refactor directory

tests/test_models/test_fusion_layers/test_point_fusion.py

@@ -7,7 +7,7 @@ CommandLine:
 
 import torch
 
-from mmdet3d.models.fusion_layers import PointFusion
+from mmdet3d.models.layers.fusion_layers import PointFusion
 
 
 def test_sample_single():

tests/test_models/test_fusion_layers/test_vote_fusion.py

@@ -7,7 +7,7 @@ CommandLine:
 
 import torch
 
-from mmdet3d.models.fusion_layers import VoteFusion
+from mmdet3d.models.layers.fusion_layers import VoteFusion
 
 
 def test_vote_fusion():

tests/test_models/test_losses.py

@@ -98,8 +98,8 @@ def test_chamfer_disrance():
 
 
 def test_paconv_regularization_loss():
+    from mmdet3d.models.layers import PAConv, PAConvCUDA
     from mmdet3d.models.losses import PAConvRegularizationLoss
-    from mmdet3d.ops import PAConv, PAConvCUDA
 
     class ToyModel(nn.Module):
 

tests/test_models/test_preprocessors/test_data_preprocessor.py

@@ -3,8 +3,8 @@ from unittest import TestCase
 
 import torch
 
-from mmdet3d.core import Det3DDataSample
 from mmdet3d.models.data_preprocessors import Det3DDataPreprocessor
+from mmdet3d.structures import Det3DDataSample
 
 
 class TestDet3DDataPreprocessor(TestCase):
@@ -33,7 +33,7 @@ class TestDet3DDataPreprocessor(TestCase):
         processor = Det3DDataPreprocessor(mean=[0, 0, 0], std=[1, 1, 1])
 
         points = torch.randn((5000, 3))
-        image = torch.randint(0, 256, (3, 11, 10))
+        image = torch.randint(0, 256, (3, 11, 10)).float()
         inputs_dict = dict(points=points, img=image)
 
         data = [{'inputs': inputs_dict, 'data_sample': Det3DDataSample()}]

tests/test_models/test_voxel_encoders/test_voxel_generator.py

 # Copyright (c) OpenMMLab. All rights reserved.
 import numpy as np
 
-from mmdet3d.core.voxel.voxel_generator import VoxelGenerator
+from mmdet3d.models.task_modules.voxel import VoxelGenerator
 
 
 def test_voxel_generator():

tests/utils/data_utils.py

 # Copyright (c) OpenMMLab. All rights reserved.
 import numpy as np
 
-from mmdet3d.core import LiDARInstance3DBoxes
 # create a dummy `results` to test the pipeline
 from mmdet3d.datasets import LoadAnnotations3D, LoadPointsFromFile
+from mmdet3d.structures import LiDARInstance3DBoxes
 
 
 def create_dummy_data_info(with_ann=True):

tests/utils/model_utils.py

@@ -7,8 +7,9 @@ import numpy as np
 import torch
 from mmengine import InstanceData
 
-from mmdet3d.core import (CameraInstance3DBoxes, DepthInstance3DBoxes,
-                          Det3DDataSample, LiDARInstance3DBoxes, PointData)
+from mmdet3d.structures import (CameraInstance3DBoxes, DepthInstance3DBoxes,
+                                Det3DDataSample, LiDARInstance3DBoxes,
+                                PointData)
 
 
 def _setup_seed(seed):
@@ -104,7 +105,6 @@ def _create_detector_inputs(seed=0,
         points = torch.rand([num_points, points_feat_dim])
     else:
         points = None
-
     if with_img:
         if isinstance(img_size, tuple):
             img = torch.rand(3, img_size[0], img_size[1])
@@ -132,8 +132,9 @@ def _create_detector_inputs(seed=0,
     gt_instance = InstanceData()
     gt_instance.labels = torch.randint(0, num_classes, [num_gt_instance])
     gt_instance.bboxes = torch.rand(num_gt_instance, 4)
-    gt_instance.bboxes[:, 2:] = \
-        gt_instance.bboxes[:, :2] + gt_instance.bboxes[:, 2:]
+    gt_instance.bboxes[:,
+                       2:] = gt_instance.bboxes[:, :2] + gt_instance.bboxes[:,
+                                                                            2:]
 
     data_sample.gt_instances = gt_instance
     data_sample.gt_pts_seg = PointData()

tools/analysis_tools/benchmark.py

@@ -7,7 +7,7 @@ from mmcv import Config
 from mmcv.parallel import MMDataParallel
 from mmcv.runner import load_checkpoint, wrap_fp16_model
 
-from mmdet3d.datasets import build_dataloader, build_dataset
+from mmdet3d.datasets import build_dataset
 from mmdet3d.models import build_detector
 from tools.misc.fuse_conv_bn import fuse_module
 
@@ -41,6 +41,11 @@ def main():
     # build the dataloader
     # TODO: support multiple images per gpu (only minor changes are needed)
     dataset = build_dataset(cfg.data.test)
+
+    # TODO fix this
+    def build_dataloader():
+        pass
+
     data_loader = build_dataloader(
         dataset,
         samples_per_gpu=1,

tools/data_converter/create_gt_database.py

@@ -9,9 +9,9 @@ from mmcv.ops import roi_align
 from pycocotools import mask as maskUtils
 from pycocotools.coco import COCO
 
-from mmdet3d.core.bbox import box_np_ops as box_np_ops
 from mmdet3d.datasets import build_dataset
-from mmdet.core.evaluation.bbox_overlaps import bbox_overlaps
+from mmdet3d.structures.ops import box_np_ops as box_np_ops
+from mmdet.evaluation import bbox_overlaps
 
 
 def _poly2mask(mask_ann, img_h, img_w):

tools/data_converter/kitti_converter.py

@@ -6,7 +6,8 @@ import mmcv
 import numpy as np
 from nuscenes.utils.geometry_utils import view_points
 
-from mmdet3d.core.bbox import box_np_ops, points_cam2img
+from mmdet3d.structures import points_cam2img
+from mmdet3d.structures.ops import box_np_ops
 from .kitti_data_utils import WaymoInfoGatherer, get_kitti_image_info
 from .nuscenes_converter import post_process_coords
 

tools/data_converter/nuscenes_converter.py

@@ -11,8 +11,8 @@ from nuscenes.utils.geometry_utils import view_points
 from pyquaternion import Quaternion
 from shapely.geometry import MultiPoint, box
 
-from mmdet3d.core.bbox import points_cam2img
 from mmdet3d.datasets import NuScenesDataset
+from mmdet3d.structures import points_cam2img
 
 nus_categories = ('car', 'truck', 'trailer', 'bus', 'construction_vehicle',
                   'bicycle', 'motorcycle', 'pedestrian', 'traffic_cone',

tools/data_converter/update_infos_to_v2.py

@@ -16,9 +16,9 @@ import mmcv
 import numpy as np
 from nuscenes.nuscenes import NuScenes
 
-from mmdet3d.core.bbox import points_cam2img
 from mmdet3d.datasets.convert_utils import get_2d_boxes
 from mmdet3d.datasets.utils import convert_quaternion_to_matrix
+from mmdet3d.structures import points_cam2img
 
 
 def get_empty_instance():

tools/deployment/mmdet3d_handler.py

@@ -7,7 +7,7 @@ import torch
 from ts.torch_handler.base_handler import BaseHandler
 
 from mmdet3d.apis import inference_detector, init_model
-from mmdet3d.core.points import get_points_type
+from mmdet3d.structures.points import get_points_type
 
 
 class MMdet3dHandler(BaseHandler):

tools/misc/fuse_conv_bn.py

@@ -12,7 +12,7 @@ def fuse_conv_bn(conv, bn):
     """During inference, the functionary of batch norm layers is turned off but
     only the mean and var alone channels are used, which exposes the chance to
     fuse it with the preceding conv layers to save computations and simplify
-    network structures."""
+    network bboxes_3d."""
     conv_w = conv.weight
     conv_b = conv.bias if conv.bias is not None else torch.zeros_like(
         bn.running_mean)

tools/update_data_coords.py

@@ -5,7 +5,7 @@ from os import path as osp
 import mmcv
 import numpy as np
 
-from mmdet3d.core.bbox import limit_period
+from mmdet3d.structures import limit_period
 
 
 def update_sunrgbd_infos(root_dir, out_dir, pkl_files):