Merge branch 'main' into adapt

.circleci/config.yml

@@ -0,0 +1,122 @@
+version: 2.1
+
+jobs:
+  lint:
+    docker:
+      - image: cimg/python:3.7.4
+    steps:
+      - checkout
+      - run:
+          name: Install pre-commit hook
+          command: |
+            sudo apt-add-repository ppa:brightbox/ruby-ng -y
+            sudo apt-get update
+            sudo apt-get install -y ruby2.7
+            pip install pre-commit
+            pre-commit install
+      - run:
+          name: Linting
+          command: pre-commit run --all-files
+      - run:
+          name: Check docstring coverage
+          command: |
+            pip install interrogate
+            interrogate -v --ignore-init-method --ignore-module --ignore-nested-functions --ignore-regex "__repr__" --fail-under 80 mmengine
+
+  build_cpu:
+    parameters:
+      # The python version must match available image tags in
+      # https://circleci.com/developer/images/image/cimg/python
+      python:
+        type: string
+        default: "3.7.4"
+      torch:
+        type: string
+      torchvision:
+        type: string
+    docker:
+      - image: cimg/python:<< parameters.python >>
+    resource_class: large
+    steps:
+      - checkout
+      - run:
+          name: Upgrade pip
+          command: |
+            python -V
+            python -m pip install pip --upgrade
+            python -m pip --version
+      - run:
+          name: Install PyTorch
+          command: python -m pip install torch==<< parameters.torch >>+cpu torchvision==<< parameters.torchvision >>+cpu -f https://download.pytorch.org/whl/torch_stable.html
+      - run:
+          name: Install mmcv-full
+          command: python -m pip install mmcv-full -f https://download.openmmlab.com/mmcv/dist/cpu/torch1.8.0/index.html
+      - run:
+          name: Install mmengine dependencies
+          command: python -m pip install -r requirements.txt
+      - run:
+          name: Build and install
+          command: python -m pip install -e .
+      - run:
+          name: Run unit tests
+          command: python -m pytest tests/
+
+  build_cu102:
+    machine:
+      image: ubuntu-1604-cuda-10.1:201909-23  # the actual version of cuda is 10.2
+    resource_class: gpu.nvidia.small
+    steps:
+      - checkout
+      - run:
+          # https://github.com/pytorch/vision/issues/2921
+          name: Install dependency of torchvision when using pyenv
+          command: sudo apt-get install -y liblzma-dev
+      - run:
+          # python3.7 should be re-installed due to the issue https://github.com/pytorch/vision/issues/2921
+          name: Select python3.7
+          command: |
+            pyenv uninstall -f 3.7.0
+            pyenv install 3.7.0
+            pyenv global 3.7.0
+      - run:
+          name: Upgrade pip
+          command: |
+            python -V
+            python -m pip install pip --upgrade
+            python -m pip --version
+      - run:
+          name: Install PyTorch
+          command: python -m pip install torch==1.8.1+cu102 torchvision==0.9.1+cu102 -f https://download.pytorch.org/whl/torch_stable.html
+      - run:
+          name: Install mmcv-full
+          command: python -m pip install mmcv-full -f https://download.openmmlab.com/mmcv/dist/cu102/torch1.8.0/index.html
+      - run:
+          name: Install mmengine dependencies
+          command: python -m pip install -r requirements.txt
+      - run:
+          name: Build and install
+          command: python -m pip install -e .
+      - run:
+          name: Run unit tests
+          command: |
+            python -m coverage run --branch --source mmengine -m pytest tests/
+            python -m coverage xml
+            python -m coverage report -m
+
+workflows:
+  unit_tests:
+    jobs:
+      - lint
+      - build_cpu:
+          name: build_cpu_th1.8_py3.7
+          torch: 1.8.0
+          torchvision: 0.9.0
+          requires:
+            - lint
+      - hold:
+          type: approval # <<< This key-value pair will set your workflow to a status of "On Hold"
+          requires:
+            - build_cpu_th1.8_py3.7
+      - build_cu102:
+          requires:
+            - hold

docs/zh_cn/tutorials/registry.md

@@ -311,11 +311,15 @@ from mmcls.models import MODELS
 model = MODELS.build(cfg=dict(type='mmdet.RetinaNet'))
 ```
 
-调用兄弟节点的模块需要指定在 `type` 中指定 `scope` 前缀，如果不想指定，我们可以将 `build` 方法中的 `default_scope` 参数设置为 'mmdet'，它会将 `default_scope` 对应的 `registry` 作为当前 `Registry` 并调用 `build` 方法。
+调用非本节点的模块需要指定在 `type` 中指定 `scope` 前缀，如果不想指定，我们可以创建一个全局变量 `default_scope` 并将 `scope_name` 设置为 'mmdet'，`Registry` 会将 `scope_name` 对应的 `registry` 作为当前 `Registry` 并调用 `build` 方法。
 
 ```python
-from mmcls.models import MODELS
+from mmengine.registry import DefaultScope, MODELS
-model = MODELS.build(cfg=dict(type='RetinaNet'), default_scope='mmdet')
+
+# 调用注册在 mmdet 中的 RetinaNet
+default_scope = DefaultScope.get_instance(
+            'my_experiment', scope_name='mmdet')
+model = MODELS.build(cfg=dict(type='RetinaNet'))
 ```
 
 注册器除了支持两层结构，三层甚至更多层结构也是支持的。
@@ -325,7 +329,7 @@ model = MODELS.build(cfg=dict(type='RetinaNet'), default_scope='mmdet')
 `DetPlus` 中定义了模块 `MetaNet`，
 
 ```python
-from mmengine.model import Registry
+from mmengine.registry import Registry
 from mmdet.model import MODELS as MMDET_MODELS
 MODELS = Registry('model', parent=MMDET_MODELS, scope='det_plus')
 
@@ -354,6 +358,10 @@ model = MODELS.build(cfg=dict(type='mmcls.ResNet'))
 from mmcls.models import MODELS
 # 需要注意前缀的顺序，'detplus.mmdet.ResNet' 是不正确的
 model = MODELS.build(cfg=dict(type='mmdet.detplus.MetaNet'))
-# 当然，更简单的方法是直接设置 default_scope
+
+# 如果希望默认从 detplus 构建模型，设置可以 default_scope
+from mmengine.registry import DefaultScope
+default_scope = DefaultScope.get_instance(
+            'my_experiment', scope_name='detplus')
 model = MODELS.build(cfg=dict(type='MetaNet', default_scope='detplus'))
 ```

mmengine/evaluator/builder.py

@@ -1,5 +1,5 @@
 # Copyright (c) OpenMMLab. All rights reserved.
-from typing import Optional, Union
+from typing import Union
 
 from ..registry import EVALUATORS
 from .base import BaseEvaluator
@@ -7,9 +7,7 @@ from .composed_evaluator import ComposedEvaluator
 
 
 def build_evaluator(
-    cfg: Union[dict, list],
+        cfg: Union[dict, list]) -> Union[BaseEvaluator, ComposedEvaluator]:
-    default_scope: Optional[str] = None
-) -> Union[BaseEvaluator, ComposedEvaluator]:
     """Build function of evaluator.
 
     When the evaluator config is a list, it will automatically build composed
@@ -18,16 +16,12 @@ def build_evaluator(
     Args:
         cfg (dict | list): Config of evaluator. When the config is a list, it
             will automatically build composed evaluators.
-        default_scope (str, optional): The ``default_scope`` is used to
-            reset the current registry. Defaults to None.
 
     Returns:
         BaseEvaluator or ComposedEvaluator: The built evaluator.
     """
     if isinstance(cfg, list):
-        evaluators = [
+        evaluators = [EVALUATORS.build(_cfg) for _cfg in cfg]
-            EVALUATORS.build(_cfg, default_scope=default_scope) for _cfg in cfg
-        ]
         return ComposedEvaluator(evaluators=evaluators)
     else:
-        return EVALUATORS.build(cfg, default_scope=default_scope)
+        return EVALUATORS.build(cfg)

mmengine/optim/optimizer/builder.py

@@ -1,7 +1,7 @@
 # Copyright (c) OpenMMLab. All rights reserved.
 import copy
 import inspect
-from typing import List, Optional
+from typing import List
 
 import torch
 import torch.nn as nn
@@ -30,10 +30,7 @@ def register_torch_optimizers() -> List[str]:
 TORCH_OPTIMIZERS = register_torch_optimizers()
 
 
-def build_optimizer(
+def build_optimizer(model: nn.Module, cfg: dict) -> torch.optim.Optimizer:
-        model: nn.Module,
-        cfg: dict,
-        default_scope: Optional[str] = None) -> torch.optim.Optimizer:
     """Build function of optimizer.
 
     If ``constructor`` is set in the ``cfg``, this method will build an
@@ -58,7 +55,6 @@ def build_optimizer(
         dict(
             type=constructor_type,
             optimizer_cfg=optimizer_cfg,
-            paramwise_cfg=paramwise_cfg),
+            paramwise_cfg=paramwise_cfg))
-        default_scope=default_scope)
+    optimizer = optim_constructor(model)
-    optimizer = optim_constructor(model, default_scope=default_scope)
     return optimizer

mmengine/optim/optimizer/default_constructor.py

@@ -241,9 +241,7 @@ class DefaultOptimizerConstructor:
                 prefix=child_prefix,
                 is_dcn_module=is_dcn_module)
 
-    def __call__(self,
+    def __call__(self, model: nn.Module) -> torch.optim.Optimizer:
-                 model: nn.Module,
-                 default_scope: Optional[str] = None) -> torch.optim.Optimizer:
         if hasattr(model, 'module'):
             model = model.module
 
@@ -251,11 +249,11 @@ class DefaultOptimizerConstructor:
         # if no paramwise option is specified, just use the global setting
         if not self.paramwise_cfg:
             optimizer_cfg['params'] = model.parameters()
-            return OPTIMIZERS.build(optimizer_cfg, default_scope=default_scope)
+            return OPTIMIZERS.build(optimizer_cfg)
 
         # set param-wise lr and weight decay recursively
         params: List = []
         self.add_params(params, model)
         optimizer_cfg['params'] = params
 
-        return OPTIMIZERS.build(optimizer_cfg, default_scope=default_scope)
+        return OPTIMIZERS.build(optimizer_cfg)

mmengine/registry/default_scope.py

@@ -25,8 +25,6 @@ class DefaultScope(ManagerMixin):
         >>> DefaultScope.get_instance('task', scope_name='mmdet')
         >>> # Get default scope globally.
         >>> scope_name = DefaultScope.get_instance('task').scope_name
-        >>> # build model from cfg.
-        >>> model = MODELS.build(model_cfg, default_scope=scope_name)
     """
 
     def __init__(self, name: str, scope_name: str):

mmengine/registry/registry.py

@@ -7,6 +7,7 @@ from typing import Any, Dict, List, Optional, Tuple, Type, Union
 
 from ..config import Config, ConfigDict
 from ..utils import is_seq_of
+from .default_scope import DefaultScope
 
 
 def build_from_cfg(
@@ -354,19 +355,13 @@ class Registry:
 
         return None
 
-    def build(self,
+    def build(self, *args, **kwargs) -> Any:
-              *args,
-              default_scope: Optional[str] = None,
-              **kwargs) -> Any:
         """Build an instance.
 
-        Build an instance by calling :attr:`build_func`. If
+        Build an instance by calling :attr:`build_func`. If the global
-        :attr:`default_scope` is given, :meth:`build` will firstly get the
+        variable default scope (:obj:`DefaultScope`) exists ,
-        responding registry and then call its own :meth:`build`.
+        :meth:`build` will firstly get the responding registry and then call
-
+        its own :meth:`build`.
-        Args:
-            default_scope (str, optional): The ``default_scope`` is used to
-                reset the current registry. Defaults to None.
 
         Examples:
             >>> from mmengine import Registry
@@ -379,9 +374,11 @@ class Registry:
             >>> cfg = dict(type='ResNet', depth=50)
             >>> model = MODELS.build(cfg)
         """
+        # get the global default scope
+        default_scope = DefaultScope.get_current_instance()
         if default_scope is not None:
             root = self._get_root_registry()
-            registry = root._search_child(default_scope)
+            registry = root._search_child(default_scope.scope_name)
             if registry is None:
                 # if `default_scope` can not be found, fallback to use self
                 warnings.warn(

mmengine/runner/runner.py

@@ -675,8 +675,7 @@ class Runner:
         if isinstance(model, nn.Module):
             return model
         elif isinstance(model, dict):
-            return MODELS.build(
+            return MODELS.build(model)
-                model, default_scope=self.default_scope.scope_name)
         else:
             raise TypeError('model should be a nn.Module object or dict, '
                             f'but got {model}')
@@ -726,9 +725,7 @@ class Runner:
                     model = model.cuda()
         else:
             model = MODEL_WRAPPERS.build(
-                model_wrapper_cfg,
+                model_wrapper_cfg, default_args=dict(model=self.model))
-                default_scope=self.default_scope.scope_name,
-                default_args=dict(model=self.model))
 
         return model
 
@@ -750,10 +747,7 @@ class Runner:
         if isinstance(optimizer, Optimizer):
             return optimizer
         elif isinstance(optimizer, dict):
-            optimizer = build_optimizer(
+            optimizer = build_optimizer(self.model, optimizer)
-                self.model,
-                optimizer,
-                default_scope=self.default_scope.scope_name)
             return optimizer
         else:
             raise TypeError('optimizer should be an Optimizer object or dict, '
@@ -801,7 +795,6 @@ class Runner:
                 param_schedulers.append(
                     PARAM_SCHEDULERS.build(
                         _scheduler,
-                        default_scope=self.default_scope.scope_name,
                         default_args=dict(optimizer=self.optimizer)))
             else:
                 raise TypeError(
@@ -837,9 +830,7 @@ class Runner:
         if isinstance(evaluator, (BaseEvaluator, ComposedEvaluator)):
             return evaluator
         elif isinstance(evaluator, dict) or is_list_of(evaluator, dict):
-            return build_evaluator(
+            return build_evaluator(evaluator)  # type: ignore
-                evaluator,
-                default_scope=self.default_scope.scope_name)  # type: ignore
         else:
             raise TypeError(
                 'evaluator should be one of dict, list of dict, BaseEvaluator '
@@ -880,8 +871,7 @@ class Runner:
         # build dataset
         dataset_cfg = dataloader_cfg.pop('dataset')
         if isinstance(dataset_cfg, dict):
-            dataset = DATASETS.build(
+            dataset = DATASETS.build(dataset_cfg)
-                dataset_cfg, default_scope=self.default_scope.scope_name)
         else:
             # fallback to raise error in dataloader
             # if `dataset_cfg` is not a valid type
@@ -891,9 +881,7 @@ class Runner:
         sampler_cfg = dataloader_cfg.pop('sampler')
         if isinstance(sampler_cfg, dict):
             sampler = DATA_SAMPLERS.build(
-                sampler_cfg,
+                sampler_cfg, default_args=dict(dataset=dataset))
-                default_scope=self.default_scope.scope_name,
-                default_args=dict(dataset=dataset))
         else:
             # fallback to raise error in dataloader
             # if `sampler_cfg` is not a valid type
@@ -961,7 +949,6 @@ class Runner:
         if 'type' in loop_cfg:
             loop = LOOPS.build(
                 loop_cfg,
-                default_scope=self.default_scope.scope_name,
                 default_args=dict(
                     runner=self, dataloader=self.train_dataloader))
         else:
@@ -1012,7 +999,6 @@ class Runner:
         if 'type' in loop_cfg:
             loop = LOOPS.build(
                 loop_cfg,
-                default_scope=self.default_scope.scope_name,
                 default_args=dict(
                     runner=self,
                     dataloader=self.val_dataloader,
@@ -1059,7 +1045,6 @@ class Runner:
         if 'type' in loop_cfg:
             loop = LOOPS.build(
                 loop_cfg,
-                default_scope=self.default_scope.scope_name,
                 default_args=dict(
                     runner=self,
                     dataloader=self.test_dataloader,

tests/test_dist/test_dist.py

@@ -6,6 +6,7 @@ from unittest.mock import patch
 
 import pytest
 import torch
+import torch.distributed as torch_dist
 import torch.multiprocessing as mp
 
 import mmengine.dist as dist
@@ -108,9 +109,16 @@ def init_process(rank, world_size, functions, backend='gloo'):
     os.environ['MASTER_ADDR'] = '127.0.0.1'
     os.environ['MASTER_PORT'] = '29505'
     os.environ['RANK'] = str(rank)
-    dist.init_dist('pytorch', backend, rank=rank, world_size=world_size)
 
-    device = 'cpu' if backend == 'gloo' else 'cuda'
+    if backend == 'nccl':
+        num_gpus = torch.cuda.device_count()
+        torch.cuda.set_device(rank % num_gpus)
+        device = 'cuda'
+    else:
+        device = 'cpu'
+
+    torch_dist.init_process_group(
+        backend=backend, rank=rank, world_size=world_size)
 
     for func in functions:
         func(device)

tests/test_dist/test_utils.py

@@ -55,7 +55,13 @@ def init_process(rank, world_size, functions, backend='gloo'):
     os.environ['MASTER_ADDR'] = '127.0.0.1'
     os.environ['MASTER_PORT'] = '29501'
     os.environ['RANK'] = str(rank)
-    dist.init_dist('pytorch', backend, rank=rank, world_size=world_size)
+
+    if backend == 'nccl':
+        num_gpus = torch.cuda.device_count()
+        torch.cuda.set_device(rank % num_gpus)
+
+    torch_dist.init_process_group(
+        backend=backend, rank=rank, world_size=world_size)
     dist.init_local_group(0, world_size)
 
     for func in functions:

tests/test_evaluator/test_base_evaluator.py

@@ -79,10 +79,9 @@ def generate_test_results(size, batch_size, pred, label):
     bs_residual = size % batch_size
     for i in range(num_batch):
         bs = bs_residual if i == num_batch - 1 else batch_size
-        data_batch = [(np.zeros(
+        data_batch = [(np.zeros((3, 10, 10)), BaseDataElement(label=label))
-            (3, 10, 10)), BaseDataElement(data={'label': label}))
                       for _ in range(bs)]
-        predictions = [BaseDataElement(data={'pred': pred}) for _ in range(bs)]
+        predictions = [BaseDataElement(pred=pred) for _ in range(bs)]
         yield (data_batch, predictions)
 
 

tests/test_registry/test_registry.py

@@ -1,8 +1,10 @@
 # Copyright (c) OpenMMLab. All rights reserved.
+import time
+
 import pytest
 
 from mmengine.config import Config, ConfigDict  # type: ignore
-from mmengine.registry import Registry, build_from_cfg
+from mmengine.registry import DefaultScope, Registry, build_from_cfg
 
 
 class TestRegistry:
@@ -342,11 +344,15 @@ class TestRegistry:
 
         # test `default_scope`
         # switch the current registry to another registry
-        dog = LITTLE_HOUNDS.build(b_cfg, default_scope='mid_hound')
+        DefaultScope.get_instance(
+            f'test-{time.time()}', scope_name='mid_hound')
+        dog = LITTLE_HOUNDS.build(b_cfg)
         assert isinstance(dog, Beagle)
 
         # `default_scope` can not be found
-        dog = MID_HOUNDS.build(b_cfg, default_scope='scope-not-found')
+        DefaultScope.get_instance(
+            f'test2-{time.time()}', scope_name='scope-not-found')
+        dog = MID_HOUNDS.build(b_cfg)
         assert isinstance(dog, Beagle)
 
     def test_repr(self):