Update Lightning Lite examples

.azure/gpu-tests-lite.yml

@@ -20,6 +20,8 @@ pr:
       - "release/*"
   paths:
     include:
+      - "examples/lite/**"
+      - "examples/run_lite_examples.sh"
       - ".azure/gpu-tests-lite.yml"
       - "requirements/lite/**"
       - "src/lightning_lite/**"
@@ -64,7 +66,7 @@ jobs:
     - bash: |
         set -e
         CUDA_VERSION_MM=$(python -c "import torch ; print(''.join(map(str, torch.version.cuda.split('.')[:2])))")
-        pip install -e .[dev,strategies] --find-links https://download.pytorch.org/whl/cu${CUDA_VERSION_MM}/torch_stable.html
+        pip install -e .[dev,strategies,examples] --find-links https://download.pytorch.org/whl/cu${CUDA_VERSION_MM}/torch_stable.html
         pip list
       env:
         PACKAGE_NAME: "lite"
@@ -108,3 +110,12 @@ jobs:
         testResultsFiles: '$(Build.StagingDirectory)/test-results.xml'
         testRunTitle: '$(Agent.OS) - $(Build.DefinitionName) - Python $(python.version)'
       condition: succeededOrFailed()
+
+    - script: |
+        # In order to run the examples, we need to substitute the meta package imports with the standalone package
+        python ../.actions/assistant.py copy_replace_imports --source_dir="./lite" --source_import="lightning.lite" --target_import="lightning_lite.lite"
+        set -e
+        bash run_lite_examples.sh --accelerator=cuda --devices=1
+        bash run_lite_examples.sh --accelerator=cuda --devices=2 --strategy ddp
+      workingDirectory: examples
+      displayName: 'Testing: Lite examples'

.azure/gpu-tests-pytorch.yml

@@ -21,12 +21,9 @@ pr:
   paths:
     include:
       - ".azure/gpu-tests-pytorch.yml"
-      - "examples/run_ddp_examples.sh"
-      - "examples/convert_from_pt_to_pl/**"
       - "examples/run_pl_examples.sh"
       - "examples/pl_basics/backbone_image_classifier.py"
       - "examples/pl_basics/autoencoder.py"
-      - "examples/pl_loops/mnist_lite.py"
       - "examples/pl_fault_tolerant/automatic.py"
       - "examples/test_pl_examples.py"
       - "examples/pl_integrations/dali_image_classifier.py"
@@ -169,7 +166,6 @@ jobs:
 
     - script: |
         set -e
-        bash run_ddp_examples.sh
         bash run_pl_examples.sh --trainer.accelerator=gpu --trainer.devices=1
         bash run_pl_examples.sh --trainer.accelerator=gpu --trainer.devices=2 --trainer.strategy=ddp
         bash run_pl_examples.sh --trainer.accelerator=gpu --trainer.devices=2 --trainer.strategy=ddp --trainer.precision=16

.github/checkgroup.yml

@@ -9,12 +9,9 @@ subprojects:
   - id: "pytorch_lightning"
     paths:
       # all examples don't need to be added because they aren't used in CI, but these are
-      - "examples/run_ddp_examples.sh"
-      - "examples/convert_from_pt_to_pl/**"
       - "examples/run_pl_examples.sh"
       - "examples/pl_basics/backbone_image_classifier.py"
       - "examples/pl_basics/autoencoder.py"
-      - "examples/pl_loops/mnist_lite.py"
       - "examples/pl_fault_tolerant/automatic.py"
       - "examples/test_pl_examples.py"
       - "examples/pl_integrations/dali_image_classifier.py"
@@ -212,6 +209,8 @@ subprojects:
 
   - id: "lightning_lite: Tests"
     paths:
+      - "examples/lite/**"
+      - "examples/run_lite_examples.sh"
       - "tests/tests_lite/**"
     checks:
       - "lite-cpu (macOS-11, lite, 3.9, 1.11)"

examples/README.md

@@ -11,23 +11,19 @@ ______________________________________________________________________
 
 ______________________________________________________________________
 
-## MNIST Examples
+## Lightning Lite Examples
 
-5 MNIST examples showing how to gradually convert from pure PyTorch to PyTorch Lightning.
+We show how to accelerate your PyTorch code with [Lightning Lite](https://pytorch-lightning.readthedocs.io/en/latest/starter/lightning_lite.html) with minimal code changes.
+You stay in full control of the training loop.
 
-The transition through [LightningLite](https://pytorch-lightning.readthedocs.io/en/latest/starter/lightning_lite.html) from pure PyTorch is optional, but it might be helpful to learn about it.
-
-- [MNIST with vanilla PyTorch](convert_from_pt_to_pl/image_classifier_1_pytorch.py)
-- [MNIST with LightningLite](convert_from_pt_to_pl/image_classifier_2_lite.py)
-- [MNIST LightningLite to LightningModule](convert_from_pt_to_pl/image_classifier_3_lite_to_lightning_module.py)
-- [MNIST with LightningModule](convert_from_pt_to_pl/image_classifier_4_lightning_module.py)
-- [MNIST with LightningModule + LightningDataModule](convert_from_pt_to_pl/image_classifier_5_lightning_datamodule.py)
+- [MNIST with vanilla PyTorch](lite/image_classifier_1_pytorch.py)
+- [MNIST with Lightning Lite](lite/image_classifier_2_lite.py)
 
 ______________________________________________________________________
 
-## Basic Examples
+## Lightning Trainer Examples
 
-In this folder, we have 2 simple examples:
+In this folder, we have 2 simple examples that showcase the power of the Lightning Trainer.
 
 - [Image Classifier](pl_basics/backbone_image_classifier.py) (trains arbitrary datasets with arbitrary backbones).
 - [Image Classifier + DALI](convert_from_pt_to_pl/image_classifier_4_dali.py) (defines the model inside the `LightningModule`).