[Refactor] rename `CLASSES` and `PALETTE` to `classes` and `palette` in dataset metainfo (#1932)

* rame CLASS and PALETTE to class and palette

* change mmcv-full to mmcv

* fix comments

configs/centerpoint/centerpoint_voxel01_second_secfpn_8xb4-cyclic-20e_nus-3d.py

@@ -127,7 +127,7 @@ train_dataloader = dict(
             data_root=data_root,
             ann_file='nuscenes_infos_train.pkl',
             pipeline=train_pipeline,
-            metainfo=dict(CLASSES=class_names),
+            metainfo=dict(classes=class_names),
             test_mode=False,
             data_prefix=data_prefix,
             use_valid_flag=True,
@@ -135,8 +135,8 @@ train_dataloader = dict(
             # and box_type_3d='Depth' in sunrgbd and scannet dataset.
             box_type_3d='LiDAR')))
 test_dataloader = dict(
-    dataset=dict(pipeline=test_pipeline, metainfo=dict(CLASSES=class_names)))
+    dataset=dict(pipeline=test_pipeline, metainfo=dict(classes=class_names)))
 val_dataloader = dict(
-    dataset=dict(pipeline=test_pipeline, metainfo=dict(CLASSES=class_names)))
+    dataset=dict(pipeline=test_pipeline, metainfo=dict(classes=class_names)))
 
 train_cfg = dict(val_interval=20)

configs/groupfree3d/groupfree3d_head-L12-O256_4xb8_scannet-seg.py

@@ -78,7 +78,7 @@ class_names = ('cabinet', 'bed', 'chair', 'sofa', 'table', 'door', 'window',
                'refrigerator', 'showercurtrain', 'toilet', 'sink', 'bathtub',
                'garbagebin')
 
-metainfo = dict(CLASSES=class_names)
+metainfo = dict(classes=class_names)
 
 train_pipeline = [
     dict(

configs/groupfree3d/groupfree3d_head-L6-O256_4xb8_scannet-seg.py

@@ -77,7 +77,7 @@ class_names = ('cabinet', 'bed', 'chair', 'sofa', 'table', 'door', 'window',
                'refrigerator', 'showercurtrain', 'toilet', 'sink', 'bathtub',
                'garbagebin')
 
-metainfo = dict(CLASSES=class_names)
+metainfo = dict(classes=class_names)
 
 train_pipeline = [
     dict(

configs/groupfree3d/groupfree3d_w2x-head-L12-O256_4xb8_scannet-seg.py

@@ -93,7 +93,7 @@ class_names = ('cabinet', 'bed', 'chair', 'sofa', 'table', 'door', 'window',
                'refrigerator', 'showercurtrain', 'toilet', 'sink', 'bathtub',
                'garbagebin')
 
-metainfo = dict(CLASSES=class_names)
+metainfo = dict(classes=class_names)
 
 train_pipeline = [
     dict(

configs/groupfree3d/groupfree3d_w2x-head-L12-O512_4xb8_scannet-seg.py

@@ -94,7 +94,7 @@ class_names = ('cabinet', 'bed', 'chair', 'sofa', 'table', 'door', 'window',
                'refrigerator', 'showercurtrain', 'toilet', 'sink', 'bathtub',
                'garbagebin')
 
-metainfo = dict(CLASSES=class_names)
+metainfo = dict(classes=class_names)
 
 train_pipeline = [
     dict(

configs/imvoxelnet/imvoxelnet_8xb4_kitti-3d-car.py

@@ -81,7 +81,7 @@ data_root = 'data/kitti/'
 class_names = ['Car']
 input_modality = dict(use_lidar=False, use_camera=True)
 point_cloud_range = [0, -39.68, -3, 69.12, 39.68, 1]
-metainfo = dict(CLASSES=class_names)
+metainfo = dict(classes=class_names)
 
 # file_client_args = dict(backend='disk')
 # Uncomment the following if use ceph or other file clients.

configs/mvxnet/mvxnet_fpn_dv_second_secfpn_8xb2-80e_kitti-3d-3class.py

@@ -142,7 +142,7 @@ model = dict(
 dataset_type = 'KittiDataset'
 data_root = 'data/kitti/'
 class_names = ['Pedestrian', 'Cyclist', 'Car']
-metainfo = dict(CLASSES=class_names)
+metainfo = dict(classes=class_names)
 input_modality = dict(use_lidar=True, use_camera=True)
 train_pipeline = [
     dict(type='LoadPointsFromFile', coord_type='LIDAR', load_dim=4, use_dim=4),

configs/parta2/PartA2_hv_secfpn_8xb2-cyclic-80e_kitti-3d-3class.py

@@ -84,7 +84,7 @@ train_dataloader = dict(
             data_prefix=dict(pts='training/velodyne_reduced'),
             pipeline=train_pipeline,
             modality=input_modality,
-            metainfo=dict(CLASSES=class_names),
+            metainfo=dict(classes=class_names),
             box_type_3d='LiDAR',
             test_mode=False)))
 test_dataloader = dict(
@@ -100,7 +100,7 @@ test_dataloader = dict(
         data_prefix=dict(pts='training/velodyne_reduced'),
         pipeline=test_pipeline,
         modality=input_modality,
-        metainfo=dict(CLASSES=class_names),
+        metainfo=dict(classes=class_names),
         box_type_3d='LiDAR',
         test_mode=True))
 val_dataloader = dict(
@@ -116,7 +116,7 @@ val_dataloader = dict(
         data_prefix=dict(pts='training/velodyne_reduced'),
         pipeline=eval_pipeline,
         modality=input_modality,
-        metainfo=dict(CLASSES=class_names),
+        metainfo=dict(classes=class_names),
         box_type_3d='LiDAR',
         test_mode=True))
 val_evaluator = dict(

configs/parta2/PartA2_hv_secfpn_8xb2-cyclic-80e_kitti-3d-car.py

@@ -135,8 +135,8 @@ test_pipeline = [
 train_dataloader = dict(
     dataset=dict(
         dataset=dict(
-            pipeline=train_pipeline, metainfo=dict(CLASSES=class_names))))
+            pipeline=train_pipeline, metainfo=dict(classes=class_names))))
 test_dataloader = dict(
-    dataset=dict(pipeline=test_pipeline, metainfo=dict(CLASSES=class_names)))
-val_dataloader = dict(dataset=dict(metainfo=dict(CLASSES=class_names)))
+    dataset=dict(pipeline=test_pipeline, metainfo=dict(classes=class_names)))
+val_dataloader = dict(dataset=dict(metainfo=dict(classes=class_names)))
 find_unused_parameters = True

configs/point_rcnn/point-rcnn_8xb2_kitti-3d-3class.py

@@ -7,7 +7,7 @@ _base_ = [
 dataset_type = 'KittiDataset'
 data_root = 'data/kitti/'
 class_names = ['Pedestrian', 'Cyclist', 'Car']
-metainfo = dict(CLASSES=class_names)
+metainfo = dict(classes=class_names)
 point_cloud_range = [0, -40, -3, 70.4, 40, 1]
 input_modality = dict(use_lidar=True, use_camera=False)
 

configs/pointpillars/pointpillars_hv_secfpn_8xb6-160e_kitti-3d-3class.py

@@ -8,7 +8,7 @@ point_cloud_range = [0, -39.68, -3, 69.12, 39.68, 1]
 # dataset settings
 data_root = 'data/kitti/'
 class_names = ['Pedestrian', 'Cyclist', 'Car']
-metainfo = dict(CLASSES=class_names)
+metainfo = dict(classes=class_names)
 
 # PointPillars adopted a different sampling strategies among classes
 db_sampler = dict(

configs/pointpillars/pointpillars_hv_secfpn_8xb6-160e_kitti-3d-car.py

@@ -4,7 +4,7 @@ _base_ = './pointpillars_hv_secfpn_8xb6-160e_kitti-3d-3class.py'
 dataset_type = 'KittiDataset'
 data_root = 'data/kitti/'
 class_names = ['Car']
-metainfo = dict(CLASSES=class_names)
+metainfo = dict(classes=class_names)
 
 point_cloud_range = [0, -39.68, -3, 69.12, 39.68, 1]
 

configs/second/second_hv_secfpn_sbn-all_16xb2-2x_waymoD5-3d-3class.py

@@ -8,7 +8,7 @@ _base_ = [
 dataset_type = 'WaymoDataset'
 data_root = 'data/waymo/kitti_format/'
 class_names = ['Car', 'Pedestrian', 'Cyclist']
-metainfo = dict(CLASSES=class_names)
+metainfo = dict(classes=class_names)
 
 point_cloud_range = [-76.8, -51.2, -2, 76.8, 51.2, 4]
 input_modality = dict(use_lidar=True, use_camera=False)

configs/ssn/ssn_hv_secfpn_sbn-all_16xb2-2x_nus-3d.py

@@ -56,11 +56,11 @@ test_pipeline = [
 train_dataloader = dict(
     batch_size=2,
     num_workers=4,
-    dataset=dict(pipeline=train_pipeline, metainfo=dict(CLASSES=class_names)))
+    dataset=dict(pipeline=train_pipeline, metainfo=dict(classes=class_names)))
 test_dataloader = dict(
-    dataset=dict(pipeline=test_pipeline, metainfo=dict(CLASSES=class_names)))
+    dataset=dict(pipeline=test_pipeline, metainfo=dict(classes=class_names)))
 val_dataloader = dict(
-    dataset=dict(pipeline=test_pipeline, metainfo=dict(CLASSES=class_names)))
+    dataset=dict(pipeline=test_pipeline, metainfo=dict(classes=class_names)))
 
 # model settings
 model = dict(

docs/en/advanced_guides/customize_dataset.md

@@ -220,7 +220,7 @@ class MyDataset(Det3DDataset):
 
     # replace with all the classes in customized pkl info file
     METAINFO = {
-       'CLASSES': ('Pedestrian', 'Cyclist', 'Car')
+       'classes': ('Pedestrian', 'Cyclist', 'Car')
     }
 
     def parse_ann_info(self, info):
@@ -268,7 +268,7 @@ data_root = 'data/custom/'
 class_names = ['Pedestrian', 'Cyclist', 'Car']  # replace with your dataset class
 point_cloud_range = [0, -40, -3, 70.4, 40, 1]  # adjust according to your dataset
 input_modality = dict(use_lidar=True, use_camera=False)
-metainfo = dict(CLASSES=class_names)
+metainfo = dict(classes=class_names)
 
 train_pipeline = [
     dict(

docs/en/advanced_guides/pure_point_cloud_dataset.md

@@ -143,7 +143,7 @@ from .custom_3d import Custom3DDataset
 
 @DATASETS.register_module()
 class MyDataset(Custom3DDataset):
-    CLASSES = ('cabinet', 'bed', 'chair', 'sofa', 'table', 'door', 'window',
+    classes = ('cabinet', 'bed', 'chair', 'sofa', 'table', 'door', 'window',
                'bookshelf', 'picture', 'counter', 'desk', 'curtain',
                'refrigerator', 'showercurtrain', 'toilet', 'sink', 'bathtub',
                'garbagebin')

docs/en/user_guides/config.md

@@ -119,7 +119,7 @@ data_root = 'data/kitti/'
 class_names = ['Pedestrian', 'Cyclist', 'Car']
 point_cloud_range = [0, -39.68, -3, 69.12, 39.68, 1]
 input_modality = dict(use_lidar=True, use_camera=False)
-metainfo = dict(CLASSES=['Pedestrian', 'Cyclist', 'Car'])
+metainfo = dict(classes=['Pedestrian', 'Cyclist', 'Car'])
 db_sampler = dict(
     data_root='data/kitti/',
     info_path='data/kitti/kitti_dbinfos_train.pkl',
@@ -250,7 +250,7 @@ train_dataloader = dict(
             ],
             modality=dict(use_lidar=True, use_camera=False),
             test_mode=False,
-            metainfo=dict(CLASSES=['Pedestrian', 'Cyclist', 'Car']),
+            metainfo=dict(classes=['Pedestrian', 'Cyclist', 'Car']),
             box_type_3d='LiDAR')))
 val_dataloader = dict(
     batch_size=1,
@@ -289,7 +289,7 @@ val_dataloader = dict(
         ],
         modality=dict(use_lidar=True, use_camera=False),
         test_mode=True,
-        metainfo=dict(CLASSES=['Pedestrian', 'Cyclist', 'Car']),
+        metainfo=dict(classes=['Pedestrian', 'Cyclist', 'Car']),
         box_type_3d='LiDAR'))
 test_dataloader = dict(
     batch_size=1,
@@ -328,7 +328,7 @@ test_dataloader = dict(
         ],
         modality=dict(use_lidar=True, use_camera=False),
         test_mode=True,
-        metainfo=dict(CLASSES=['Pedestrian', 'Cyclist', 'Car']),
+        metainfo=dict(classes=['Pedestrian', 'Cyclist', 'Car']),
         box_type_3d='LiDAR'))
 ```
 

mmdet3d/datasets/dataset_wrappers.py

@@ -19,8 +19,8 @@ class CBGSDataset(object):
     def __init__(self, dataset):
         self.dataset = DATASETS.build(dataset)
         self.metainfo = self.dataset.metainfo
-        self.CLASSES = self.metainfo['CLASSES']
-        self.cat2id = {name: i for i, name in enumerate(self.CLASSES)}
+        self.classes = self.metainfo['classes']
+        self.cat2id = {name: i for i, name in enumerate(self.classes)}
         self.sample_indices = self._get_sample_indices()
         # self.dataset.data_infos = self.data_infos
         if hasattr(self.dataset, 'flag'):
@@ -54,7 +54,7 @@ class CBGSDataset(object):
 
         sample_indices = []
 
-        frac = 1.0 / len(self.CLASSES)
+        frac = 1.0 / len(self.classes)
         ratios = [frac / v for v in class_distribution.values()]
         for cls_inds, ratio in zip(list(class_sample_idxs.values()), ratios):
             sample_indices += np.random.choice(cls_inds,

mmdet3d/datasets/det3d_dataset.py

@@ -102,29 +102,29 @@ class Det3DDataset(BaseDataset):
 
         self.box_type_3d, self.box_mode_3d = get_box_type(box_type_3d)
 
-        if metainfo is not None and 'CLASSES' in metainfo:
-            # we allow to train on subset of self.METAINFO['CLASSES']
+        if metainfo is not None and 'classes' in metainfo:
+            # we allow to train on subset of self.METAINFO['classes']
             # map unselected labels to -1
             self.label_mapping = {
                 i: -1
-                for i in range(len(self.METAINFO['CLASSES']))
+                for i in range(len(self.METAINFO['classes']))
             }
             self.label_mapping[-1] = -1
-            for label_idx, name in enumerate(metainfo['CLASSES']):
-                ori_label = self.METAINFO['CLASSES'].index(name)
+            for label_idx, name in enumerate(metainfo['classes']):
+                ori_label = self.METAINFO['classes'].index(name)
                 self.label_mapping[ori_label] = label_idx
 
-            self.num_ins_per_cat = {name: 0 for name in metainfo['CLASSES']}
+            self.num_ins_per_cat = {name: 0 for name in metainfo['classes']}
         else:
             self.label_mapping = {
                 i: i
-                for i in range(len(self.METAINFO['CLASSES']))
+                for i in range(len(self.METAINFO['classes']))
             }
             self.label_mapping[-1] = -1
 
             self.num_ins_per_cat = {
                 name: 0
-                for name in self.METAINFO['CLASSES']
+                for name in self.METAINFO['classes']
             }
 
         super().__init__(
@@ -255,7 +255,7 @@ class Det3DDataset(BaseDataset):
             ann_info['instances'] = info['instances']
 
             for label in ann_info['gt_labels_3d']:
-                cat_name = self.metainfo['CLASSES'][label]
+                cat_name = self.metainfo['classes'][label]
                 self.num_ins_per_cat[cat_name] += 1
 
         return ann_info
@@ -336,11 +336,11 @@ class Det3DDataset(BaseDataset):
         """
         ori_num_per_cat = dict()
         for label in old_labels:
-            cat_name = self.metainfo['CLASSES'][label]
+            cat_name = self.metainfo['classes'][label]
             ori_num_per_cat[cat_name] = ori_num_per_cat.get(cat_name, 0) + 1
         new_num_per_cat = dict()
         for label in new_labels:
-            cat_name = self.metainfo['CLASSES'][label]
+            cat_name = self.metainfo['classes'][label]
             new_num_per_cat[cat_name] = new_num_per_cat.get(cat_name, 0) + 1
         content_show = [['category', 'new number', 'ori number']]
         for cat_name, num in ori_num_per_cat.items():

mmdet3d/datasets/kitti2d_dataset.py

@@ -36,7 +36,7 @@ class Kitti2DDataset(Det3DDataset):
             Defaults to False.
     """
 
-    CLASSES = ('car', 'pedestrian', 'cyclist')
+    classes = ('car', 'pedestrian', 'cyclist')
     """
     Annotation format:
     [
@@ -90,7 +90,7 @@ class Kitti2DDataset(Det3DDataset):
         self.data_infos = mmengine.load(ann_file)
         self.cat2label = {
             cat_name: i
-            for i, cat_name in enumerate(self.CLASSES)
+            for i, cat_name in enumerate(self.classes)
         }
         return self.data_infos
 
@@ -122,7 +122,7 @@ class Kitti2DDataset(Det3DDataset):
         difficulty = annos['difficulty']
 
         # remove classes that is not needed
-        selected = self.keep_arrays_by_name(gt_names, self.CLASSES)
+        selected = self.keep_arrays_by_name(gt_names, self.classes)
         gt_bboxes = gt_bboxes[selected]
         gt_names = gt_names[selected]
         difficulty = difficulty[selected]
@@ -215,7 +215,7 @@ class Kitti2DDataset(Det3DDataset):
         """
         from mmdet3d.structures.ops.transforms import bbox2result_kitti2d
         sample_idx = [info['image']['image_idx'] for info in self.data_infos]
-        result_files = bbox2result_kitti2d(outputs, self.CLASSES, sample_idx,
+        result_files = bbox2result_kitti2d(outputs, self.classes, sample_idx,
                                            out)
         return result_files
 
@@ -237,5 +237,5 @@ class Kitti2DDataset(Det3DDataset):
                                        ]), 'KITTI data set only evaluate bbox'
         gt_annos = [info['annos'] for info in self.data_infos]
         ap_result_str, ap_dict = kitti_eval(
-            gt_annos, result_files, self.CLASSES, eval_types=['bbox'])
+            gt_annos, result_files, self.classes, eval_types=['bbox'])
         return ap_result_str, ap_dict