tfm.vision.layers.MultilevelDetectionGenerator

Generates detected boxes with scores and classes for one-stage detector.

apply_nms A bool of whether or not apply non maximum suppression. If False, the decoded boxes and their scores are returned.
pre_nms_top_k An int of the number of top scores proposals to be kept before applying NMS.
pre_nms_score_threshold A float of the score threshold to apply before applying NMS. Proposals whose scores are below this threshold are thrown away.
nms_iou_threshold A float in [0, 1], the NMS IoU threshold.
max_num_detections An int of the final number of total detections to generate.
nms_version A string of batched, v1 or v2 specifies NMS version
use_cpu_nms A bool of whether or not enforce NMS to run on CPU.
soft_nms_sigma A float representing the sigma parameter for Soft NMS. When soft_nms_sigma=0.0, we fall back to standard NMS.
tflite_post_processing_config An optional dictionary containing post-processing parameters used for TFLite custom NMS op.
pre_nms_top_k_sharding_block For v3 (edge tpu friendly) NMS, avoids creating long axis for pre_nms_top_k. Will do top_k in shards of size [num_classes, pre_nms_top_k_sharding_block * boxes_per_location]
nms_v3_refinements For v3 (edge tpu friendly) NMS, sets how close result should be to standard NMS. When None, 2 is used. Here is some experimental deviations for different refinement values: if == 0, AP is reduced 1.0%, AR is reduced 5% on COCO if == 1, AP is reduced 0.2%, AR is reduced 2% on COCO if == 2, AP is reduced <0.1%, AR is reduced <1% on COCO
return_decoded A bool of whether to return decoded boxes before NMS regardless of whether apply_nms is True or not.
use_class_agnostic_nms A bool of whether non max suppression is operated on all the boxes using max scores across all classes.
box_coder_weights An optional list of 4 positive floats to scale y, x, h, and w when encoding box coordinates. If set to None, does not perform scaling. For Faster RCNN, the open-source implementation recommends using [10.0, 10.0, 5.0, 5.0].
**kwargs Additional keyword arguments passed to Layer.

ValueError If use_class_agnostic_nms is required by nms_version is not specified as v2.

Methods

call

This is where the layer's logic lives.

The call() method may not create state (except in its first invocation, wrapping the creation of variables or other resources in tf.init_scope()). It is recommended to create state, including tf.Variable instances and nested Layer instances, in __init__(), or in the build() method that is called automatically before call() executes for the first time.

Args
inputs Input tensor, or dict/list/tuple of input tensors. The first positional inputs argument is subject to special rules:

  • inputs must be explicitly passed. A layer cannot have zero arguments, and inputs cannot be provided via the default value of a keyword argument.
  • NumPy array or Python scalar values in inputs get cast as tensors.
  • Keras mask metadata is only collected from inputs.
  • Layers are built (build(input_shape) method) using shape info from inputs only.
  • input_spec compatibility is only checked against inputs.
  • Mixed precision input casting is only applied to inputs. If a layer has tensor arguments in *args or **kwargs, their casting behavior in mixed precision should be handled manually.
  • The SavedModel input specification is generated using inputs only.
  • Integration with various ecosystem packages like TFMOT, TFLite, TF.js, etc is only supported for inputs and not for tensors in positional and keyword arguments.
*args Additional positional arguments. May contain tensors, although this is not recommended, for the reasons above.
**kwargs Additional keyword arguments. May contain tensors, although this is not recommended, for the reasons above. The following optional keyword arguments are reserved:
  • training: Boolean scalar tensor of Python boolean indicating whether the call is meant for training or inference.
  • mask: Boolean input mask. If the layer's call() method takes a mask argument, its default value will be set to the mask generated for inputs by the previous layer (if input did come from a layer that generated a corresponding mask, i.e. if it came from a Keras layer with masking support).
  • Returns
    A tensor or list/tuple of tensors.