mmcv Config使用之BEVFormer中的配置文件使用

最新推荐文章于 2025-08-12 23:13:02 发布

信雪神话

最新推荐文章于 2025-08-12 23:13:02 发布

阅读量116

点赞数 2

CC 4.0 BY-SA版权

分类专栏： AutomaticDrive 文章标签：深度学习人工智能算法

本文链接：https://blog.csdn.net/hookie1990/article/details/149069142

AutomaticDrive 专栏收录该内容

4 篇文章

订阅专栏

mmcv.Config.fromfile() 是 MMCV 提供的一个配置加载工具，它会将 .py 配置文件中的变量解析为一个 Config 对象。这个对象允许你通过点号语法（如 cfg.data.train）访问嵌套结构。

from mmcv import Config
cfg = Config.fromfile('path/to/config.py')
作用：读取 config.py 文件中所有定义的变量，并构建成一个 Config 对象。
你可以像访问字典一样访问这些变量：
print(cfg.data.train)print(cfg.model.type)

BEVFormer 工程里train.py有如下代码片段

args = parse_args() #导入配置文件路径 默认为BEVFormer/projects/configs/bevformer/bevformer_tiny.py
cfg = Config.fromfile(args.config)

args.config 默认为bevformer_tiny.py，即将bevformer_tiny.py定义的变量构建一个config对象。

train.py中代码：

if len(cfg.workflow) == 2:
        val_dataset = copy.deepcopy(cfg.data.val)
        # in case we use a dataset wrapper
        if 'dataset' in cfg.data.train:
            val_dataset.pipeline = cfg.data.train.dataset.pipeline
        else:
            val_dataset.pipeline = cfg.data.train.pipeline

找到bevformer_tiny.py定义代码片段:

data = dict(
    samples_per_gpu=1,
    workers_per_gpu=8,
    train=dict(
        type=dataset_type,
        data_root=data_root,
        ann_file=data_root + 'robotruck_infos_train.pkl',
        pipeline=train_pipeline,
        classes=class_names,
        modality=input_modality,
        test_mode=False,
        use_valid_flag=True,
        bev_size=(bev_h_, bev_w_),
        queue_length=queue_length,
        # we use box_type_3d='LiDAR' in kitti and nuscenes dataset
        # and box_type_3d='Depth' in sunrgbd and scannet dataset.
        box_type_3d='LiDAR'),
    val=dict(type=dataset_type,
             data_root=data_root,
             ann_file=data_root + 'robotruck_infos_val.pkl',
             pipeline=test_pipeline,  bev_size=(bev_h_, bev_w_),
             classes=class_names, modality=input_modality, samples_per_gpu=1),
    test=dict(type=dataset_type,
              data_root=data_root,
              ann_file=data_root + 'robotruck_infos_val.pkl',
              pipeline=test_pipeline, bev_size=(bev_h_, bev_w_),
              classes=class_names, modality=input_modality),
    shuffler_sampler=dict(type='DistributedGroupSampler'),
    nonshuffler_sampler=dict(type='DistributedSampler')
)

data.train无dataset字段，因此val_dataset.pipeline = cfg.data.train.pipeline，此处引用的pipeline即train_pipeline，即:

train_pipeline = [
    dict(type='LoadMultiViewImageFromFiles', to_float32=True),
    dict(type='PhotoMetricDistortionMultiViewImage'),
    dict(type='LoadAnnotations3D', with_bbox_3d=True, with_label_3d=True, with_attr_label=False),
    dict(type='ObjectRangeFilter', point_cloud_range=point_cloud_range),
    dict(type='ObjectNameFilter', classes=class_names),
    dict(type='RandomScaleImageMultiViewImage', scales=[0.3]),
    dict(type='NormalizeMultiviewImage', **img_norm_cfg),
    dict(type='PadMultiViewImage', size_divisor=32),
    dict(type='DefaultFormatBundle3D', class_names=class_names),
    dict(type='CustomCollect3D', keys=['gt_bboxes_3d', 'gt_labels_3d', 'img'])
]