Fix some typos (#4858)

* Fix some typos

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

* More

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

* More

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

* More

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

docs/changelog.md

@@ -393,7 +393,7 @@ Function `get_subset_by_classes` in dataset is refactored and only filters out i
 - Implement FCOS training tricks (#2935)
 - Use BaseDenseHead as base class for anchor-base heads (#2963)
 - Add `with_cp` for BasicBlock (#2891)
-- Add `stem_channles` argument for ResNet (#2954)
+- Add `stem_channels` argument for ResNet (#2954)
 
 **Improvements**
 

docs/tutorials/config.md

@@ -290,7 +290,7 @@ test_pipeline = [
         flip=False,  # Whether to flip images during testing
         transforms=[
             dict(type='Resize',  # Use resize augmentation
-                 keep_ratio=True),  # Whether to keep the ratio between height and width, the img_scale set here will be supressed by the img_scale set above.
+                 keep_ratio=True),  # Whether to keep the ratio between height and width, the img_scale set here will be suppressed by the img_scale set above.
             dict(type='RandomFlip'),  # Thought RandomFlip is added in pipeline, it is not used because flip=False
             dict(
                 type='Normalize',  # Normalization config, the values are from img_norm_cfg

mmdet/core/bbox/assigners/atss_assigner.py

@@ -48,7 +48,7 @@ class ATSSAssigner(BaseAssigner):
         4. get corresponding iou for the these candidates, and compute the
            mean and std, set mean + std as the iou threshold
         5. select these candidates whose iou are greater than or equal to
-           the threshold as postive
+           the threshold as positive
         6. limit the positive sample's center in gt
 
 

mmdet/core/bbox/assigners/base_assigner.py

@@ -6,5 +6,4 @@ class BaseAssigner(metaclass=ABCMeta):
 
     @abstractmethod
     def assign(self, bboxes, gt_bboxes, gt_bboxes_ignore=None, gt_labels=None):
-        """Assign boxes to either a ground truth boxe or a negative boxes."""
-        pass
+        """Assign boxes to either a ground truth boxes or a negative boxes."""

mmdet/core/bbox/assigners/hungarian_assigner.py

@@ -18,7 +18,7 @@ class HungarianAssigner(BaseAssigner):
 
     This class computes an assignment between the targets and the predictions
     based on the costs. The costs are weighted sum of three components:
-    classfication cost, regression L1 cost and regression iou cost. The
+    classification cost, regression L1 cost and regression iou cost. The
     targets don't include the no_object, so generally there are more
     predictions than targets. After the one-to-one matching, the un-matched
     are treated as backgrounds. Thus each query prediction will be assigned

mmdet/core/bbox/assigners/max_iou_assigner.py

@@ -182,7 +182,7 @@ class MaxIoUAssigner(BaseAssigner):
         assigned_gt_inds[pos_inds] = argmax_overlaps[pos_inds] + 1
 
         if self.match_low_quality:
-            # Low-quality matching will overwirte the assigned_gt_inds assigned
+            # Low-quality matching will overwrite the assigned_gt_inds assigned
             # in Step 3. Thus, the assigned gt might not be the best one for
             # prediction.
             # For example, if bbox A has 0.9 and 0.8 iou with GT bbox 1 & 2,

mmdet/core/bbox/coder/base_bbox_coder.py

@@ -10,10 +10,8 @@ class BaseBBoxCoder(metaclass=ABCMeta):
     @abstractmethod
     def encode(self, bboxes, gt_bboxes):
         """Encode deltas between bboxes and ground truth boxes."""
-        pass
 
     @abstractmethod
     def decode(self, bboxes, bboxes_pred):
         """Decode the predicted bboxes according to prediction and base
         boxes."""
-        pass

mmdet/core/bbox/coder/bucketing_bbox_coder.py

@@ -10,7 +10,7 @@ from .base_bbox_coder import BaseBBoxCoder
 
 @BBOX_CODERS.register_module()
 class BucketingBBoxCoder(BaseBBoxCoder):
-    """Bucketing BBox Coder for Side-Aware Bounday Localization (SABL).
+    """Bucketing BBox Coder for Side-Aware Boundary Localization (SABL).
 
     Boundary Localization with Bucketing and Bucketing Guided Rescoring
     are implemented here.
@@ -255,7 +255,7 @@ def bbox2bucket(proposals,
         bucket_cls_d_weights
     ],
                                    dim=-1)
-    # ignore second nearest buckets for cls if necessay
+    # ignore second nearest buckets for cls if necessary
     if cls_ignore_neighbor:
         bucket_cls_weights = (~((bucket_cls_weights == 1) &
                                 (bucket_labels == 0))).float()

mmdet/core/bbox/iou_calculators/iou2d_calculator.py

@@ -88,7 +88,7 @@ def bbox_overlaps(bboxes1, bboxes2, mode='iou', is_aligned=False, eps=1e-6):
     """
 
     assert mode in ['iou', 'iof', 'giou'], f'Unsupported mode {mode}'
-    # Either the boxes are empty or the length of boxes's last dimenstion is 4
+    # Either the boxes are empty or the length of boxes' last dimension is 4
     assert (bboxes1.size(-1) == 4 or bboxes1.size(0) == 0)
     assert (bboxes2.size(-1) == 4 or bboxes2.size(0) == 0)
 

mmdet/core/bbox/match_costs/match_cost.py

@@ -136,7 +136,7 @@ class ClassificationCost(object):
         # Following the official DETR repo, contrary to the loss that
         # NLL is used, we approximate it in 1 - cls_score[gt_label].
         # The 1 is a constant that doesn't change the matching,
-        # so it can be ommitted.
+        # so it can be omitted.
         cls_score = cls_pred.softmax(-1)
         cls_cost = -cls_score[:, gt_labels]
         return cls_cost * self.weight
@@ -179,6 +179,6 @@ class IoUCost(object):
         # overlaps: [num_bboxes, num_gt]
         overlaps = bbox_overlaps(
             bboxes, gt_bboxes, mode=self.iou_mode, is_aligned=False)
-        # The 1 is a constant that doesn't change the matching, so ommitted.
+        # The 1 is a constant that doesn't change the matching, so omitted.
         iou_cost = -overlaps
         return iou_cost * self.weight

mmdet/core/evaluation/class_names.py

@@ -94,7 +94,7 @@ dataset_aliases = {
     'imagenet_det': ['det', 'imagenet_det', 'ilsvrc_det'],
     'imagenet_vid': ['vid', 'imagenet_vid', 'ilsvrc_vid'],
     'coco': ['coco', 'mscoco', 'ms_coco'],
-    'wider_face': ['WIDERFaceDataset', 'wider_face', 'WDIERFace'],
+    'wider_face': ['WIDERFaceDataset', 'wider_face', 'WIDERFace'],
     'cityscapes': ['cityscapes']
 }
 

mmdet/core/evaluation/mean_ap.py

@@ -124,7 +124,7 @@ def tpfp_imagenet(det_bboxes,
             # find best overlapped available gt
             for j in range(num_gts):
                 # different from PASCAL VOC: allow finding other gts if the
-                # best overlaped ones are already matched by other det bboxes
+                # best overlapped ones are already matched by other det bboxes
                 if gt_covered[j]:
                     continue
                 elif ious[i, j] >= iou_thrs[j] and ious[i, j] > max_iou:

mmdet/core/mask/structures.py

@@ -23,7 +23,6 @@ class BaseInstanceMasks(metaclass=ABCMeta):
         Returns:
             BaseInstanceMasks: The rescaled masks.
         """
-        pass
 
     @abstractmethod
     def resize(self, out_shape, interpolation='nearest'):
@@ -36,7 +35,6 @@ class BaseInstanceMasks(metaclass=ABCMeta):
         Returns:
             BaseInstanceMasks: The resized masks.
         """
-        pass
 
     @abstractmethod
     def flip(self, flip_direction='horizontal'):
@@ -48,7 +46,6 @@ class BaseInstanceMasks(metaclass=ABCMeta):
         Returns:
             BaseInstanceMasks: The flipped masks.
         """
-        pass
 
     @abstractmethod
     def pad(self, out_shape, pad_val):
@@ -61,7 +58,6 @@ class BaseInstanceMasks(metaclass=ABCMeta):
         Returns:
             BaseInstanceMasks: The padded masks.
         """
-        pass
 
     @abstractmethod
     def crop(self, bbox):
@@ -73,7 +69,6 @@ class BaseInstanceMasks(metaclass=ABCMeta):
         Return:
             BaseInstanceMasks: The cropped masks.
         """
-        pass
 
     @abstractmethod
     def crop_and_resize(self,
@@ -99,18 +94,15 @@ class BaseInstanceMasks(metaclass=ABCMeta):
         Return:
             BaseInstanceMasks: the cropped and resized masks.
         """
-        pass
 
     @abstractmethod
     def expand(self, expanded_h, expanded_w, top, left):
         """see :class:`Expand`."""
-        pass
 
     @property
     @abstractmethod
     def areas(self):
         """ndarray: areas of each instance."""
-        pass
 
     @abstractmethod
     def to_ndarray(self):
@@ -119,7 +111,6 @@ class BaseInstanceMasks(metaclass=ABCMeta):
         Return:
             ndarray: Converted masks in the format of ndarray.
         """
-        pass
 
     @abstractmethod
     def to_tensor(self, dtype, device):
@@ -132,7 +123,6 @@ class BaseInstanceMasks(metaclass=ABCMeta):
         Returns:
             Tensor: Converted masks in the format of Tensor.
         """
-        pass
 
     @abstractmethod
     def translate(self,
@@ -154,7 +144,6 @@ class BaseInstanceMasks(metaclass=ABCMeta):
         Returns:
             Translated masks.
         """
-        pass
 
     def shear(self,
               out_shape,
@@ -176,7 +165,6 @@ class BaseInstanceMasks(metaclass=ABCMeta):
         Returns:
             ndarray: Sheared masks.
         """
-        pass
 
     @abstractmethod
     def rotate(self, out_shape, angle, center=None, scale=1.0, fill_val=0):
@@ -195,7 +183,6 @@ class BaseInstanceMasks(metaclass=ABCMeta):
         Returns:
             Rotated masks.
         """
-        pass
 
 
 class BitmapMasks(BaseInstanceMasks):

mmdet/datasets/custom.py

@@ -262,7 +262,6 @@ class CustomDataset(Dataset):
 
     def format_results(self, results, **kwargs):
         """Place holder to format result to dataset specific output."""
-        pass
 
     def evaluate(self,
                  results,

mmdet/datasets/pipelines/auto_augment.py

@@ -634,7 +634,7 @@ class Translate(object):
                 min_y = np.maximum(0, min_y + offset)
                 max_y = np.minimum(h, max_y + offset)
 
-            # the boxs translated outside of image will be filtered along with
+            # the boxes translated outside of image will be filtered along with
             # the corresponding masks, by invoking ``_filter_invalid``.
             results[key] = np.concatenate([min_x, min_y, max_x, max_y],
                                           axis=-1)

mmdet/datasets/pipelines/transforms.py

@@ -120,7 +120,7 @@ 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 \

mmdet/models/backbones/detectors_resnet.py

@@ -149,7 +149,7 @@ class ResLayer(nn.Sequential):
                  rfp_inplanes=None,
                  **kwargs):
         self.block = block
-        assert downsample_first, f'downsampel_first={downsample_first} is ' \
+        assert downsample_first, f'downsample_first={downsample_first} is ' \
                                  'not supported in DetectoRS'
 
         downsample = None

mmdet/models/backbones/hrnet.py

@@ -527,7 +527,7 @@ 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:

mmdet/models/backbones/regnet.py

@@ -21,7 +21,7 @@ class RegNet(ResNet):
             - wm (float): quantization parameter to quantize the width
             - depth (int): depth of the backbone
             - group_w (int): width of group
-            - bot_mul (float): bottleneck ratio, i.e. expansion of bottlneck.
+            - bot_mul (float): bottleneck ratio, i.e. expansion of bottleneck.
         strides (Sequence[int]): Strides of the first block of each stage.
         base_channels (int): Base channels after stem layer.
         in_channels (int): Number of input image channels. Default: 3.
@@ -252,7 +252,7 @@ class RegNet(ResNet):
 
     @staticmethod
     def quantize_float(number, divisor):
-        """Converts a float to closest non-zero int divisible by divior.
+        """Converts a float to closest non-zero int divisible by divisor.
 
         Args:
             number (int): Original number to be quantized.

mmdet/models/dense_heads/anchor_free_head.py

@@ -197,7 +197,7 @@ class AnchorFreeHead(BaseDenseHead, BBoxTestMixin):
         return multi_apply(self.forward_single, feats)[:2]
 
     def forward_single(self, x):
-        """Forward features of a single scale levle.
+        """Forward features of a single scale level.
 
         Args:
             x (Tensor): FPN feature maps of the specified stride.
@@ -274,7 +274,7 @@ class AnchorFreeHead(BaseDenseHead, BBoxTestMixin):
 
     @abstractmethod
     def get_targets(self, points, gt_bboxes_list, gt_labels_list):
-        """Compute regression, classification and centerss targets for points
+        """Compute regression, classification and centerness targets for points
         in multiple images.
 
         Args:

mmdet/models/dense_heads/atss_head.py

@@ -612,7 +612,7 @@ class ATSSHead(AnchorHead):
                     image with shape (N, 4).
                 bbox_weights (Tensor): BBox weights of all anchors in the
                     image with shape (N, 4)
-                pos_inds (Tensor): Indices of postive anchor with shape
+                pos_inds (Tensor): Indices of positive anchor with shape
                     (num_pos,).
                 neg_inds (Tensor): Indices of negative anchor with shape
                     (num_neg,).

mmdet/models/dense_heads/cascade_rpn_head.py

@@ -53,7 +53,7 @@ class AdaptiveConv(nn.Module):
         assert kernel_size == 3, 'Adaptive conv only supports kernels 3'
         if self.adapt_type == 'offset':
             assert stride == 1 and padding == 1 and groups == 1, \
-                'Addptive conv offset mode only supports padding: {1}, ' \
+                'Adaptive conv offset mode only supports padding: {1}, ' \
                 f'stride: {1}, groups: {1}'
             self.conv = DeformConv2d(
                 in_channels,
@@ -99,7 +99,7 @@ class StageCascadeRPNHead(RPNHead):
         in_channels (int): Number of channels in the input feature map.
         anchor_generator (dict): anchor generator config.
         adapt_cfg (dict): adaptation config.
-        bridged_feature (bool, optional): wheater update rpn feature.
+        bridged_feature (bool, optional): whether update rpn feature.
             Default: False.
         with_cls (bool, optional): wheather use classification branch.
             Default: True.
@@ -471,7 +471,7 @@ class StageCascadeRPNHead(RPNHead):
             num_total_samples = num_total_pos + num_total_neg
         else:
             # 200 is hard-coded average factor,
-            # which follows guilded anchoring.
+            # which follows guided anchoring.
             num_total_samples = sum([label.numel()
                                      for label in labels_list]) / 200.0
 

mmdet/models/dense_heads/corner_head.py

@@ -805,7 +805,7 @@ class CornerHead(BaseDenseHead):
         return feat
 
     def _local_maximum(self, heat, kernel=3):
-        """Extract local maximum pixel with given kernal.
+        """Extract local maximum pixel with given kernel.
 
         Args:
             heat (Tensor): Target heatmap.

mmdet/models/dense_heads/fcos_head.py

@@ -16,7 +16,7 @@ class FCOSHead(AnchorFreeHead):
     """Anchor-free head used in `FCOS <https://arxiv.org/abs/1904.01355>`_.
 
     The FCOS head does not use anchor boxes. Instead bounding boxes are
-    predicted at each pixel and a centerness measure is used to supress
+    predicted at each pixel and a centerness measure is used to suppress
     low-quality predictions.
     Here norm_on_bbox, centerness_on_reg, dcn_on_last_conv are training
     tricks used in official repo, which will bring remarkable mAP gains
@@ -116,14 +116,14 @@ class FCOSHead(AnchorFreeHead):
                 bbox_preds (list[Tensor]): Box energies / deltas for each \
                     scale level, each is a 4D-tensor, the channel number is \
                     num_points * 4.
-                centernesses (list[Tensor]): Centerss for each scale level, \
+                centernesses (list[Tensor]): centerness for each scale level, \
                     each is a 4D-tensor, the channel number is num_points * 1.
         """
         return multi_apply(self.forward_single, feats, self.scales,
                            self.strides)
 
     def forward_single(self, x, scale, stride):
-        """Forward features of a single scale levle.
+        """Forward features of a single scale level.
 
         Args:
             x (Tensor): FPN feature maps of the specified stride.
@@ -171,7 +171,7 @@ class FCOSHead(AnchorFreeHead):
             bbox_preds (list[Tensor]): Box energies / deltas for each scale
                 level, each is a 4D-tensor, the channel number is
                 num_points * 4.
-            centernesses (list[Tensor]): Centerss for each scale level, each
+            centernesses (list[Tensor]): centerness for each scale level, each
                 is a 4D-tensor, the channel number is num_points * 1.
             gt_bboxes (list[Tensor]): Ground truth bboxes for each image with
                 shape (num_gts, 4) in [tl_x, tl_y, br_x, br_y] format.
@@ -466,7 +466,7 @@ class FCOSHead(AnchorFreeHead):
         return points
 
     def get_targets(self, points, gt_bboxes_list, gt_labels_list):
-        """Compute regression, classification and centerss targets for points
+        """Compute regression, classification and centerness targets for points
         in multiple images.
 
         Args:

mmdet/models/dense_heads/fsaf_head.py

@@ -394,7 +394,7 @@ class FSAFHead(RetinaHead):
                 - cls_loss: Reduced corrected classification loss. Scalar.
                 - reg_loss: Reduced corrected regression loss. Scalar.
                 - pos_flags (Tensor): Corrected bool tensor indicating the
-                  final postive anchors. Shape: (num_anchors, ).
+                  final positive anchors. Shape: (num_anchors, ).
         """
         loc_weight = torch.ones_like(reg_loss)
         cls_weight = torch.ones_like(cls_loss)

mmdet/models/dense_heads/gfl_head.py

@@ -575,7 +575,7 @@ class GFLHead(AnchorHead):
                     image with shape (N, 4).
                 bbox_weights (Tensor): BBox weights of all anchors in the
                     image with shape (N, 4).
-                pos_inds (Tensor): Indices of postive anchor with shape
+                pos_inds (Tensor): Indices of positive anchor with shape
                     (num_pos,).
                 neg_inds (Tensor): Indices of negative anchor with shape
                     (num_neg,).

mmdet/models/dense_heads/paa_head.py

@@ -385,7 +385,7 @@ class PAAHead(ATSSHead):
                 - pos_inds_temp (Tensor): Indices of positive samples.
                 - ignore_inds_temp (Tensor): Indices of ignore samples.
         """
-        # The implementation is (c) in Fig.3 in origin paper intead of (b).
+        # The implementation is (c) in Fig.3 in origin paper instead of (b).
         # You can refer to issues such as
         # https://github.com/kkhoot/PAA/issues/8 and
         # https://github.com/kkhoot/PAA/issues/9.
@@ -632,9 +632,9 @@ class PAAHead(ATSSHead):
                     after voting, with shape (num_anchors,).
         """
         candidate_mask = mlvl_nms_scores > score_thr
-        candidate_mask_nozeros = candidate_mask.nonzero()
-        candidate_inds = candidate_mask_nozeros[:, 0]
-        candidate_labels = candidate_mask_nozeros[:, 1]
+        candidate_mask_nonzeros = candidate_mask.nonzero()
+        candidate_inds = candidate_mask_nonzeros[:, 0]
+        candidate_labels = candidate_mask_nonzeros[:, 1]
         candidate_bboxes = mlvl_bboxes[candidate_inds]
         candidate_scores = mlvl_nms_scores[candidate_mask]
         det_bboxes_voted = []

mmdet/models/dense_heads/reppoints_head.py

@@ -167,7 +167,7 @@ class RepPointsHead(AnchorFreeHead):
 
         :param pts: the input points sets (fields), each points
             set (fields) is represented as 2n scalar.
-        :param y_first: if y_fisrt=True, the point set is represented as
+        :param y_first: if y_first=True, the point set is represented as
             [y1, x1, y2, x2 ... yn, xn], otherwise the point set is
             represented as [x1, y1, x2, y2 ... xn, yn].
         :return: each points set is converting to a bbox [x1, y1, x2, y2].

mmdet/models/dense_heads/transformer_head.py

@@ -582,7 +582,7 @@ class TransformerHead(AnchorFreeHead):
                 [nb_dec, bs, num_query, 4].
             img_metas (list[dict]): Meta information of each image.
             rescale (bool, optional): If True, return boxes in original
-                image space. Defalut False.
+                image space. Default False.
 
         Returns:
             list[list[Tensor, Tensor]]: Each item in result_list is 2-tuple. \
@@ -593,7 +593,7 @@ class TransformerHead(AnchorFreeHead):
                 the corresponding box.
         """
         # NOTE defaultly only using outputs from the last feature level,
-        # and only the ouputs from the last decoder layer is used.
+        # and only the outputs from the last decoder layer is used.
         cls_scores = all_cls_scores_list[-1][-1]
         bbox_preds = all_bbox_preds_list[-1][-1]
 

mmdet/models/detectors/sparse_rcnn.py

@@ -46,7 +46,7 @@ class SparseRCNN(TwoStageDetector):
 
         assert proposals is None, 'Sparse R-CNN does not support' \
                                   ' external proposals'
-        assert gt_masks is None, 'Sparse R-CNN does not instance segmenntaion'
+        assert gt_masks is None, 'Sparse R-CNN does not instance segmentation'
 
         x = self.extract_feat(img)
         proposal_boxes, proposal_features, imgs_whwh = \

mmdet/models/detectors/yolact.py

@@ -25,7 +25,7 @@ class YOLACT(SingleStageDetector):
         self.init_segm_mask_weights()
 
     def init_segm_mask_weights(self):
-        """Initialize weights of the YOLACT semg head and YOLACT mask head."""
+        """Initialize weights of the YOLACT segm head and YOLACT mask head."""
         self.segm_head.init_weights()
         self.mask_head.init_weights()
 

mmdet/models/losses/gaussian_focal_loss.py

@@ -41,7 +41,7 @@ class GaussianFocalLoss(nn.Module):
 
     Args:
         alpha (float): Power of prediction.
-        gamma (float): Power of target for negtive samples.
+        gamma (float): Power of target for negative samples.
         reduction (str): Options are "none", "mean" and "sum".
         loss_weight (float): Loss weight of current loss.
     """

mmdet/models/necks/bfp.py

@@ -8,7 +8,7 @@ from ..builder import NECKS
 
 @NECKS.register_module()
 class BFP(nn.Module):
-    """BFP (Balanced Feature Pyrmamids)
+    """BFP (Balanced Feature Pyramids)
 
     BFP takes multi-level features as inputs and gather them into a single one,
     then refine the gathered feature and scatter the refined results to

mmdet/models/necks/fpn_carafe.py

@@ -10,7 +10,7 @@ class FPN_CARAFE(nn.Module):
     """FPN_CARAFE is a more flexible implementation of FPN. It allows more
     choice for upsample methods during the top-down pathway.
 
-    It can reproduce the preformance of ICCV 2019 paper
+    It can reproduce the performance of ICCV 2019 paper
     CARAFE: Content-Aware ReAssembly of FEatures
     Please refer to https://arxiv.org/abs/1905.02188 for more details.
 

mmdet/models/necks/hrfpn.py

@@ -9,7 +9,7 @@ from ..builder import NECKS
 
 @NECKS.register_module()
 class HRFPN(nn.Module):
-    """HRFPN (High Resolution Feature Pyrmamids)
+    """HRFPN (High Resolution Feature Pyramids)
 
     paper: `High-Resolution Representations for Labeling Pixels and Regions
     <https://arxiv.org/abs/1904.04514>`_.

mmdet/models/roi_heads/base_roi_head.py

@@ -81,7 +81,6 @@ class BaseRoIHead(nn.Module, metaclass=ABCMeta):
                       gt_masks=None,
                       **kwargs):
         """Forward function during training."""
-        pass
 
     async def async_simple_test(self, x, img_meta, **kwargs):
         """Asynchronized test function."""
@@ -95,7 +94,6 @@ class BaseRoIHead(nn.Module, metaclass=ABCMeta):
                     rescale=False,
                     **kwargs):
         """Test without augmentation."""
-        pass
 
     def aug_test(self, x, proposal_list, img_metas, rescale=False, **kwargs):
         """Test with augmentations.
@@ -103,4 +101,3 @@ class BaseRoIHead(nn.Module, metaclass=ABCMeta):
         If rescale is False, then returned bboxes and masks will fit the scale
         of imgs[0].
         """
-        pass

mmdet/models/roi_heads/bbox_heads/sabl_head.py

@@ -244,13 +244,13 @@ class SABLHead(nn.Module):
         reg_fy = torch.transpose(reg_fy, 1, 2)
         return reg_fx.contiguous(), reg_fy.contiguous()
 
-    def reg_pred(self, x, offfset_fcs, cls_fcs):
-        """Predict bucketing esimation (cls_pred) and fine regression (offset
+    def reg_pred(self, x, offset_fcs, cls_fcs):
+        """Predict bucketing estimation (cls_pred) and fine regression (offset
         pred) with side-aware features."""
         x_offset = x.view(-1, self.reg_in_channels)
         x_cls = x.view(-1, self.reg_in_channels)
 
-        for fc in offfset_fcs:
+        for fc in offset_fcs:
             x_offset = self.relu(fc(x_offset))
         for fc in cls_fcs:
             x_cls = self.relu(fc(x_cls))

mmdet/models/roi_heads/mask_heads/fcn_mask_head.py

@@ -304,7 +304,7 @@ class FCNMaskHead(nn.Module):
 
 
 def _do_paste_mask(masks, boxes, img_h, img_w, skip_empty=True):
-    """Paste instance masks acoording to boxes.
+    """Paste instance masks according to boxes.
 
     This implementation is modified from
     https://github.com/facebookresearch/detectron2/

mmdet/models/roi_heads/mask_heads/mask_point_head.py

@@ -49,7 +49,7 @@ class MaskPointHead(nn.Module):
         super().__init__()
         self.num_fcs = num_fcs
         self.in_channels = in_channels
-        self.fc_channles = fc_channels
+        self.fc_channels = fc_channels
         self.num_classes = num_classes
         self.class_agnostic = class_agnostic
         self.coarse_pred_each_layer = coarse_pred_each_layer

mmdet/models/roi_heads/sparse_roi_head.py

@@ -190,10 +190,10 @@ class SparseRoIHead(CascadeRoIHead):
             cls_pred_list = bbox_results['detach_cls_score_list']
             proposal_list = bbox_results['detach_proposal_list']
             for i in range(num_imgs):
-                normolize_bbox_ccwh = bbox_xyxy_to_cxcywh(proposal_list[i] /
+                normalize_bbox_ccwh = bbox_xyxy_to_cxcywh(proposal_list[i] /
                                                           imgs_whwh[i])
                 assign_result = self.bbox_assigner[stage].assign(
-                    normolize_bbox_ccwh, cls_pred_list[i], gt_bboxes[i],
+                    normalize_bbox_ccwh, cls_pred_list[i], gt_bboxes[i],
                     gt_labels[i], img_metas[i])
                 sampling_result = self.bbox_sampler[stage].sample(
                     assign_result, proposal_list[i], gt_bboxes[i])

mmdet/models/utils/transformer.py

@@ -109,7 +109,7 @@ class FFN(nn.Module):
             `MultiheadAttention`.
         feedforward_channels (int): The hidden dimension of FFNs.
         num_fcs (int, optional): The number of fully-connected layers in
-            FFNs. Defaluts to 2.
+            FFNs. Defaults to 2.
         act_cfg (dict, optional): The activation config for FFNs.
         dropout (float, optional): Probability of an element to be
             zeroed. Default 0.0.
@@ -179,7 +179,7 @@ class TransformerEncoderLayer(nn.Module):
         order (tuple[str]): The order for encoder layer. Valid examples are
             ('selfattn', 'norm', 'ffn', 'norm') and ('norm', 'selfattn',
             'norm', 'ffn'). Default ('selfattn', 'norm', 'ffn', 'norm').
-        act_cfg (dict): The activation config for FFNs. Defalut ReLU.
+        act_cfg (dict): The activation config for FFNs. Default ReLU.
         norm_cfg (dict): Config dict for normalization layer. Default
             layer normalization.
         num_fcs (int): The number of fully-connected layers for FFNs.
@@ -280,7 +280,7 @@ class TransformerDecoderLayer(nn.Module):
             ('selfattn', 'norm', 'multiheadattn', 'norm', 'ffn', 'norm') and
             ('norm', 'selfattn', 'norm', 'multiheadattn', 'norm', 'ffn').
             Default the former.
-        act_cfg (dict): Same as `TransformerEncoderLayer`. Defalut ReLU.
+        act_cfg (dict): Same as `TransformerEncoderLayer`. Default ReLU.
         norm_cfg (dict): Config dict for normalization layer. Default
             layer normalization.
         num_fcs (int): The number of fully-connected layers in FFNs.
@@ -412,7 +412,7 @@ class TransformerEncoder(nn.Module):
         feedforward_channels (int): Same as `TransformerEncoderLayer`.
         dropout (float): Same as `TransformerEncoderLayer`. Default 0.0.
         order (tuple[str]): Same as `TransformerEncoderLayer`.
-        act_cfg (dict): Same as `TransformerEncoderLayer`. Defalut ReLU.
+        act_cfg (dict): Same as `TransformerEncoderLayer`. Default ReLU.
         norm_cfg (dict): Same as `TransformerEncoderLayer`. Default
             layer normalization.
         num_fcs (int): Same as `TransformerEncoderLayer`. Default 2.
@@ -496,7 +496,7 @@ class TransformerDecoder(nn.Module):
         feedforward_channels (int): Same as `TransformerDecoderLayer`.
         dropout (float): Same as `TransformerDecoderLayer`. Default 0.0.
         order (tuple[str]): Same as `TransformerDecoderLayer`.
-        act_cfg (dict): Same as `TransformerDecoderLayer`. Defalut ReLU.
+        act_cfg (dict): Same as `TransformerDecoderLayer`. Default ReLU.
         norm_cfg (dict): Same as `TransformerDecoderLayer`. Default
             layer normalization.
         num_fcs (int): Same as `TransformerDecoderLayer`. Default 2.
@@ -621,7 +621,7 @@ class Transformer(nn.Module):
             encoder and decoder.
         dropout (float): Probability of an element to be zeroed. Default 0.0.
         act_cfg (dict): Activation config for FFNs used in both encoder
-            and decoder. Defalut ReLU.
+            and decoder. Default ReLU.
         norm_cfg (dict): Config dict for normalization used in both encoder
             and decoder. Default layer normalization.
         num_fcs (int): The number of fully-connected layers in FFNs, which is

tools/analysis_tools/robustness_eval.py

@@ -89,7 +89,7 @@ def get_coco_style_results(filename,
             print(f'Mean Performance under Corruption [mPC] ({task})')
             print_coco_results(mPC)
         if 'rPC' in prints:
-            print(f'Realtive Performance under Corruption [rPC] ({task})')
+            print(f'Relative Performance under Corruption [rPC] ({task})')
             print_coco_results(rPC)
     else:
         if 'P' in prints:
@@ -146,7 +146,7 @@ def get_voc_style_results(filename, prints='mPC', aggregate='benchmark'):
         print('Mean Performance under Corruption [mPC] in AP50 = '
               f'{np.mean(mPC):0.3f}')
     if 'rPC' in prints:
-        print('Realtive Performance under Corruption [rPC] in % = '
+        print('Relative Performance under Corruption [rPC] in % = '
               f'{np.mean(rPC) * 100:0.1f}')
 
     return np.mean(results, axis=2, keepdims=True)

tools/analysis_tools/test_robustness.py

@@ -367,7 +367,7 @@ def main():
                 mmcv.dump(aggregated_results, eval_results_filename)
 
     if rank == 0:
-        # print filan results
+        # print final results
         print('\nAggregated results:')
         prints = args.final_prints
         aggregate = args.final_prints_aggregate

tools/dataset_converters/cityscapes.py

@@ -140,7 +140,7 @@ def main():
     for split, json_name in set_name.items():
         print(f'Converting {split} into {json_name}')
         with mmcv.Timer(
-                print_tmpl='It tooks {}s to convert Cityscapes annotation'):
+                print_tmpl='It took {}s to convert Cityscapes annotation'):
             files = collect_files(
                 osp.join(img_dir, split), osp.join(gt_dir, split))
             image_infos = collect_annotations(files, nproc=args.nproc)