Update azure-gpu-test.yml (#1669)

Author: rasbt

.github/azure-gpu-test-with-thunder.yml

@@ -0,0 +1,72 @@
+name: GPU tests with Thunder
+
+trigger:
+  branches:
+    include:
+      - "main"
+      - "wip"
+
+pr:
+  branches:
+    include:
+      - "main"
+      - "wip"
+
+jobs:
+  - job: testing
+    timeoutInMinutes: "30"
+    cancelTimeoutInMinutes: "2"
+    pool: "lit-rtx-3090"
+    variables:
+      DEVICES: $( python -c 'print("$(Agent.Name)".split("_")[-1])' )
+      CI: "true"
+    container:
+      image: "pytorchlightning/pytorch_lightning:base-cuda-py3.10-torch2.2-cuda12.1.0"
+      options: "--gpus=all --shm-size=8gb"
+    workspace:
+      clean: all
+    steps:
+
+    - bash: |
+        echo "##vso[task.setvariable variable=CUDA_VISIBLE_DEVICES]$(DEVICES)"
+      displayName: 'set env. vars'
+
+    - bash: |
+        echo $(DEVICES)
+        echo $CUDA_VISIBLE_DEVICES
+        whereis nvidia
+        nvidia-smi
+        which python && which pip
+        python --version
+        pip --version
+        pip list
+      displayName: "Image info & NVIDIA"
+
+    - script: |
+        pip install --upgrade pip
+        pip install '.[all,test]'
+      displayName: 'Install dependencies'
+
+    - script: |
+        pip uninstall -y torchvision torchaudio
+        pip install --pre 'nvfuser-cu121[torch]' --extra-index-url https://pypi.nvidia.com
+      displayName: 'Install PyTorch nightly'
+
+    - bash: |
+        set -e
+        pip list
+        python -c "import torch ; mgpu = torch.cuda.device_count() ; assert mgpu == 2, f'GPU: {mgpu}'"
+      displayName: "Env details"
+
+    - bash: pytest -v --disable-pytest-warnings --strict-markers --color=yes
+      displayName: 'Ordinary tests'
+      env:
+        PL_RUN_CUDA_TESTS: "1"
+      timeoutInMinutes: "5"
+
+    - bash: bash run_standalone_tests.sh
+      workingDirectory: tests
+      env:
+        PL_RUN_CUDA_TESTS: "1"
+      displayName: "Standalone tests"
+      timeoutInMinutes: "10"

.github/azure-gpu-test.yml

@@ -1,3 +1,5 @@
+name: GPU tests
+
 trigger:
   branches:
     include:
@@ -43,20 +45,16 @@ jobs:
     - script: |
         pip install --upgrade pip
         pip install '.[all,test]'
+        pip install -U torch torchvision torchaudio
       displayName: 'Install dependencies'
 
-    - script: |
-        pip uninstall -y torchvision torchaudio
-        pip install --pre 'nvfuser-cu121[torch]' --extra-index-url https://pypi.nvidia.com
-      displayName: 'Install PyTorch nightly'
-
     - bash: |
         set -e
         pip list
         python -c "import torch ; mgpu = torch.cuda.device_count() ; assert mgpu == 2, f'GPU: {mgpu}'"
       displayName: "Env details"
 
-    - bash: pytest -v --disable-pytest-warnings --strict-markers --color=yes
+    - bash: pytest -v --disable-pytest-warnings --strict-markers --color=yes --ignore-glob="tests/test_thunder*" --ignore="tests/test_unsloth_executor.py"
       displayName: 'Ordinary tests'
       env:
         PL_RUN_CUDA_TESTS: "1"

.github/workflows/cpu-tests.yml

@@ -4,7 +4,7 @@ on:
   push:
     branches: [main, wip]
   pull_request:
-    branches: [main, "carmocca/*", wip]
+    branches: [main, wip]
 
 concurrency:
   group: ${{ github.workflow }}-${{ github.ref }}-${{ github.head_ref }}