Merge branch 'master' into cgebbe-wandb-all-checkpoints

Author: Borda

.actions/assistant.py

@@ -18,10 +18,11 @@
 import shutil
 import tempfile
 import urllib.request
+from collections.abc import Iterable, Iterator, Sequence
 from itertools import chain
 from os.path import dirname, isfile
 from pathlib import Path
-from typing import Any, Dict, Iterable, Iterator, List, Optional, Sequence, Tuple
+from typing import Any, Optional
 
 from packaging.requirements import Requirement
 from packaging.version import Version
@@ -127,7 +128,7 @@ def _parse_requirements(lines: Iterable[str]) -> Iterator[_RequirementWithCommen
         pip_argument = None
 
 
-def load_requirements(path_dir: str, file_name: str = "base.txt", unfreeze: str = "all") -> List[str]:
+def load_requirements(path_dir: str, file_name: str = "base.txt", unfreeze: str = "all") -> list[str]:
     """Loading requirements from a file.
 
     >>> path_req = os.path.join(_PROJECT_ROOT, "requirements")
@@ -222,7 +223,7 @@ def _load_aggregate_requirements(req_dir: str = "requirements", freeze_requireme
         fp.writelines([ln + os.linesep for ln in requires] + [os.linesep])
 
 
-def _retrieve_files(directory: str, *ext: str) -> List[str]:
+def _retrieve_files(directory: str, *ext: str) -> list[str]:
     all_files = []
     for root, _, files in os.walk(directory):
         for fname in files:
@@ -232,7 +233,7 @@ def _retrieve_files(directory: str, *ext: str) -> List[str]:
     return all_files
 
 
-def _replace_imports(lines: List[str], mapping: List[Tuple[str, str]], lightning_by: str = "") -> List[str]:
+def _replace_imports(lines: list[str], mapping: list[tuple[str, str]], lightning_by: str = "") -> list[str]:
     """Replace imports of standalone package to lightning.
 
     >>> lns = [
@@ -320,7 +321,7 @@ def copy_replace_imports(
             fo.writelines(lines)
 
 
-def create_mirror_package(source_dir: str, package_mapping: Dict[str, str]) -> None:
+def create_mirror_package(source_dir: str, package_mapping: dict[str, str]) -> None:
     """Create a mirror package with adjusted imports."""
     # replace imports and copy the code
     mapping = package_mapping.copy()
@@ -480,6 +481,25 @@ def convert_version2nightly(ver_file: str = "src/version.info") -> None:
         with open(ver_file, "w") as fo:
             fo.write(version + os.linesep)
 
+    @staticmethod
+    def generate_docker_tags(
+        release_version: str,
+        python_version: str,
+        torch_version: str,
+        cuda_version: str,
+        docker_project: str = "pytorchlightning/pytorch_lightning",
+        add_latest: bool = False,
+    ) -> None:
+        """Generate docker tags for the given versions."""
+        tags = [f"latest-py{python_version}-torch{torch_version}-cuda{cuda_version}"]
+        if release_version:
+            tags += [f"{release_version}-py{python_version}-torch{torch_version}-cuda{cuda_version}"]
+        if add_latest:
+            tags += ["latest"]
+
+        tags = [f"{docker_project}:{tag}" for tag in tags]
+        print(",".join(tags))
+
 
 if __name__ == "__main__":
     import jsonargparse

.actions/requirements.txt

@@ -1,3 +1,3 @@
-jsonargparse >=4.16.0, <4.28.0
+jsonargparse >=4.16.0, <=4.35.0
 requests
 packaging

.azure/gpu-benchmarks.yml

@@ -46,7 +46,7 @@ jobs:
     variables:
       DEVICES: $( python -c 'print("$(Agent.Name)".split("_")[-1])' )
     container:
-      image: "pytorchlightning/pytorch_lightning:base-cuda-py3.12-torch2.4-cuda12.1.0"
+      image: "pytorchlightning/pytorch_lightning:base-cuda-py3.12-torch2.5-cuda12.1.0"
       options: "--gpus=all --shm-size=32g"
     strategy:
       matrix:
@@ -75,7 +75,9 @@ jobs:
           pip list
         displayName: "Image info & NVIDIA"
 
-      - bash: pip install -e .[dev] --find-links ${TORCH_URL}
+      - bash: |
+          pip install -e .[dev] --find-links ${TORCH_URL}
+          pip install setuptools==75.6.0 jsonargparse==4.35.0
         env:
           FREEZE_REQUIREMENTS: "1"
         displayName: "Install package"
@@ -106,5 +108,6 @@ jobs:
         condition: and(succeeded(), eq(variables['PACKAGE_NAME'], 'fabric'))
         env:
           PL_RUN_CUDA_TESTS: "1"
+          PL_RUN_STANDALONE_TESTS: "1"
         displayName: "Testing: fabric standalone tasks"
         timeoutInMinutes: "10"

.azure/gpu-tests-fabric.yml

@@ -56,11 +56,14 @@ jobs:
       options: "--gpus=all --shm-size=2gb  -v /var/tmp:/var/tmp"
     strategy:
       matrix:
+        "Fabric | oldest":
+          image: "pytorchlightning/pytorch_lightning:base-cuda-py3.10-torch2.1-cuda12.1.1"
+          PACKAGE_NAME: "fabric"
         "Fabric | latest":
-          image: "pytorchlightning/pytorch_lightning:base-cuda-py3.11-torch2.3-cuda12.1.0"
+          image: "pytorchlightning/pytorch_lightning:base-cuda-py3.12-torch2.6-cuda12.4.1"
           PACKAGE_NAME: "fabric"
         "Lightning | latest":
-          image: "pytorchlightning/pytorch_lightning:base-cuda-py3.12-torch2.4-cuda12.1.0"
+          image: "pytorchlightning/pytorch_lightning:base-cuda-py3.12-torch2.6-cuda12.4.1"
           PACKAGE_NAME: "lightning"
     workspace:
       clean: all
@@ -77,9 +80,8 @@ jobs:
         displayName: "set env. vars"
       - bash: |
           echo "##vso[task.setvariable variable=TORCH_URL]https://download.pytorch.org/whl/test/cu${CUDA_VERSION_MM}"
-          echo "##vso[task.setvariable variable=TORCHVISION_URL]https://download.pytorch.org/whl/test/cu124/torchvision-0.19.0%2Bcu124-cp${PYTHON_VERSION_MM}-cp${PYTHON_VERSION_MM}-linux_x86_64.whl"
         condition: endsWith(variables['Agent.JobName'], 'future')
-        displayName: "set env. vars 4 future"
+        displayName: "extend env. vars 4 future"
 
       - bash: |
           echo $(DEVICES)
@@ -105,14 +107,17 @@ jobs:
         displayName: "Adjust dependencies"
 
       - bash: |
+          set -e
           extra=$(python -c "print({'lightning': 'fabric-'}.get('$(PACKAGE_NAME)', ''))")
-          pip install -e ".[${extra}dev]" pytest-timeout -U --find-links="${TORCH_URL}" --find-links="${TORCHVISION_URL}"
+          pip install -e ".[${extra}dev]" pytest-timeout -U --extra-index-url="${TORCH_URL}"
+          pip install setuptools==75.6.0 jsonargparse==4.35.0
         displayName: "Install package & dependencies"
 
       - bash: |
           set -e
           python requirements/collect_env_details.py
           python -c "import torch ; mgpu = torch.cuda.device_count() ; assert mgpu == 2, f'GPU: {mgpu}'"
+          python requirements/pytorch/check-avail-extras.py
           python -c "import bitsandbytes"
         displayName: "Env details"
 
@@ -134,15 +139,17 @@ jobs:
         condition: and(succeeded(), eq(variables['PACKAGE_NAME'], 'fabric'))
         displayName: "Adjust tests & examples"
 
-      - bash: python -m coverage run --source ${COVERAGE_SOURCE} -m pytest . -v --durations=50
-        workingDirectory: tests/tests_fabric/
+      - bash: python -m coverage run --source ${COVERAGE_SOURCE} -m pytest tests_fabric/ -v --durations=50
+        workingDirectory: tests/
         displayName: "Testing: fabric standard"
         timeoutInMinutes: "10"
 
-      - bash: bash ../run_standalone_tests.sh "."
-        workingDirectory: tests/tests_fabric/
+      - bash: |
+          wget https://raw.githubusercontent.com/Lightning-AI/utilities/main/scripts/run_standalone_tests.sh
+          bash ./run_standalone_tests.sh "tests_fabric"
+        workingDirectory: tests/
         env:
-          PL_STANDALONE_TESTS_SOURCE: $(COVERAGE_SOURCE)
+          PL_RUN_STANDALONE_TESTS: "1"
         displayName: "Testing: fabric standalone"
         timeoutInMinutes: "10"
 
@@ -157,7 +164,7 @@ jobs:
           ./codecov --token=$(CODECOV_TOKEN) --commit=$(Build.SourceVersion) \
             --flags=gpu,pytest,${COVERAGE_SOURCE} --name="GPU-coverage" --env=linux,azure
           ls -l
-        workingDirectory: tests/tests_fabric/
+        workingDirectory: tests/
         displayName: "Statistics"
 
       - script: |

.azure/gpu-tests-pytorch.yml

@@ -49,11 +49,14 @@ jobs:
     cancelTimeoutInMinutes: "2"
     strategy:
       matrix:
+        "PyTorch | oldest":
+          image: "pytorchlightning/pytorch_lightning:base-cuda-py3.10-torch2.1-cuda12.1.1"
+          PACKAGE_NAME: "pytorch"
         "PyTorch | latest":
-          image: "pytorchlightning/pytorch_lightning:base-cuda-py3.11-torch2.3-cuda12.1.0"
+          image: "pytorchlightning/pytorch_lightning:base-cuda-py3.12-torch2.6-cuda12.4.1"
           PACKAGE_NAME: "pytorch"
         "Lightning | latest":
-          image: "pytorchlightning/pytorch_lightning:base-cuda-py3.12-torch2.4-cuda12.1.0"
+          image: "pytorchlightning/pytorch_lightning:base-cuda-py3.12-torch2.6-cuda12.4.1"
           PACKAGE_NAME: "lightning"
     pool: lit-rtx-3090
     variables:
@@ -81,9 +84,8 @@ jobs:
         displayName: "set env. vars"
       - bash: |
           echo "##vso[task.setvariable variable=TORCH_URL]https://download.pytorch.org/whl/test/cu${CUDA_VERSION_MM}"
-          echo "##vso[task.setvariable variable=TORCHVISION_URL]https://download.pytorch.org/whl/test/cu124/torchvision-0.19.0%2Bcu124-cp${PYTHON_VERSION_MM}-cp${PYTHON_VERSION_MM}-linux_x86_64.whl"
         condition: endsWith(variables['Agent.JobName'], 'future')
-        displayName: "set env. vars 4 future"
+        displayName: "extend env. vars 4 future"
 
       - bash: |
           echo $(DEVICES)
@@ -109,8 +111,10 @@ jobs:
         displayName: "Adjust dependencies"
 
       - bash: |
+          set -e
           extra=$(python -c "print({'lightning': 'pytorch-'}.get('$(PACKAGE_NAME)', ''))")
-          pip install -e ".[${extra}dev]" pytest-timeout -U --find-links="${TORCH_URL}" --find-links="${TORCHVISION_URL}"
+          pip install -e ".[${extra}dev]" pytest-timeout -U --extra-index-url="${TORCH_URL}"
+          pip install setuptools==75.6.0 jsonargparse==4.35.0
         displayName: "Install package & dependencies"
 
       - bash: pip uninstall -y lightning
@@ -155,16 +159,18 @@ jobs:
           ls -l checkpoints/
         displayName: "Get legacy checkpoints"
 
-      - bash: python -m coverage run --source ${COVERAGE_SOURCE} -m pytest -v --durations=50
-        workingDirectory: tests/tests_pytorch
+      - bash: python -m coverage run --source ${COVERAGE_SOURCE} -m pytest tests_pytorch/ -v --durations=50
+        workingDirectory: tests/
         displayName: "Testing: PyTorch standard"
         timeoutInMinutes: "35"
 
-      - bash: bash ../run_standalone_tests.sh "."
-        workingDirectory: tests/tests_pytorch
+      - bash: |
+          wget https://raw.githubusercontent.com/Lightning-AI/utilities/main/scripts/run_standalone_tests.sh
+          bash ./run_standalone_tests.sh "tests_pytorch"
+        workingDirectory: tests/
         env:
           PL_USE_MOCKED_MNIST: "1"
-          PL_STANDALONE_TESTS_SOURCE: $(COVERAGE_SOURCE)
+          PL_RUN_STANDALONE_TESTS: "1"
         displayName: "Testing: PyTorch standalone tests"
         timeoutInMinutes: "35"
 

.github/CODEOWNERS

@@ -5,39 +5,16 @@
 # the repo. Unless a later match takes precedence,
 # @global-owner1 and @global-owner2 will be requested for
 # review when someone opens a pull request.
-* @lantiga @borda @tchaton @awaelchli @justusschock
-
-# CI/CD and configs
-/.actions/                  @borda @ethanwharris @justusschock
-/.github/                   @borda @ethanwharris @justusschock
-/.azure/                    @borda @ethanwharris @justusschock
-/dockers/                   @borda @ethanwharris @justusschock
-*.yml                       @borda @ethanwharris @justusschock
+* @lantiga @borda @tchaton @justusschock @ethanwharris
 
 # Docs
-/docs/                                      @lantiga @borda @awaelchli
-/docs/*/conf.py                             @borda @awaelchli
 /.github/*.md                               @williamfalcon @lantiga @borda
-/.github/ISSUE_TEMPLATE/                    @borda @tchaton @awaelchli
-/docs/source-fabric/conf.py                 @borda @awaelchli
-/docs/source-fabric/index.rst               @awaelchli @lantiga
-/docs/source-pytorch/conf.py                @borda @awaelchli
+/docs/source-fabric/index.rst               @williamfalcon @lantiga
 /docs/source-pytorch/index.rst              @williamfalcon @lantiga
 /docs/source-pytorch/levels                 @williamfalcon @lantiga
 
-# PyTorch Lightning
-/src/lightning/pytorch                      @lantiga @borda @tchaton @awaelchli @justusschock
-
-# Lightning Data
-/src/lightning/data/                        @tchaton @lantiga
-
-# Lightning Fabric
-/src/lightning/fabric                       @lantiga @borda @tchaton @awaelchli @justusschock
-
 /.github/CODEOWNERS                  @williamfalcon
 /SECURITY.md                         @williamfalcon @lantiga
 /README.md                           @williamfalcon @lantiga
-/setup.py                            @williamfalcon @borda
-/src/pytorch_lightning/__about__.py  @williamfalcon @borda
-/src/lightning_fabric/__about__.py   @williamfalcon @borda @awaelchli
-/src/*/__setup__.py                  @borda @justusschock
+/src/pytorch_lightning/__about__.py  @williamfalcon @lantiga @borda
+/src/lightning_fabric/__about__.py   @williamfalcon @lantiga @borda

.github/CONTRIBUTING.md

@@ -182,14 +182,14 @@ We welcome any useful contribution! For your convenience here's a recommended wo
 1. Use tags in PR name for the following cases:
 
    - **\[blocked by #<number>\]** if your work is dependent on other PRs.
-   - **\[wip\]** when you start to re-edit your work, mark it so no one will accidentally merge it in meantime.
+   - **[wip]** when you start to re-edit your work, mark it so no one will accidentally merge it in meantime.
 
 ### Question & Answer
 
 #### How can I help/contribute?
 
 All types of contributions are welcome - reporting bugs, fixing documentation, adding test cases, solving issues, and preparing bug fixes.
-To get started with code contributions, look for issues marked with the label [good first issue](https://github.com/Lightning-AI/lightning/issues?q=is%3Aopen+is%3Aissue+label%3A%22good+first+issue%22) or chose something close to your domain with the label [help wanted](https://github.com/Lightning-AI/lightning/issues?q=is%3Aopen+is%3Aissue+label%3A%22help+wanted%22). Before coding, make sure that the issue description is clear and comment on the issue so that we can assign it to you (or simply self-assign if you can).
+To get started with code contributions, look for issues marked with the label [good first issue](https://github.com/Lightning-AI/pytorch-lightning/issues?q=is%3Aopen+is%3Aissue+label%3A%22good+first+issue%22) or chose something close to your domain with the label [help wanted](https://github.com/Lightning-AI/pytorch-lightning/issues?q=is%3Aopen+is%3Aissue+label%3A%22help+wanted%22). Before coding, make sure that the issue description is clear and comment on the issue so that we can assign it to you (or simply self-assign if you can).
 
 #### Is there a recommendation for branch names?
 

.github/ISSUE_TEMPLATE/1_bug_report.yaml

@@ -37,6 +37,7 @@ body:
         - "v2.2"
         - "v2.3"
         - "v2.4"
+        - "v2.5"
         - "master"
     validations:
       required: true
@@ -91,8 +92,8 @@ body:
           <summary>Current environment</summary>
 
         ```
-        #- PyTorch Lightning Version (e.g., 2.4.0):
-        #- PyTorch Version (e.g., 2.4):
+        #- PyTorch Lightning Version (e.g., 2.5.0):
+        #- PyTorch Version (e.g., 2.5):
         #- Python version (e.g., 3.12):
         #- OS (e.g., Linux):
         #- CUDA/cuDNN version:

.github/actions/pip-wheels/action.yml

@@ -46,8 +46,8 @@ runs:
       run: |
         # cat requirements.dump
         pip wheel -r requirements.dump --prefer-binary \
-          --wheel-dir=.wheels \
-          -f ${{ inputs.torch-url }} -f ${{ inputs.wheel-dir }}
+          --wheel-dir=".wheels" \
+          --extra-index-url=${{ inputs.torch-url }} -f ${{ inputs.wheel-dir }}
         ls -lh .wheels/
       shell: bash
 

.github/advanced-issue-labeler.yml

@@ -16,5 +16,7 @@ policy:
             keys: ["v2_3", "v2.3", "2.3.x"]
           - name: "ver: 2.4.x"
             keys: ["v2_4", "v2.4", "2.4.x"]
-          - name: "ver: 2.4.x"
+          - name: "ver: 2.5.x"
+            keys: ["v2_5", "v2.5", "2.5.x"]
+          - name: "ver: 2.5.x"
             keys: ["master"]