tlc.integration.super_gradients.callbacks.pipeline_params

Utility dataclass for storing SuperGradients inference pipeline parameters.

Module Contents

Classes

Class

Description

PipelineParams

Parameters for model prediction pipelines across detection and pose models.

API

class PipelineParams

Parameters for model prediction pipelines across detection and pose models.

iou: float | None = None

IoU threshold for NMS.

  • Supported by: detection models (e.g., YoloBase, CustomizableDetector, PPYoloE, SlidingWindowInferenceDetectionWrapper) and YoloNASPose.

  • Default: model-specific default or None to use model default.

conf: float | None = None

Confidence threshold; predictions below are discarded.

  • Supported by: detection models and pose models (YoloNASPose, DEKRPoseEstimationModel).

  • Default: model-specific default or None to use model default.

fuse_model: bool = True

Fuse layers to improve performance (at the cost of memory).

  • Supported by: detection, pose, classification, segmentation.

  • Default: True.

skip_image_resizing: bool = False

Skip image resizing in the processor (stride-constrained auto-padding).

  • Supported by: detection, pose, classification, segmentation.

  • Default: False.

nms_top_k: int | None = None

Top-K candidates considered before NMS.

  • Supported by: detection models exposing full NMS controls (CustomizableDetector, PPYoloE, SlidingWindowInferenceDetectionWrapper).

  • Default: model-specific default or None to use model default.

max_predictions: int | None = None

Max detections returned per image after NMS.

  • Supported by: detection models exposing full NMS controls (CustomizableDetector, PPYoloE, SlidingWindowInferenceDetectionWrapper).

  • Default: model-specific default or None to use model default.

multi_label_per_box: bool | None = None

Allow multiple class labels per anchor.

  • Supported by: detection models that support multi-label decoding (e.g., PPYoloE, some CustomizableDetector heads).

  • Default: model-specific default or None to use model default.

class_agnostic_nms: bool | None = None

Apply NMS across classes (True) or per-class (False).

  • Supported by: detection models exposing class-agnostic NMS.

  • Default: model-specific default or None to use model default.

pre_nms_max_predictions: int | None = None

Top-K candidates kept before NMS in pose decoding.

  • Supported by: YoloNASPose.

  • Default: model-specific default or None to use model default.

post_nms_max_predictions: int | None = None

Top-K predictions retained after NMS in pose decoding.

  • Supported by: YoloNASPose.

  • Default: model-specific default or None to use model default.

fp16: bool = True

Use mixed precision for inference.

  • Supported by: most pipelines.

  • Default: True.