efficientdet_d3(pretrained: bool = False, overrides: object = {})EfficientDet-D3 ().
Compound-scaled variant: EfficientNet-B3 backbone, BiFPN 160 channels with 6 repeats, head depth 4, 896x896 input. Approximately 12.0M parameters; COCO test-dev mAP of 45.8% (paper Table 2).
Model Size
Parameters
pretrainedbool= FalseReserved for future pretrained-weight loading. Currently ignored.
**overridesobject= {}Keyword overrides forwarded into
EfficientDetConfig.Returns
EfficientDetForObjectDetectionDetector with the D3 configuration applied.
Notes
See Tan et al., 2020 (arXiv:1911.09070), Table 2.
Examples
>>> import lucid
>>> from lucid.models.vision.efficientdet import efficientdet_d3
>>> model = efficientdet_d3()
>>> x = lucid.randn(1, 3, 896, 896)
>>> out = model(x)
>>> out.pred_boxes.shape[-1]
4