Revert "Revert "Docker automation fix""

mlcommons · Jul 27, 2023 · 9ef40d2 · 9ef40d2
1 parent db37df4
commit 9ef40d2
Show file tree

Hide file tree

Showing 56 changed files with 6,669 additions and 1,702 deletions.
diff --git a/.github/workflows/CI.yml b/.github/workflows/CI.yml
@@ -6,11 +6,13 @@ jobs:
   fastmri:
     runs-on: ubuntu-latest
     steps:
-    - uses: actions/checkout@v2
+    - uses: actions/checkout@v3
     - name: Set up Python 3.9
-      uses: actions/setup-python@v2
+      uses: actions/setup-python@v4
       with:
         python-version: 3.9
+        cache: 'pip' # Cache pip dependencies\.
+        cache-dependency-path: '**/setup.py'
     - name: Install Modules and Run
       run: |
         pip install .[jax_cpu]
@@ -22,11 +24,13 @@ jobs:
   wmt_jax:
     runs-on: ubuntu-latest
     steps:
-    - uses: actions/checkout@v2
+    - uses: actions/checkout@v3
     - name: Set up Python 3.9
-      uses: actions/setup-python@v2
+      uses: actions/setup-python@v4
       with:
         python-version: 3.9
+        cache: 'pip' # Cache pip dependencies\.
+        cache-dependency-path: '**/setup.py'
     - name: Install Modules and Run
       run: |
         pip install .[jax_cpu]
@@ -37,11 +41,13 @@ jobs:
   wmt_pytorch:
     runs-on: ubuntu-latest
     steps:
-    - uses: actions/checkout@v2
+    - uses: actions/checkout@v3
     - name: Set up Python 3.9
-      uses: actions/setup-python@v2
+      uses: actions/setup-python@v4
       with:
         python-version: 3.9
+        cache: 'pip' # Cache pip dependencies\.
+        cache-dependency-path: '**/setup.py'
     - name: Install Modules and Run
       run: |
         pip install .[jax_cpu]
@@ -52,11 +58,13 @@ jobs:
   imagenet_jax:
     runs-on: ubuntu-latest
     steps:
-    - uses: actions/checkout@v2
+    - uses: actions/checkout@v3
     - name: Set up Python 3.9
-      uses: actions/setup-python@v2
+      uses: actions/setup-python@v4
       with:
         python-version: 3.9
+        cache: 'pip' # Cache pip dependencies\.
+        cache-dependency-path: '**/setup.py'
     - name: Install Modules and Run
       run: |
         pip install .[jax_cpu]
@@ -68,11 +76,13 @@ jobs:
   imagenet_pytorch:
     runs-on: ubuntu-latest
     steps:
-    - uses: actions/checkout@v2
+    - uses: actions/checkout@v3
     - name: Set up Python 3.9
-      uses: actions/setup-python@v2
+      uses: actions/setup-python@v4
       with:
         python-version: 3.9
+        cache: 'pip' # Cache pip dependencies\.
+        cache-dependency-path: '**/setup.py'
     - name: Install Modules and Run
       run: |
         pip install .[jax_cpu]
@@ -85,11 +95,13 @@ jobs:
   criteo_jax:
     runs-on: ubuntu-latest
     steps:
-    - uses: actions/checkout@v2
+    - uses: actions/checkout@v3
     - name: Set up Python 3.9
-      uses: actions/setup-python@v2
+      uses: actions/setup-python@v4
       with:
         python-version: 3.9
+        cache: 'pip' # Cache pip dependencies\.
+        cache-dependency-path: '**/setup.py'
     - name: Install Modules and Run
       run: |
         pip install .[jax_cpu]
@@ -100,11 +112,13 @@ jobs:
   criteo_pytorch:
     runs-on: ubuntu-latest
     steps:
-    - uses: actions/checkout@v2
+    - uses: actions/checkout@v3
     - name: Set up Python 3.9
-      uses: actions/setup-python@v2
+      uses: actions/setup-python@v4
       with:
         python-version: 3.9
+        cache: 'pip' # Cache pip dependencies\.
+        cache-dependency-path: '**/setup.py'
     - name: Install Modules and Run
       run: |
         pip install .[jax_cpu]
@@ -115,11 +129,13 @@ jobs:
   speech_jax:
     runs-on: ubuntu-latest
     steps:
-    - uses: actions/checkout@v2
+    - uses: actions/checkout@v3
     - name: Set up Python 3.9
-      uses: actions/setup-python@v2
+      uses: actions/setup-python@v4
       with:
         python-version: 3.9
+        cache: 'pip' # Cache pip dependencies\.
+        cache-dependency-path: '**/setup.py'
     - name: Install Modules and Run
       run: |
         pip install .[jax_cpu]
@@ -131,11 +147,13 @@ jobs:
   speech_pytorch:
     runs-on: ubuntu-latest
     steps:
-    - uses: actions/checkout@v2
+    - uses: actions/checkout@v3
     - name: Set up Python 3.9
-      uses: actions/setup-python@v2
+      uses: actions/setup-python@v4
       with:
         python-version: 3.9
+        cache: 'pip' # Cache pip dependencies\.
+        cache-dependency-path: '**/setup.py'
     - name: Install Modules and Run
       run: |
         pip install .[jax_cpu]
@@ -147,11 +165,13 @@ jobs:
   ogbg:
     runs-on: ubuntu-latest
     steps:
-    - uses: actions/checkout@v2
+    - uses: actions/checkout@v3
     - name: Set up Python 3.9
-      uses: actions/setup-python@v2
+      uses: actions/setup-python@v4
       with:
         python-version: 3.9
+        cache: 'pip' # Cache pip dependencies\.
+        cache-dependency-path: '**/setup.py'
     - name: Install Modules and Run
       run: |
         pip install .[jax_cpu]
@@ -163,22 +183,44 @@ jobs:
   pytest:
     runs-on: ubuntu-latest
     steps:
-    - uses: actions/checkout@v2
+    - uses: actions/checkout@v3
     - name: Set up Python 3.9
-      uses: actions/setup-python@v2
+      uses: actions/setup-python@v4
       with:
         python-version: 3.9
+        cache: 'pip' # Cache pip dependencies\.
+        cache-dependency-path: '**/setup.py'
     - name: Install pytest
       run: |
         python -m pip install --upgrade pip
         pip install pytest
         pip install .[full]
         pip install .[jax_cpu]
         pip install .[pytorch_cpu]
-    - name: Run pytest
+    - name: Run pytest tests
       run: |
         pytest -vx tests/version_test.py
         pytest -vx tests/test_num_params.py
         pytest -vx tests/test_param_shapes.py
         pytest -vx tests/test_param_types.py
-        pytest -vx tests/test_ssim.py
+        pytest -vx tests/test_ssim.py
+  pytest-baselines:
+    runs-on: ubuntu-latest
+    steps:
+    - uses: actions/checkout@v3
+    - name: Set up Python 3.9
+      uses: actions/setup-python@v4
+      with:
+        python-version: 3.9
+        cache: 'pip' # Cache pip dependencies\.
+        cache-dependency-path: '**/setup.py'
+    - name: Install pytest
+      run: |
+        python -m pip install --upgrade pip
+        pip install pytest
+        pip install .[full]
+        pip install .[jax_cpu]
+        pip install .[pytorch_cpu]
+    - name: Run baseline tests
+      run: |
+        pytest --verbosity=1 tests/test_baselines.py
diff --git a/.github/workflows/regression_tests.yml b/.github/workflows/regression_tests.yml
@@ -0,0 +1,136 @@
+name: Containerized Regression Tests
+
+on: 
+  pull_request:
+    branches:
+      - 'main' 
+
+jobs:
+  fastmri_jax:
+    runs-on: self-hosted
+    steps:
+    - uses: actions/checkout@v2
+    - name: Run containerized workload
+      run: |
+        docker pull us-central1-docker.pkg.dev/training-algorithms-external/mlcommons-docker-repo/algoperf_jax_dev 
+        docker run  -v $HOME/data/:/data/ -v $HOME/experiment_runs/:/experiment_runs -v $HOME/experiment_runs/logs:/logs --gpus all --ipc=host us-central1-docker.pkg.dev/training-algorithms-external/mlcommons-docker-repo/algoperf_jax_dev -d fastmri -f jax -s baselines/adamw/jax/submission.py -w fastmri -t baselines/adamw/tuning_search_space.json -e tests/regression_tests/adamw -m 10 -c False -o True -r false
+  imagenet_resnet_jax:
+    runs-on: self-hosted
+    steps:
+    - uses: actions/checkout@v2
+    - name: Run containerized workload
+      run: |
+        docker pull us-central1-docker.pkg.dev/training-algorithms-external/mlcommons-docker-repo/algoperf_jax_dev 
+        docker run  -v $HOME/data/:/data/ -v $HOME/experiment_runs/:/experiment_runs -v $HOME/experiment_runs/logs:/logs --gpus all --ipc=host us-central1-docker.pkg.dev/training-algorithms-external/mlcommons-docker-repo/algoperf_jax_dev -d imagenet -f jax -s baselines/adamw/jax/submission.py -w imagenet_resnet -t baselines/adamw/tuning_search_space.json -e tests/regression_tests/adamw -m 10 -c False -o True -r false
+  imagenet_vit_jax:
+    runs-on: self-hosted
+    steps:
+    - uses: actions/checkout@v2
+    - name: Run containerized workload
+      run: |
+        docker pull us-central1-docker.pkg.dev/training-algorithms-external/mlcommons-docker-repo/algoperf_jax_dev 
+        docker run  -v $HOME/data/:/data/ -v $HOME/experiment_runs/:/experiment_runs -v $HOME/experiment_runs/logs:/logs --gpus all --ipc=host us-central1-docker.pkg.dev/training-algorithms-external/mlcommons-docker-repo/algoperf_jax_dev -d imagenet -f jax -s baselines/adamw/jax/submission.py -w imagenet_vit -t baselines/adamw/tuning_search_space.json -e tests/regression_tests/adamw -m 10 -c False -o True -r false
+  ogbg_jax:
+    runs-on: self-hosted
+    steps:
+    - uses: actions/checkout@v2
+    - name: Run containerized workload
+      run: |
+        docker pull us-central1-docker.pkg.dev/training-algorithms-external/mlcommons-docker-repo/algoperf_jax_dev 
+        docker run  -v $HOME/data/:/data/ -v $HOME/experiment_runs/:/experiment_runs -v $HOME/experiment_runs/logs:/logs --gpus all --ipc=host us-central1-docker.pkg.dev/training-algorithms-external/mlcommons-docker-repo/algoperf_jax_dev -d ogbg -f jax -s baselines/adamw/jax/submission.py -w ogbg -t baselines/adamw/tuning_search_space.json -e tests/regression_tests/adamw -m 10 -c False -o True -r false
+  criteo_jax:
+    runs-on: self-hosted
+    steps:
+    - uses: actions/checkout@v2
+    - name: Run containerized workload
+      run: |
+        docker pull us-central1-docker.pkg.dev/training-algorithms-external/mlcommons-docker-repo/algoperf_jax_dev 
+        docker run  -v $HOME/data/:/data/ -v $HOME/experiment_runs/:/experiment_runs -v $HOME/experiment_runs/logs:/logs --gpus all --ipc=host us-central1-docker.pkg.dev/training-algorithms-external/mlcommons-docker-repo/algoperf_jax_dev -d criteo1tb -f jax -s baselines/adamw/jax/submission.py -w criteo1tb -t baselines/adamw/tuning_search_space.json -e tests/regression_tests/adamw -m 10 -c False -o True -r false
+  librispeech_conformer_jax:
+    runs-on: self-hosted
+    steps:
+    - uses: actions/checkout@v2
+    - name: Run containerized workload
+      run: |
+        docker pull us-central1-docker.pkg.dev/training-algorithms-external/mlcommons-docker-repo/algoperf_jax_dev 
+        docker run  -v $HOME/data/:/data/ -v $HOME/experiment_runs/:/experiment_runs -v $HOME/experiment_runs/logs:/logs --gpus all --ipc=host us-central1-docker.pkg.dev/training-algorithms-external/mlcommons-docker-repo/algoperf_jax_dev -d librispeech -f jax -s baselines/adamw/jax/submission.py -w librispeech_conformer -t baselines/adamw/tuning_search_space.json -e tests/regression_tests/adamw -m 10 -c False -o True -r false
+  librispeech_deepspeech_jax:
+    runs-on: self-hosted
+    steps:
+    - uses: actions/checkout@v2
+    - name: Run containerized workload
+      run: |
+        docker pull us-central1-docker.pkg.dev/training-algorithms-external/mlcommons-docker-repo/algoperf_jax_dev 
+        docker run  -v $HOME/data/:/data/ -v $HOME/experiment_runs/:/experiment_runs -v $HOME/experiment_runs/logs:/logs --gpus all --ipc=host us-central1-docker.pkg.dev/training-algorithms-external/mlcommons-docker-repo/algoperf_jax_dev -d librispeech -f jax -s baselines/adamw/jax/submission.py -w librispeech_deepspeech -t baselines/adamw/tuning_search_space.json -e tests/regression_tests/adamw -m 10 -c False -o True -r false
+  wmt_jax:
+    runs-on: self-hosted
+    steps:
+    - uses: actions/checkout@v2
+    - name: Run containerized workload
+      run: |
+        docker pull us-central1-docker.pkg.dev/training-algorithms-external/mlcommons-docker-repo/algoperf_jax_dev 
+        docker run  -v $HOME/data/:/data/ -v $HOME/experiment_runs/:/experiment_runs -v $HOME/experiment_runs/logs:/logs --gpus all --ipc=host us-central1-docker.pkg.dev/training-algorithms-external/mlcommons-docker-repo/algoperf_jax_dev -d wmt -f jax -s baselines/adamw/jax/submission.py -w wmt -t baselines/adamw/tuning_search_space.json -e tests/regression_tests/adamw -m 10 -c False -o True -r false     
+  fastmri_pytorch:
+    runs-on: self-hosted
+    steps:
+    - uses: actions/checkout@v2
+    - name: Run containerized workload
+      run: |
+        docker pull us-central1-docker.pkg.dev/training-algorithms-external/mlcommons-docker-repo/algoperf_pytorch_dev 
+        docker run  -v $HOME/data/:/data/ -v $HOME/experiment_runs/:/experiment_runs -v $HOME/experiment_runs/logs:/logs --gpus all --ipc=host us-central1-docker.pkg.dev/training-algorithms-external/mlcommons-docker-repo/algoperf_pytorch_dev -d fastmri -f pytorch -s baselines/adamw/pytorch/submission.py -w fastmri -t baselines/adamw/tuning_search_space.json -e tests/regression_tests/adamw -m 10 -c False -o True -r false
+  imagenet_resnet_pytorch:
+    runs-on: self-hosted
+    steps:
+    - uses: actions/checkout@v2
+    - name: Run containerized workload
+      run: |
+        docker pull us-central1-docker.pkg.dev/training-algorithms-external/mlcommons-docker-repo/algoperf_pytorch_dev 
+        docker run  -v $HOME/data/:/data/ -v $HOME/experiment_runs/:/experiment_runs -v $HOME/experiment_runs/logs:/logs --gpus all --ipc=host us-central1-docker.pkg.dev/training-algorithms-external/mlcommons-docker-repo/algoperf_pytorch_dev -d imagenet -f pytorch -s baselines/adamw/pytorch/submission.py -w imagenet_resnet -t baselines/adamw/tuning_search_space.json -e tests/regression_tests/adamw -m 10 -c False -o True -r false
+  imagenet_vit_pytorch:
+    runs-on: self-hosted
+    steps:
+    - uses: actions/checkout@v2
+    - name: Run containerized workload
+      run: |
+        docker pull us-central1-docker.pkg.dev/training-algorithms-external/mlcommons-docker-repo/algoperf_pytorch_dev 
+        docker run  -v $HOME/data/:/data/ -v $HOME/experiment_runs/:/experiment_runs -v $HOME/experiment_runs/logs:/logs --gpus all --ipc=host us-central1-docker.pkg.dev/training-algorithms-external/mlcommons-docker-repo/algoperf_pytorch_dev -d imagenet -f pytorch -s baselines/adamw/pytorch/submission.py -w imagenet_vit -t baselines/adamw/tuning_search_space.json -e tests/regression_tests/adamw -m 10 -c False -o True -r false
+  ogbg_pytorch:
+    runs-on: self-hosted
+    steps:
+    - uses: actions/checkout@v2
+    - name: Run containerized workload
+      run: |
+        docker pull us-central1-docker.pkg.dev/training-algorithms-external/mlcommons-docker-repo/algoperf_pytorch_dev 
+        docker run  -v $HOME/data/:/data/ -v $HOME/experiment_runs/:/experiment_runs -v $HOME/experiment_runs/logs:/logs --gpus all --ipc=host us-central1-docker.pkg.dev/training-algorithms-external/mlcommons-docker-repo/algoperf_pytorch_dev -d ogbg -f pytorch -s baselines/adamw/pytorch/submission.py -w ogbg -t baselines/adamw/tuning_search_space.json -e tests/regression_tests/adamw -m 10 -c False -o True -r false
+  criteo_pytorch:
+    runs-on: self-hosted
+    steps:
+    - uses: actions/checkout@v2
+    - name: Run containerized workload
+      run: |
+        docker pull us-central1-docker.pkg.dev/training-algorithms-external/mlcommons-docker-repo/algoperf_pytorch_dev 
+        docker run  -v $HOME/data/:/data/ -v $HOME/experiment_runs/:/experiment_runs -v $HOME/experiment_runs/logs:/logs --gpus all --ipc=host us-central1-docker.pkg.dev/training-algorithms-external/mlcommons-docker-repo/algoperf_pytorch_dev -d criteo1tb -f pytorch -s baselines/adamw/pytorch/submission.py -w criteo1tb -t baselines/adamw/tuning_search_space.json -e tests/regression_tests/adamw -m 10 -c False -o True -r false
+  librispeech_conformer_pytorch:
+    runs-on: self-hosted
+    steps:
+    - uses: actions/checkout@v2
+    - name: Run containerized workload
+      run: |
+        docker pull us-central1-docker.pkg.dev/training-algorithms-external/mlcommons-docker-repo/algoperf_pytorch_dev 
+        docker run  -v $HOME/data/:/data/ -v $HOME/experiment_runs/:/experiment_runs -v $HOME/experiment_runs/logs:/logs --gpus all --ipc=host us-central1-docker.pkg.dev/training-algorithms-external/mlcommons-docker-repo/algoperf_pytorch_dev -d librispeech -f pytorch -s baselines/adamw/pytorch/submission.py -w librispeech_conformer -t baselines/adamw/tuning_search_space.json -e tests/regression_tests/adamw -m 10 -c False -o True -r false
+  librispeech_deepspeech_pytorch:
+    runs-on: self-hosted
+    steps:
+    - uses: actions/checkout@v2
+    - name: Run containerized workload
+      run: |
+        docker pull us-central1-docker.pkg.dev/training-algorithms-external/mlcommons-docker-repo/algoperf_pytorch_dev 
+        docker run  -v $HOME/data/:/data/ -v $HOME/experiment_runs/:/experiment_runs -v $HOME/experiment_runs/logs:/logs --gpus all --ipc=host us-central1-docker.pkg.dev/training-algorithms-external/mlcommons-docker-repo/algoperf_pytorch_dev -d librispeech -f pytorch -s baselines/adamw/pytorch/submission.py -w librispeech_deepspeech -t baselines/adamw/tuning_search_space.json -e tests/regression_tests/adamw -m 10 -c False -o True -r false
+  wmt_pytorch:
+    runs-on: self-hosted
+    steps:
+    - uses: actions/checkout@v2
+    - name: Run containerized workload
+      run: |
+        docker pull us-central1-docker.pkg.dev/training-algorithms-external/mlcommons-docker-repo/algoperf_pytorch_dev 
+        docker run  -v $HOME/data/:/data/ -v $HOME/experiment_runs/:/experiment_runs -v $HOME/experiment_runs/logs:/logs --gpus all --ipc=host us-central1-docker.pkg.dev/training-algorithms-external/mlcommons-docker-repo/algoperf_pytorch_dev -d wmt -f pytorch -s baselines/adamw/pytorch/submission.py -w wmt -t baselines/adamw/tuning_search_space.json -e tests/regression_tests/adamw -m 10 -c False -o True -r false