.yaml

@@ -15,40 +15,40 @@ init:
 runtime:
   metricsCollectors:
     - kind: StdOut
-      image: docker.io/kubeflowkatib/file-metrics-collector:latest
+      image: ghcr.io/kubeflow/katib/file-metrics-collector:v0.18.0
     - kind: File
-      image: docker.io/kubeflowkatib/file-metrics-collector:latest
+      image: ghcr.io/kubeflow/katib/file-metrics-collector:v0.18.0
     - kind: TensorFlowEvent
-      image: docker.io/kubeflowkatib/tfevent-metrics-collector:latest
+      image: ghcr.io/kubeflow/katib/tfevent-metrics-collector:v0.18.0
       resources:
         limits:
           memory: 1Gi
   suggestions:
     - algorithmName: random
-      image: docker.io/kubeflowkatib/suggestion-hyperopt:latest
+      image: ghcr.io/kubeflow/katib/suggestion-hyperopt:v0.18.0
     - algorithmName: tpe
-      image: docker.io/kubeflowkatib/suggestion-hyperopt:latest
+      image: ghcr.io/kubeflow/katib/suggestion-hyperopt:v0.18.0
     - algorithmName: grid
-      image: docker.io/kubeflowkatib/suggestion-optuna:latest
+      image: ghcr.io/kubeflow/katib/suggestion-optuna:v0.18.0
     - algorithmName: hyperband
-      image: docker.io/kubeflowkatib/suggestion-hyperband:latest
+      image: ghcr.io/kubeflow/katib/suggestion-hyperband:v0.18.0
     - algorithmName: bayesianoptimization
-      image: docker.io/kubeflowkatib/suggestion-skopt:latest
+      image: ghcr.io/kubeflow/katib/suggestion-skopt:v0.18.0
     - algorithmName: cmaes
-      image: docker.io/kubeflowkatib/suggestion-goptuna:latest
+      image: ghcr.io/kubeflow/katib/suggestion-goptuna:v0.18.0
     - algorithmName: sobol
-      image: docker.io/kubeflowkatib/suggestion-goptuna:latest
+      image: ghcr.io/kubeflow/katib/suggestion-goptuna:v0.18.0
     - algorithmName: multivariate-tpe
-      image: docker.io/kubeflowkatib/suggestion-optuna:latest
+      image: ghcr.io/kubeflow/katib/suggestion-optuna:v0.18.0
     - algorithmName: enas
-      image: docker.io/kubeflowkatib/suggestion-enas:latest
+      image: ghcr.io/kubeflow/katib/suggestion-enas:v0.18.0
       resources:
         limits:
           memory: 400Mi
     - algorithmName: darts
-      image: docker.io/kubeflowkatib/suggestion-darts:latest
+      image: ghcr.io/kubeflow/katib/suggestion-darts:v0.18.0
     - algorithmName: pbt
-      image: docker.io/kubeflowkatib/suggestion-pbt:latest
+      image: ghcr.io/kubeflow/katib/suggestion-pbt:v0.18.0
       persistentVolumeClaimSpec:
         accessModes:
           - ReadWriteMany
@@ -57,4 +57,4 @@ runtime:
             storage: 5Gi
   earlyStoppings:
     - algorithmName: medianstop
-      image: docker.io/kubeflowkatib/earlystopping-medianstop:latest
+      image: ghcr.io/kubeflow/katib/earlystopping-medianstop:v0.18.0

manifests/v1beta1/installs/katib-standalone-postgres/kustomization.yaml

@@ -18,15 +18,15 @@ resources:
   # Katib webhooks.
   - ../../components/webhook/
 images:
-  - name: docker.io/kubeflowkatib/katib-controller
-    newName: docker.io/kubeflowkatib/katib-controller
-    newTag: latest
-  - name: docker.io/kubeflowkatib/katib-db-manager
-    newName: docker.io/kubeflowkatib/katib-db-manager
-    newTag: latest
-  - name: docker.io/kubeflowkatib/katib-ui
-    newName: docker.io/kubeflowkatib/katib-ui
-    newTag: latest
+  - name: ghcr.io/kubeflow/katib/katib-controller
+    newName: ghcr.io/kubeflow/katib/katib-controller
+    newTag: v0.18.0
+  - name: ghcr.io/kubeflow/katib/katib-db-manager
+    newName: ghcr.io/kubeflow/katib/katib-db-manager
+    newTag: v0.18.0
+  - name: ghcr.io/kubeflow/katib/katib-ui
+    newName: ghcr.io/kubeflow/katib/katib-ui
+    newTag: v0.18.0
 patchesJson6902:
   - target:
       group: apps

manifests/v1beta1/installs/katib-standalone/katib-config.yaml

@@ -1,4 +1,3 @@
----
 apiVersion: config.kubeflow.org/v1beta1
 kind: KatibConfig
 init:
@@ -15,40 +14,40 @@ init:
 runtime:
   metricsCollectors:
     - kind: StdOut
-      image: docker.io/kubeflowkatib/file-metrics-collector:latest
+      image: ghcr.io/kubeflow/katib/file-metrics-collector:v0.18.0
     - kind: File
-      image: docker.io/kubeflowkatib/file-metrics-collector:latest
+      image: ghcr.io/kubeflow/katib/file-metrics-collector:v0.18.0
     - kind: TensorFlowEvent
-      image: docker.io/kubeflowkatib/tfevent-metrics-collector:latest
+      image: ghcr.io/kubeflow/katib/tfevent-metrics-collector:v0.18.0
       resources:
         limits:
           memory: 1Gi
   suggestions:
     - algorithmName: random
-      image: docker.io/kubeflowkatib/suggestion-hyperopt:latest
+      image: ghcr.io/kubeflow/katib/suggestion-hyperopt:v0.18.0
     - algorithmName: tpe
-      image: docker.io/kubeflowkatib/suggestion-hyperopt:latest
+      image: ghcr.io/kubeflow/katib/suggestion-hyperopt:v0.18.0
     - algorithmName: grid
-      image: docker.io/kubeflowkatib/suggestion-optuna:latest
+      image: ghcr.io/kubeflow/katib/suggestion-optuna:v0.18.0
     - algorithmName: hyperband
-      image: docker.io/kubeflowkatib/suggestion-hyperband:latest
+      image: ghcr.io/kubeflow/katib/suggestion-hyperband:v0.18.0
     - algorithmName: bayesianoptimization
-      image: docker.io/kubeflowkatib/suggestion-skopt:latest
+      image: ghcr.io/kubeflow/katib/suggestion-skopt:v0.18.0
     - algorithmName: cmaes
-      image: docker.io/kubeflowkatib/suggestion-goptuna:latest
+      image: ghcr.io/kubeflow/katib/suggestion-goptuna:v0.18.0
     - algorithmName: sobol
-      image: docker.io/kubeflowkatib/suggestion-goptuna:latest
+      image: ghcr.io/kubeflow/katib/suggestion-goptuna:v0.18.0
     - algorithmName: multivariate-tpe
-      image: docker.io/kubeflowkatib/suggestion-optuna:latest
+      image: ghcr.io/kubeflow/katib/suggestion-optuna:v0.18.0
     - algorithmName: enas
-      image: docker.io/kubeflowkatib/suggestion-enas:latest
+      image: ghcr.io/kubeflow/katib/suggestion-enas:v0.18.0
       resources:
         limits:
           memory: 400Mi
     - algorithmName: darts
-      image: docker.io/kubeflowkatib/suggestion-darts:latest
+      image: ghcr.io/kubeflow/katib/suggestion-darts:v0.18.0
     - algorithmName: pbt
-      image: docker.io/kubeflowkatib/suggestion-pbt:latest
+      image: ghcr.io/kubeflow/katib/suggestion-pbt:v0.18.0
       persistentVolumeClaimSpec:
         accessModes:
           - ReadWriteMany
@@ -57,4 +56,4 @@ runtime:
             storage: 5Gi
   earlyStoppings:
     - algorithmName: medianstop
-      image: docker.io/kubeflowkatib/earlystopping-medianstop:latest
+      image: ghcr.io/kubeflow/katib/earlystopping-medianstop:v0.18.0

manifests/v1beta1/installs/katib-standalone/kustomization.yaml

@@ -18,15 +18,15 @@ resources:
   # Katib webhooks.
   - ../../components/webhook/
 images:
-  - name: docker.io/kubeflowkatib/katib-controller
-    newName: docker.io/kubeflowkatib/katib-controller
-    newTag: latest
-  - name: docker.io/kubeflowkatib/katib-db-manager
-    newName: docker.io/kubeflowkatib/katib-db-manager
-    newTag: latest
-  - name: docker.io/kubeflowkatib/katib-ui
-    newName: docker.io/kubeflowkatib/katib-ui
-    newTag: latest
+  - name: ghcr.io/kubeflow/katib/katib-controller
+    newName: ghcr.io/kubeflow/katib/katib-controller
+    newTag: v0.18.0
+  - name: ghcr.io/kubeflow/katib/katib-db-manager
+    newName: ghcr.io/kubeflow/katib/katib-db-manager
+    newTag: v0.18.0
+  - name: ghcr.io/kubeflow/katib/katib-ui
+    newName: ghcr.io/kubeflow/katib/katib-ui
+    newTag: v0.18.0
 configMapGenerator:
   - name: katib-config
     behavior: create

manifests/v1beta1/installs/katib-with-kubeflow/kustomization.yaml

@@ -9,15 +9,15 @@ resources:
   - ui-virtual-service.yaml
   - istio-authorizationpolicy.yaml
 images:
-  - name: docker.io/kubeflowkatib/katib-controller
-    newName: docker.io/kubeflowkatib/katib-controller
-    newTag: latest
-  - name: docker.io/kubeflowkatib/katib-db-manager
-    newName: docker.io/kubeflowkatib/katib-db-manager
-    newTag: latest
-  - name: docker.io/kubeflowkatib/katib-ui
-    newName: docker.io/kubeflowkatib/katib-ui
-    newTag: latest
+  - name: ghcr.io/kubeflow/katib/katib-controller
+    newName: ghcr.io/kubeflow/katib/katib-controller
+    newTag: v0.18.0
+  - name: ghcr.io/kubeflow/katib/katib-db-manager
+    newName: ghcr.io/kubeflow/katib/katib-db-manager
+    newTag: v0.18.0
+  - name: ghcr.io/kubeflow/katib/katib-ui
+    newName: ghcr.io/kubeflow/katib/katib-ui
+    newTag: v0.18.0
 
 patchesStrategicMerge:
   - patches/remove-namespace.yaml

pkg/controller.v1beta1/experiment/experiment_controller_test.go

@@ -483,7 +483,7 @@ func newFakeInstance() *experimentsv1beta1.Experiment {
 					Containers: []corev1.Container{
 						{
 							Name:  primaryContainer,
-							Image: "docker.io/kubeflowkatib/pytorch-mnist-cpu",
+							Image: "ghcr.io/kubeflow/katib/pytorch-mnist-cpu",
 							Command: []string{
 								"python3",
 								"/opt/pytorch-mnist/mnist.py",
@@ -619,7 +619,7 @@ func newFakeBatchJob() *batchv1.Job {
 					Containers: []corev1.Container{
 						{
 							Name:  primaryContainer,
-							Image: "docker.io/kubeflowkatib/pytorch-mnist-cpu",
+							Image: "ghcr.io/kubeflow/katib/pytorch-mnist-cpu",
 							Command: []string{
 								"python3",
 								"/opt/pytorch-mnist/mnist.py",

pkg/controller.v1beta1/experiment/manifest/generator_test.go

@@ -61,7 +61,7 @@ func TestGetRunSpecWithHP(t *testing.T) {
 					Containers: []v1.Container{
 						{
 							Name:  "training-container",
-							Image: "docker.io/kubeflowkatib/pytorch-mnist-cpu",
+							Image: "ghcr.io/kubeflow/katib/pytorch-mnist-cpu",
 							Command: []string{
 								"python3",
 								"/opt/pytorch-mnist/mnist.py",
@@ -170,7 +170,7 @@ spec:
     spec:
       containers:
         - name: training-container
-          image: docker.io/kubeflowkatib/pytorch-mnist-cpu
+          image: ghcr.io/kubeflow/katib/pytorch-mnist-cpu
           command:
             - "python3"
             - "/opt/pytorch-mnist/mnist.py"
@@ -186,7 +186,7 @@ spec:
     spec:
       containers:
         - name: training-container
-          image: docker.io/kubeflowkatib/pytorch-mnist-cpu
+          image: ghcr.io/kubeflow/katib/pytorch-mnist-cpu
           command:
             - python3
             - /opt/pytorch-mnist/mnist.py
@@ -207,7 +207,7 @@ spec:
     spec:
       containers:
         - name: training-container
-          image: docker.io/kubeflowkatib/pytorch-mnist-cpu
+          image: ghcr.io/kubeflow/katib/pytorch-mnist-cpu
           command:
             - "python3"
             - "/opt/pytorch-mnist/mnist.py"
@@ -337,7 +337,7 @@ func newFakeInstance() *experimentsv1beta1.Experiment {
 					Containers: []v1.Container{
 						{
 							Name:  "training-container",
-							Image: "docker.io/kubeflowkatib/pytorch-mnist-cpu",
+							Image: "ghcr.io/kubeflow/katib/pytorch-mnist-cpu",
 							Command: []string{
 								"python3",
 								"/opt/pytorch-mnist/mnist.py",

pkg/controller.v1beta1/trial/trial_controller_test.go

@@ -440,7 +440,7 @@ func newFakeTrialBatchJob(mcType commonv1beta1.CollectorKind, trialName string)
 					Containers: []corev1.Container{
 						{
 							Name:  primaryContainer,
-							Image: "docker.io/kubeflowkatib/pytorch-mnist-cpu",
+							Image: "ghcr.io/kubeflow/katib/pytorch-mnist-cpu",
 							Command: []string{
 								"python3",
 								"/opt/pytorch-mnist/mnist.py",

pkg/metricscollector/v1beta1/tfevent-metricscollector/tfevent_loader.py

@@ -30,11 +30,23 @@ import api_pb2
 import rfc3339
 import tensorflow as tf
 from tensorboard.backend.event_processing.event_accumulator import EventAccumulator
-from tensorboard.backend.event_processing.tag_types import TENSORS
+from tensorboard.backend.event_processing.tag_types import SCALARS, TENSORS
 
 from pkg.metricscollector.v1beta1.common import const
 
 
+def _should_consider(tag: str, metric_name: str, tfefile: str) -> bool:
+    tfefile_parent_dir = (
+        os.path.dirname(metric_name)
+        if len(metric_name.split("/")) >= 2
+        else os.path.dirname(tfefile)
+    )
+    basedir_name = os.path.dirname(tfefile)
+    return tag.startswith(metric_name.split("/")[-1]) and basedir_name.endswith(
+        tfefile_parent_dir
+    )
+
+
 class TFEventFileParser:
     def __init__(self, metric_names):
         self.metric_names = metric_names
@@ -47,31 +59,36 @@ class TFEventFileParser:
 
     def parse_summary(self, tfefile):
         metric_logs = []
-        event_accumulator = EventAccumulator(tfefile, size_guidance={TENSORS: 0})
+        event_accumulator = EventAccumulator(
+            tfefile, size_guidance={SCALARS: 0, TENSORS: 0}
+        )
         event_accumulator.Reload()
-        for tag in event_accumulator.Tags()[TENSORS]:
+        tags = event_accumulator.Tags()
+        for tag in tags[TENSORS]:
             for m in self.metric_names:
-                tfefile_parent_dir = (
-                    os.path.dirname(m)
-                    if len(m.split("/")) >= 2
-                    else os.path.dirname(tfefile)
-                )
-                basedir_name = os.path.dirname(tfefile)
-                if not tag.startswith(m.split("/")[-1]) or not basedir_name.endswith(
-                    tfefile_parent_dir
-                ):
-                    continue
-
-                for tensor in event_accumulator.Tensors(tag):
-                    ml = api_pb2.MetricLog(
-                        time_stamp=rfc3339.rfc3339(
-                            datetime.fromtimestamp(tensor.wall_time)
-                        ),
-                        metric=api_pb2.Metric(
-                            name=m, value=str(tf.make_ndarray(tensor.tensor_proto))
-                        ),
-                    )
-                    metric_logs.append(ml)
+                if _should_consider(tag, m, tfefile):
+                    for tensor in event_accumulator.Tensors(tag):
+                        ml = api_pb2.MetricLog(
+                            time_stamp=rfc3339.rfc3339(
+                                datetime.fromtimestamp(tensor.wall_time)
+                            ),
+                            metric=api_pb2.Metric(
+                                name=m, value=str(tf.make_ndarray(tensor.tensor_proto))
+                            ),
+                        )
+                        metric_logs.append(ml)
+        # support old-style tensorboard metrics too
+        for tag in tags[SCALARS]:
+            for m in self.metric_names:
+                if _should_consider(tag, m, tfefile):
+                    for scalar in event_accumulator.Scalars(tag):
+                        ml = api_pb2.MetricLog(
+                            time_stamp=rfc3339.rfc3339(
+                                datetime.fromtimestamp(scalar.wall_time)
+                            ),
+                            metric=api_pb2.Metric(name=m, value=str(scalar.value)),
+                        )
+                        metric_logs.append(ml)
 
         return metric_logs
 

pkg/ui/v1beta1/frontend/cypress/fixtures/trial-template.json

@@ -8,15 +8,15 @@
           "Templates": [
             {
               "Path": "defaultTrialTemplate.yaml",
-              "Yaml": "apiVersion: batch/v1\nkind: Job\nspec:\n  template:\n    spec:\n      containers:\n        - name: training-container\n          image: docker.io/kubeflowkatib/pytorch-mnist:v1beta1-45c5727\n          command:\n            - \"python3\"\n            - \"/opt/pytorch-mnist/mnist.py\"\n            - \"--epochs=1\"\n            - \"--lr=${trialParameters.learningRate}\"\n            - \"--momentum=${trialParameters.momentum}\"\n      restartPolicy: Never"
+              "Yaml": "apiVersion: batch/v1\nkind: Job\nspec:\n  template:\n    spec:\n      containers:\n        - name: training-container\n          image: ghcr.io/kubeflow/katib/pytorch-mnist:v1beta1-45c5727\n          command:\n            - \"python3\"\n            - \"/opt/pytorch-mnist/mnist.py\"\n            - \"--epochs=1\"\n            - \"--lr=${trialParameters.learningRate}\"\n            - \"--momentum=${trialParameters.momentum}\"\n      restartPolicy: Never"
             },
             {
               "Path": "enasCPUTemplate",
-              "Yaml": "apiVersion: batch/v1\nkind: Job\nspec:\n  template:\n    spec:\n      containers:\n        - name: training-container\n          image: docker.io/kubeflowkatib/enas-cnn-cifar10-cpu:v1beta1-45c5727\n          command:\n            - python3\n            - -u\n            - RunTrial.py\n            - --num_epochs=1\n            - \"--architecture=\\\"${trialParameters.neuralNetworkArchitecture}\\\"\"\n            - \"--nn_config=\\\"${trialParameters.neuralNetworkConfig}\\\"\"\n      restartPolicy: Never"
+              "Yaml": "apiVersion: batch/v1\nkind: Job\nspec:\n  template:\n    spec:\n      containers:\n        - name: training-container\n          image: ghcr.io/kubeflow/katib/enas-cnn-cifar10-cpu:v1beta1-45c5727\n          command:\n            - python3\n            - -u\n            - RunTrial.py\n            - --num_epochs=1\n            - \"--architecture=\\\"${trialParameters.neuralNetworkArchitecture}\\\"\"\n            - \"--nn_config=\\\"${trialParameters.neuralNetworkConfig}\\\"\"\n      restartPolicy: Never"
             },
             {
               "Path": "pytorchJobTemplate",
-              "Yaml": "apiVersion: \"kubeflow.org/v1\"\nkind: PyTorchJob\nspec:\n  pytorchReplicaSpecs:\n    Master:\n      replicas: 1\n      restartPolicy: OnFailure\n      template:\n        spec:\n          containers:\n            - name: pytorch\n              image: docker.io/kubeflowkatib/pytorch-mnist:v1beta1-45c5727\n              imagePullPolicy: Always\n              command:\n                - \"python3\"\n                - \"/opt/pytorch-mnist/mnist.py\"\n                - \"--epochs=1\"\n                - \"--lr=${trialParameters.learningRate}\"\n                - \"--momentum=${trialParameters.momentum}\"\n    Worker:\n      replicas: 2\n      restartPolicy: OnFailure\n      template:\n        spec:\n          containers:\n            - name: pytorch\n              image: docker.io/kubeflowkatib/pytorch-mnist:v1beta1-45c5727\n              imagePullPolicy: Always\n              command:\n                - \"python3\"\n                - \"/opt/pytorch-mnist/mnist.py\"\n                - \"--epochs=1\"\n                - \"--lr=${trialParameters.learningRate}\"\n                - \"--momentum=${trialParameters.momentum}\""
+              "Yaml": "apiVersion: \"kubeflow.org/v1\"\nkind: PyTorchJob\nspec:\n  pytorchReplicaSpecs:\n    Master:\n      replicas: 1\n      restartPolicy: OnFailure\n      template:\n        spec:\n          containers:\n            - name: pytorch\n              image: ghcr.io/kubeflow/katib/pytorch-mnist:v1beta1-45c5727\n              imagePullPolicy: Always\n              command:\n                - \"python3\"\n                - \"/opt/pytorch-mnist/mnist.py\"\n                - \"--epochs=1\"\n                - \"--lr=${trialParameters.learningRate}\"\n                - \"--momentum=${trialParameters.momentum}\"\n    Worker:\n      replicas: 2\n      restartPolicy: OnFailure\n      template:\n        spec:\n          containers:\n            - name: pytorch\n              image: ghcr.io/kubeflow/katib/pytorch-mnist:v1beta1-45c5727\n              imagePullPolicy: Always\n              command:\n                - \"python3\"\n                - \"/opt/pytorch-mnist/mnist.py\"\n                - \"--epochs=1\"\n                - \"--lr=${trialParameters.learningRate}\"\n                - \"--momentum=${trialParameters.momentum}\""
             }
           ]
         }

pkg/util/v1beta1/katibconfig/config_test.go

@@ -407,7 +407,7 @@ init:
 runtime:
   suggestions:
   - algorithmName: random
-    image: docker.io/kubeflowkatib/suggestion-hyperopt:latest
+    image: ghcr.io/kubeflow/katib/suggestion-hyperopt:latest
 `), os.FileMode(0600)); err != nil {
 		t.Fatal(err)
 	}

pkg/webhook/v1beta1/experiment/validator/validator_test.go

@@ -1433,7 +1433,7 @@ func newFakeBatchJob() *batchv1.Job {
 					Containers: []v1.Container{
 						{
 							Name:  "training-container",
-							Image: "docker.io/kubeflowkatib/pytorch-mnist-cpu",
+							Image: "ghcr.io/kubeflow/katib/pytorch-mnist-cpu",
 							Command: []string{
 								"python3",
 								"--epochs=1",

scripts/v1beta1/release.sh

@@ -68,7 +68,7 @@ fi
 
 # ------------------ Change image tag ------------------
 # Change Katib image tags to the new release tag.
-make update-images OLD_PREFIX="docker.io/kubeflowkatib/" NEW_PREFIX="docker.io/kubeflowkatib/" TAG="${TAG}"
+make update-images OLD_PREFIX="ghcr.io/kubeflow/katib/" NEW_PREFIX="ghcr.io/kubeflow/katib/" TAG="${TAG}"
 
 # ------------------ Publish Katib SDK ------------------
 # Remove first "v" for the SDK version.

scripts/v1beta1/update-images.sh

@@ -28,8 +28,8 @@
 # 5. Katib Trial training containers
 #
 # Run ./scripts/v1beta1/update-images.sh <OLD_PREFIX> <NEW_PREFIX> <TAG> to execute it.
-# For example, to update images from: docker.io/kubeflowkatib/ to: docker.io/private/ registry with tag: v0.12.0, run:
-# ./scripts/v1beta1/update-images.sh docker.io/kubeflowkatib/ docker.io/private/ v0.12.0
+# For example, to update images from: ghcr.io/kubeflow/katib/ to: ghcr.io/private/ registry with tag: v0.12.0, run:
+# ./scripts/v1beta1/update-images.sh ghcr.io/kubeflow/katib/ ghcr.io/private/ v0.12.0
 
 set -o errexit
 set -o pipefail
@@ -42,8 +42,8 @@ TAG=${3:-""}
 if [[ -z "$OLD_PREFIX" || -z "$NEW_PREFIX" || -z "$TAG" ]]; then
   echo "Image old prefix, new prefix, and tag must be set"
   echo -e "Usage: $0 <OLD_PREFIX> <NEW_PREFIX> <TAG>\n" 1>&2
-  echo "For example, to update images from: docker.io/kubeflowkatib/ to: docker.io/private/ registry with tag: v0.12.0, run:"
-  echo "$0 docker.io/kubeflowkatib/ docker.io/private/ v0.12.0"
+  echo "For example, to update images from: ghcr.io/kubeflow/katib/ to: ghcr.io/private/ registry with tag: v0.12.0, run:"
+  echo "$0 ghcr.io/kubeflow/katib/ ghcr.io/private/ v0.12.0"
   exit 1
 fi
 

sdk/python/v1beta1/kubeflow/katib/api/katib_client_test.py

@@ -100,7 +100,7 @@ def generate_trial_template() -> V1beta1TrialTemplate:
                     "containers": [
                         {
                             "name": "training-container",
-                            "image": "docker.io/kubeflowkatib/pytorch-mnist-cpu:v0.14.0",
+                            "image": "ghcr.io/kubeflow/katib/pytorch-mnist-cpu:v0.14.0",
                             "command": [
                                 "python3",
                                 "/opt/pytorch-mnist/mnist.py",

sdk/python/v1beta1/setup.py

@@ -56,7 +56,7 @@ if os.path.exists(katib_grpc_svc_file):
 
 setuptools.setup(
     name="kubeflow-katib",
-    version="0.17.0",
+    version="0.18.0",
     author="Kubeflow Authors",
     author_email="premnath.vel@gmail.com",
     license="Apache License Version 2.0",

test/e2e/v1beta1/scripts/aws/setup-katib.sh

@@ -32,7 +32,7 @@ kubectl version
 kubectl cluster-info
 
 # Update Katib images with the current PULL SHA.
-make update-images OLD_PREFIX="docker.io/kubeflowkatib/" NEW_PREFIX="${ECR_REGISTRY}/${REPO_NAME}/v1beta1/" TAG="${PULL_PULL_SHA}"
+make update-images OLD_PREFIX="ghcr.io/kubeflow/katib/" NEW_PREFIX="${ECR_REGISTRY}/${REPO_NAME}/v1beta1/" TAG="${PULL_PULL_SHA}"
 
 echo -e "\n The Katib will be deployed with the following configs"
 cat "manifests/v1beta1/installs/katib-standalone/kustomization.yaml"

test/e2e/v1beta1/scripts/gh-actions/build-load.sh

@@ -30,7 +30,7 @@ TUNE_API=${2:-false}
 TRIAL_IMAGES=${3:-""}
 EXPERIMENTS=${4:-""}
 
-REGISTRY="docker.io/kubeflowkatib"
+REGISTRY="ghcr.io/kubeflow/katib"
 TAG="e2e-test"
 VERSION="v1beta1"
 CMD_PREFIX="cmd"

test/e2e/v1beta1/scripts/gh-actions/setup-katib.sh

@@ -30,7 +30,7 @@ TRAINING_OPERATOR_VERSION="v1.9.0"
 echo "Start to install Katib"
 
 # Update Katib images with `e2e-test`.
-cd ../../../../../ && make update-images OLD_PREFIX="docker.io/kubeflowkatib/" NEW_PREFIX="docker.io/kubeflowkatib/" TAG="$E2E_TEST_IMAGE_TAG" && cd -
+cd ../../../../../ && make update-images OLD_PREFIX="ghcr.io/kubeflow/katib/" NEW_PREFIX="ghcr.io/kubeflow/katib/" TAG="$E2E_TEST_IMAGE_TAG" && cd -
 
 # first declare the which kustomization file to use, by default use mysql.
 KUSTOMIZATION_FILE="../../../../../manifests/v1beta1/installs/katib-standalone/kustomization.yaml"

test/e2e/v1beta1/testdata/invalid-experiment.yaml

@@ -40,7 +40,7 @@ spec:
           spec:
             containers:
               - name: training-container
-                image: docker.io/kubeflowkatib/pytorch-mnist-cpu:latest
+                image: ghcr.io/kubeflow/katib/pytorch-mnist-cpu:v0.18.0
                 command:
                   - "python3"
                   - "/opt/pytorch-mnist/mnist.py"

test/e2e/v1beta1/testdata/valid-experiment.yaml

@@ -40,7 +40,7 @@ spec:
           spec:
             containers:
               - name: training-container
-                image: docker.io/kubeflowkatib/pytorch-mnist-cpu:latest
+                image: ghcr.io/kubeflow/katib/pytorch-mnist-cpu:v0.18.0
                 command:
                   - "python3"
                   - "/opt/pytorch-mnist/mnist.py"

test/unit/v1beta1/metricscollector/test_tfevent_metricscollector.py

@@ -13,10 +13,19 @@
 # limitations under the License.
 
 import os
+import tempfile
 import unittest
 
+import tensorboardX
 import utils
 
+METRIC_DIR_NAMES = ("train", "test")
+METRIC_NAMES = ("accuracy", "loss")
+QUALIFIED_METRIC_NAMES = tuple(
+    f"{dir}/{metric}"
+    for dir in METRIC_DIR_NAMES
+    for metric in METRIC_NAMES
+)
 
 class TestTFEventMetricsCollector(unittest.TestCase):
     def test_parse_file(self):
@@ -24,24 +33,47 @@ class TestTFEventMetricsCollector(unittest.TestCase):
         current_dir = os.path.dirname(os.path.abspath(__file__))
         logs_dir = os.path.join(current_dir, "testdata/tfevent-metricscollector/logs")
 
-        # Metric format is "{{dirname}}/{{metrics name}}"
-        metric_names = ["train/accuracy", "train/loss", "test/loss", "test/accuracy"]
-        metric_logs = utils.get_metric_logs(logs_dir, metric_names)
+
+        metric_logs = utils.get_metric_logs(logs_dir, QUALIFIED_METRIC_NAMES)
         self.assertEqual(20, len(metric_logs))
 
         for log in metric_logs:
             actual = log["metric"]["name"]
-            self.assertIn(actual, metric_names)
+            self.assertIn(actual, QUALIFIED_METRIC_NAMES)
 
-        # Metric format is "{{metrics name}}"
-        metric_names = ["accuracy", "loss"]
-        metrics_file_dir = os.path.join(logs_dir, "train")
-        metric_logs = utils.get_metric_logs(metrics_file_dir, metric_names)
-        self.assertEqual(10, len(metric_logs))
+        train_metric_logs = utils.get_metric_logs(
+            os.path.join(logs_dir, "train"), METRIC_NAMES)
+        self.assertEqual(10, len(train_metric_logs))
+
+        for log in train_metric_logs:
+            actual = log["metric"]["name"]
+            self.assertIn(actual, METRIC_NAMES)
+
+    def test_parse_file_with_tensorboardX(self):
+        logs_dir = tempfile.mkdtemp()
+        num_iters = 3
+
+        for dir_name in METRIC_DIR_NAMES:
+            with tensorboardX.SummaryWriter(os.path.join(logs_dir, dir_name)) as writer:
+                for metric_name in METRIC_NAMES:
+                    for iter in range(num_iters):
+                        writer.add_scalar(metric_name, 0.1, iter)
+
+
+        metric_logs = utils.get_metric_logs(logs_dir, QUALIFIED_METRIC_NAMES)
+        self.assertEqual(num_iters * len(QUALIFIED_METRIC_NAMES), len(metric_logs))
 
         for log in metric_logs:
             actual = log["metric"]["name"]
-            self.assertIn(actual, metric_names)
+            self.assertIn(actual, QUALIFIED_METRIC_NAMES)
+
+        train_metric_logs = utils.get_metric_logs(
+            os.path.join(logs_dir, "train"), METRIC_NAMES)
+        self.assertEqual(num_iters * len(METRIC_NAMES), len(train_metric_logs))
+
+        for log in train_metric_logs:
+            actual = log["metric"]["name"]
+            self.assertIn(actual, METRIC_NAMES)
 
 
 if __name__ == '__main__':

test/unit/v1beta1/requirements.txt

@@ -1,3 +1,4 @@
 grpcio-testing==1.64.1
 pytest==7.2.0
+tensorboardX==2.6.2.2
 kubeflow-training[huggingface]==1.9.0