Skip to content

CenterNet

CenterNet class

dlf.models.centernet.CenterNet(
    feature_extractor,
    num_classes=12,
    score_threshold=0.1,
    pooling_nms=True,
    dropout_rate=None,
    max_objects=100,
    weight_decay=None,
    model_weights=None,
    summary=False,
    optimizer=None,
    loss=None,
    **kwargs
)

A implementation of CenterNet for object detection

Aliases

  • CenterNet
  • center_net
  • centernet

Arguments

  • feature_extractor: dict. A feature extractor
  • num_classes: int. Number of classes to detect
  • score_threshold: float, optional. Minimum classification score to be a valid box
  • pooling_nms: bool, optional. If true, max pooling is used as non maximum suppression. Defaults to False.
  • dropout_rate: float, optional. Dropout rate if None no dropout is used. Defaults to None.
  • max_objects: int, optional. max number of objects to detect. Defaults to 100.
  • weight_decay: float, optional. Weight decay factor. Defaults to None.
  • model_weights: str, optional. Path to the pretrained model weights. Defaults to None.
  • summary: bool, optional. If true a summary of the model will be printed to stdout. Defaults to False.
  • optimizer: list of dict, optional. Name of optimizer used for training.
  • loss: list of dict, optional. List of loss objects to build for this model. Defaults to None.

YAML Configuration

model:
    centernet:
        feature_extractor:
        model_name: ResNet50
        input_shape:
            - &width 512
            - &height 512
            - 3
        weights: "imagenet"
        num_classes: 6
        weight_decay: 0.0005
        summary: True
        loss:
            centernetloss:

        optimizer:
        - Adam: # lr for simclr unsupervised
            learning_rate:
                PolynomialDecay:
                initial_learning_rate: 0.001
                decay_steps: 195312
                end_learning_rate: 0.00001

References