Merge branch 'add-model-docstring' into 'master'

Reformat docstrings in code

See merge request open-mmlab/mmdet.3d!115

mmdet3d/models/dense_heads/vote_head.py

 import numpy as np
 import torch
-import torch.nn as nn
-import torch.nn.functional as F
 from mmcv.cnn import ConvModule
+from torch import nn as nn
+from torch.nn import functional as F
 
 from mmdet3d.core.post_processing import aligned_3d_nms
 from mmdet3d.models.builder import build_loss

mmdet3d/models/detectors/base.py

-import os.path as osp
+from os import path as osp
 
 from mmdet3d.core import Box3DMode, show_result
 from mmdet.models.detectors import BaseDetector
 
 
 class Base3DDetector(BaseDetector):
-    """Base class for detectors"""
+    """Base class for detectors."""
 
     def forward_test(self, points, img_metas, img=None, **kwargs):
         """
@@ -42,14 +42,15 @@ class Base3DDetector(BaseDetector):
             return self.aug_test(points, img_metas, img, **kwargs)
 
     def forward(self, return_loss=True, **kwargs):
-        """
-        Calls either forward_train or forward_test depending on whether
-        return_loss=True. Note this setting will change the expected inputs.
-        When `return_loss=True`, img and img_metas are single-nested (i.e.
+        """Calls either forward_train or forward_test depending on whether
+        return_loss=True.
+
+        Note this setting will change the expected inputs. When
+        `return_loss=True`, img and img_metas are single-nested (i.e.
         torch.Tensor and list[dict]), and when `resturn_loss=False`, img and
-        img_metas should be double nested
-        (i.e.  list[torch.Tensor], list[list[dict]]), with the outer list
-        indicating test time augmentations.
+        img_metas should be double nested (i.e.  list[torch.Tensor],
+        list[list[dict]]), with the outer list indicating test time
+        augmentations.
         """
         if return_loss:
             return self.forward_train(**kwargs)

mmdet3d/models/detectors/dynamic_voxelnet.py

 import torch
-import torch.nn.functional as F
+from torch.nn import functional as F
 
 from mmdet.models import DETECTORS
 from .voxelnet import VoxelNet
@@ -7,8 +7,8 @@ from .voxelnet import VoxelNet
 
 @DETECTORS.register_module()
 class DynamicVoxelNet(VoxelNet):
-    """VoxelNet using `dynamic voxelization
-    <https://arxiv.org/abs/1910.06528>`_."""
+    r"""VoxelNet using `dynamic voxelization <https://arxiv.org/abs/1910.06528>`_.
+    """
 
     def __init__(self,
                  voxel_layer,
@@ -33,7 +33,7 @@ class DynamicVoxelNet(VoxelNet):
         )
 
     def extract_feat(self, points, img_metas):
-        """Extract features from points"""
+        """Extract features from points."""
         voxels, coors = self.voxelize(points)
         voxel_features, feature_coors = self.voxel_encoder(voxels, coors)
         batch_size = coors[-1, 0].item() + 1
@@ -45,7 +45,14 @@ class DynamicVoxelNet(VoxelNet):
 
     @torch.no_grad()
     def voxelize(self, points):
-        """Apply dynamic voxelization to points"""
+        """Apply dynamic voxelization to points.
+
+        Args:
+            points (list[torch.Tensor]): Points of each sample.
+
+        Returns:
+            tuple[torch.Tensor]: Concatenated points and coordinates.
+        """
         coors = []
         # dynamic voxelization only provide a coors mapping
         for res in points:

mmdet3d/models/detectors/mvx_faster_rcnn.py

 import torch
-import torch.nn.functional as F
+from torch.nn import functional as F
 
 from mmdet.models import DETECTORS
 from .mvx_two_stage import MVXTwoStageDetector
@@ -7,6 +7,7 @@ from .mvx_two_stage import MVXTwoStageDetector
 
 @DETECTORS.register_module()
 class MVXFasterRCNN(MVXTwoStageDetector):
+    """Multi-modality VoxelNet using Faster R-CNN."""
 
     def __init__(self, **kwargs):
         super(MVXFasterRCNN, self).__init__(**kwargs)
@@ -14,12 +15,21 @@ class MVXFasterRCNN(MVXTwoStageDetector):
 
 @DETECTORS.register_module()
 class DynamicMVXFasterRCNN(MVXTwoStageDetector):
+    """Multi-modality VoxelNet using Faster R-CNN and dynamic voxelization."""
 
     def __init__(self, **kwargs):
         super(DynamicMVXFasterRCNN, self).__init__(**kwargs)
 
     @torch.no_grad()
     def voxelize(self, points):
+        """Apply dynamic voxelization to points.
+
+        Args:
+            points (list[torch.Tensor]): Points of each sample.
+
+        Returns:
+            tuple[torch.Tensor]: Concatenated points and coordinates.
+        """
         coors = []
         # dynamic voxelization only provide a coors mapping
         for res in points:
@@ -34,6 +44,7 @@ class DynamicMVXFasterRCNN(MVXTwoStageDetector):
         return points, coors_batch
 
     def extract_pts_feat(self, points, img_feats, img_metas):
+        """Extract point features."""
         if not self.with_pts_bbox:
             return None
         voxels, coors = self.voxelize(points)

mmdet3d/models/detectors/mvx_two_stage.py

-import os.path as osp
-
 import torch
-import torch.nn as nn
-import torch.nn.functional as F
+from os import path as osp
+from torch import nn as nn
+from torch.nn import functional as F
 
 from mmdet3d.core import (Box3DMode, bbox3d2result, merge_aug_bboxes_3d,
                           show_result)
@@ -15,6 +14,7 @@ from .base import Base3DDetector
 
 @DETECTORS.register_module()
 class MVXTwoStageDetector(Base3DDetector):
+    """Base class of Multi-modality VoxelNet."""
 
     def __init__(self,
                  pts_voxel_layer=None,
@@ -69,6 +69,7 @@ class MVXTwoStageDetector(Base3DDetector):
         self.init_weights(pretrained=pretrained)
 
     def init_weights(self, pretrained=None):
+        """Initialize model weights."""
         super(MVXTwoStageDetector, self).init_weights(pretrained)
         if pretrained is None:
             img_pretrained = None
@@ -99,59 +100,72 @@ class MVXTwoStageDetector(Base3DDetector):
 
     @property
     def with_img_shared_head(self):
+        """bool: Whether the detector has a shared head in image branch."""
         return hasattr(self,
                        'img_shared_head') and self.img_shared_head is not None
 
     @property
     def with_pts_bbox(self):
+        """bool: Whether the detector has a 3D box head."""
         return hasattr(self,
                        'pts_bbox_head') and self.pts_bbox_head is not None
 
     @property
     def with_img_bbox(self):
+        """bool: Whether the detector has a 2D image box head."""
         return hasattr(self,
                        'img_bbox_head') and self.img_bbox_head is not None
 
     @property
     def with_img_backbone(self):
+        """bool: Whether the detector has a 2D image backbone."""
         return hasattr(self, 'img_backbone') and self.img_backbone is not None
 
     @property
     def with_pts_backbone(self):
+        """bool: Whether the detector has a 3D backbone."""
         return hasattr(self, 'pts_backbone') and self.pts_backbone is not None
 
     @property
     def with_fusion(self):
+        """bool: Whether the detector has a fusion layer."""
         return hasattr(self,
                        'pts_fusion_layer') and self.fusion_layer is not None
 
     @property
     def with_img_neck(self):
+        """bool: Whether the detector has a neck in image branch."""
         return hasattr(self, 'img_neck') and self.img_neck is not None
 
     @property
     def with_pts_neck(self):
+        """bool: Whether the detector has a neck in 3D detector branch."""
         return hasattr(self, 'pts_neck') and self.pts_neck is not None
 
     @property
     def with_img_rpn(self):
+        """bool: Whether the detector has a 2D RPN in image detector branch."""
         return hasattr(self, 'img_rpn_head') and self.img_rpn_head is not None
 
     @property
     def with_img_roi_head(self):
+        """bool: Whether the detector has a RoI Head in image branch."""
         return hasattr(self, 'img_roi_head') and self.img_roi_head is not None
 
     @property
     def with_voxel_encoder(self):
+        """bool: Whether the detector has a voxel encoder."""
         return hasattr(self,
                        'voxel_encoder') and self.voxel_encoder is not None
 
     @property
     def with_middle_encoder(self):
+        """bool: Whether the detector has a middle encoder."""
         return hasattr(self,
                        'middle_encoder') and self.middle_encoder is not None
 
     def extract_img_feat(self, img, img_metas):
+        """Extract features of images."""
         if self.with_img_backbone:
             if img.dim() == 5 and img.size(0) == 1:
                 img.squeeze_()
@@ -166,6 +180,7 @@ class MVXTwoStageDetector(Base3DDetector):
         return img_feats
 
     def extract_pts_feat(self, pts, img_feats, img_metas):
+        """Extract features of points."""
         if not self.with_pts_bbox:
             return None
         voxels, num_points, coors = self.voxelize(pts)
@@ -179,12 +194,22 @@ class MVXTwoStageDetector(Base3DDetector):
         return x
 
     def extract_feat(self, points, img, img_metas):
+        """Extract features from images and points."""
         img_feats = self.extract_img_feat(img, img_metas)
         pts_feats = self.extract_pts_feat(points, img_feats, img_metas)
         return (img_feats, pts_feats)
 
     @torch.no_grad()
     def voxelize(self, points):
+        """Apply dynamic voxelization to points.
+
+        Args:
+            points (list[torch.Tensor]): Points of each sample.
+
+        Returns:
+            tuple[torch.Tensor]: Concatenated points, number of points
+                per voxel, and coordinates.
+        """
         voxels, coors, num_points = [], [], []
         for res in points:
             res_voxels, res_coors, res_num_points = self.pts_voxel_layer(res)
@@ -208,8 +233,33 @@ class MVXTwoStageDetector(Base3DDetector):
                       gt_labels=None,
                       gt_bboxes=None,
                       img=None,
-                      bboxes=None,
+                      proposals=None,
                       gt_bboxes_ignore=None):
+        """Forward training function.
+
+        Args:
+            points (list[torch.Tensor], optional): Points of each sample.
+                Defaults to None.
+            img_metas (list[dict], optional): Meta information of each sample.
+                Defaults to None.
+            gt_bboxes_3d (list[:obj:`BaseInstance3DBoxes`], optional):
+                Ground truth 3D boxes. Defaults to None.
+            gt_labels_3d (list[torch.Tensor], optional): Ground truth labels
+                of 3D boxes. Defaults to None.
+            gt_labels (list[torch.Tensor], optional): Ground truth labels
+                of 2D boxes in images. Defaults to None.
+            gt_bboxes (list[torch.Tensor], optional): Ground truth 2D boxes in
+                images. Defaults to None.
+            img (torch.Tensor optional): Images of each sample with shape
+                (N, C, H, W). Defaults to None.
+            proposals ([list[torch.Tensor], optional): Predicted proposals
+                used for training Fast RCNN. Defaults to None.
+            gt_bboxes_ignore (list[torch.Tensor], optional): Ground truth
+                2D boxes in images to be ignored. Defaults to None.
+
+        Returns:
+            dict: Losses of different branches.
+        """
         img_feats, pts_feats = self.extract_feat(
             points, img=img, img_metas=img_metas)
         losses = dict()
@@ -225,8 +275,7 @@ class MVXTwoStageDetector(Base3DDetector):
                 gt_bboxes=gt_bboxes,
                 gt_labels=gt_labels,
                 gt_bboxes_ignore=gt_bboxes_ignore,
-                bboxes=bboxes,
-            )
+                proposals=proposals)
             losses.update(losses_img)
         return losses
 

mmdet3d/models/detectors/parta2.py

 import torch
-import torch.nn.functional as F
+from torch.nn import functional as F
 
 from mmdet3d.ops import Voxelization
 from mmdet.models import DETECTORS
@@ -9,7 +9,7 @@ from .two_stage import TwoStage3DDetector
 
 @DETECTORS.register_module()
 class PartA2(TwoStage3DDetector):
-    """Part-A2 detector
+    """Part-A2 detector.
 
     Please refer to the `paper <https://arxiv.org/abs/1907.03670>`_
     """

mmdet3d/models/detectors/single_stage.py

-import torch.nn as nn
+from torch import nn as nn
 
 from mmdet.models import DETECTORS, build_backbone, build_head, build_neck
 from .base import Base3DDetector
@@ -6,7 +6,7 @@ from .base import Base3DDetector
 
 @DETECTORS.register_module()
 class SingleStage3DDetector(Base3DDetector):
-    """SingleStage3DDetector
+    """SingleStage3DDetector.
 
     This class serves as a base class for single-stage 3D detectors.
 

mmdet3d/models/detectors/two_stage.py

@@ -4,11 +4,11 @@ from .base import Base3DDetector
 
 @DETECTORS.register_module()
 class TwoStage3DDetector(Base3DDetector, TwoStageDetector):
-    """Base class of two-stage 3D detector
+    """Base class of two-stage 3D detector.
 
     It inherits original ``:class:TwoStageDetector`` and
-    ``:class:Base3DDetector``. This class could serve as a base class for
-    all two-stage 3D detectors.
+    ``:class:Base3DDetector``. This class could serve as a base class for all
+    two-stage 3D detectors.
     """
 
     def __init__(self, **kwargs):

mmdet3d/models/detectors/voxelnet.py

 import torch
-import torch.nn.functional as F
+from torch.nn import functional as F
 
 from mmdet3d.core import bbox3d2result, merge_aug_bboxes_3d
 from mmdet3d.ops import Voxelization
@@ -34,7 +34,7 @@ class VoxelNet(SingleStage3DDetector):
         self.middle_encoder = builder.build_middle_encoder(middle_encoder)
 
     def extract_feat(self, points, img_metas):
-        """Extract features from points"""
+        """Extract features from points."""
         voxels, num_points, coors = self.voxelize(points)
         voxel_features = self.voxel_encoder(voxels, num_points, coors)
         batch_size = coors[-1, 0].item() + 1
@@ -46,7 +46,7 @@ class VoxelNet(SingleStage3DDetector):
 
     @torch.no_grad()
     def voxelize(self, points):
-        """Apply hard voxelization to points"""
+        """Apply hard voxelization to points."""
         voxels, coors, num_points = [], [], []
         for res in points:
             res_voxels, res_coors, res_num_points = self.voxel_layer(res)

mmdet3d/models/fusion_layers/point_fusion.py

 import torch
-import torch.nn as nn
-import torch.nn.functional as F
 from mmcv.cnn import ConvModule, xavier_init
+from torch import nn as nn
+from torch.nn import functional as F
 
 from ..registry import FUSION_LAYERS
 
@@ -23,7 +23,7 @@ def point_sample(
     padding_mode='zeros',
     align_corners=True,
 ):
-    """Obtain image features using points
+    """Obtain image features using points.
 
     Args:
         img_features (Tensor): 1xCxHxW image features
@@ -113,7 +113,7 @@ def point_sample(
 
 @FUSION_LAYERS.register_module()
 class PointFusion(nn.Module):
-    """Fuse image features from multi-scale features
+    """Fuse image features from multi-scale features.
 
     Args:
         img_channels (list[int] | int): Channels of image features.
@@ -225,7 +225,7 @@ class PointFusion(nn.Module):
                 xavier_init(m, distribution='uniform')
 
     def forward(self, img_feats, pts, pts_feats, img_metas):
-        """Forward function
+        """Forward function.
 
         Args:
             img_feats (list[Tensor]): img features

mmdet3d/models/losses/chamfer_distance.py

 import torch
-import torch.nn as nn
+from torch import nn as nn
 from torch.nn.functional import l1_loss, mse_loss, smooth_l1_loss
 
 from mmdet.models.builder import LOSSES

mmdet3d/models/middle_encoders/pillar_scatter.py

@@ -23,7 +23,7 @@ class PointPillarsScatter(nn.Module):
         self.in_channels = in_channels
 
     def forward(self, voxel_features, coors, batch_size=None):
-        """Foraward function to scatter features"""
+        """Foraward function to scatter features."""
         # TODO: rewrite the function in a batch manner
         # no need to deal with different batch cases
         if batch_size is not None:
@@ -32,7 +32,7 @@ class PointPillarsScatter(nn.Module):
             return self.forward_single(voxel_features, coors)
 
     def forward_single(self, voxel_features, coors):
-        """Scatter features of single sample
+        """Scatter features of single sample.
 
         Args:
             voxel_features (torch.Tensor): Voxel features in shape (N, M, C).
@@ -56,7 +56,7 @@ class PointPillarsScatter(nn.Module):
         return [canvas]
 
     def forward_batch(self, voxel_features, coors, batch_size):
-        """Scatter features of single sample
+        """Scatter features of single sample.
 
         Args:
             voxel_features (torch.Tensor): Voxel features in shape (N, M, C).

mmdet3d/models/middle_encoders/sparse_encoder.py

-import torch.nn as nn
+from torch import nn as nn
 
-import mmdet3d.ops.spconv as spconv
 from mmdet3d.ops import make_sparse_convmodule
+from mmdet3d.ops import spconv as spconv
 from ..registry import MIDDLE_ENCODERS
 
 
 @MIDDLE_ENCODERS.register_module()
 class SparseEncoder(nn.Module):
-    """Sparse encoder for Second
+    """Sparse encoder for Second.
 
     See https://arxiv.org/abs/1907.03670 for more detials.
 
@@ -81,7 +81,7 @@ class SparseEncoder(nn.Module):
             conv_type='SparseConv3d')
 
     def forward(self, voxel_features, coors, batch_size):
-        """Forward of SparseEncoder
+        """Forward of SparseEncoder.
 
         Args:
             voxel_features (torch.float32): shape [N, C]
@@ -113,7 +113,7 @@ class SparseEncoder(nn.Module):
         return spatial_features
 
     def make_encoder_layers(self, make_block, norm_cfg, in_channels):
-        """make encoder layers using sparse convs
+        """make encoder layers using sparse convs.
 
         Args:
             make_block (method): a bounded function to build blocks

mmdet3d/models/middle_encoders/sparse_unet.py

 import torch
-import torch.nn as nn
+from torch import nn as nn
 
-import mmdet3d.ops.spconv as spconv
 from mmdet3d.ops import SparseBasicBlock, make_sparse_convmodule
+from mmdet3d.ops import spconv as spconv
 from ..registry import MIDDLE_ENCODERS
 
 
 @MIDDLE_ENCODERS.register_module()
 class SparseUNet(nn.Module):
-    """SparseUNet for PartA^2
+    """SparseUNet for PartA^2.
 
     See https://arxiv.org/abs/1907.03670 for more detials.
 
@@ -92,7 +92,7 @@ class SparseUNet(nn.Module):
             conv_type='SparseConv3d')
 
     def forward(self, voxel_features, coors, batch_size):
-        """Forward of SparseUNet
+        """Forward of SparseUNet.
 
         Args:
             voxel_features (torch.float32): shape [N, C]
@@ -184,7 +184,7 @@ class SparseUNet(nn.Module):
         return x
 
     def make_encoder_layers(self, make_block, norm_cfg, in_channels):
-        """make encoder layers using sparse convs
+        """make encoder layers using sparse convs.
 
         Args:
             make_block (method): a bounded function to build blocks
@@ -230,7 +230,7 @@ class SparseUNet(nn.Module):
         return out_channels
 
     def make_decoder_layers(self, make_block, norm_cfg, in_channels):
-        """make decoder layers using sparse convs
+        """make decoder layers using sparse convs.
 
         Args:
             make_block (method): a bounded function to build blocks

mmdet3d/models/model_utils/vote_module.py

 import torch
-import torch.nn as nn
 from mmcv.cnn import ConvModule
+from torch import nn as nn
 
 from mmdet3d.models.builder import build_loss
 

mmdet3d/models/necks/second_fpn.py

 import torch
-import torch.nn as nn
 from mmcv.cnn import (build_norm_layer, build_upsample_layer, constant_init,
                       is_norm, kaiming_init)
+from torch import nn as nn
 
 from mmdet.models import NECKS
 
 
 @NECKS.register_module()
 class SECONDFPN(nn.Module):
-    """FPN used in SECOND/PointPillars/PartA2/MVXNet
+    """FPN used in SECOND/PointPillars/PartA2/MVXNet.
 
     Args:
         in_channels (list[int]): Input channels of multi-scale feature maps
@@ -46,7 +46,7 @@ class SECONDFPN(nn.Module):
         self.deblocks = nn.ModuleList(deblocks)
 
     def init_weights(self):
-        """Initialize weights of FPN"""
+        """Initialize weights of FPN."""
         for m in self.modules():
             if isinstance(m, nn.Conv2d):
                 kaiming_init(m)
@@ -54,7 +54,7 @@ class SECONDFPN(nn.Module):
                 constant_init(m, 1)
 
     def forward(self, x):
-        """Forward function
+        """Forward function.
 
         Args:
             x (torch.Tensor): 4D Tensor in (N, C, H, W) shape.

mmdet3d/models/roi_heads/base_3droi_head.py

 from abc import ABCMeta, abstractmethod
-
-import torch.nn as nn
+from torch import nn as nn
 
 
 class Base3DRoIHead(nn.Module, metaclass=ABCMeta):
-    """Base class for 3d RoIHeads"""
+    """Base class for 3d RoIHeads."""
 
     def __init__(self,
                  bbox_head=None,
@@ -51,7 +50,7 @@ class Base3DRoIHead(nn.Module, metaclass=ABCMeta):
 
     @abstractmethod
     def init_assigner_sampler(self):
-        """Initialize assigner and sampler"""
+        """Initialize assigner and sampler."""
         pass
 
     @abstractmethod
@@ -63,7 +62,7 @@ class Base3DRoIHead(nn.Module, metaclass=ABCMeta):
                       gt_labels,
                       gt_bboxes_ignore=None,
                       **kwargs):
-        """Forward function during training
+        """Forward function during training.
 
         Args:
             x (dict): Contains features from the first stage.

mmdet3d/models/roi_heads/bbox_heads/parta2_bbox_head.py

 import numpy as np
 import torch
-import torch.nn as nn
 from mmcv.cnn import ConvModule, normal_init, xavier_init
+from torch import nn as nn
 
-import mmdet3d.ops.spconv as spconv
 from mmdet3d.core.bbox.structures import (LiDARInstance3DBoxes,
                                           rotation_3d_in_axis, xywhr2xyxyr)
 from mmdet3d.models.builder import build_loss
 from mmdet3d.ops import make_sparse_convmodule
+from mmdet3d.ops import spconv as spconv
 from mmdet3d.ops.iou3d.iou3d_utils import nms_gpu, nms_normal_gpu
 from mmdet.core import build_bbox_coder, multi_apply
 from mmdet.models import HEADS
@@ -223,7 +223,7 @@ class PartA2BboxHead(nn.Module):
         self.init_weights()
 
     def init_weights(self):
-        """Initialize weights of the bbox head"""
+        """Initialize weights of the bbox head."""
         for m in self.modules():
             if isinstance(m, (nn.Conv2d, nn.Conv1d)):
                 xavier_init(m, distribution='uniform')
@@ -556,7 +556,7 @@ class PartA2BboxHead(nn.Module):
                         nms_thr,
                         input_meta,
                         use_rotate_nms=True):
-        """Multi-class NMS for box head
+        """Multi-class NMS for box head.
 
         Note:
             This function has large overlap with the `box3d_multiclass_nms`

mmdet3d/models/roi_heads/mask_heads/pointwise_semantic_head.py

 import torch
-import torch.nn as nn
-import torch.nn.functional as F
+from torch import nn as nn
+from torch.nn import functional as F
 
 from mmdet3d.core.bbox.structures import rotation_3d_in_axis
 from mmdet3d.models.builder import build_loss
@@ -57,7 +57,6 @@ class PointwiseSemanticHead(nn.Module):
 
         Returns:
             dict: part features, segmentation and part predictions.
-
         """
         seg_preds = self.seg_cls_layer(x)  # (N, 1)
         part_preds = self.seg_reg_layer(x)  # (N, 3)
@@ -73,7 +72,8 @@ class PointwiseSemanticHead(nn.Module):
             seg_preds=seg_preds, part_preds=part_preds, part_feats=part_feats)
 
     def get_targets_single(self, voxel_centers, gt_bboxes_3d, gt_labels_3d):
-        """generate segmentation and part prediction targets for a single sample
+        """generate segmentation and part prediction targets for a single
+        sample.
 
         Args:
             voxel_centers (torch.Tensor): shape [voxel_num, 3],
@@ -120,7 +120,7 @@ class PointwiseSemanticHead(nn.Module):
         return seg_targets, part_targets
 
     def get_targets(self, voxels_dict, gt_bboxes_3d, gt_labels_3d):
-        """generate segmentation and part prediction targets
+        """generate segmentation and part prediction targets.
 
         Args:
             voxel_centers (torch.Tensor): shape [voxel_num, 3],

mmdet3d/models/roi_heads/part_aggregation_roi_head.py

-import torch.nn.functional as F
+from torch.nn import functional as F
 
 from mmdet3d.core import AssignResult
 from mmdet3d.core.bbox import bbox3d2result, bbox3d2roi
@@ -10,7 +10,7 @@ from .base_3droi_head import Base3DRoIHead
 
 @HEADS.register_module()
 class PartAggregationROIHead(Base3DRoIHead):
-    """Part aggregation roi head for PartA2
+    """Part aggregation roi head for PartA2.
 
     Args:
         semantic_head (ConfigDict): Config of semantic head.
@@ -44,21 +44,21 @@ class PartAggregationROIHead(Base3DRoIHead):
         self.init_assigner_sampler()
 
     def init_weights(self, pretrained):
-        """Initialize weights, skip since ``PartAggregationROIHead``
-        does not need to initialize weights"""
+        """Initialize weights, skip since ``PartAggregationROIHead`` does not
+        need to initialize weights."""
         pass
 
     def init_mask_head(self):
-        """Initialize mask head, skip since ``PartAggregationROIHead``
-        does not have one."""
+        """Initialize mask head, skip since ``PartAggregationROIHead`` does not
+        have one."""
         pass
 
     def init_bbox_head(self, bbox_head):
-        """Initialize box head"""
+        """Initialize box head."""
         self.bbox_head = build_head(bbox_head)
 
     def init_assigner_sampler(self):
-        """Initialize assigner and sampler"""
+        """Initialize assigner and sampler."""
         self.bbox_assigner = None
         self.bbox_sampler = None
         if self.train_cfg:
@@ -78,7 +78,7 @@ class PartAggregationROIHead(Base3DRoIHead):
 
     def forward_train(self, feats_dict, voxels_dict, img_metas, proposal_list,
                       gt_bboxes_3d, gt_labels_3d):
-        """Training forward function of PartAggregationROIHead
+        """Training forward function of PartAggregationROIHead.
 
         Args:
             feats_dict (dict): Contains features from the first stage.
@@ -116,7 +116,7 @@ class PartAggregationROIHead(Base3DRoIHead):
 
     def simple_test(self, feats_dict, voxels_dict, img_metas, proposal_list,
                     **kwargs):
-        """Simple testing forward function of PartAggregationROIHead
+        """Simple testing forward function of PartAggregationROIHead.
 
         Note:
             This function assumes that the batch size is 1
@@ -216,7 +216,7 @@ class PartAggregationROIHead(Base3DRoIHead):
         return bbox_results
 
     def _assign_and_sample(self, proposal_list, gt_bboxes_3d, gt_labels_3d):
-        """Assign and sample proposals for training
+        """Assign and sample proposals for training.
 
         Args:
             proposal_list (list[dict]): Proposals produced by RPN.
@@ -290,7 +290,7 @@ class PartAggregationROIHead(Base3DRoIHead):
 
     def _semantic_forward_train(self, x, voxels_dict, gt_bboxes_3d,
                                 gt_labels_3d):
-        """Train semantic head
+        """Train semantic head.
 
         Args:
             x (torch.Tensor): Point-wise semantic features for segmentation