Fix typo: upsampe_cfg -> upsample_cfg (#449)

* Fix typo: upsampe_cfg -> upsample_cfg

Signed-off-by: lizz <lizz@sensetime.com>

* convoluton -> convolution

Signed-off-by: lizz <lizz@sensetime.com>

* more

Signed-off-by: lizz <lizz@sensetime.com>

* ok

Signed-off-by: lizz <lizz@sensetime.com>

mmseg/apis/test.py

@@ -42,7 +42,7 @@ def single_gpu_test(model,
     Args:
         model (nn.Module): Model to be tested.
         data_loader (utils.data.Dataloader): Pytorch data loader.
-        show (bool): Whether show results during infernece. Default: False.
+        show (bool): Whether show results during inference. Default: False.
         out_dir (str, optional): If specified, the results will be dumped into
             the directory to save output results.
         efficient_test (bool): Whether save the results as local numpy files to

mmseg/core/evaluation/metrics.py

@@ -212,7 +212,7 @@ def eval_metrics(results,
      Returns:
          float: Overall accuracy on all images.
          ndarray: Per category accuracy, shape (num_classes, ).
-         ndarray: Per category evalution metrics, shape (num_classes, ).
+         ndarray: Per category evaluation metrics, shape (num_classes, ).
     """
     if isinstance(metrics, str):
         metrics = [metrics]

mmseg/core/seg/sampler/base_pixel_sampler.py

@@ -10,4 +10,3 @@ class BasePixelSampler(metaclass=ABCMeta):
     @abstractmethod
     def sample(self, seg_logit, seg_label):
         """Placeholder for sample function."""
-        pass

mmseg/datasets/custom.py

@@ -214,8 +214,8 @@ class CustomDataset(Dataset):
             idx (int): Index of data.
 
         Returns:
-            dict: Testing data after pipeline with new keys intorduced by
+            dict: Testing data after pipeline with new keys introduced by
-                piepline.
+                pipeline.
         """
 
         img_info = self.img_infos[idx]
@@ -225,7 +225,6 @@ class CustomDataset(Dataset):
 
     def format_results(self, results, **kwargs):
         """Place holder to format result to dataset specific output."""
-        pass
 
     def get_gt_seg_maps(self, efficient_test=False):
         """Get ground truth segmentation maps for evaluation."""

mmseg/datasets/pipelines/transforms.py

@@ -14,7 +14,7 @@ class Resize(object):
     contains the key "scale", then the scale in the input dict is used,
     otherwise the specified scale in the init method is used.
 
-    ``img_scale`` can be Nong, a tuple (single-scale) or a list of tuple
+    ``img_scale`` can be None, a tuple (single-scale) or a list of tuple
     (multi-scale). There are 4 multiscale modes:
 
     - ``ratio_range is not None``:
@@ -89,7 +89,7 @@ class Resize(object):
         Args:
             img_scales (list[tuple]): Images scale range for sampling.
                 There must be two tuples in img_scales, which specify the lower
-                and uper bound of image scales.
+                and upper bound of image scales.
 
         Returns:
             (tuple, None): Returns a tuple ``(img_scale, None)``, where

mmseg/models/backbones/cgnet.py

@@ -13,7 +13,7 @@ from ..builder import BACKBONES
 class GlobalContextExtractor(nn.Module):
     """Global Context Extractor for CGNet.
 
-    This class is employed to refine the joFint feature of both local feature
+    This class is employed to refine the joint feature of both local feature
     and surrounding context.
 
     Args:
@@ -357,7 +357,7 @@ class CGNet(nn.Module):
             raise TypeError('pretrained must be a str or None')
 
     def train(self, mode=True):
-        """Convert the model into training mode whill keeping the normalization
+        """Convert the model into training mode will keeping the normalization
         layer freezed."""
         super(CGNet, self).train(mode)
         if mode and self.norm_eval:

mmseg/models/backbones/hrnet.py

@@ -545,7 +545,7 @@ class HRNet(nn.Module):
         return y_list
 
     def train(self, mode=True):
-        """Convert the model into training mode whill keeping the normalization
+        """Convert the model into training mode will keeping the normalization
         layer freezed."""
         super(HRNet, self).train(mode)
         if mode and self.norm_eval:

mmseg/models/backbones/mobilenet_v3.py

@@ -19,7 +19,7 @@ class MobileNetV3(nn.Module):
     <https://ieeexplore.ieee.org/document/9008835>`_.
 
     Args:
-        arch (str): Architechture of mobilnetv3, from {'small', 'large'}.
+        arch (str): Architecture of mobilnetv3, from {'small', 'large'}.
             Default: 'small'.
         conv_cfg (dict): Config dict for convolution layer.
             Default: None, which means using conv2d.
@@ -28,13 +28,13 @@ class MobileNetV3(nn.Module):
         out_indices (tuple[int]): Output from which layer.
             Default: (0, 1, 12).
         frozen_stages (int): Stages to be frozen (all param fixed).
-            Defualt: -1, which means not freezing any parameters.
+            Default: -1, which means not freezing any parameters.
         norm_eval (bool): Whether to set norm layers to eval mode, namely,
             freeze running stats (mean and var). Note: Effect on Batch Norm
             and its variants only. Default: False.
         with_cp (bool): Use checkpoint or not. Using checkpoint will save
             some memory while slowing down the training speed.
-            Defualt: False.
+            Default: False.
     """
     # Parameters to build each block:
     #     [kernel size, mid channels, out channels, with_se, act type, stride]

mmseg/models/backbones/unet.py

@@ -35,7 +35,7 @@ class BasicConvBlock(nn.Module):
             Default: dict(type='BN').
         act_cfg (dict | None): Config dict for activation layer in ConvModule.
             Default: dict(type='ReLU').
-        dcn (bool): Use deformable convoluton in convolutional layer or not.
+        dcn (bool): Use deformable convolution in convolutional layer or not.
             Default: None.
         plugins (dict): plugins for convolutional layers. Default: None.
     """
@@ -171,7 +171,7 @@ class InterpConv(nn.Module):
         kernel_size (int): Kernel size of the convolutional layer. Default: 1.
         stride (int): Stride of the convolutional layer. Default: 1.
         padding (int): Padding of the convolutional layer. Default: 1.
-        upsampe_cfg (dict): Interpolation config of the upsample layer.
+        upsample_cfg (dict): Interpolation config of the upsample layer.
             Default: dict(
                 scale_factor=2, mode='bilinear', align_corners=False).
     """
@@ -188,7 +188,7 @@ class InterpConv(nn.Module):
                  kernel_size=1,
                  stride=1,
                  padding=0,
-                 upsampe_cfg=dict(
+                 upsample_cfg=dict(
                      scale_factor=2, mode='bilinear', align_corners=False)):
         super(InterpConv, self).__init__()
 
@@ -202,7 +202,7 @@ class InterpConv(nn.Module):
             conv_cfg=conv_cfg,
             norm_cfg=norm_cfg,
             act_cfg=act_cfg)
-        upsample = nn.Upsample(**upsampe_cfg)
+        upsample = nn.Upsample(**upsample_cfg)
         if conv_first:
             self.interp_upsample = nn.Sequential(conv, upsample)
         else:
@@ -232,17 +232,17 @@ class UNet(nn.Module):
         strides (Sequence[int 1 | 2]): Strides of each stage in encoder.
             len(strides) is equal to num_stages. Normally the stride of the
             first stage in encoder is 1. If strides[i]=2, it uses stride
-            convolution to downsample in the correspondance encoder stage.
+            convolution to downsample in the correspondence encoder stage.
             Default: (1, 1, 1, 1, 1).
         enc_num_convs (Sequence[int]): Number of convolutional layers in the
-            convolution block of the correspondance encoder stage.
+            convolution block of the correspondence encoder stage.
             Default: (2, 2, 2, 2, 2).
         dec_num_convs (Sequence[int]): Number of convolutional layers in the
-            convolution block of the correspondance decoder stage.
+            convolution block of the correspondence decoder stage.
             Default: (2, 2, 2, 2).
         downsamples (Sequence[int]): Whether use MaxPool to downsample the
             feature map after the first stage of encoder
-            (stages: [1, num_stages)). If the correspondance encoder stage use
+            (stages: [1, num_stages)). If the correspondence encoder stage use
             stride convolution (strides[i]=2), it will never use MaxPool to
             downsample, even downsamples[i-1]=True.
             Default: (True, True, True, True).
@@ -263,14 +263,14 @@ class UNet(nn.Module):
         norm_eval (bool): Whether to set norm layers to eval mode, namely,
             freeze running stats (mean and var). Note: Effect on Batch Norm
             and its variants only. Default: False.
-        dcn (bool): Use deformable convoluton in convolutional layer or not.
+        dcn (bool): Use deformable convolution in convolutional layer or not.
             Default: None.
         plugins (dict): plugins for convolutional layers. Default: None.
 
     Notice:
-        The input image size should be devisible by the whole downsample rate
+        The input image size should be divisible by the whole downsample rate
         of the encoder. More detail of the whole downsample rate can be found
-        in UNet._check_input_devisible.
+        in UNet._check_input_divisible.
 
     """
 
@@ -373,7 +373,7 @@ class UNet(nn.Module):
             in_channels = base_channels * 2**i
 
     def forward(self, x):
-        self._check_input_devisible(x)
+        self._check_input_divisible(x)
         enc_outs = []
         for enc in self.encoder:
             x = enc(x)
@@ -395,7 +395,7 @@ class UNet(nn.Module):
                 if isinstance(m, _BatchNorm):
                     m.eval()
 
-    def _check_input_devisible(self, x):
+    def _check_input_divisible(self, x):
         h, w = x.shape[-2:]
         whole_downsample_rate = 1
         for i in range(1, self.num_stages):
@@ -403,7 +403,7 @@ class UNet(nn.Module):
                 whole_downsample_rate *= 2
         assert (h % whole_downsample_rate == 0) \
             and (w % whole_downsample_rate == 0),\
-            f'The input image size {(h, w)} should be devisible by the whole '\
+            f'The input image size {(h, w)} should be divisible by the whole '\
             f'downsample rate {whole_downsample_rate}, when num_stages is '\
             f'{self.num_stages}, strides is {self.strides}, and downsamples '\
             f'is {self.downsamples}.'

mmseg/models/decode_heads/apc_head.py

@@ -13,7 +13,7 @@ class ACM(nn.Module):
 
     Args:
         pool_scale (int): Pooling scale used in Adaptive Context
-            Module to extract region fetures.
+            Module to extract region features.
         fusion (bool): Add one conv to fuse residual feature.
         in_channels (int): Input channels.
         channels (int): Channels after modules, before conv_seg.

mmseg/models/decode_heads/dm_head.py

@@ -59,15 +59,15 @@ class DCM(nn.Module):
 
     def forward(self, x):
         """Forward function."""
-        generted_filter = self.filter_gen_conv(
+        generated_filter = self.filter_gen_conv(
             F.adaptive_avg_pool2d(x, self.filter_size))
         x = self.input_redu_conv(x)
         b, c, h, w = x.shape
         # [1, b * c, h, w], c = self.channels
         x = x.view(1, b * c, h, w)
         # [b * c, 1, filter_size, filter_size]
-        generted_filter = generted_filter.view(b * c, 1, self.filter_size,
+        generated_filter = generated_filter.view(b * c, 1, self.filter_size,
-                                               self.filter_size)
+                                                 self.filter_size)
         pad = (self.filter_size - 1) // 2
         if (self.filter_size - 1) % 2 == 0:
             p2d = (pad, pad, pad, pad)
@@ -75,7 +75,7 @@ class DCM(nn.Module):
             p2d = (pad + 1, pad, pad + 1, pad)
         x = F.pad(input=x, pad=p2d, mode='constant', value=0)
         # [1, b * c, h, w]
-        output = F.conv2d(input=x, weight=generted_filter, groups=b * c)
+        output = F.conv2d(input=x, weight=generated_filter, groups=b * c)
         # [b, c, h, w]
         output = output.view(b, c, h, w)
         if self.norm is not None:

mmseg/models/decode_heads/gc_head.py

@@ -17,7 +17,7 @@ class GCHead(FCNHead):
         pooling_type (str): The pooling type of context aggregation.
             Options are 'att', 'avg'. Default: 'avg'.
         fusion_types (tuple[str]): The fusion type for feature fusion.
-            Options are 'channel_add', 'channel_mul'. Defautl: ('channel_add',)
+            Options are 'channel_add', 'channel_mul'. Default: ('channel_add',)
     """
 
     def __init__(self,

mmseg/models/losses/lovasz_loss.py

@@ -132,7 +132,7 @@ def lovasz_softmax_flat(probs, labels, classes='present', class_weight=None):
         probs (torch.Tensor): [P, C], class probabilities at each prediction
             (between 0 and 1).
         labels (torch.Tensor): [P], ground truth labels (between 0 and C - 1).
-        classes (str | list[int], optional): Classes choosed to calculate loss.
+        classes (str | list[int], optional): Classes chosen to calculate loss.
             'all' for all classes, 'present' for classes present in labels, or
             a list of classes to average. Default: 'present'.
         class_weight (list[float], optional): The weight for each class.
@@ -183,7 +183,7 @@ def lovasz_softmax(probs,
             prediction (between 0 and 1).
         labels (torch.Tensor): [B, H, W], ground truth labels (between 0 and
             C - 1).
-        classes (str | list[int], optional): Classes choosed to calculate loss.
+        classes (str | list[int], optional): Classes chosen to calculate loss.
             'all' for all classes, 'present' for classes present in labels, or
             a list of classes to average. Default: 'present'.
         per_image (bool, optional): If per_image is True, compute the loss per
@@ -232,7 +232,7 @@ class LovaszLoss(nn.Module):
     Args:
         loss_type (str, optional): Binary or multi-class loss.
             Default: 'multi_class'. Options are "binary" and "multi_class".
-        classes (str | list[int], optional): Classes choosed to calculate loss.
+        classes (str | list[int], optional): Classes chosen to calculate loss.
             'all' for all classes, 'present' for classes present in labels, or
             a list of classes to average. Default: 'present'.
         per_image (bool, optional): If per_image is True, compute the loss per

mmseg/models/utils/inverted_residual.py

@@ -1,5 +1,5 @@
 from mmcv.cnn import ConvModule
-from torch import nn as nn
+from torch import nn
 from torch.utils import checkpoint as cp
 
 from .se_layer import SELayer
@@ -101,10 +101,10 @@ class InvertedResidualV3(nn.Module):
         in_channels (int): The input channels of this Module.
         out_channels (int): The output channels of this Module.
         mid_channels (int): The input channels of the depthwise convolution.
-        kernel_size (int): The kernal size of the depthwise convolution.
+        kernel_size (int): The kernel size of the depthwise convolution.
             Default: 3.
         stride (int): The stride of the depthwise convolution. Default: 1.
-        se_cfg (dict): Config dict for se layer. Defaul: None, which means no
+        se_cfg (dict): Config dict for se layer. Default: None, which means no
             se layer.
         with_expand_conv (bool): Use expand conv or not. If set False,
             mid_channels must be the same with in_channels. Default: True.

mmseg/models/utils/se_layer.py

@@ -15,10 +15,10 @@ class SELayer(nn.Module):
         conv_cfg (None or dict): Config dict for convolution layer.
             Default: None, which means using conv2d.
         act_cfg (dict or Sequence[dict]): Config dict for activation layer.
-            If act_cfg is a dict, two activation layers will be configurated
+            If act_cfg is a dict, two activation layers will be configured
             by this dict. If act_cfg is a sequence of dicts, the first
-            activation layer will be configurated by the first dict and the
+            activation layer will be configured by the first dict and the
-            second activation layer will be configurated by the second dict.
+            second activation layer will be configured by the second dict.
             Default: (dict(type='ReLU'), dict(type='HSigmoid', bias=3.0,
             divisor=6.0)).
     """

mmseg/models/utils/up_conv_block.py

@@ -36,7 +36,7 @@ class UpConvBlock(nn.Module):
             high-level feature map is the same as that of skip feature map
             (low-level feature map from encoder), it does not need upsample the
             high-level feature map and the upsample_cfg is None.
-        dcn (bool): Use deformable convoluton in convolutional layer or not.
+        dcn (bool): Use deformable convolution in convolutional layer or not.
             Default: None.
         plugins (dict): plugins for convolutional layers. Default: None.
     """

mmseg/ops/encoding.py

@@ -1,5 +1,5 @@
 import torch
-from torch import nn as nn
+from torch import nn
 from torch.nn import functional as F
 
 
@@ -43,14 +43,14 @@ class Encoding(nn.Module):
         return scaled_l2_norm
 
     @staticmethod
-    def aggregate(assigment_weights, x, codewords):
+    def aggregate(assignment_weights, x, codewords):
         num_codes, channels = codewords.size()
         reshaped_codewords = codewords.view((1, 1, num_codes, channels))
         batch_size = x.size(0)
 
         expanded_x = x.unsqueeze(2).expand(
             (batch_size, x.size(1), num_codes, channels))
-        encoded_feat = (assigment_weights.unsqueeze(3) *
+        encoded_feat = (assignment_weights.unsqueeze(3) *
                         (expanded_x - reshaped_codewords)).sum(dim=1)
         return encoded_feat
 
@@ -61,10 +61,10 @@ class Encoding(nn.Module):
         # [batch_size, height x width, channels]
         x = x.view(batch_size, self.channels, -1).transpose(1, 2).contiguous()
         # assignment_weights: [batch_size, channels, num_codes]
-        assigment_weights = F.softmax(
+        assignment_weights = F.softmax(
             self.scaled_l2(x, self.codewords, self.scale), dim=2)
         # aggregate
-        encoded_feat = self.aggregate(assigment_weights, x, self.codewords)
+        encoded_feat = self.aggregate(assignment_weights, x, self.codewords)
         return encoded_feat
 
     def __repr__(self):

tests/test_models/test_unet.py

@@ -171,7 +171,8 @@ def test_interp_conv():
         64,
         32,
         conv_first=False,
-        upsampe_cfg=dict(scale_factor=2, mode='bilinear', align_corners=False))
+        upsample_cfg=dict(
+            scale_factor=2, mode='bilinear', align_corners=False))
     x = torch.randn(1, 64, 128, 128)
     x_out = block(x)
     assert isinstance(block.interp_upsample[0], nn.Upsample)
@@ -184,7 +185,7 @@ def test_interp_conv():
         64,
         32,
         conv_first=False,
-        upsampe_cfg=dict(scale_factor=2, mode='nearest'))
+        upsample_cfg=dict(scale_factor=2, mode='nearest'))
     x = torch.randn(1, 64, 128, 128)
     x_out = block(x)
     assert isinstance(block.interp_upsample[0], nn.Upsample)
@@ -255,7 +256,7 @@ def test_up_conv_block():
         32,
         upsample_cfg=dict(
             type='InterpConv',
-            upsampe_cfg=dict(
+            upsample_cfg=dict(
                 scale_factor=2, mode='bilinear', align_corners=False)))
     skip_x = torch.randn(1, 32, 256, 256)
     x = torch.randn(1, 64, 128, 128)
@@ -285,7 +286,7 @@ def test_up_conv_block():
         dilation=3,
         upsample_cfg=dict(
             type='InterpConv',
-            upsampe_cfg=dict(
+            upsample_cfg=dict(
                 scale_factor=2, mode='bilinear', align_corners=False)))
     skip_x = torch.randn(1, 32, 256, 256)
     x = torch.randn(1, 64, 128, 128)
@@ -347,7 +348,7 @@ def test_unet():
         UNet(3, 64, 5, plugins=plugins)
 
     with pytest.raises(AssertionError):
-        # Check whether the input image size can be devisible by the whole
+        # Check whether the input image size can be divisible by the whole
         # downsample rate of the encoder. The whole downsample rate of this
         # case is 8.
         unet = UNet(
@@ -364,7 +365,7 @@ def test_unet():
         unet(x)
 
     with pytest.raises(AssertionError):
-        # Check whether the input image size can be devisible by the whole
+        # Check whether the input image size can be divisible by the whole
         # downsample rate of the encoder. The whole downsample rate of this
         # case is 16.
         unet = UNet(
@@ -381,7 +382,7 @@ def test_unet():
         unet(x)
 
     with pytest.raises(AssertionError):
-        # Check whether the input image size can be devisible by the whole
+        # Check whether the input image size can be divisible by the whole
         # downsample rate of the encoder. The whole downsample rate of this
         # case is 8.
         unet = UNet(
@@ -398,7 +399,7 @@ def test_unet():
         unet(x)
 
     with pytest.raises(AssertionError):
-        # Check whether the input image size can be devisible by the whole
+        # Check whether the input image size can be divisible by the whole
         # downsample rate of the encoder. The whole downsample rate of this
         # case is 8.
         unet = UNet(
@@ -415,7 +416,7 @@ def test_unet():
         unet(x)
 
     with pytest.raises(AssertionError):
-        # Check whether the input image size can be devisible by the whole
+        # Check whether the input image size can be divisible by the whole
         # downsample rate of the encoder. The whole downsample rate of this
         # case is 32.
         unet = UNet(