nucleus.annotation_uploader

AnnotationUploader

This is a helper class not intended for direct use. Please use dataset.annotate

PredictionUploader

This is a helper class not intended for direct use. Please use dataset.annotate
class nucleus.annotation_uploader.AnnotationUploader(dataset_id, client)

This is a helper class not intended for direct use. Please use dataset.annotate or dataset.upload_predictions.

This class is purely a helper class for implementing dataset.annotate/dataset.predict.

Parameters:
static check_for_duplicate_ids(annotations)

Do not allow annotations to have the same (annotation_id, reference_id) tuple

Parameters:

annotations (Iterable[nucleus.annotation.Annotation])

get_form_data_and_file_pointers_fn(segmentations, update)

Defines a function to be called on each retry.

File pointers are also returned so whoever calls this function can appropriately close the files. This is intended for use with a FormDataContextHandler in order to make form data requests.

Parameters:
upload(annotations, batch_size=5000, update=False, remote_files_per_upload_request=20, local_files_per_upload_request=10, trained_slice_id=None)

For more details on parameters and functionality, see dataset.annotate.

Parameters:

  • annotations (Iterable[nucleus.annotation.Annotation])

  • batch_size (int)

  • update (bool)

  • remote_files_per_upload_request (int)

  • local_files_per_upload_request (int)

  • trained_slice_id (Optional[str])

class nucleus.annotation_uploader.PredictionUploader(client, dataset_id=None, model_id=None, model_run_id=None, trained_slice_id=None)

This is a helper class not intended for direct use. Please use dataset.annotate or dataset.upload_predictions.

This class is purely a helper class for implementing dataset.annotate/dataset.predict.

Parameters:

  • client (nucleus.NucleusClient)

  • dataset_id (Optional[str])

  • model_id (Optional[str])

  • model_run_id (Optional[str])

  • trained_slice_id (Optional[str])

static check_for_duplicate_ids(annotations)

Do not allow annotations to have the same (annotation_id, reference_id) tuple

Parameters:

annotations (Iterable[nucleus.annotation.Annotation])

get_form_data_and_file_pointers_fn(segmentations, update)

Defines a function to be called on each retry.

File pointers are also returned so whoever calls this function can appropriately close the files. This is intended for use with a FormDataContextHandler in order to make form data requests.

Parameters:
upload(annotations, batch_size=5000, update=False, remote_files_per_upload_request=20, local_files_per_upload_request=10, trained_slice_id=None)

For more details on parameters and functionality, see dataset.annotate.

Parameters:

  • annotations (Iterable[nucleus.annotation.Annotation])

  • batch_size (int)

  • update (bool)

  • remote_files_per_upload_request (int)

  • local_files_per_upload_request (int)

  • trained_slice_id (Optional[str])