Merge branch 'master' into bugfix/3827_test_ddp_error

Author: gianscarpe

.circleci/config.yml

@@ -36,7 +36,7 @@ references:
       command: |
         gcloud --quiet auth configure-docker
         #cd dockers/tpu-tests
-        docker build --tag "$GCR_IMAGE_PATH:$CIRCLE_WORKFLOW_JOB_ID" -f ./dockers/tpu-tests/Dockerfile --build-arg "PYTHON_VERSION=$PYTHON_VER" .
+        docker build --tag "$GCR_IMAGE_PATH:$CIRCLE_WORKFLOW_JOB_ID" -f ./dockers/tpu-tests/Dockerfile --build-arg "PYTHON_VERSION=$PYTHON_VER" --build-arg "PYTORCH_VERSION=$XLA_VER" .
         docker push "$GCR_IMAGE_PATH:$CIRCLE_WORKFLOW_JOB_ID"
 
   deploy_cluster: &deploy_cluster
@@ -45,6 +45,7 @@ references:
      command: |
        go get github.com/google/go-jsonnet/cmd/jsonnet
        export PATH=$PATH:$HOME/go/bin
+       python -c "fname = 'dockers/tpu-tests/tpu_test_cases.jsonnet' ; fff = open(fname).read().replace('pytorch-VERSION', 'pytorch-$XLA_VER') ; open(fname, 'w').write(fff)"
        job_name=$(jsonnet -J ml-testing-accelerators/ dockers/tpu-tests/tpu_test_cases.jsonnet --ext-str image=$GCR_IMAGE_PATH --ext-str image-tag=$CIRCLE_WORKFLOW_JOB_ID | kubectl create -f -)
        job_name=${job_name#job.batch/}
        job_name=${job_name% created}
@@ -102,6 +103,7 @@ jobs:
       - image: circleci/python:3.7
     environment:
       - PYTHON_VER: << parameters.python >>
+      - XLA_VER: 1.7
       - MAX_CHECKS: 240
       - CHECK_SPEEP: 5
     steps:

.drone.yml

@@ -32,9 +32,12 @@ steps:
     - pip --version
     - nvidia-smi
     - pip install -r ./requirements/devel.txt --upgrade-strategy only-if-needed -v --no-cache-dir
+    # when Image has defined CUDa version we can switch to this package spec "nvidia-dali-cuda${CUDA_VERSION%%.*}0"
+    # todo: temprarl fix till https://github.com/PyTorchLightning/pytorch-lightning/pull/4922 is resolved
+    - pip install --extra-index-url https://developer.download.nvidia.com/compute/redist "nvidia-dali-cuda100<0.27" --upgrade-strategy only-if-needed
     - pip list
-    - coverage run --source pytorch_lightning -m pytest pytorch_lightning tests -v --color=yes --durations=25 # --flake8
-    - python -m pytest benchmarks pl_examples -v --color=yes --maxfail=2 --durations=0 # --flake8
+    - coverage run --source pytorch_lightning -m pytest pytorch_lightning tests -v --durations=25 # --flake8
+    - python -m pytest benchmarks pl_examples -v --maxfail=2 --durations=0 # --flake8
     #- cd docs; make doctest; make coverage
     - coverage report
     # see: https://docs.codecov.io/docs/merging-reports

.github/CODEOWNERS

@@ -12,4 +12,16 @@
 /tests/metrics/*              @teddykoker @ananyahjha93 @justusschock
 /docs/source/metrics.rst      @teddykoker @ananyahjha93 @justusschock
 
+# API
+/pytorch_lightning/callbacks/base.py  @williamfalcon
+/pytorch_lightning/core/datamodule.py @williamfalcon
+/pytorch_lightning/trainer/trainer.py @williamfalcon
+/pytorch_lightning/core/hooks.py     @williamfalcon
+/pytorch_lightning/core/lightning.py  @williamfalcon
 
+
+# accelerators
+/pytorch_lightning/accelerators/*  @williamfalcon
+
+# owners
+/pytorch_lightning/.github/CODEOWNERS @williamfalcon

.github/CONTRIBUTING.md

@@ -112,8 +112,8 @@ In case you adding new dependencies, make sure that they are compatible with the
 
 ### Coding Style
 
-1. Use f-strings for output formation (except logging when we stay with lazy `logging.info("Hello %s!`, name);
-2. Black code formatter is used using `pre-commit` hook.
+1. Use f-strings for output formation (except logging when we stay with lazy `logging.info("Hello %s!", name)`.
+2. Black code formatter is used using a `pre-commit` hook.
 
 ### Documentation
 
@@ -182,10 +182,10 @@ python -m pip install ".[dev, examples]"
 python -m pip install pre-commit
 ```
 
-You can run the full test-case in your terminal via this bash script:
+You can run the full test-case in your terminal via this make script:
 
 ```bash
-bash .run_local_tests.sh
+make test
 ```
 
 Note: if your computer does not have multi-GPU nor TPU these tests are skipped.
@@ -196,7 +196,7 @@ This is useful if you do not test against all required dependency versions.
 **Docker:** Another option is utilize the [pytorch lightning cuda base docker image](https://hub.docker.com/repository/docker/pytorchlightning/pytorch_lightning/tags?page=1&name=cuda). You can then run:
 
 ```bash
-python -m pytest pytorch_lightning tests pl_examples -v --color=yes --flake8
+python -m pytest pytorch_lightning tests pl_examples -v --flake8
 ```
 
 ### Pull Request
@@ -276,6 +276,57 @@ We welcome any useful contribution! For your convenience here's a recommended wo
    git push -f
    ```
 
+4. **How to add new tests**
+
+We are using [pytest](https://docs.pytest.org/en/stable/) in Pytorch Lightning.
+
+Here are tutorials:
+* (recommended) [Visual Testing with pytest](https://www.youtube.com/playlist?list=PLCTHcU1KoD99Rim2tzg-IhYY2iu9FFvNo) from JetBrains on YouTube
+* [Effective Python Testing With Pytest](https://realpython.com/pytest-python-testing/) article on realpython.com
+
+Here is the process to create a new test
+
+* 0. Optional: Follow tutorials !
+* 1. Find a file in tests/ which match what you want to test. If none, create one.
+* 2. Use this template to get started !
+* 3. Use `BoringModel and derivates to test out your code`.
+
+```python
+# TEST SHOULD BE IN YOUR FILE: tests/..../...py
+
+# RUN OUR TEST WITH: pytest tests/..../...py::test_explain_what_is_being_tested --verbose --capture=no
+
+# TEST CODE TEMPLATE
+
+# pytest decorator
+# @pytest.mark.skipif(not torch.cuda.is_available(), reason="test requires GPU machine")
+def test_explain_what_is_being_tested(tmpdir):
+    """
+    Test description about text reason to be
+    """
+
+    # os.environ["PL_DEV_DEBUG"] = '1' optional. When activated, you can use internal trainer.dev_debugger
+
+    class ExtendedModel(BoringModel):
+        ...
+
+    model = ExtendedModel()
+
+    # BoringModel is a functional model. You might want to set methods to None to test your behaviour
+    # Example: model.training_step_end = None
+
+    trainer = Trainer(
+        default_root_dir=tmpdir, # will save everything within a tmpdir generated for this test
+        ...
+    )
+    trainer.fit(model)
+    result = trainer.test()
+
+    # assert the behaviour is correct.
+    assert ...
+    assert ...
+```
+
 ### Bonus Workflow Tip
 
 If you don't want to remember all the commands above every time you want to push some code/setup a Lightning Dev environment on a new VM, you can set up bash aliases for some common commands. You can add these to one of your `~/.bashrc`, `~/.zshrc`, or `~/.bash_aliases` files.

.github/PULL_REQUEST_TEMPLATE.md

@@ -4,6 +4,8 @@
 Please include a summary of the change and which issue is fixed.
  Please also include relevant motivation and context.
  List any dependencies that are required for this change.
+
+If we didn't discuss your PR in Github issues there's a high chance it will not be merged.
 -->
 
 Fixes # (issue)
@@ -17,13 +19,17 @@ Fixes # (issue)
 - [ ] Did you verify new and existing tests pass locally with your changes?
 - [ ] If you made a notable change (that affects users), did you update the [CHANGELOG](https://github.com/PyTorchLightning/pytorch-lightning/blob/master/CHANGELOG.md)?
 
-<!-- For CHANGELOG separate each item in unreleased section by a blank line to reduce collisions -->
+<!-- For CHANGELOG separate each item in the unreleased section by a blank line to reduce collisions -->
 
 ## PR review
- - [ ] Is this pull request ready for review? (if not, please submit in draft mode)
+Anyone in the community is free to review the PR once the tests have passed.
+Before you start reviewing make sure you have read [Review guidelines](https://github.com/PyTorchLightning/pytorch-lightning/wiki/Review-guidelines). In short, see the following bullet-list:  
 
-Anyone in the community is free to review the PR once the tests have passed.    
-If we didn't discuss your PR in Github issues there's a high chance it will not be merged.
+ - [ ] Is this pull request ready for review? (if not, please submit in draft mode)
+ - [ ] Check that all items from **Before submitting** are resolved
+ - [ ] Make sure the title is self-explanatory and the description concisely explains the PR
+ - [ ] Add labels and milestones (and optionally projects) to the PR so it can be classified; _Bugfixes should be including in bug-fix release milestones (m.f.X) and features should be included in (m.X.b) releases._
+ 
 
 ## Did you have fun?
 Make sure you had fun coding 🙃

.github/stale.yml

@@ -40,3 +40,10 @@ pulls:
 staleLabel: "won't fix"
 # Limit the number of actions per hour, from 1-30. Default is 30
 limitPerRun: 10
+
+# Set to true to ignore issues in a project (defaults to false)
+exemptProjects: true
+# Set to true to ignore issues in a milestone (defaults to false)
+exemptMilestones: true
+# Set to true to ignore issues with an assignee (defaults to false)
+exemptAssignees: true

.github/workflows/ci_dockers.yml

@@ -40,7 +40,7 @@ jobs:
       fail-fast: false
       matrix:
         python_version: [3.7]
-        xla_version: [1.6, "nightly"]
+        xla_version: [1.6, 1.7, "nightly"]
     steps:
       - name: Checkout
         uses: actions/checkout@v2
@@ -66,6 +66,9 @@ jobs:
       fail-fast: false
       matrix:
         include:
+          # todo: see notes in Dockerfile
+          #- python_version: 3.7
+          #  pytorch_version: 1.8
           - python_version: 3.8
             pytorch_version: 1.7
           - python_version: 3.7
@@ -110,9 +113,8 @@ jobs:
             pytorch_version: 1.4
           - python_version: 3.7
             pytorch_version: 1.7
-          # TODO
-          # - python_version: 3.7
-          #   pytorch_version: 1.8
+          - python_version: 3.7
+            pytorch_version: 1.8
     steps:
       - name: Checkout
         uses: actions/checkout@v2

.github/workflows/ci_test-base.yml

@@ -69,7 +69,7 @@ jobs:
     - name: Test Package [only]
       run: |
         # NOTE: run coverage on tests does not propagare faler status for Win, https://github.com/nedbat/coveragepy/issues/1003
-        coverage run --source pytorch_lightning -m pytest pytorch_lightning -v --color=yes --junitxml=junit/test-results-${{ runner.os }}-${{ matrix.python-version }}-${{ matrix.requires }}.xml
+        coverage run --source pytorch_lightning -m pytest pytorch_lightning -v --junitxml=junit/test-results-${{ runner.os }}-${{ matrix.python-version }}-${{ matrix.requires }}.xml
 
     - name: Upload pytest test results
       uses: actions/upload-artifact@v2

.github/workflows/ci_test-conda.yml

@@ -16,7 +16,7 @@ jobs:
       matrix:
         # os: [ubuntu-20.04]
         python-version: [3.7]
-        pytorch-version: [1.3, 1.4, 1.5, 1.6, 1.7]
+        pytorch-version: [1.3, 1.4, 1.5, 1.6, 1.7, 1.8]
 
     # Timeout: https://stackoverflow.com/a/59076067/4521646
     timeout-minutes: 35
@@ -41,7 +41,7 @@ jobs:
     - name: Tests
       run: |
         # NOTE: run coverage on tests does not propagare faler status for Win, https://github.com/nedbat/coveragepy/issues/1003
-        python -m pytest pytorch_lightning tests -v --color=yes --durations=0 --junitxml=junit/test-results-${{ runner.os }}-${{ matrix.python-version }}-${{ matrix.requires }}.xml
+        python -m pytest pytorch_lightning tests -v --durations=50 --junitxml=junit/test-results-${{ runner.os }}-${{ matrix.python-version }}-${{ matrix.requires }}.xml
         # coverage report
       shell: bash -l {0}
 

.github/workflows/ci_test-full.yml

@@ -64,6 +64,8 @@ jobs:
         python -c "fname = 'requirements/extra.txt' ; req = open(fname).read().replace('>=', '==') ; open(fname, 'w').write(req)"
         python -c "fname = 'requirements/test.txt' ; req = open(fname).read().replace('>=', '==') ; open(fname, 'w').write(req)"
         python -c "fname = 'requirements/examples.txt' ; req = open(fname).read().replace('>=', '==') ; open(fname, 'w').write(req)"
+        # remove Fairscale from requirements
+        python -c "fname = 'requirements/extra.txt' ; lines = [line for line in open(fname).readlines() if 'fairscale' not in line] ; open(fname, 'w').writelines(lines)"
 
     # Note: This uses an internal pip API and may not always work
     # https://github.com/actions/cache/blob/master/examples.md#multiple-oss-in-a-workflow
@@ -119,7 +121,7 @@ jobs:
     - name: Tests
       run: |
         # NOTE: do not include coverage report here, see: https://github.com/nedbat/coveragepy/issues/1003
-        coverage run --source pytorch_lightning -m pytest pytorch_lightning tests pl_examples -v --color=yes --durations=0 --junitxml=junit/test-results-${{ runner.os }}-${{ matrix.python-version }}-${{ matrix.requires }}.xml
+        coverage run --source pytorch_lightning -m pytest pytorch_lightning tests pl_examples -v --durations=50 --junitxml=junit/test-results-${{ runner.os }}-${{ matrix.python-version }}-${{ matrix.requires }}.xml
 
     - name: Upload pytest test results
       uses: actions/upload-artifact@v2