add config refactor example

tools/cfgs/dataset_configs/modifiers/waymo/multiframe.yaml

@@ -0,0 +1,64 @@
+SEQUENCE_CONFIG:
+    ENABLED: True
+    SAMPLE_OFFSET: [-3, 0]
+
+TRAIN_WITH_SPEED: True 
+
+
+DATA_AUGMENTOR:
+    AUG_CONFIG_LIST:
+        - NAME: gt_sampling
+          USE_ROAD_PLANE: False
+          DB_INFO_PATH:
+              - waymo_processed_data_v0_5_0_waymo_dbinfos_train_sampled_1_multiframe_-4_to_0.pkl
+
+          USE_SHARED_MEMORY: False  # set it to True to speed up (it costs about 50GB? shared memory)
+          DB_DATA_PATH:
+              - waymo_processed_data_v0_5_0_gt_database_train_sampled_1_multiframe_-4_to_0_global.npy
+
+          PREPARE: {
+             filter_by_min_points: ['Vehicle:5', 'Pedestrian:5', 'Cyclist:5'],
+             filter_by_difficulty: [-1],
+          }
+
+          SAMPLE_GROUPS: ['Vehicle:15', 'Pedestrian:10', 'Cyclist:10']
+          NUM_POINT_FEATURES: 6
+          REMOVE_EXTRA_WIDTH: [0.0, 0.0, 0.0]
+          LIMIT_WHOLE_SCENE: True
+
+          FILTER_OBJ_POINTS_BY_TIMESTAMP: True 
+          TIME_RANGE: [0.3, 0.0]  # 0.3s-0.0s indicates 4 frames 
+
+        - NAME: random_world_flip
+          ALONG_AXIS_LIST: ['x', 'y']
+
+        - NAME: random_world_rotation
+          WORLD_ROT_ANGLE: [-0.78539816, 0.78539816]
+
+        - NAME: random_world_scaling
+          WORLD_SCALE_RANGE: [0.95, 1.05]
+
+
+POINT_FEATURE_ENCODING:
+    used_feature_list: ['x', 'y', 'z', 'intensity', 'elongation', 'timestamp']
+    src_feature_list: ['x', 'y', 'z', 'intensity', 'elongation', 'timestamp']
+
+
+DATA_PROCESSOR:
+    - NAME: mask_points_and_boxes_outside_range
+      REMOVE_OUTSIDE_BOXES: True
+      USE_CENTER_TO_FILTER: True
+
+    - NAME: shuffle_points
+      SHUFFLE_ENABLED: {
+        'train': True,
+        'test': True
+      }
+
+    - NAME: transform_points_to_voxels
+      VOXEL_SIZE: [0.1, 0.1, 0.15]
+      MAX_POINTS_PER_VOXEL: 5
+      MAX_NUMBER_OF_VOXELS: {
+        'train': 180000,
+        'test': 400000
+      }

tools/cfgs/waymo_models/modifiers/mppnet/16frames.yaml

@@ -0,0 +1,16 @@
+DATA_CONFIG:
+    SEQUENCE_CONFIG:
+        SAMPLE_OFFSET: [-15,0]
+
+MODEL:
+    ROI_HEAD:
+        TRANS_INPUT: 64
+
+        ROI_GRID_POOL:
+            MLPS: [[64,64]]
+            POOL_RADIUS: [0.8]
+            NSAMPLE: [16]
+
+        Transformer:
+            num_frames: 16
+            sequence_stride: 4

tools/cfgs/waymo_models/modifiers/mppnet/4frames.yaml

@@ -0,0 +1,15 @@
+DATA_CONFIG:
+    SEQUENCE_CONFIG:
+        SAMPLE_OFFSET: [-3,0]
+
+MODEL:
+    ROI_HEAD:
+        TRANS_INPUT: 256
+
+        ROI_GRID_POOL:
+            MLPS: [[128,128], [128,128]]
+            POOL_RADIUS: [0.8, 1.6]
+            NSAMPLE: [16, 16]
+
+        Transformer:
+            num_frames: 4

tools/cfgs/waymo_models/modifiers/mppnet/e2e_memorybank_inference.yaml

@@ -0,0 +1,93 @@
+DATA_CONFIG:
+    _BASE_CONFIG_: cfgs/dataset_configs/modifiers/waymo/multiframe.yaml
+    _TRANSFORM:
+    - name: remove_element
+      target: ROI_BOXES_PATH
+    - name: remove_element
+      target: USE_PREDBOX
+
+    SAMPLED_INTERVAL: {
+        'train': 5,
+        'test': 1
+    }
+
+    POINT_FEATURE_ENCODING:
+        encoding_type: absolute_coordinates_encoding
+        used_feature_list: ['x', 'y', 'z', 'intensity', 'elongation', 'time']
+        src_feature_list: ['x', 'y', 'z', 'intensity', 'elongation', 'time']
+
+MODEL:
+    NAME: MPPNetE2E
+
+    VFE:
+        NAME: DynMeanVFE
+
+    BACKBONE_3D:
+        NAME: VoxelResBackBone8x
+
+    MAP_TO_BEV:
+        NAME: HeightCompression
+        NUM_BEV_FEATURES: 256
+
+    BACKBONE_2D:
+        NAME: BaseBEVBackbone 
+        NUM_FRAME: 2
+        LAYER_NUMS: [5, 5]
+        LAYER_STRIDES: [1, 2]
+        NUM_FILTERS: [128, 256]
+        UPSAMPLE_STRIDES: [1, 2]
+        NUM_UPSAMPLE_FILTERS: [256, 256]
+
+    DENSE_HEAD:
+        NAME: CenterHead
+        CLASS_AGNOSTIC: False
+
+        CLASS_NAMES_EACH_HEAD: [
+            ['Vehicle', 'Pedestrian', 'Cyclist']
+        ]
+
+        SHARED_CONV_CHANNEL: 64
+        USE_BIAS_BEFORE_NORM: True
+        NUM_HM_CONV: 2
+        SEPARATE_HEAD_CFG:
+            HEAD_ORDER: ['center', 'center_z', 'dim', 'rot','vel']
+            HEAD_DICT: {
+                'center': {'out_channels': 2, 'num_conv': 2},
+                'center_z': {'out_channels': 1, 'num_conv': 2},
+                'dim': {'out_channels': 3, 'num_conv': 2},
+                'rot': {'out_channels': 2, 'num_conv': 2},
+                'vel': {'out_channels': 2, 'num_conv': 2},
+
+            }
+
+        TARGET_ASSIGNER_CONFIG:
+            FEATURE_MAP_STRIDE: 8
+            NUM_MAX_OBJS: 500
+            GAUSSIAN_OVERLAP: 0.1
+            MIN_RADIUS: 2
+
+        LOSS_CONFIG:
+            LOSS_WEIGHTS: {
+                'cls_weight': 1.0,
+                'loc_weight': 2.0,
+                'code_weights': [1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.2, 0.2]
+            }
+
+        POST_PROCESSING:
+            SCORE_THRESH: 0.1
+            POST_CENTER_LIMIT_RANGE: [-75.2, -75.2, -2, 75.2, 75.2, 4]
+            MAX_OBJ_PER_SAMPLE: 500
+            NMS_CONFIG:
+                NMS_TYPE: nms_gpu
+                NMS_THRESH: 0.7
+                NMS_PRE_MAXSIZE: 4096
+                NMS_POST_MAXSIZE: 500
+
+    ROI_HEAD:
+        NAME: MPPNetHeadE2E
+        USE_BOX_ENCODING:
+            NORM_T0: True
+            ALL_YAW_T0: True
+
+        Transformer:
+            sequence_stride: 1 #16frame using 4

tools/cfgs/waymo_models/mppnet.yaml

@@ -0,0 +1,170 @@
+CLASS_NAMES: ['Vehicle', 'Pedestrian', 'Cyclist']
+
+DATA_CONFIG:
+    _BASE_CONFIG_: cfgs/dataset_configs/waymo_dataset.yaml
+    PROCESSED_DATA_TAG: 'waymo_processed_data_v0_5_0'
+
+    SAMPLED_INTERVAL: {
+        'train': 1,
+        'test': 1
+    }
+    FILTER_EMPTY_BOXES_FOR_TRAIN: True
+    DISABLE_NLZ_FLAG_ON_POINTS: True
+
+    SEQUENCE_CONFIG:
+        ENABLED: True
+        SAMPLE_OFFSET: [-3,0]
+
+    USE_PREDBOX: True
+    ROI_BOXES_PATH: {
+        'train': '../output/xxxxx/train/result.pkl',  # example: predicted boxes of RPN in training set
+        'test': '../output/xxxxx/val/result.pkl',  # example: predicted boxes of RPN in evalulation set
+    }
+
+    DATA_AUGMENTOR:
+        DISABLE_AUG_LIST: [ 'placeholder' ]
+        AUG_CONFIG_LIST:
+
+            -   NAME: random_world_flip
+                ALONG_AXIS_LIST: [ 'x', 'y' ]
+
+            -   NAME: random_world_rotation
+                WORLD_ROT_ANGLE: [ -0.78539816, 0.78539816 ]
+
+            -   NAME: random_world_scaling
+                WORLD_SCALE_RANGE: [ 0.95, 1.05 ]
+
+    DATA_PROCESSOR:
+        -   NAME: mask_points_and_boxes_outside_range
+            REMOVE_OUTSIDE_BOXES: True
+
+        -   NAME: shuffle_points
+            SHUFFLE_ENABLED: {
+                'train': True,
+                'test': True
+            }
+
+    POINT_FEATURE_ENCODING: {
+        encoding_type: absolute_coordinates_encoding,
+        used_feature_list: ['x', 'y', 'z', 'intensity', 'elongation', 'time'],
+        src_feature_list: ['x', 'y', 'z', 'intensity', 'elongation', 'time'],
+        }
+
+
+MODEL:
+    NAME: MPPNet
+
+    ROI_HEAD:
+        NAME: MPPNetHead
+        TRANS_INPUT: 256
+        CLASS_AGNOSTIC: True
+        USE_BOX_ENCODING:
+            ENABLED: True 
+        AVG_STAGE1_SCORE: True
+        USE_TRAJ_EMPTY_MASK: True
+        USE_AUX_LOSS: True
+        IOU_WEIGHT: [0.5,0.4]
+
+
+
+        ROI_GRID_POOL:
+            GRID_SIZE: 4
+            MLPS: [[128,128], [128,128]]
+            POOL_RADIUS: [0.8, 1.6]
+            NSAMPLE: [16, 16]
+            POOL_METHOD: max_pool
+
+
+        _TRANSFORM:
+            name: copy
+            src: TRANS_INPUT
+            dest: Transformer.hidden_dim
+        Transformer:
+            num_lidar_points: 128
+            num_proxy_points: 64  # GRID_SIZE*GRID_SIZE*GRID_SIZE
+            pos_hidden_dim: 64
+            enc_layers: 3
+            dim_feedforward: 512
+            dropout: 0.1
+            nheads: 4
+            pre_norm: False
+            num_frames: 4
+            num_groups: 4
+            use_grid_pos:
+                enabled: True
+                init_type: index
+
+            use_mlp_mixer:
+                enabled: True
+                hidden_dim: 16
+
+        TARGET_CONFIG:
+            BOX_CODER: ResidualCoder
+            ROI_PER_IMAGE: 96
+            FG_RATIO: 0.5
+            REG_AUG_METHOD: single
+            ROI_FG_AUG_TIMES: 10
+            RATIO: 0.2
+            USE_ROI_AUG: True
+            USE_TRAJ_AUG:
+                ENABLED: True
+                THRESHOD: 0.8
+            SAMPLE_ROI_BY_EACH_CLASS: True
+            CLS_SCORE_TYPE: roi_iou
+
+            CLS_FG_THRESH: 0.75
+            CLS_BG_THRESH: 0.25
+            CLS_BG_THRESH_LO: 0.1
+            HARD_BG_RATIO: 0.8
+
+            REG_FG_THRESH: 0.55
+
+        LOSS_CONFIG:
+            CLS_LOSS: BinaryCrossEntropy
+            REG_LOSS: smooth-l1
+            CORNER_LOSS_REGULARIZATION: True
+            LOSS_WEIGHTS: {
+                'rcnn_cls_weight': 1.0,
+                'rcnn_reg_weight': 1.0,
+                'rcnn_corner_weight': 2.0,
+                'traj_reg_weight': [2.0, 2.0, 2.0],
+                'code_weights': [1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0]
+            }
+
+    POST_PROCESSING:
+        RECALL_THRESH_LIST: [0.3, 0.5, 0.7]
+        SCORE_THRESH: 0.1
+        OUTPUT_RAW_SCORE: False
+        SAVE_BBOX: False
+        EVAL_METRIC: waymo
+        NOT_APPLY_NMS_FOR_VEL: True
+
+        NMS_CONFIG:
+            MULTI_CLASSES_NMS: False
+            NMS_TYPE: nms_gpu
+            NMS_THRESH: 0.7
+            NMS_PRE_MAXSIZE: 4096
+            NMS_POST_MAXSIZE: 500
+
+
+OPTIMIZATION:
+    BATCH_SIZE_PER_GPU: 2
+    NUM_EPOCHS: 6
+
+    OPTIMIZER: adam_onecycle
+    LR: 0.003
+    WEIGHT_DECAY: 0.01
+    MOMENTUM: 0.9
+
+    MOMS: [0.95, 0.85]
+    PCT_START: 0.4
+    DIV_FACTOR: 10
+    DECAY_STEP_LIST: [35, 45]
+    LR_DECAY: 0.1
+    LR_CLIP: 0.0000001
+
+    LR_WARMUP: False
+    WARMUP_EPOCH: 1
+
+    GRAD_NORM_CLIP: 10
+