package google.cloud.datalabeling.v1beta1
Get desktop application:
View/edit binary Protocol Buffers messages
Creates dataset. If success return a Dataset resource.
message
data_labeling_service.proto:327Request message for CreateDataset.
string parent = 1
Required. Dataset resource parent, format: projects/{project_id}
optional dataset = 2
Required. The dataset to be created.
Gets dataset by resource name.
message
data_labeling_service.proto:337Request message for GetDataSet.
string name = 1
Required. Dataset resource name, format: projects/{project_id}/datasets/{dataset_id}
Lists datasets under a project. Pagination is supported.
message
data_labeling_service.proto:344Request message for ListDataset.
string parent = 1
Required. Dataset resource parent, format: projects/{project_id}
string filter = 2
Optional. Filter on dataset is not supported at this moment.
int32 page_size = 3
Optional. Requested page size. Server may return fewer results than requested. Default value is 100.
string page_token = 4
Optional. A token identifying a page of results for the server to return. Typically obtained by [ListDatasetsResponse.next_page_token][google.cloud.datalabeling.v1beta1.ListDatasetsResponse.next_page_token] of the previous [DataLabelingService.ListDatasets] call. Returns the first page if empty.
message
data_labeling_service.proto:365Results of listing datasets within a project.
repeated datasets = 1
The list of datasets to return.
string next_page_token = 2
A token to retrieve next page of results.
Deletes a dataset by resource name.
message
data_labeling_service.proto:374Request message for DeleteDataset.
string name = 1
Required. Dataset resource name, format: projects/{project_id}/datasets/{dataset_id}
Imports data into dataset based on source locations defined in request. It can be called multiple times for the same dataset. Each dataset can only have one long running operation running on it. For example, no labeling task (also long running operation) can be started while importing is still ongoing. Vice versa.
message
data_labeling_service.proto:381Request message for ImportData API.
string name = 1
Required. Dataset resource name, format: projects/{project_id}/datasets/{dataset_id}
optional input_config = 2
Required. Specify the input source of the data.
string user_email_address = 3
Email of the user who started the import task and should be notified by email. If empty no notification will be sent.
Exports data and annotations from dataset.
message
data_labeling_service.proto:395Request message for ExportData API.
string name = 1
Required. Dataset resource name, format: projects/{project_id}/datasets/{dataset_id}
string annotated_dataset = 2
Required. Annotated dataset resource name. DataItem in Dataset and their annotations in specified annotated dataset will be exported. It's in format of projects/{project_id}/datasets/{dataset_id}/annotatedDatasets/ {annotated_dataset_id}
string filter = 3
Optional. Filter is not supported at this moment.
optional output_config = 4
Required. Specify the output destination.
string user_email_address = 5
Email of the user who started the export task and should be notified by email. If empty no notification will be sent.
Gets a data item in a dataset by resource name. This API can be called after data are imported into dataset.
message
data_labeling_service.proto:419Request message for GetDataItem.
string name = 1
Required. The name of the data item to get, format: projects/{project_id}/datasets/{dataset_id}/dataItems/{data_item_id}
Lists data items in a dataset. This API can be called after data are imported into dataset. Pagination is supported.
message
data_labeling_service.proto:426Request message for ListDataItems.
string parent = 1
Required. Name of the dataset to list data items, format: projects/{project_id}/datasets/{dataset_id}
string filter = 2
Optional. Filter is not supported at this moment.
int32 page_size = 3
Optional. Requested page size. Server may return fewer results than requested. Default value is 100.
string page_token = 4
Optional. A token identifying a page of results for the server to return. Typically obtained by [ListDataItemsResponse.next_page_token][google.cloud.datalabeling.v1beta1.ListDataItemsResponse.next_page_token] of the previous [DataLabelingService.ListDataItems] call. Return first page if empty.
message
data_labeling_service.proto:447Results of listing data items in a dataset.
repeated data_items = 1
The list of data items to return.
string next_page_token = 2
A token to retrieve next page of results.
Gets an annotated dataset by resource name.
message
data_labeling_service.proto:456Request message for GetAnnotatedDataset.
string name = 1
Required. Name of the annotated dataset to get, format: projects/{project_id}/datasets/{dataset_id}/annotatedDatasets/ {annotated_dataset_id}
Lists annotated datasets for a dataset. Pagination is supported.
message
data_labeling_service.proto:464Request message for ListAnnotatedDatasets.
string parent = 1
Required. Name of the dataset to list annotated datasets, format: projects/{project_id}/datasets/{dataset_id}
string filter = 2
Optional. Filter is not supported at this moment.
int32 page_size = 3
Optional. Requested page size. Server may return fewer results than requested. Default value is 100.
string page_token = 4
Optional. A token identifying a page of results for the server to return. Typically obtained by [ListAnnotatedDatasetsResponse.next_page_token][google.cloud.datalabeling.v1beta1.ListAnnotatedDatasetsResponse.next_page_token] of the previous [DataLabelingService.ListAnnotatedDatasets] call. Return first page if empty.
message
data_labeling_service.proto:485Results of listing annotated datasets for a dataset.
repeated annotated_datasets = 1
The list of annotated datasets to return.
string next_page_token = 2
A token to retrieve next page of results.
Deletes an annotated dataset by resource name.
message
data_labeling_service.proto:494Request message for DeleteAnnotatedDataset.
string name = 1
Required. Name of the annotated dataset to delete, format: projects/{project_id}/datasets/{dataset_id}/annotatedDatasets/ {annotated_dataset_id}
Starts a labeling task for image. The type of image labeling task is configured by feature in the request.
message
data_labeling_service.proto:502Request message for starting an image labeling task.
oneof request_config
Required. Config for labeling tasks. The type of request config must match the selected feature.
image_classification_config = 4
Configuration for image classification task. One of image_classification_config, bounding_poly_config, polyline_config and segmentation_config are required.
bounding_poly_config = 5
Configuration for bounding box and bounding poly task. One of image_classification_config, bounding_poly_config, polyline_config and segmentation_config are required.
polyline_config = 6
Configuration for polyline task. One of image_classification_config, bounding_poly_config, polyline_config and segmentation_config are required.
segmentation_config = 7
Configuration for segmentation task. One of image_classification_config, bounding_poly_config, polyline_config and segmentation_config are required.
string parent = 1
Required. Name of the dataset to request labeling task, format: projects/{project_id}/datasets/{dataset_id}
optional basic_config = 2
Required. Basic human annotation config.
LabelImageRequest.Feature feature = 3
Required. The type of image labeling task.
Starts a labeling task for video. The type of video labeling task is configured by feature in the request.
message
data_labeling_service.proto:566Request message for LabelVideo.
oneof request_config
Required. Config for labeling tasks. The type of request config must match the selected feature.
video_classification_config = 4
Configuration for video classification task. One of video_classification_config, object_detection_config, object_tracking_config and event_config is required.
object_detection_config = 5
Configuration for video object detection task. One of video_classification_config, object_detection_config, object_tracking_config and event_config is required.
object_tracking_config = 6
Configuration for video object tracking task. One of video_classification_config, object_detection_config, object_tracking_config and event_config is required.
event_config = 7
Configuration for video event task. One of video_classification_config, object_detection_config, object_tracking_config and event_config is required.
string parent = 1
Required. Name of the dataset to request labeling task, format: projects/{project_id}/datasets/{dataset_id}
optional basic_config = 2
Required. Basic human annotation config.
LabelVideoRequest.Feature feature = 3
Required. The type of video labeling task.
Starts a labeling task for text. The type of text labeling task is configured by feature in the request.
message
data_labeling_service.proto:620Request message for LabelText.
oneof request_config
Required. Config for labeling tasks. The type of request config must match the selected feature.
text_classification_config = 4
Configuration for text classification task. One of text_classification_config and text_entity_extraction_config is required.
text_entity_extraction_config = 5
Configuration for entity extraction task. One of text_classification_config and text_entity_extraction_config is required.
string parent = 1
Required. Name of the data set to request labeling task, format: projects/{project_id}/datasets/{dataset_id}
optional basic_config = 2
Required. Basic human annotation config.
LabelTextRequest.Feature feature = 6
Required. The type of text labeling task.
Gets an example by resource name, including both data and annotation.
message
data_labeling_service.proto:658Request message for GetExample
string name = 1
Required. Name of example, format: projects/{project_id}/datasets/{dataset_id}/annotatedDatasets/ {annotated_dataset_id}/examples/{example_id}
string filter = 2
Optional. An expression for filtering Examples. Filter by annotation_spec.display_name is supported. Format "annotation_spec.display_name = {display_name}"
Lists examples in an annotated dataset. Pagination is supported.
message
data_labeling_service.proto:671Request message for ListExamples.
string parent = 1
Required. Example resource parent.
string filter = 2
Optional. An expression for filtering Examples. For annotated datasets that have annotation spec set, filter by annotation_spec.display_name is supported. Format "annotation_spec.display_name = {display_name}"
int32 page_size = 3
Optional. Requested page size. Server may return fewer results than requested. Default value is 100.
string page_token = 4
Optional. A token identifying a page of results for the server to return. Typically obtained by [ListExamplesResponse.next_page_token][google.cloud.datalabeling.v1beta1.ListExamplesResponse.next_page_token] of the previous [DataLabelingService.ListExamples] call. Return first page if empty.
message
data_labeling_service.proto:694Results of listing Examples in and annotated dataset.
repeated examples = 1
The list of examples to return.
string next_page_token = 2
A token to retrieve next page of results.
Creates an annotation spec set by providing a set of labels.
message
data_labeling_service.proto:703Request message for CreateAnnotationSpecSet.
string parent = 1
Required. AnnotationSpecSet resource parent, format: projects/{project_id}
optional annotation_spec_set = 2
Required. Annotation spec set to create. Annotation specs must be included. Only one annotation spec will be accepted for annotation specs with same display_name.
Gets an annotation spec set by resource name.
message
data_labeling_service.proto:715Request message for GetAnnotationSpecSet.
string name = 1
Required. AnnotationSpecSet resource name, format: projects/{project_id}/annotationSpecSets/{annotation_spec_set_id}
Lists annotation spec sets for a project. Pagination is supported.
message
data_labeling_service.proto:722Request message for ListAnnotationSpecSets.
string parent = 1
Required. Parent of AnnotationSpecSet resource, format: projects/{project_id}
string filter = 2
Optional. Filter is not supported at this moment.
int32 page_size = 3
Optional. Requested page size. Server may return fewer results than requested. Default value is 100.
string page_token = 4
Optional. A token identifying a page of results for the server to return. Typically obtained by [ListAnnotationSpecSetsResponse.next_page_token][google.cloud.datalabeling.v1beta1.ListAnnotationSpecSetsResponse.next_page_token] of the previous [DataLabelingService.ListAnnotationSpecSets] call. Return first page if empty.
message
data_labeling_service.proto:743Results of listing annotation spec set under a project.
repeated annotation_spec_sets = 1
The list of annotation spec sets.
string next_page_token = 2
A token to retrieve next page of results.
Deletes an annotation spec set by resource name.
message
data_labeling_service.proto:752Request message for DeleteAnnotationSpecSet.
string name = 1
Required. AnnotationSpec resource name, format: `projects/{project_id}/annotationSpecSets/{annotation_spec_set_id}`.
Creates an instruction for how data should be labeled.
message
data_labeling_service.proto:759Request message for CreateInstruction.
string parent = 1
Required. Instruction resource parent, format: projects/{project_id}
optional instruction = 2
Required. Instruction of how to perform the labeling task.
Gets an instruction by resource name.
message
data_labeling_service.proto:769Request message for GetInstruction.
string name = 1
Required. Instruction resource name, format: projects/{project_id}/instructions/{instruction_id}
Lists instructions for a project. Pagination is supported.
message
data_labeling_service.proto:783Request message for ListInstructions.
string parent = 1
Required. Instruction resource parent, format: projects/{project_id}
string filter = 2
Optional. Filter is not supported at this moment.
int32 page_size = 3
Optional. Requested page size. Server may return fewer results than requested. Default value is 100.
string page_token = 4
Optional. A token identifying a page of results for the server to return. Typically obtained by [ListInstructionsResponse.next_page_token][google.cloud.datalabeling.v1beta1.ListInstructionsResponse.next_page_token] of the previous [DataLabelingService.ListInstructions] call. Return first page if empty.
message
data_labeling_service.proto:804Results of listing instructions under a project.
repeated instructions = 1
The list of Instructions to return.
string next_page_token = 2
A token to retrieve next page of results.
Deletes an instruction object by resource name.
message
data_labeling_service.proto:776Request message for DeleteInstruction.
string name = 1
Required. Instruction resource name, format: projects/{project_id}/instructions/{instruction_id}
Gets an evaluation by resource name.
message
data_labeling_service.proto:813Request message for GetEvaluation.
string name = 1
Required. Name of the evaluation. Format: 'projects/{project_id}/datasets/{dataset_id}/evaluations/{evaluation_id}'
Searchs evaluations within a project. Supported filter: evaluation_job, evaluation_time.
message
data_labeling_service.proto:820Request message for SearchEvaluation.
string parent = 1
Required. Evaluation search parent. Format: projects/{project_id}
string filter = 2
Optional. Support filtering by model id, job state, start and end time. Format: "evaluation_job.evaluation_job_id = {evaluation_job_id} AND evaluation_job.evaluation_job_run_time_start = {timestamp} AND evaluation_job.evaluation_job_run_time_end = {timestamp} AND annotation_spec.display_name = {display_name}"
int32 page_size = 3
Optional. Requested page size. Server may return fewer results than requested. Default value is 100.
string page_token = 4
Optional. A token identifying a page of results for the server to return. Typically obtained by [SearchEvaluationsResponse.next_page_token][google.cloud.datalabeling.v1beta1.SearchEvaluationsResponse.next_page_token] of the previous [DataLabelingService.SearchEvaluations] call. Return first page if empty.
message
data_labeling_service.proto:846Results of searching evaluations.
repeated evaluations = 1
The list of evaluations to return.
string next_page_token = 2
A token to retrieve next page of results.
Searchs example comparisons in evaluation, in format of examples of both ground truth and prediction(s). It is represented as a search with evaluation id.
message
data_labeling_service.proto:855Request message of SearchExampleComparisons.
string parent = 1
Required. Name of the Evaluation resource to search example comparison from. Format: projects/{project_id}/datasets/{dataset_id}/evaluations/{evaluation_id}
int32 page_size = 2
Optional. Requested page size. Server may return fewer results than requested. Default value is 100.
string page_token = 3
Optional. A token identifying a page of results for the server to return. Typically obtained by [SearchExampleComparisons.next_page_token][] of the previous [DataLabelingService.SearchExampleComparisons] call. Return first page if empty.
message
data_labeling_service.proto:874Results of searching example comparisons.
repeated example_comparisons = 1
string next_page_token = 2
A token to retrieve next page of results.
Creates an evaluation job.
message
data_labeling_service.proto:890Request message for CreateEvaluationJob.
string parent = 1
Required. Evaluation job resource parent, format: projects/{project_id}.
optional job = 2
Required. The evaluation job to create.
Updates an evaluation job.
message
data_labeling_service.proto:900Request message for UpdateEvaluationJob.
optional evaluation_job = 1
Required. Evaluation job that is going to be updated.
optional update_mask = 2
Optional. Mask for which field in evaluation_job should be updated.
Gets an evaluation job by resource name.
message
data_labeling_service.proto:909Request message for GetEvaluationJob.
string name = 1
Required. Name of the evaluation job. Format: 'projects/{project_id}/evaluationJobs/{evaluation_job_id}'
Pauses an evaluation job. Pausing a evaluation job that is already in PAUSED state will be a no-op.
message
data_labeling_service.proto:916Request message for PauseEvaluationJob.
string name = 1
Required. Name of the evaluation job that is going to be paused. Format: 'projects/{project_id}/evaluationJobs/{evaluation_job_id}'
Resumes a paused evaluation job. Deleted evaluation job can't be resumed. Resuming a running evaluation job will be a no-op.
message
data_labeling_service.proto:923Request message ResumeEvaluationJob.
string name = 1
Required. Name of the evaluation job that is going to be resumed. Format: 'projects/{project_id}/evaluationJobs/{evaluation_job_id}'
Stops and deletes an evaluation job.
message
data_labeling_service.proto:930Request message DeleteEvaluationJob.
string name = 1
Required. Name of the evaluation job that is going to be deleted. Format: 'projects/{project_id}/evaluationJobs/{evaluation_job_id}'
Lists all evaluation jobs within a project with possible filters. Pagination is supported.
message
data_labeling_service.proto:937Request message for ListEvaluationJobs.
string parent = 1
Required. Evaluation resource parent. Format: "projects/{project_id}"
string filter = 2
Optional. Only support filter by model id and job state. Format: "evaluation_job.model_id = {model_id} AND evaluation_job.state = {EvaluationJob::State}"
int32 page_size = 3
Optional. Requested page size. Server may return fewer results than requested. Default value is 100.
string page_token = 4
Optional. A token identifying a page of results for the server to return. Typically obtained by [ListEvaluationJobs.next_page_token][] of the previous [DataLabelingService.ListEvaluationJobs] call. Return first page if empty.
message
data_labeling_service.proto:960Results for listing evaluation jobs.
repeated evaluation_jobs = 1
The list of evaluation jobs to return.
string next_page_token = 2
A token to retrieve next page of results.
message
dataset.proto:187AnnotatedDataset is a set holding annotations for data in a Dataset. Each labeling task will generate an AnnotatedDataset under the Dataset that the task is requested for.
Used as response type in: DataLabelingService.GetAnnotatedDataset
Used as field type in:
string name = 1
Output only. AnnotatedDataset resource name in format of: projects/{project_id}/datasets/{dataset_id}/annotatedDatasets/ {annotated_dataset_id}
string display_name = 2
Output only. The display name of the AnnotatedDataset. It is specified in HumanAnnotationConfig when user starts a labeling task. Maximum of 64 characters.
string description = 9
Output only. The description of the AnnotatedDataset. It is specified in HumanAnnotationConfig when user starts a labeling task. Maximum of 10000 characters.
AnnotationSource annotation_source = 3
Output only. Source of the annotation.
AnnotationType annotation_type = 8
Output only. Type of the annotation. It is specified when starting labeling task.
int64 example_count = 4
Output only. Number of examples in the annotated dataset.
int64 completed_example_count = 5
Output only. Number of examples that have annotation in the annotated dataset.
optional label_stats = 6
Output only. Per label statistics.
optional create_time = 7
Output only. Time the AnnotatedDataset was created.
optional metadata = 10
Output only. Additional information about AnnotatedDataset.
repeated string blocking_resources = 11
Output only. The names of any related resources that are blocking changes to the annotated dataset.
message
dataset.proto:242Metadata on AnnotatedDataset.
Used in:
oneof annotation_request_config
Specific request configuration used when requesting the labeling task.
image_classification_config = 2
Configuration for image classification task.
bounding_poly_config = 3
Configuration for image bounding box and bounding poly task.
polyline_config = 4
Configuration for image polyline task.
segmentation_config = 5
Configuration for image segmentation task.
video_classification_config = 6
Configuration for video classification task.
object_detection_config = 7
Configuration for video object detection task.
object_tracking_config = 8
Configuration for video object tracking task.
event_config = 9
Configuration for video event labeling task.
text_classification_config = 10
Configuration for text classification task.
text_entity_extraction_config = 11
Configuration for text entity extraction task.
optional human_annotation_config = 1
HumanAnnotationConfig used when requesting the human labeling task for this AnnotatedDataset.
message
annotation.proto:40Annotation for Example. Each example may have one or more annotations. For example in image classification problem, each image might have one or more labels. We call labels binded with this image an Annotation.
Used in:
string name = 1
Output only. Unique name of this annotation, format is: projects/{project_id}/datasets/{dataset_id}/annotatedDatasets/{annotated_dataset}/examples/{example_id}/annotations/{annotation_id}
AnnotationSource annotation_source = 2
Output only. The source of the annotation.
optional annotation_value = 3
Output only. This is the actual annotation value, e.g classification, bounding box values are stored here.
optional annotation_metadata = 4
Output only. Annotation metadata, including information like votes for labels.
AnnotationSentiment annotation_sentiment = 6
Output only. Sentiment for this annotation.
message
annotation.proto:322Additional information associated with the annotation.
Used in:
optional operator_metadata = 2
Metadata related to human labeling.
Used in:
ANNOTATION_SENTIMENT_UNSPECIFIED = 0
NEGATIVE = 1
This annotation describes negatively about the data.
POSITIVE = 2
This label describes positively about the data.
Specifies where is the answer from.
Used in: ,
ANNOTATION_SOURCE_UNSPECIFIED = 0
OPERATOR = 3
Answer is provided by a human contributor.
message
annotation_spec_set.proto:51Container of information related to one annotation spec.
Used in: , , , , , , , , , , , ,
string display_name = 1
Required. The display name of the AnnotationSpec. Maximum of 64 characters.
string description = 2
Optional. User-provided description of the annotation specification. The description can be up to 10000 characters long.
message
annotation_spec_set.proto:29AnnotationSpecSet is a collection of label definitions. For example, in image classification tasks, we define a set of labels, this set is called AnnotationSpecSet. AnnotationSpecSet is immutable upon creation.
Used as response type in: DataLabelingService.CreateAnnotationSpecSet, DataLabelingService.GetAnnotationSpecSet
Used as field type in: ,
string name = 1
Output only. AnnotationSpecSet resource name, format: projects/{project_id}/annotationSpecSets/{annotation_spec_set_id}
string display_name = 2
Required. The display name for AnnotationSpecSet defined by user. Maximum of 64 characters.
string description = 3
Optional. User-provided description of the annotation specification set. The description can be up to 10000 characters long.
repeated annotation_specs = 4
Required. The actual spec set defined by the users.
repeated string blocking_resources = 5
Output only. The names of any related resources that are blocking changes to the annotation spec set.
Used in: , ,
ANNOTATION_TYPE_UNSPECIFIED = 0
IMAGE_CLASSIFICATION_ANNOTATION = 1
Classification annotations in an image.
IMAGE_BOUNDING_BOX_ANNOTATION = 2
Bounding box annotations in an image.
IMAGE_ORIENTED_BOUNDING_BOX_ANNOTATION = 13
Oriented bounding box. The box does not have to be parallel to horizontal line.
IMAGE_BOUNDING_POLY_ANNOTATION = 10
Bounding poly annotations in an image.
IMAGE_POLYLINE_ANNOTATION = 11
Polyline annotations in an image.
IMAGE_SEGMENTATION_ANNOTATION = 12
Segmentation annotations in an image.
VIDEO_SHOTS_CLASSIFICATION_ANNOTATION = 3
Classification annotations in video shots.
VIDEO_OBJECT_TRACKING_ANNOTATION = 4
Video object tracking annotation.
VIDEO_OBJECT_DETECTION_ANNOTATION = 5
Video object detection annotation.
VIDEO_EVENT_ANNOTATION = 6
Video event annotation.
TEXT_CLASSIFICATION_ANNOTATION = 8
Classification for text.
TEXT_ENTITY_EXTRACTION_ANNOTATION = 9
Entity extraction for text.
GENERAL_CLASSIFICATION_ANNOTATION = 14
General classification.
message
annotation.proto:116Annotation value for an example.
Used in:
oneof value_type
image_classification_annotation = 1
Annotation value for image classification case.
image_bounding_poly_annotation = 2
Annotation value for image bounding box, oriented bounding box and polygon cases.
image_polyline_annotation = 8
Annotation value for image polyline cases. Polyline here is different from BoundingPoly. It is formed by line segments connected to each other but not closed form(Bounding Poly). The line segments can cross each other.
image_segmentation_annotation = 9
Annotation value for image segmentation.
text_classification_annotation = 3
Annotation value for text classification case.
text_entity_extraction_annotation = 10
Annotation value for text entity extraction case.
video_classification_annotation = 4
Annotation value for video classification case.
video_object_tracking_annotation = 5
Annotation value for video object detection and tracking case.
video_event_annotation = 6
Annotation value for video event case.
message
evaluation_job.proto:147Records a failed attempt.
Used in:
optional attempt_time = 1
repeated partial_failures = 2
message
dataset.proto:127The BigQuery location for the input content.
Used in:
string input_uri = 1
Required. BigQuery URI to a table, up to 2000 characters long. Accepted forms: BigQuery gs path e.g. bq://projectId.bqDatasetId.bqTableId
message
evaluation.proto:67Options regarding evaluation between bounding boxes.
Used in:
float iou_threshold = 1
Minimize IoU required to consider 2 bounding boxes are matched.
message
annotation.proto:179A bounding polygon in the image.
Used in: ,
repeated vertices = 1
The bounding polygon vertices.
Config for image bounding poly (and bounding box) human labeling task.
Used in: , ,
string annotation_spec_set = 1
Required. Annotation spec set resource name.
string instruction_message = 2
Optional. Instruction message showed on contributors UI.
message
dataset.proto:111Metadata for classification annotations.
Used in:
bool is_multi_label = 1
Whether the classification task is multi-label or not.
message
evaluation.proto:81Used in:
optional pr_curve = 1
Precision-recall curve.
optional confusion_matrix = 2
message
evaluation.proto:148Confusion matrix of the model running the classification. Not applicable when label filtering is specified in evaluation option.
Used in:
repeated row = 1
message
evaluation.proto:149Used in:
optional annotation_spec = 1
The predicted annotation spec.
int32 item_count = 2
Number of items being predicted as this label.
message
evaluation.proto:158A row in the confusion matrix.
Used in:
optional annotation_spec = 1
the original annotation spec of this row.
repeated entries = 2
Info describing predicted label distribution.
message
operations.proto:221Metadata of a CreateInstruction operation.
string instruction = 1
The name of the created Instruction. projects/{project_id}/instructions/{instruction_id}
repeated partial_failures = 2
Partial failures encountered. E.g. single files that couldn't be read. Status details field will contain standard GCP error details.
optional create_time = 3
Timestamp when create instruction request was created.
message
instruction.proto:74Instruction from a CSV file.
Used in:
string gcs_file_uri = 1
CSV file for the instruction. Only gcs path is allowed.
message
dataset.proto:166DataItem is a piece of data, without annotation. For example, an image.
Used as response type in: DataLabelingService.GetDataItem
Used as field type in:
oneof payload
Output only.
image_payload = 2
The image payload, a container of the image bytes/uri.
text_payload = 3
The text payload, a container of text content.
video_payload = 4
The video payload, a container of the video uri.
string name = 1
Output only. Name of the data item, in format of: projects/{project_id}/datasets/{dataset_id}/dataItems/{data_item_id}
Used in: ,
DATA_TYPE_UNSPECIFIED = 0
IMAGE = 1
VIDEO = 2
TEXT = 4
GENERAL_DATA = 6
message
dataset.proto:44Dataset is the resource to hold your data. You can request multiple labeling tasks for a dataset while each one will generate an AnnotatedDataset.
Used as response type in: DataLabelingService.CreateDataset, DataLabelingService.GetDataset
Used as field type in: ,
string name = 1
Output only. Dataset resource name, format is: projects/{project_id}/datasets/{dataset_id}
string display_name = 2
Required. The display name of the dataset. Maximum of 64 characters.
string description = 3
Optional. User-provided description of the annotation specification set. The description can be up to 10000 characters long.
optional create_time = 4
Output only. Time the dataset is created.
repeated input_configs = 5
Output only. This is populated with the original input configs where ImportData is called. It is available only after the clients import data to this dataset.
repeated string blocking_resources = 6
Output only. The names of any related resources that are blocking changes to the dataset.
int64 data_item_count = 7
Output only. The number of data items in the dataset.
message
evaluation.proto:31Describes an evaluation between 2 annotated datasets. Created by an evaluation plan.
Used as response type in: DataLabelingService.GetEvaluation
Used as field type in:
string name = 1
Resource name of an evaluation. Format: 'projects/{project_id}/datasets/{dataset_id}/evaluations/{evaluation_id}'
optional config = 2
Options used in evaluation plan for creating the evaluation.
optional evaluation_job_run_time = 3
Output only. Timestamp when the evaluation plan triggered this evaluation flow.
optional create_time = 4
Output only. Timestamp when this model evaluation was created.
optional evaluation_metrics = 5
Output only. Metrics of the evaluation.
AnnotationType annotation_type = 6
Type of the annotation to compute metrics for in the groundtruth and annotation labeled dataset. Required for creation.
int64 evaluated_item_count = 7
Output only. Count of items in groundtruth dataset included in this evaluation. Will be unset if annotation type is not applicable.
message
evaluation.proto:59Used in: ,
oneof vertical_option
Vertical specific options for general metrics.
bounding_box_evaluation_options = 1
message
evaluation_job.proto:33Defines an evaluation job that is triggered periodically to generate evaluations.
Used as response type in: DataLabelingService.CreateEvaluationJob, DataLabelingService.GetEvaluationJob, DataLabelingService.UpdateEvaluationJob
Used as field type in: , ,
string name = 1
Format: 'projects/{project_id}/evaluationJobs/{evaluation_job_id}'
string description = 2
Description of the job. The description can be up to 25000 characters long.
EvaluationJob.State state = 3
string schedule = 4
Describes the schedule on which the job will be executed. Minimum schedule unit is 1 day. The schedule can be either of the following types: * [Crontab](http://en.wikipedia.org/wiki/Cron#Overview) * English-like [schedule](https: //cloud.google.com/scheduler/docs/configuring/cron-job-schedules)
string model_version = 5
The versioned model that is being evaluated here. Only one job is allowed for each model name. Format: 'projects/*/models/*/versions/*'
optional evaluation_job_config = 6
Detailed config for running this eval job.
string annotation_spec_set = 7
Name of the AnnotationSpecSet.
bool label_missing_ground_truth = 8
If a human annotation should be requested when some data don't have ground truth.
repeated attempts = 9
Output only. Any attempts with errors happening in evaluation job runs each time will be recorded here incrementally.
optional create_time = 10
Timestamp when this evaluation job was created.
State of the job.
Used in:
STATE_UNSPECIFIED = 0
SCHEDULED = 1
RUNNING = 2
PAUSED = 3
STOPPED = 4
message
evaluation_job.proto:137Used in:
string email = 1
Required. Email of the user who will be receiving the alert.
double min_acceptable_mean_average_precision = 2
If a single evaluation run's aggregate mean average precision is lower than this threshold, the alert will be triggered.
message
evaluation_job.proto:90Used in:
oneof human_annotation_request_config
config specific to different supported human annotation use cases.
image_classification_config = 4
bounding_poly_config = 5
video_classification_config = 6
object_detection_config = 7
text_classification_config = 8
object_tracking_config = 12
optional input_config = 1
Input config for data, gcs_source in the config will be the root path for data. Data should be organzied chronically under that path.
optional evaluation_config = 2
Config used to create evaluation.
optional human_annotation_config = 3
map<string, string> bigquery_import_keys = 9
Mappings between reserved keys for bigquery import and customized tensor names. Key is the reserved key, value is tensor name in the bigquery table. Different annotation type has different required key mapping. See user manual for more details: https: //docs.google.com/document/d/1bg1meMIBGY // 9I5QEoFoHSX6u9LsZQYBSmPt6E9SxqHZc/edit#heading=h.tfyjhxhvsqem
int32 example_count = 10
Max number of examples to collect in each period.
double example_sample_percentage = 11
Percentage of examples to collect in each period. 0.1 means 10% of total examples will be collected, and 0.0 means no collection.
optional evaluation_job_alert_config = 13
Alert config for the evaluation job. The alert will be triggered when its criteria is met.
message
evaluation.proto:72Used in:
oneof metrics
Common metrics covering most genernal cases.
classification_metrics = 1
object_detection_metrics = 2
Config for video event human labeling task.
Used in: ,
repeated string annotation_spec_sets = 1
Required. The list of annotation spec set resource name. Similar to video classification, we support selecting event from multiple AnnotationSpecSet at the same time.
message
dataset.proto:283An Example is a piece of data and its annotation. For example, an image with label "house".
Used as response type in: DataLabelingService.GetExample
Used as field type in: ,
oneof payload
Output only. The data part of Example.
image_payload = 2
The image payload, a container of the image bytes/uri.
text_payload = 6
The text payload, a container of the text content.
video_payload = 7
The video payload, a container of the video uri.
string name = 1
Output only. Name of the example, in format of: projects/{project_id}/datasets/{dataset_id}/annotatedDatasets/ {annotated_dataset_id}/examples/{example_id}
repeated annotations = 5
Output only. Annotations for the piece of data in Example. One piece of data can have multiple annotations.
message
operations.proto:77Metadata of an ExportData operation.
string dataset = 1
Output only. The name of dataset to be exported. "projects/*/datasets/*/Datasets/*"
repeated partial_failures = 2
Output only. Partial failures encountered. E.g. single files that couldn't be read. Status details field will contain standard GCP error details.
optional create_time = 3
Output only. Timestamp when export dataset request was created.
message
operations.proto:43Response used for ExportDataset longrunning operation.
string dataset = 1
Ouptut only. The name of dataset. "projects/*/datasets/*/Datasets/*"
int32 total_count = 2
Output only. Total number of examples requested to export
int32 export_count = 3
Output only. Number of examples exported successfully.
optional label_stats = 4
Output only. Statistic infos of labels in the exported dataset.
optional output_config = 5
Output only. output_config in the ExportData request.
message
dataset.proto:149Export destination of the data.Only gcs path is allowed in output_uri.
Used in:
string output_uri = 1
Required. The output uri of destination file.
string mime_type = 2
Required. The format of the gcs destination. Only "text/csv" and "application/json" are supported.
message
dataset.proto:160Export folder destination of the data.
Used in:
string output_folder_uri = 1
Required. Cloud Storage directory to export data to.
message
dataset.proto:117Source of the Cloud Storage file to be imported.
Used in:
string input_uri = 1
Required. The input URI of source file. This must be a Cloud Storage path (`gs://...`).
string mime_type = 2
Required. The format of the source file. Only "text/csv" is supported.
message
human_annotation_config.proto:41Configuration for how human labeling task should be done.
Used in: , , , , , , , , , , , , , , , ,
string instruction = 1
Required except for LabelAudio case. Instruction resource name.
string annotated_dataset_display_name = 2
Required. A human-readable name for AnnotatedDataset defined by users. Maximum of 64 characters .
string annotated_dataset_description = 3
Optional. A human-readable description for AnnotatedDataset. The description can be up to 10000 characters long.
string label_group = 4
Optional. A human-readable label used to logically group labeling tasks. This string must match the regular expression `[a-zA-Z\\d_-]{0,128}`.
string language_code = 5
Optional. The Language of this question, as a [BCP-47](https://www.rfc-editor.org/rfc/bcp/bcp47.txt). Default value is en-US. Only need to set this when task is language related. For example, French text classification or Chinese audio transcription.
int32 replica_count = 6
Optional. Replication of questions. Each question will be sent to up to this number of contributors to label. Aggregated answers will be returned. Default is set to 1. For image related labeling, valid values are 1, 3, 5.
optional question_duration = 7
Optional. Maximum duration for contributors to answer a question. Default is 1800 seconds.
repeated string contributor_emails = 9
Optional. If you want your own labeling contributors to manage and work on this labeling request, you can set these contributors here. We will give them access to the question types in crowdcompute. Note that these emails must be registered in crowdcompute worker UI: https://crowd-compute.appspot.com/
string user_email_address = 10
Email of the user who started the labeling task and should be notified by email. If empty no notification will be sent.
message
annotation.proto:192Image bounding poly annotation. It represents a polygon including bounding box in the image.
Used in:
oneof bounded_area
The region of the polygon. If it is a bounding box, it is guaranteed to be four points.
bounding_poly = 2
normalized_bounding_poly = 3
optional annotation_spec = 1
Label of object in this bounding polygon.
message
annotation.proto:152Image classification annotation definition.
Used in:
optional annotation_spec = 1
Label of image.
message
human_annotation_config.proto:88Config for image classification human labeling task.
Used in: , ,
string annotation_spec_set = 1
Required. Annotation spec set resource name.
bool allow_multi_label = 2
Optional. If allow_multi_label is true, contributors are able to choose multiple labels for one image.
StringAggregationType answer_aggregation_type = 3
Optional. The type of how to aggregate answers.
message
data_payloads.proto:29Container of information about an image.
Used in: ,
string mime_type = 1
Image format.
bytes image_thumbnail = 2
A byte string of a thumbnail image.
string image_uri = 3
Image uri from the user bucket.
string signed_uri = 4
Signed uri of the image file in the service bucket.
message
annotation.proto:218A polyline for the image annotation.
Used in:
oneof poly
polyline = 2
normalized_polyline = 3
optional annotation_spec = 1
Label of this polyline.
message
annotation.proto:230Image segmentation annotation.
Used in:
map<string, > annotation_colors = 1
The mapping between rgb color and annotation spec. The key is the rgb color represented in format of rgb(0, 0, 0). The value is the AnnotationSpec.
string mime_type = 2
Image format.
bytes image_bytes = 3
A byte string of a full image's color map.
message
operations.proto:62Metadata of an ImportData operation.
string dataset = 1
Ouptut only. The name of imported dataset. "projects/*/datasets/*"
repeated partial_failures = 2
Output only. Partial failures encountered. E.g. single files that couldn't be read. Status details field will contain standard GCP error details.
optional create_time = 3
Output only. Timestamp when import dataset request was created.
message
operations.proto:31Response used for ImportData longrunning operation.
string dataset = 1
Ouptut only. The name of imported dataset.
int32 total_count = 2
Output only. Total number of examples requested to import
int32 import_count = 3
Output only. Number of examples imported successfully.
message
dataset.proto:73The configuration of input data, including data type, location, etc.
Used in: , ,
oneof data_type_metadata
Optional. The metadata associated with each data type.
text_metadata = 6
Required for text import, as language code must be specified.
oneof source
Required. Where the data is from.
gcs_source = 2
Source located in Cloud Storage.
bigquery_source = 5
DataType data_type = 1
Required. Data type must be specifed when user tries to import data.
AnnotationType annotation_type = 3
Optional. If input contains annotation, user needs to specify the type and metadata of the annotation when creating it as an annotated dataset.
optional classification_metadata = 4
Optional. Metadata about annotations in the input. Each annotation type may have different metadata. Metadata for classification problem.
message
instruction.proto:34Instruction of how to perform the labeling task for human operators. Currently two types of instruction are supported - CSV file and PDF. One of the two types instruction must be provided. CSV file is only supported for image classification task. Instructions for other task should be provided as PDF. For image classification, CSV and PDF can be provided at the same time.
Used as response type in: DataLabelingService.GetInstruction
Used as field type in: ,
string name = 1
Output only. Instruction resource name, format: projects/{project_id}/instructions/{instruction_id}
string display_name = 2
Required. The display name of the instruction. Maximum of 64 characters.
string description = 3
Optional. User-provided description of the instruction. The description can be up to 10000 characters long.
optional create_time = 4
Output only. Creation time of instruction.
optional update_time = 5
Output only. Last update time of instruction.
DataType data_type = 6
Required. The data type of this instruction.
optional csv_instruction = 7
One of CSV or PDF instruction is required. Instruction from a CSV file, such as for classification task. The CSV file should have exact two columns, in the following format: * The first column is labeled data, such as an image reference, text. * The second column is comma separated labels associated with data.
optional pdf_instruction = 9
One of CSV or PDF instruction is required. Instruction from a PDF document. The PDF should be in a Cloud Storage bucket.
repeated string blocking_resources = 10
Output only. The names of any related resources that are blocking changes to the instruction.
message
operations.proto:155Details of a LabelImageBoundingBox operation metadata.
Used in:
optional basic_config = 1
Basic human annotation config used in labeling request.
message
operations.proto:167Details of LabelImageBoundingPoly operation metadata.
Used in:
optional basic_config = 1
Basic human annotation config used in labeling request.
message
operations.proto:149Metadata of a LabelImageClassification operation.
Used in:
optional basic_config = 1
Basic human annotation config used in labeling request.
message
operations.proto:161Details of a LabelImageOrientedBoundingBox operation metadata.
Used in:
optional basic_config = 1
Basic human annotation config.
message
operations.proto:173Details of LabelImagePolyline operation metadata.
Used in:
optional basic_config = 1
Basic human annotation config used in labeling request.
Image labeling task feature.
Used in:
FEATURE_UNSPECIFIED = 0
CLASSIFICATION = 1
Label whole image with one or more of labels.
BOUNDING_BOX = 2
Label image with bounding boxes for labels.
ORIENTED_BOUNDING_BOX = 6
Label oriented bounding box. The box does not have to be parallel to horizontal line.
BOUNDING_POLY = 3
Label images with bounding poly. A bounding poly is a plane figure that is bounded by a finite chain of straight line segments closing in a loop.
POLYLINE = 4
Label images with polyline. Polyline is formed by connected line segments which are not in closed form.
SEGMENTATION = 5
Label images with segmentation. Segmentation is different from bounding poly since it is more fine-grained, pixel level annotation.
message
operations.proto:179Details of a LabelImageSegmentation operation metadata.
Used in:
optional basic_config = 1
Basic human annotation config.
message
operations.proto:93Metadata of a labeling operation, such as LabelImage or LabelVideo. Next tag: 18
oneof details
Ouptut only. Details of specific label operation.
image_classification_details = 3
Details of label image classification operation.
image_bounding_box_details = 4
Details of label image bounding box operation.
image_bounding_poly_details = 11
Details of label image bounding poly operation.
image_oriented_bounding_box_details = 14
Details of label image oriented bounding box operation.
image_polyline_details = 12
Details of label image polyline operation.
image_segmentation_details = 15
Details of label image segmentation operation.
video_classification_details = 5
Details of label video classification operation.
video_object_detection_details = 6
Details of label video object detection operation.
video_object_tracking_details = 7
Details of label video object tracking operation.
video_event_details = 8
Details of label video event operation.
text_classification_details = 9
Details of label text classification operation.
text_entity_extraction_details = 13
Details of label text entity extraction operation.
int32 progress_percent = 1
Output only. Progress of label operation. Range: [0, 100].
repeated partial_failures = 2
Output only. Partial failures encountered. E.g. single files that couldn't be read. Status details field will contain standard GCP error details.
optional create_time = 16
Output only. Timestamp when labeling request was created.
message
dataset.proto:232Statistics about annotation specs.
Used in: ,
map<string, int64> example_count = 1
Map of each annotation spec's example count. Key is the annotation spec name and value is the number of examples for that annotation spec. If the annotated dataset does not have annotation spec, the map will return a pair where the key is empty string and value is the total number of annotations.
message
operations.proto:209Details of a LabelTextClassification operation metadata.
Used in:
optional basic_config = 1
Basic human annotation config used in labeling request.
message
operations.proto:215Details of a LabelTextEntityExtraction operation metadata.
Used in:
optional basic_config = 1
Basic human annotation config used in labeling request.
Text labeling task feature.
Used in:
FEATURE_UNSPECIFIED = 0
TEXT_CLASSIFICATION = 1
Label text content to one of more labels.
TEXT_ENTITY_EXTRACTION = 2
Label entities and their span in text.
message
operations.proto:185Details of a LabelVideoClassification operation metadata.
Used in:
optional basic_config = 1
Basic human annotation config used in labeling request.
message
operations.proto:203Details of a LabelVideoEvent operation metadata.
Used in:
optional basic_config = 1
Basic human annotation config used in labeling request.
message
operations.proto:191Details of a LabelVideoObjectDetection operation metadata.
Used in:
optional basic_config = 1
Basic human annotation config used in labeling request.
message
operations.proto:197Details of a LabelVideoObjectTracking operation metadata.
Used in:
optional basic_config = 1
Basic human annotation config used in labeling request.
Video labeling task feature.
Used in:
FEATURE_UNSPECIFIED = 0
CLASSIFICATION = 1
Label whole video or video segment with one or more labels.
OBJECT_DETECTION = 2
Label objects with bounding box on image frames extracted from the video.
OBJECT_TRACKING = 3
Label and track objects in video.
EVENT = 4
Label the range of video for the specified events.
message
annotation.proto:185Normalized bounding polygon.
Used in: ,
repeated normalized_vertices = 1
The bounding polygon normalized vertices.
message
annotation.proto:212Normalized polyline.
Used in:
repeated normalized_vertices = 1
The normalized polyline vertices.
message
annotation.proto:170A vertex represents a 2D point in the image. NOTE: the normalized vertex coordinates are relative to the original image and range from 0 to 1.
Used in: ,
float x = 1
X coordinate.
float y = 2
Y coordinate.
Config for video object detection human labeling task. Object detection will be conducted on the images extracted from the video, and those objects will be labeled with bounding boxes. User need to specify the number of images to be extracted per second as the extraction frame rate.
Used in: , ,
string annotation_spec_set = 1
Required. Annotation spec set resource name.
double extraction_frame_rate = 3
Required. Number of frames per second to be extracted from the video.
message
evaluation.proto:88Used in:
optional pr_curve = 1
Precision-recall curve.
Config for video object tracking human labeling task.
Used in: , ,
string annotation_spec_set = 1
Required. Annotation spec set resource name.
message
annotation.proto:288Video frame level annotation for object detection and tracking.
Used in:
oneof bounded_area
The bounding box location of this object track for the frame.
bounding_poly = 1
normalized_bounding_poly = 2
optional time_offset = 3
The time offset of this frame relative to the beginning of the video.
message
annotation.proto:328General information useful for labels coming from contributors.
Used in:
float score = 1
Confidence score corresponding to a label. For examle, if 3 contributors have answered the question and 2 of them agree on the final label, the confidence score will be 0.67 (2/3).
int32 total_votes = 2
The total number of contributors that answer this question.
int32 label_votes = 3
The total number of contributors that choose this label.
repeated string comments = 4
Comments from contributors.
message
dataset.proto:134The configuration of output data.
Used in: ,
oneof destination
Required. Location to output data to.
gcs_destination = 1
Output to a file in Cloud Storage. Should be used for labeling output other thanimage segmentation.
gcs_folder_destination = 2
Output to a folder in Cloud Storage. Should be used for image segmentation labeling output.
message
instruction.proto:80Instruction from a PDF file.
Used in:
string gcs_file_uri = 1
PDF file for the instruction. Only gcs path is allowed.
message
annotation.proto:206A line with multiple line segments.
Used in:
repeated vertices = 1
The polyline vertices.
Config for image polyline human labeling task.
Used in: ,
string annotation_spec_set = 1
Required. Annotation spec set resource name.
string instruction_message = 2
Optional. Instruction message showed on contributors UI.
message
evaluation.proto:93Used in: ,
optional annotation_spec = 1
PR curve against which annotation spec. Could be empty.
float area_under_curve = 2
Area under precision recall curve.
repeated confidence_metrics_entries = 3
entries to draw PR graph.
float mean_average_precision = 4
mean average prcision of this curve.
message
evaluation.proto:94Used in:
float confidence_threshold = 1
Threshold used for this entry, for example, IoU threshold for bounding box problem, or detection threshold for classification.
float recall = 2
Recall value.
float precision = 3
Precision value.
float f1_score = 4
Harmonic mean of recall and precision.
float recall_at1 = 5
Recall value for entries with label that has highest score.
float precision_at1 = 6
Precision value for entries with label that has highest score.
float f1_score_at1 = 7
The harmonic mean of [recall_at1][google.cloud.datalabeling.v1beta1.PrCurve.ConfidenceMetricsEntry.recall_at1] and [precision_at1][google.cloud.datalabeling.v1beta1.PrCurve.ConfidenceMetricsEntry.precision_at1].
float recall_at5 = 8
Recall value for entries with label that has highest 5 scores.
float precision_at5 = 9
Precision value for entries with label that has highest 5 scores.
float f1_score_at5 = 10
The harmonic mean of [recall_at5][google.cloud.datalabeling.v1beta1.PrCurve.ConfidenceMetricsEntry.recall_at5] and [precision_at5][google.cloud.datalabeling.v1beta1.PrCurve.ConfidenceMetricsEntry.precision_at5].
message
data_labeling_service.proto:877Example comparisons containing annotation comparison between groundtruth and predictions.
Used in:
optional ground_truth_example = 1
repeated model_created_examples = 2
Config for image segmentation
Used in: ,
string annotation_spec_set = 1
Required. Annotation spec set resource name. format: projects/{project_id}/annotationSpecSets/{annotation_spec_set_id}
string instruction_message = 2
Instruction message showed on labelers UI.
Config for setting up sentiments.
Used in:
bool enable_label_sentiment_selection = 1
If set to true, contributors will have the option to select sentiment of the label they selected, to mark it as negative or positive label. Default is false.
message
annotation.proto:259Start and end position in a sequence (e.g. text segment).
Used in:
int32 start = 1
Start position (inclusive).
int32 end = 2
End position (exclusive).
Used in:
STRING_AGGREGATION_TYPE_UNSPECIFIED = 0
MAJORITY_VOTE = 1
Majority vote to aggregate answers.
UNANIMOUS_VOTE = 2
Unanimous answers will be adopted.
NO_AGGREGATION = 3
Preserve all answers by crowd compute.
message
annotation.proto:244Text classification annotation.
Used in:
optional annotation_spec = 1
Label of the text.
Config for text classification human labeling task.
Used in: , ,
bool allow_multi_label = 1
Optional. If allow_multi_label is true, contributors are able to choose multiple labels for one text segment.
string annotation_spec_set = 2
Required. Annotation spec set resource name.
optional sentiment_config = 3
Optional. Configs for sentiment selection.
message
annotation.proto:250Text entity extraction annotation.
Used in:
optional annotation_spec = 1
Label of the text entities.
optional sequential_segment = 2
Position of the entity.
Config for text entity extraction human labeling task.
Used in: ,
string annotation_spec_set = 1
Required. Annotation spec set resource name.
message
dataset.proto:103Metadata for the text.
Used in:
string language_code = 1
The language of this text, as a [BCP-47](https://www.rfc-editor.org/rfc/bcp/bcp47.txt). Default value is en-US.
message
data_payloads.proto:44Container of information about a piece of text.
Used in: ,
string text_content = 1
Text content.
message
annotation.proto:268A time period inside of an example that has a time dimension (e.g. video).
Used in: , ,
optional start_time_offset = 1
Start of the time segment (inclusive), represented as the duration since the example start.
optional end_time_offset = 2
End of the time segment (exclusive), represented as the duration since the example start.
message
annotation.proto:159A vertex represents a 2D point in the image. NOTE: the vertex coordinates are in the same scale as the original image.
Used in: ,
int32 x = 1
X coordinate.
int32 y = 2
Y coordinate.
message
annotation.proto:279Video classification annotation.
Used in:
optional time_segment = 1
The time segment of the video to which the annotation applies.
optional annotation_spec = 2
Label of the segment specified by time_segment.
Config for video classification human labeling task. Currently two types of video classification are supported: 1. Assign labels on the entire video. 2. Split the video into multiple video clips based on camera shot, and assign labels on each video clip.
Used in: , ,
repeated annotation_spec_set_configs = 1
Required. The list of annotation spec set configs. Since watching a video clip takes much longer time than an image, we support label with multiple AnnotationSpecSet at the same time. Labels in each AnnotationSpecSet will be shown in a group to contributors. Contributors can select one or more (depending on whether to allow multi label) from each group.
bool apply_shot_detection = 2
Optional. Option to apply shot detection on the video.
Annotation spec set with the setting of allowing multi labels or not.
Used in:
string annotation_spec_set = 1
Required. Annotation spec set resource name.
bool allow_multi_label = 2
Optional. If allow_multi_label is true, contributors are able to choose multiple labels from one annotation spec set.
message
annotation.proto:313Video event annotation.
Used in:
optional annotation_spec = 1
Label of the event in this annotation.
optional time_segment = 2
The time segment of the video to which the annotation applies.
message
annotation.proto:301Video object tracking annotation.
Used in:
optional annotation_spec = 1
Label of the object tracked in this annotation.
optional time_segment = 2
The time segment of the video to which object tracking applies.
repeated object_tracking_frames = 3
The list of frames where this object track appears.
message
data_payloads.proto:60Container of information of a video.
Used in: ,
string mime_type = 1
Video format.
string video_uri = 2
Video uri from the user bucket.
repeated video_thumbnails = 3
The list of video thumbnails.
float frame_rate = 4
FPS of the video.
string signed_uri = 5
Signed uri of the video file in the service bucket.
message
data_payloads.proto:50Container of information of a video thumbnail.
Used in:
bytes thumbnail = 1
A byte string of the video frame.
optional time_offset = 2
Time offset relative to the beginning of the video, corresponding to the video frame where the thumbnail has been extracted from.