Upgrade pre commit hooks (#2321)

* Upgrade the versions of pre-commit hooks

* update the versions of zh-cn.yaml

.pre-commit-config-zh-cn.yaml

 exclude: ^tests/data/
 repos:
   - repo: https://gitee.com/openmmlab/mirrors-flake8
-    rev: 3.8.3
+    rev: 5.0.4
     hooks:
       - id: flake8
   - repo: https://gitee.com/openmmlab/mirrors-isort
@@ -9,11 +9,11 @@ repos:
     hooks:
       - id: isort
   - repo: https://gitee.com/openmmlab/mirrors-yapf
-    rev: v0.30.0
+    rev: v0.32.0
     hooks:
       - id: yapf
   - repo: https://gitee.com/openmmlab/mirrors-pre-commit-hooks
-    rev: v3.1.0
+    rev: v4.3.0
     hooks:
       - id: trailing-whitespace
       - id: check-yaml
@@ -26,7 +26,7 @@ repos:
       - id: mixed-line-ending
         args: ["--fix=lf"]
   - repo: https://gitee.com/openmmlab/mirrors-codespell
-    rev: v2.1.0
+    rev: v2.2.1
     hooks:
       - id: codespell
   - repo: https://gitee.com/openmmlab/mirrors-mdformat
@@ -44,7 +44,7 @@ repos:
       - id: docformatter
         args: ["--in-place", "--wrap-descriptions", "79"]
   - repo: https://github.com/asottile/pyupgrade
-    rev: v2.32.1
+    rev: v3.0.0
     hooks:
       - id: pyupgrade
         args: ["--py36-plus"]

.pre-commit-config.yaml

 exclude: ^tests/data/
 repos:
   - repo: https://github.com/PyCQA/flake8
-    rev: 3.8.3
+    rev: 5.0.4
     hooks:
       - id: flake8
   - repo: https://github.com/PyCQA/isort
@@ -9,11 +9,11 @@ repos:
     hooks:
       - id: isort
   - repo: https://github.com/pre-commit/mirrors-yapf
-    rev: v0.30.0
+    rev: v0.32.0
     hooks:
       - id: yapf
   - repo: https://github.com/pre-commit/pre-commit-hooks
-    rev: v3.1.0
+    rev: v4.3.0
     hooks:
       - id: trailing-whitespace
       - id: check-yaml
@@ -26,7 +26,7 @@ repos:
       - id: mixed-line-ending
         args: ["--fix=lf"]
   - repo: https://github.com/codespell-project/codespell
-    rev: v2.1.0
+    rev: v2.2.1
     hooks:
       - id: codespell
   - repo: https://github.com/executablebooks/mdformat
@@ -44,7 +44,7 @@ repos:
       - id: docformatter
         args: ["--in-place", "--wrap-descriptions", "79"]
   - repo: https://github.com/asottile/pyupgrade
-    rev: v2.32.1
+    rev: v3.0.0
     hooks:
       - id: pyupgrade
         args: ["--py36-plus"]

mmcv/cnn/utils/flops_counter.py

@@ -513,7 +513,6 @@ def batch_counter_hook(module: nn.Module, input: tuple, output: Any) -> None:
 
 
 def add_batch_counter_variables_or_reset(module: nn.Module) -> None:
-
     module.__batch_counter__ = 0
 
 

mmcv/cnn/utils/weight_init.py

@@ -213,7 +213,8 @@ class ConstantInit(BaseInit):
 @INITIALIZERS.register_module(name='Xavier')
 class XavierInit(BaseInit):
     r"""Initialize module parameters with values according to the method
-    described in `Understanding the difficulty of training deep feedforward
+    described in `Understanding the difficulty of training deep feedforward.
+
     neural networks - Glorot, X. & Bengio, Y. (2010).
     <http://proceedings.mlr.press/v9/glorot10a/glorot10a.pdf>`_
 
@@ -271,7 +272,6 @@ class NormalInit(BaseInit):
             Defaults to None.
         layer (str | list[str], optional): the layer will be initialized.
             Defaults to None.
-
     """
 
     def __init__(self, mean: float = 0, std: float = 1, **kwargs):
@@ -317,7 +317,6 @@ class TruncNormalInit(BaseInit):
             Defaults to None.
         layer (str | list[str], optional): the layer will be initialized.
             Defaults to None.
-
     """
 
     def __init__(self,
@@ -401,7 +400,8 @@ class UniformInit(BaseInit):
 @INITIALIZERS.register_module(name='Kaiming')
 class KaimingInit(BaseInit):
     r"""Initialize module parameters with the values according to the method
-    described in `Delving deep into rectifiers: Surpassing human-level
+    described in `Delving deep into rectifiers: Surpassing human-level.
+
     performance on ImageNet classification - He, K. et al. (2015).
     <https://www.cv-foundation.org/openaccess/content_iccv_2015/
     papers/He_Delving_Deep_into_ICCV_2015_paper.pdf>`_
@@ -688,12 +688,12 @@ def trunc_normal_(tensor: Tensor,
                   std: float = 1.,
                   a: float = -2.,
                   b: float = 2.) -> Tensor:
-    r"""Fills the input Tensor with values drawn from a truncated
-    normal distribution. The values are effectively drawn from the
-    normal distribution :math:`\mathcal{N}(\text{mean}, \text{std}^2)`
-    with values outside :math:`[a, b]` redrawn until they are within
-    the bounds. The method used for generating the random values works
-    best when :math:`a \leq \text{mean} \leq b`.
+    r"""Fills the input Tensor with values drawn from a truncated normal
+    distribution. The values are effectively drawn from the normal distribution
+    :math:`\mathcal{N}(\text{mean}, \text{std}^2)` with values outside
+    :math:`[a, b]` redrawn until they are within the bounds. The method used
+    for generating the random values works best when :math:`a \leq \text{mean}
+    \leq b`.
 
     Modified from
     https://github.com/pytorch/pytorch/blob/master/torch/nn/init.py

mmcv/image/colorspace.py

@@ -276,7 +276,6 @@ def ycbcr2bgr(img: np.ndarray) -> np.ndarray:
 
 
 def convert_color_factory(src: str, dst: str) -> Callable:
-
     code = getattr(cv2, f'COLOR_{src.upper()}2{dst.upper()}')
 
     def convert_color(img: np.ndarray) -> np.ndarray:

mmcv/onnx/info.py

@@ -6,7 +6,6 @@ import torch
 
 
 def is_custom_op_loaded() -> bool:
-
     # Following strings of text style are from colorama package
     bright_style, reset_style = '\x1b[1m', '\x1b[0m'
     red_text, blue_text = '\x1b[31m', '\x1b[34m'

mmcv/ops/active_rotated_filter.py

@@ -47,12 +47,12 @@ class ActiveRotatedFilterFunction(Function):
     def backward(ctx, grad_out: torch.Tensor) -> Tuple[torch.Tensor, None]:
         """
         Args:
-            grad_output (torch.Tensor): The gradiant of output features
+            grad_output (torch.Tensor): The gradient of output features
                 with shape [num_output_planes * num_rotations,
                 num_input_planes * num_orientations, H, W].
 
         Returns:
-            torch.Tensor: The gradiant of input features with shape
+            torch.Tensor: The gradient of input features with shape
             [num_output_planes, num_input_planes, num_orientations, H, W].
         """
         input, indices = ctx.saved_tensors

mmcv/ops/correlation.py

@@ -116,7 +116,7 @@ class CorrelationFunction(Function):
 
 
 class Correlation(nn.Module):
-    r"""Correlation operator
+    r"""Correlation operator.
 
     This correlation operator works for optical flow correlation computation.
 

mmcv/ops/csrc/pytorch/mlu/psamask_mlu.cpp

@@ -161,7 +161,7 @@ void PSAMaskForwardMLUKernelLauncher(const int psa_type, const Tensor x,
   TORCH_CHECK(h_feature * w_feature == y_c,
               "channel of y should be the same as h_feature * w_feature");
   TORCH_CHECK(psa_type == 0 || psa_type == 1,
-              "psa_type only suppurts 'COLLECT' and 'DISTRIBUTE' currently");
+              "psa_type only supports 'COLLECT' and 'DISTRIBUTE' currently");
 
   if (x.numel() == 0) {
     CNLOG(INFO) << "skip zero-element tensor";
@@ -227,7 +227,7 @@ void PSAMaskBackwardMLUKernelLauncher(const int psa_type, const Tensor dy,
   TORCH_CHECK(h_mask * w_mask == dx_c,
               "channel of dx should be the same as h_mask * w_mask");
   TORCH_CHECK(psa_type == 0 || psa_type == 1,
-              "psa_type only suppurts 'COLLECT' and 'DISTRIBUTE' currently");
+              "psa_type only supports 'COLLECT' and 'DISTRIBUTE' currently");
 
   if (dx.numel() == 0) {
     CNLOG(INFO) << "skip zero-element tensor";

mmcv/ops/csrc/pytorch/mlu/roi_align_mlu.cpp

@@ -45,7 +45,7 @@ void ROIAlignForwardMLUKernelLauncher(Tensor input, Tensor rois, Tensor output,
               input.dim(), "D");
   TORCH_CHECK(rois.dim() == 2, "rois should be a 2d tensor, got ", rois.dim(),
               "D");
-  TORCH_CHECK(pool_mode == 1, "pool_mode only suppurts 'avg' currently");
+  TORCH_CHECK(pool_mode == 1, "pool_mode only supports 'avg' currently");
 
   auto memory_format =
       torch_mlu::cnnl::ops::get_channels_last_memory_format(input.dim());
@@ -121,7 +121,7 @@ void ROIAlignBackwardMLUKernelLauncher(Tensor grad, Tensor rois,
               "D");
   TORCH_CHECK(rois.dim() == 2, "rois should be a 2d tensor, got ", rois.dim(),
               "D");
-  TORCH_CHECK(pool_mode == 1, "pool_mode only suppurts 'avg' currently");
+  TORCH_CHECK(pool_mode == 1, "pool_mode only supports 'avg' currently");
 
   int batch_size = grad_input.size(0);
   int channels = grad_input.size(1);

mmcv/ops/rotated_feature_align.py

@@ -66,11 +66,11 @@ class RotatedFeatureAlignFunction(Function):
     def backward(ctx: Any, grad_output: torch.Tensor) -> tuple:
         """
         Args:
-            grad_output (torch.Tensor): The gradiant of output features
+            grad_output (torch.Tensor): The gradient of output features
                 with shape [N,C,H,W].
 
         Returns:
-            torch.Tensor: The gradiant of input features with shape [N,C,H,W].
+            torch.Tensor: The gradient of input features with shape [N,C,H,W].
         """
         best_rbboxes = ctx.saved_tensors[0]
         points = ctx.points

mmcv/ops/sparse_modules.py

@@ -53,10 +53,9 @@ class SparseModule(nn.Module):
 
 
 class SparseSequential(SparseModule):
-    r"""A sequential container.
-    Modules will be added to it in the order they are passed in the
-    constructor.
-    Alternatively, an ordered dict of modules can also be passed in.
+    r"""A sequential container. Modules will be added to it in the order they
+    are passed in the constructor. Alternatively, an ordered dict of modules
+    can also be passed in.
 
     To make it easier to understand, given is a small example::
 

mmcv/video/io.py

@@ -42,7 +42,7 @@ class Cache:
 class VideoReader:
     """Video class with similar usage to a list object.
 
-    This video warpper class provides convenient apis to access frames.
+    This video wrapper class provides convenient apis to access frames.
     There exists an issue of OpenCV's VideoCapture class that jumping to a
     certain frame may be inaccurate. It is fixed in this class by checking
     the position after jumping each time.

setup.cfg

@@ -23,4 +23,4 @@ default_section = THIRDPARTY
 # than "BA"
 [codespell]
 quiet-level = 3
-ignore-words-list = inout,hist,ba
+ignore-words-list = inout,hist,ba,inh,ro,tne,warmup,warpped,warpping

tests/test_cnn/test_conv_module.py

@@ -208,7 +208,6 @@ def relu_forward(self, x):
 @patch('torch.nn.BatchNorm2d.forward', bn_forward)
 @patch('torch.nn.Conv2d.forward', conv_forward)
 def test_order():
-
     with pytest.raises(AssertionError):
         # order must be a tuple
         order = ['conv', 'norm', 'act']

tests/test_cnn/test_generalized_attention.py

@@ -5,7 +5,6 @@ from mmcv.cnn.bricks import GeneralizedAttention
 
 
 def test_context_block():
-
     # test attention_type='1000'
     imgs = torch.randn(2, 16, 20, 20)
     gen_attention_block = GeneralizedAttention(16, attention_type='1000')

tests/test_cnn/test_transformer.py

@@ -14,7 +14,6 @@ from mmcv.runner import ModuleList
 
 
 def test_adaptive_padding():
-
     for padding in ('same', 'corner'):
         kernel_size = 16
         stride = 16
@@ -320,7 +319,6 @@ def test_patch_embed():
 
 
 def test_patch_merging():
-
     # Test the model with int padding
     in_c = 3
     out_c = 4

tests/test_cnn/test_wrappers.py

@@ -333,7 +333,6 @@ def test_linear(in_w, in_h, in_feature, out_feature):
 
 @patch('mmcv.cnn.bricks.wrappers.TORCH_VERSION', (1, 10))
 def test_nn_op_forward_called():
-
     for m in ['Conv2d', 'ConvTranspose2d', 'MaxPool2d']:
         with patch(f'torch.nn.{m}.forward') as nn_module_forward:
             # randn input

tests/test_device/test_ipu/test_ipu_hooks.py

@@ -58,7 +58,6 @@ class ToyModel(nn.Module):
 
 @skip_no_ipu
 def test_ipu_hook_wrapper(tmp_path):
-
     model = ToyModel()
     dummy_input = {
         'data': {

tests/test_device/test_ipu/test_ipu_model.py

@@ -172,7 +172,6 @@ def run_model(ipu_options,
 
 @skip_no_ipu
 def test_run_model():
-
     # test feature alignment not support gradientAccumulation mode
     options_cfg = dict(
         randomSeed=888,