napari_cellseg3d.code_models.workers_utils#

Several worker-related utilities for inference and training.

Classes

InferenceResult([image_id, original, ...])

Class to record results of a segmentation job.

LogSignal([parent])

Signal to send messages to be logged from another thread.

ONNXModelWrapper(file_location)

Class to replace torch model by ONNX Runtime session.

QuantileNormalization()

MONAI-style transform to normalize each image in a batch individually by quantile normalization.

QuantileNormalizationd(*args, **kwargs)

MONAI-style dict transform to normalize each image in a batch individually by quantile normalization.

RemapTensor(new_max, new_min)

Remap the values of a tensor to a new range.

Threshold([threshold])

Threshold a tensor to 0 or 1.

TqdmToLogSignal(log_func)

File-like object to redirect tqdm output to the logger widget in the GUI that self.log emits to.

TrainingReport([show_plot, epoch, ...])

Class to record results of a training job.

WeightsDownloader([log_widget])

A utility class the downloads the weights of a model when needed.