sleap.nn.inference

Inference pipelines and utilities.

class sleap.nn.inference.BottomupPredictor(bottomup_config: sleap.nn.config.training_job.TrainingJobConfig, bottomup_model: sleap.nn.model.Model, peak_threshold: float = 0.2)

classmethod from_trained_models(bottomup_model_path: str) → sleap.nn.inference.BottomupPredictor

Create predictor from saved models.

class sleap.nn.inference.MockPredictor(labels: sleap.io.dataset.Labels)

class sleap.nn.inference.Predictor

Base interface class for predictors.

class sleap.nn.inference.SingleInstancePredictor(confmap_config: sleap.nn.config.training_job.TrainingJobConfig, confmap_model: sleap.nn.model.Model, peak_threshold: float = 0.2, integral_refinement: bool = True, integral_patch_size: int = 5)

classmethod from_trained_models(confmap_model_path: str, peak_threshold: float = 0.2, integral_refinement: bool = True, integral_patch_size: int = 5) → sleap.nn.inference.SingleInstancePredictor

Create predictor from saved models.

class sleap.nn.inference.TopdownPredictor(centroid_config: Optional[sleap.nn.config.training_job.TrainingJobConfig] = None, centroid_model: Optional[sleap.nn.model.Model] = None, confmap_config: Optional[sleap.nn.config.training_job.TrainingJobConfig] = None, confmap_model: Optional[sleap.nn.model.Model] = None, batch_size: int = 1, peak_threshold: float = 0.2, integral_refinement: bool = True, integral_patch_size: int = 5)

classmethod from_trained_models(centroid_model_path: Optional[str] = None, confmap_model_path: Optional[str] = None, batch_size: int = 1, peak_threshold: float = 0.2, integral_refinement: bool = True, integral_patch_size: int = 5) → sleap.nn.inference.TopdownPredictor

Create predictor from saved models.

Parameters

• centroid_model_path – Path to centroid model folder.

• confmap_model_path – Path to topdown confidence map model folder.

Returns

An instance of TopdownPredictor with the loaded models.

One of the two models can be left as None to perform inference with ground truth data. This will only work with LabelsReader as the provider.

class sleap.nn.inference.VisualPredictor(config: sleap.nn.config.training_job.TrainingJobConfig, model: sleap.nn.model.Model)

Predictor class for generating the visual output of model.

sleap.nn.inference.find_heads_for_model_paths(paths) → Dict[str, str]

Given list of models paths, returns dict with path keyed by head name.

sleap.nn.inference.main()

CLI for running inference.

sleap.nn.inference.make_predictor_from_models(trained_model_paths: Dict[str, str], labels_path: Optional[str] = None, policy_args: Optional[dict] = None) → sleap.nn.inference.Predictor

Given dict of paths keyed by head name, returns appropriate predictor.

sleap.nn.inference.make_predictor_from_paths(paths) → sleap.nn.inference.Predictor

Builds predictor object from a list of model paths.

sleap.nn.inference.safely_generate(ds: tensorflow.python.data.ops.dataset_ops.DatasetV2, progress: bool = True)

Yields examples from dataset, catching and logging exceptions.