Merge pull request #2 from sshaoshuai/develop_gcx_release

Support PointPillar, Second, AxisAlignedTargetAssigner, MultiHead and multi-gpu testing

tools/cfgs/dataset_configs/kitti_dataset_pointpillar.yaml

+DATASET: 'KittiDataset'
+DATA_PATH: '../data/kitti'
+
+POINT_CLOUD_RANGE: [0, -39.68, -3, 69.12, 39.68, 1]
+
+DATA_SPLIT: {
+    'train': train,
+    'test': val
+}
+
+INFO_PATH: {
+    'train': [kitti_infos_train.pkl],
+    'test': [kitti_infos_val.pkl],
+}
+
+FOV_POINTS_ONLY: True
+
+
+DATA_AUGMENTOR:
+    - NAME: gt_sampling
+      USE_ROAD_PLANE: True
+      DB_INFO_PATH:
+          - kitti_dbinfos_train.pkl
+      PREPARE: {
+         filter_by_min_points: ['Car:5', 'Pedestrian:5', 'Cyclist:5'],
+         filter_by_difficulty: [-1],
+      }
+
+      SAMPLE_GROUPS: ['Car:15','Pedestrian:10', 'Cyclist:10']
+      NUM_POINT_FEATURES: 4
+      DATABASE_WITH_FAKELIDAR: True
+      REMOVE_EXTRA_WIDTH: [0.0, 0.0, 0.0]
+      LIMIT_WHOLE_SCENE: False
+
+    - NAME: random_world_flip
+      ALONG_AXIS_LIST: ['x']
+
+    - 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: {
+    encoding_type: absolute_coordinates_encoding,
+    used_feature_list: ['x', 'y', 'z', 'intensity'],
+    src_feature_list: ['x', 'y', 'z', 'intensity'],
+}
+
+
+DATA_PROCESSOR:
+    - NAME: mask_points_and_boxes_outside_range
+      REMOVE_OUTSIDE_BOXES: True
+
+    - NAME: shuffle_points
+      SHUFFLE_ENABLED: {
+        'train': True,
+        'test': False
+      }
+
+    - NAME: transform_points_to_voxels
+      VOXEL_SIZE: [0.16, 0.16, 4]
+      MAX_POINTS_PER_VOXEL: 32
+      MAX_NUMBER_OF_VOXELS: {
+        'train': 16000,
+        'test': 40000
+      }

tools/cfgs/kitti_models/PartA2.yaml

@@ -46,14 +46,14 @@ MODEL:
             {
                 'anchor_sizes': [[0.8, 0.6, 1.73]],
                 'anchor_rotations': [0, 1.57],
-                'anchor_bottom_heights': [-1.78],
+                'anchor_bottom_heights': [-0.6],
                 'align_center': False,
                 'feature_map_stride': 8
             },
             {
                 'anchor_sizes': [[1.76, 0.6, 1.73]],
                 'anchor_rotations': [0, 1.57],
-                'anchor_bottom_heights': [-1.78],
+                'anchor_bottom_heights': [-0.6],
                 'align_center': False,
                 'feature_map_stride': 8
             }

tools/cfgs/kitti_models/second.yaml

+CLASS_NAMES: ['Car', 'Pedestrian', 'Cyclist']
+
+DATA_CONFIG: 
+    _BASE_CONFIG_: cfgs/dataset_configs/kitti_dataset.yaml
+
+
+MODEL:
+    NAME: SECONDNet
+
+    VFE:
+        NAME: MeanVFE
+
+    BACKBONE_3D:
+        NAME: VoxelBackBone8x
+
+    MAP_TO_BEV:
+        NAME: HeightCompression
+        NUM_BEV_FEATURES: 256
+
+    BACKBONE_2D:
+        NAME: BaseBEVBackbone
+
+        LAYER_NUMS: [5, 5]
+        LAYER_STRIDES: [1, 2]
+        NUM_FILTERS: [128, 256]
+        UPSAMPLE_STRIDES: [1, 2]
+        NUM_UPSAMPLE_FILTERS: [256, 256]
+
+    DENSE_HEAD:
+        NAME: AnchorHeadSingle
+        CLASS_AGNOSTIC: False
+
+        USE_DIRECTION_CLASSIFIER: True
+        DIR_OFFSET: 0.78539
+        DIR_LIMIT_OFFSET: 0.0
+        NUM_DIR_BINS: 2
+
+        ANCHOR_GENERATOR_CONFIG: [
+            {
+                'anchor_sizes': [[3.9, 1.6, 1.56]],
+                'anchor_rotations': [0, 1.57],
+                'anchor_bottom_heights': [-1.78],
+                'align_center': False,
+                'feature_map_stride': 8
+            },
+            {
+                'anchor_sizes': [[0.8, 0.6, 1.73]],
+                'anchor_rotations': [0, 1.57],
+                'anchor_bottom_heights': [-0.6],
+                'align_center': False,
+                'feature_map_stride': 8
+            },
+            {
+                'anchor_sizes': [[1.76, 0.6, 1.73]],
+                'anchor_rotations': [0, 1.57],
+                'anchor_bottom_heights': [-0.6],
+                'align_center': False,
+                'feature_map_stride': 8
+            }
+        ]
+
+        TARGET_ASSIGNER_CONFIG:
+            NAME: AxisAlignedTargetAssigner
+            POS_FRACTION: -1.0
+            SAMPLE_SIZE: 512
+            MATCHED_THRESHOLDS: [0.6, 0.5, 0.5]
+            UNMATCHED_THRESHOLDS: [0.45, 0.35, 0.35]
+            NORM_BY_NUM_EXAMPLES: False
+            MATCH_HEIGHT: False
+            BOX_CODER: ResidualCoder
+
+        LOSS_CONFIG:
+            LOSS_WEIGHTS: {
+                'cls_weight': 1.0,
+                'loc_weight': 2.0,
+                'dir_weight': 0.2,
+                '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
+
+        EVAL_METRIC: kitti
+
+        NMS_CONFIG:
+            MULTI_CLASSES_NMS: False
+            NMS_TYPE: nms_gpu
+            NMS_THRESH: 0.01
+            NMS_PRE_MAXSIZE: 4096
+            NMS_POST_MAXSIZE: 500
+
+
+OPTIMIZATION:
+    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

tools/cfgs/kitti_models/second_multihead.yaml

+CLASS_NAMES: ['Car', 'Pedestrian', 'Cyclist']
+DATA_CONFIG: 
+    _BASE_CONFIG_: cfgs/dataset_configs/kitti_dataset.yaml
+
+
+MODEL:
+    NAME: SECONDNet
+
+    VFE:
+        NAME: MeanVFE
+
+    BACKBONE_3D:
+        NAME: VoxelBackBone8x
+
+    MAP_TO_BEV:
+        NAME: HeightCompression
+        NUM_BEV_FEATURES: 256
+
+    BACKBONE_2D:
+        NAME: BaseBEVBackbone
+
+        LAYER_NUMS: [5, 5]
+        LAYER_STRIDES: [1, 2]
+        NUM_FILTERS: [128, 256]
+        UPSAMPLE_STRIDES: [1, 2]
+        NUM_UPSAMPLE_FILTERS: [256, 256]
+
+    DENSE_HEAD:
+        NAME: AnchorHeadMulti
+        CLASS_AGNOSTIC: False
+
+        USE_DIRECTION_CLASSIFIER: True
+        DIR_OFFSET: 0.78539
+        DIR_LIMIT_OFFSET: 0.0
+        NUM_DIR_BINS: 2
+        
+        USE_MULTI_HEAD: True
+        ANCHOR_GENERATOR_CONFIG: [
+            {
+                'anchor_sizes': [[3.9, 1.6, 1.56]],
+                'anchor_rotations': [0, 1.57],
+                'anchor_bottom_heights': [-1.6],
+                'align_center': False,
+                'feature_map_stride': 16
+            },
+            {
+                'anchor_sizes': [[0.8, 0.6, 1.73]],
+                'anchor_rotations': [0, 1.57],
+                'anchor_bottom_heights': [-1.6],
+                'align_center': False,
+                'feature_map_stride': 8
+            },
+            {
+                'anchor_sizes': [[1.76, 0.6, 1.73]],
+                'anchor_rotations': [0, 1.57],
+                'anchor_bottom_heights': [-1.6],
+                'align_center': False,
+                'feature_map_stride': 8
+            }
+        ]
+
+        RPN_HEAD_CFGS: [
+            {
+                'head_cls_name': ['Car'],
+                'stride': 2,
+                'layer_num': 2
+            },
+            {
+                'head_cls_name': ['Pedestrian', 'Cyclist'],
+                'stride': 1,
+                'layer_num': 2
+            }, 
+
+
+        ]
+
+
+        TARGET_ASSIGNER_CONFIG:
+           NAME: AxisAlignedTargetAssigner
+           POS_FRACTION: -1.0
+           SAMPLE_SIZE: 512
+           MATCHED_THRESHOLDS: [0.6, 0.5, 0.5]
+           UNMATCHED_THRESHOLDS: [0.45, 0.35, 0.35]
+           NORM_BY_NUM_EXAMPLES: False
+           MATCH_HEIGHT: False
+           BOX_CODER: ResidualCoder
+
+        LOSS_CONFIG:
+            LOSS_WEIGHTS: {
+                'cls_weight': 1.0,
+                'loc_weight': 2.0,
+                'dir_weight': 0.2,
+                '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]
+        MULTI_CLASSES_NMS: False
+        SCORE_THRESH: 0.1
+        OUTPUT_RAW_SCORE: False
+
+        EVAL_METRIC: kitti
+
+        NMS_CONFIG:
+            MULTI_CLASSES_NMS: False
+            NMS_TYPE: nms_gpu
+            NMS_THRESH: 0.1
+            NMS_PRE_MAXSIZE: 4096
+            NMS_POST_MAXSIZE: 500
+
+
+OPTIMIZATION:
+    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

tools/eval_utils/eval_utils.py

@@ -3,7 +3,7 @@ import time
 import pickle
 import numpy as np
 import torch
-from mmpcdet.utils import common_utils
+from pcdet.utils import common_utils
 
 
 def statistics_info(cfg, ret_dict, metric, disp_dict):
@@ -38,7 +38,13 @@ def eval_one_epoch(cfg, model, dataloader, epoch_id, logger, dist_test=False, sa
 
     logger.info('*************** EPOCH %s EVALUATION *****************' % epoch_id)
     if dist_test:
-        raise NotImplementedError
+        num_gpus = torch.cuda.device_count()
+        local_rank = cfg.LOCAL_RANK % num_gpus
+        model = torch.nn.parallel.DistributedDataParallel(
+                model,
+                device_ids=[local_rank],
+                broadcast_buffers=False
+        )
     model.eval()
 
     if cfg.LOCAL_RANK == 0:
@@ -71,7 +77,8 @@ def eval_one_epoch(cfg, model, dataloader, epoch_id, logger, dist_test=False, sa
 
     if dist_test:
         rank, world_size = common_utils.get_dist_info()
-        raise NotImplementedError
+        det_annos = common_utils.merge_results_dist(det_annos, len(dataset), tmpdir=result_dir / 'tmpdir')
+        metric = common_utils.merge_results_dist([metric], world_size, tmpdir=result_dir / 'tmpdir')
 
     logger.info('*************** Performance of EPOCH %s *****************' % epoch_id)
     sec_per_example = (time.time() - start_time) / len(dataloader.dataset)
@@ -82,7 +89,10 @@ def eval_one_epoch(cfg, model, dataloader, epoch_id, logger, dist_test=False, sa
 
     ret_dict = {}
     if dist_test:
-        raise NotImplementedError
+        for key, val in metric[0].items():
+            for k in range(1, world_size):
+                metric[0][key] += metric[k][key]
+        metric = metric[0]
 
     gt_num_cnt = metric['gt_num']
     for cur_thresh in cfg.MODEL.POST_PROCESSING.RECALL_THRESH_LIST:

tools/scripts/slurm_test_mgpu.sh

+#!/usr/bin/env bash
+
+set -x
+
+PARTITION=$1
+GPUS=$2
+GPUS_PER_NODE=$GPUS
+PY_ARGS=${@:3}
+JOB_NAME=eval
+SRUN_ARGS=${SRUN_ARGS:-""}
+
+srun -p ${PARTITION} \
+    --job-name=${JOB_NAME} \
+    --gres=gpu:${GPUS_PER_NODE} \
+    --ntasks=${GPUS} \
+    --ntasks-per-node=${GPUS_PER_NODE} \
+    --kill-on-bad-exit=1 \
+    ${SRUN_ARGS} \
+    python -u test.py --launcher slurm ${PY_ARGS}
+

tools/test.py

@@ -8,14 +8,14 @@ import datetime
 import argparse
 from pathlib import Path
 import torch.distributed as dist
-from mmpcdet.datasets import build_dataloader
-from mmpcdet.models import build_network
-from mmpcdet.utils import common_utils
-from mmpcdet.config import cfg, cfg_from_list, cfg_from_yaml_file, log_config_to_file
+from pcdet.datasets import build_dataloader
+from pcdet.models import build_network
+from pcdet.utils import common_utils
+from pcdet.config import cfg, cfg_from_list, cfg_from_yaml_file, log_config_to_file
 from eval_utils import eval_utils
 
 
-def parge_config():
+def parse_config():
     parser = argparse.ArgumentParser(description='arg parser')
     parser.add_argument('--cfg_file', type=str, default=None, help='specify the config for training')
 
@@ -128,7 +128,7 @@ def repeat_eval_ckpt(model, test_loader, args, eval_output_dir, logger, ckpt_dir
 
 
 def main():
-    args, cfg = parge_config()
+    args, cfg = parse_config()
     if args.launcher == 'none':
         dist_test = False
     else:

tools/train.py

@@ -16,7 +16,7 @@ import datetime
 import glob
 
 
-def parge_config():
+def parse_config():
     parser = argparse.ArgumentParser(description='arg parser')
     parser.add_argument('--cfg_file', type=str, default=None, help='specify the config for training')
 
@@ -54,7 +54,7 @@ def parge_config():
 
 
 def main():
-    args, cfg = parge_config()
+    args, cfg = parse_config()
     if args.launcher == 'none':
         dist_train = False
     else: