Contains annotation information that is relevant toAutoML.

Annotation details for translation.

Annotation details for image object detection.

Annotation details for video object tracking.

Annotation details for text sentiment.

Output only . The resource ID of the annotation spec that this annotation pertains to. The annotation spec comes from either an ancestor dataset, or the dataset that was used to train the model in use.

A definition of an annotation spec.

Required. The name of the annotation spec to show in the interface. The name can be up to 32 characters long and must match the regexp[a-zA-Z0-9_]+. (_), and ASCII digits 0-9.

API documentation forautoml_v1beta1.types.Any class.

The data statistics of a series of ARRAY values.

Input configuration for BatchPredict Action.

The format of input depends on the ML problem of the model used forprediction. As input source the[gcs_source][google.cloud.automl.v1beta1.InputConfig.gcs_source] isexpected, unless specified otherwise.

The formats are represented in EBNF with commas being literal and withnon-terminal symbols defined near the end of this comment. The formatsare:

• For Image Classification: CSV file(s) with each line having just asingle column: GCS_FILE_PATH which leads to image of up to 30MB insize. Supported extensions: .JPEG, .GIF, .PNG. This path is treatedas the ID in the Batch predict output. Three sample rows:gs://folder/image1.jpeg gs://folder/image2.gif gs://folder/image3.png

• For Image Object Detection: CSV file(s) with each line having just asingle column: GCS_FILE_PATH which leads to image of up to 30MB insize. Supported extensions: .JPEG, .GIF, .PNG. This path is treatedas the ID in the Batch predict output. Three sample rows:gs://folder/image1.jpeg gs://folder/image2.gif gs://folder/image3.png

• For Video Classification: CSV file(s) with each line in format:GCS_FILE_PATH,TIME_SEGMENT_START,TIME_SEGMENT_ENDGCS_FILE_PATH leads to video of up to 50GB in size and up to 3hduration. Supported extensions: .MOV, .MPEG4, .MP4, .AVI.TIME_SEGMENT_START and TIME_SEGMENT_END must be within the lengthof the video, and end has to be after the start. Three sample rows:gs://folder/video1.mp4,10,40 gs://folder/video1.mp4,20,60gs://folder/vid2.mov,0,inf

• For Video Object Tracking: CSV file(s) with each line in format:GCS_FILE_PATH,TIME_SEGMENT_START,TIME_SEGMENT_ENDGCS_FILE_PATH leads to video of up to 50GB in size and up to 3hduration. Supported extensions: .MOV, .MPEG4, .MP4, .AVI.TIME_SEGMENT_START and TIME_SEGMENT_END must be within the lengthof the video, and end has to be after the start. Three sample rows:gs://folder/video1.mp4,10,240 gs://folder/video1.mp4,300,360gs://folder/vid2.mov,0,inf

• For Text Classification: CSV file(s) with each line having just asingle column: GCS_FILE_PATH | TEXT_SNIPPET Any given text filecan have size upto 128kB. Any given text snippet content must have60,000 characters or less. Three sample rows: gs://folder/text1.txt"Some text content to predict" gs://folder/text3.pdf Supported fileextensions: .txt, .pdf

• For Text Sentiment: CSV file(s) with each line having just a singlecolumn: GCS_FILE_PATH | TEXT_SNIPPET Any given text file can havesize upto 128kB. Any given text snippet content must have 500characters or less. Three sample rows: gs://folder/text1.txt "Sometext content to predict" gs://folder/text3.pdf Supported fileextensions: .txt, .pdf

• For Text Extraction .JSONL (i.e. JSON Lines) file(s) which eitherprovide text in-line or as documents (for a single BatchPredict callonly one of the these formats may be used). The in-line .JSONLfile(s) contain per line a proto that wraps a temporary user-assignedTextSnippet ID (string up to 2000 characters long) called "id", aTextSnippet proto (in json representation) and zero or moreTextFeature protos. Any given text snippet content must have 30,000characters or less, and also be UTF-8 NFC encoded (ASCII already is).The IDs provided should be unique. The document .JSONL file(s)contain, per line, a proto that wraps a Document proto withinput_config set. Only PDF documents are supported now, and eachdocument must be up to 2MB large. Any given .JSONL file must be 100MBor smaller, and no more than 20 files may be given.

• For Tables: Either[gcs_source][google.cloud.automl.v1beta1.InputConfig.gcs_source] or

[bigquery_source][google.cloud.automl.v1beta1.InputConfig.bigquery_source].GCS case: CSV file(s), each by itself 10GB or smaller and total sizemust be 100GB or smaller, where first file must have a header containingcolumn names. If the first row of a subsequent file is the same as theheader, then it is also treated as a header. All other rows containvalues for the corresponding columns. The column names must contain themodel's

[input_feature_column_specs'][google.cloud.automl.v1beta1.TablesModelMetadata.input_feature_column_specs]

[display_name-s][google.cloud.automl.v1beta1.ColumnSpec.display_name](order doesn't matter). The columns corresponding to the model's inputfeature column specs must contain values compatible with the columnspec's data types. Prediction on all the rows, i.e. the CSV lines, willbe attempted. For FORECASTING

[prediction_type][google.cloud.automl.v1beta1.TablesModelMetadata.prediction_type]:all columns having

[TIME_SERIES_AVAILABLE_PAST_ONLY][google.cloud.automl.v1beta1.ColumnSpec.ForecastingMetadata.ColumnType]type will be ignored. First three sample rows of a CSV file: "FirstName","Last Name","Dob","Addresses"

"John","Doe","1968-01-22","[{"status":"current","address":"123_First_Avenue","city":"Seattle","state":"WA","zip":"11111","numberOfYears":"1"},{"status":"previous","address":"456_Main_Street","city":"Portland","state":"OR","zip":"22222","numberOfYears":"5"}]"

"Jane","Doe","1980-10-16","[{"status":"current","address":"789_Any_Avenue","city":"Albany","state":"NY","zip":"33333","numberOfYears":"2"},{"status":"previous","address":"321_Main_Street","city":"Hoboken","state":"NJ","zip":"44444","numberOfYears":"3"}]}BigQuery case: An URI of a BigQuery table. The user data size of theBigQuery table must be 100GB or smaller. The column names must containthe model's

[input_feature_column_specs'][google.cloud.automl.v1beta1.TablesModelMetadata.input_feature_column_specs]

[display_name-s][google.cloud.automl.v1beta1.ColumnSpec.display_name](order doesn't matter). The columns corresponding to the model's inputfeature column specs must contain values compatible with the columnspec's data types. Prediction on all the rows of the table will beattempted. For FORECASTING

[prediction_type][google.cloud.automl.v1beta1.TablesModelMetadata.prediction_type]:all columns having

[TIME_SERIES_AVAILABLE_PAST_ONLY][google.cloud.automl.v1beta1.ColumnSpec.ForecastingMetadata.ColumnType]type will be ignored.

Definitions: GCS_FILE_PATH = A path to file on GCS, e.g."gs://folder/video.avi". TEXT_SNIPPET = A content of a text snippet,UTF-8 encoded, enclosed within double quotes ("") TIME_SEGMENT_START =TIME_OFFSET Expresses a beginning, inclusive, of a time segment withinan example that has a time dimension (e.g. video). TIME_SEGMENT_END =TIME_OFFSET Expresses an end, exclusive, of a time segment within anexample that has a time dimension (e.g. video). TIME_OFFSET = A numberof seconds as measured from the start of an example (e.g. video).Fractions are allowed, up to a microsecond precision. "inf" is allowedand it means the end of the example.

Errors: If any of the provided CSV files can't be parsed or if more thancertain percent of CSV rows cannot be processed then the operation failsand prediction does not happen. Regardless of overall success or failurethe per-row failures, up to a certain count cap, will be listed inOperation.metadata.partial_failures.

The Google Cloud Storage location for the input content.

Details of BatchPredict operation.

Output only. Information further describing this batch predict's output.

Output configuration for BatchPredict Action.

As destination the

[gcs_destination][google.cloud.automl.v1beta1.BatchPredictOutputConfig.gcs_destination]must be set unless specified otherwise for a domain. If gcs_destinationis set then in the given directory a new directory is created. Its namewill be "prediction--", where timestamp is in YYYY-MM-DDThh:mm:ss.sssZISO-8601 format. The contents of it depends on the ML problem thepredictions are made for.

• For Image Classification: In the created directory filesimage_classification_1.jsonl,image_classification_2.jsonl,...,\image_classification_N.jsonlwill be created, where N may be 1, and depends on the total number ofthe successfully predicted images and annotations. A single imagewill be listed only once with all its annotations, and itsannotations will never be split across files. Each .JSONL file willcontain, per line, a JSON representation of a proto that wrapsimage's "ID" : "" followed by a list of zero or moreAnnotationPayload protos (called annotations), which haveclassification detail populated. If prediction for any image failed(partially or completely), then an additionalerrors_1.jsonl,errors_2.jsonl,...,errors_N.jsonl files will be created (Ndepends on total number of failed predictions). These files will havea JSON representation of a proto that wraps the same "ID" : "" buthere followed by exactly one

`google.rpc.Status <https:%20//github.com/googleapis/googleapis/blob/master/google/rpc/status.proto>__containing onlycodeandmessage`\ fields.

• For Image Object Detection: In the created directory filesimage_object_detection_1.jsonl,image_object_detection_2.jsonl,...,\image_object_detection_N.jsonlwill be created, where N may be 1, and depends on the total number ofthe successfully predicted images and annotations. Each .JSONL filewill contain, per line, a JSON representation of a proto that wrapsimage's "ID" : "" followed by a list of zero or moreAnnotationPayload protos (called annotations), which haveimage_object_detection detail populated. A single image will belisted only once with all its annotations, and its annotations willnever be split across files. If prediction for any image failed(partially or completely), then additionalerrors_1.jsonl,errors_2.jsonl,...,errors_N.jsonl files will be created (Ndepends on total number of failed predictions). These files will havea JSON representation of a proto that wraps the same "ID" : "" buthere followed by exactly one

`google.rpc.Status <https:%20//github.com/googleapis/googleapis/blob/master/google/rpc/status.proto>__containing onlycodeandmessage`\ fields. * For VideoClassification: In the created directory a video_classification.csvfile, and a .JSON file per each video classification requested in theinput (i.e. each line in given CSV(s)), will be created.

::

    The format of video_classification.csv is:

GCS_FILE_PATH,TIME_SEGMENT_START,TIME_SEGMENT_END,JSON_FILE_NAME,STATUSwhere: GCS_FILE_PATH,TIME_SEGMENT_START,TIME_SEGMENT_END = matches1 to 1 the prediction input lines (i.e. video_classification.csv hasprecisely the same number of lines as the prediction input had.)JSON_FILE_NAME = Name of .JSON file in the output directory, whichcontains prediction responses for the video time segment. STATUS = "OK"if prediction completed successfully, or an error code with messageotherwise. If STATUS is not "OK" then the .JSON file for that line maynot exist or be empty.

::

    Each .JSON file, assuming STATUS is "OK", will contain a list of    AnnotationPayload protos in JSON format, which are the predictions    for the video time segment the file is assigned to in the    video_classification.csv. All AnnotationPayload protos will have    video_classification field set, and will be sorted by    video_classification.type field (note that the returned types are    governed by `classifaction_types` parameter in    [PredictService.BatchPredictRequest.params][]).

• For Video Object Tracking: In the created directory avideo_object_tracking.csv file will be created, and multiple filesvideo_object_trackinng_1.json,video_object_trackinng_2.json,...,video_object_trackinng_N.json, where N is the number of requestsin the input (i.e. the number of lines in given CSV(s)).

  ::

  The format of video_object_tracking.csv is:

GCS_FILE_PATH,TIME_SEGMENT_START,TIME_SEGMENT_END,JSON_FILE_NAME,STATUSwhere: GCS_FILE_PATH,TIME_SEGMENT_START,TIME_SEGMENT_END = matches1 to 1 the prediction input lines (i.e. video_object_tracking.csv hasprecisely the same number of lines as the prediction input had.)JSON_FILE_NAME = Name of .JSON file in the output directory, whichcontains prediction responses for the video time segment. STATUS = "OK"if prediction completed successfully, or an error code with messageotherwise. If STATUS is not "OK" then the .JSON file for that line maynot exist or be empty.

::

    Each .JSON file, assuming STATUS is "OK", will contain a list of    AnnotationPayload protos in JSON format, which are the predictions    for each frame of the video time segment the file is assigned to in    video_object_tracking.csv. All AnnotationPayload protos will have    video_object_tracking field set.

• For Text Classification: In the created directory filestext_classification_1.jsonl,text_classification_2.jsonl,...,\text_classification_N.jsonlwill be created, where N may be 1, and depends on the total number ofinputs and annotations found.

  ::

  Each .JSONL file will contain, per line, a JSON representation of aproto that wraps input text snippet or input text file and a list ofzero or more AnnotationPayload protos (called annotations), whichhave classification detail populated. A single text snippet or filewill be listed only once with all its annotations, and itsannotations will never be split across files.If prediction for any text snippet or file failed (partially orcompletely), then additional `errors_1.jsonl`, `errors_2.jsonl`,...,`errors_N.jsonl` files will be created (N depends on total number offailed predictions). These files will have a JSON representation of aproto that wraps input text snippet or input text file followed byexactly one

`google.rpc.Status <https:%20//github.com/googleapis/googleapis/blob/master/google/rpc/status.proto>__containing onlycodeandmessage`.

• For Text Sentiment: In the created directory filestext_sentiment_1.jsonl,text_sentiment_2.jsonl,...,\text_sentiment_N.jsonl will becreated, where N may be 1, and depends on the total number of inputsand annotations found.

  ::

  Each .JSONL file will contain, per line, a JSON representation of aproto that wraps input text snippet or input text file and a list ofzero or more AnnotationPayload protos (called annotations), whichhave text_sentiment detail populated. A single text snippet or filewill be listed only once with all its annotations, and itsannotations will never be split across files.If prediction for any text snippet or file failed (partially orcompletely), then additional `errors_1.jsonl`, `errors_2.jsonl`,...,`errors_N.jsonl` files will be created (N depends on total number offailed predictions). These files will have a JSON representation of aproto that wraps input text snippet or input text file followed byexactly one

`google.rpc.Status <https:%20//github.com/googleapis/googleapis/blob/master/google/rpc/status.proto>__containing onlycodeandmessage`.

• For Text Extraction: In the created directory filestext_extraction_1.jsonl,text_extraction_2.jsonl,...,\text_extraction_N.jsonl will becreated, where N may be 1, and depends on the total number of inputsand annotations found. The contents of these .JSONL file(s) depend onwhether the input used inline text, or documents. If input wasinline, then each .JSONL file will contain, per line, a JSONrepresentation of a proto that wraps given in request text snippet's"id" (if specified), followed by input text snippet, and a list ofzero or more AnnotationPayload protos (called annotations), whichhave text_extraction detail populated. A single text snippet will belisted only once with all its annotations, and its annotations willnever be split across files. If input used documents, then each.JSONL file will contain, per line, a JSON representation of a protothat wraps given in request document proto, followed by its OCR-edrepresentation in the form of a text snippet, finally followed by alist of zero or more AnnotationPayload protos (called annotations),which have text_extraction detail populated and refer, via theirindices, to the OCR-ed text snippet. A single document (and its textsnippet) will be listed only once with all its annotations, and itsannotations will never be split across files. If prediction for anytext snippet failed (partially or completely), then additionalerrors_1.jsonl,errors_2.jsonl,...,errors_N.jsonl fileswill be created (N depends on total number of failed predictions).These files will have a JSON representation of a proto that wrapseither the "id" : "" (in case of inline) or the document proto (incase of document) but here followed by exactly one

`google.rpc.Status <https:%20//github.com/googleapis/googleapis/blob/master/google/rpc/status.proto>__containing onlycodeandmessage`.

• For Tables: Output depends on whether

[gcs_destination][google.cloud.automl.v1beta1.BatchPredictOutputConfig.gcs_destination]or

[bigquery_destination][google.cloud.automl.v1beta1.BatchPredictOutputConfig.bigquery_destination]is set (either is allowed). GCS case: In the created directory filestables_1.csv,tables_2.csv,...,tables_N.csv will becreated, where N may be 1, and depends on the total number of thesuccessfully predicted rows. For all CLASSIFICATION

[prediction_type-s][google.cloud.automl.v1beta1.TablesModelMetadata.prediction_type]:Each .csv file will contain a header, listing all columns'

[display_name-s][google.cloud.automl.v1beta1.ColumnSpec.display_name]given on input followed by M target column names in the format of

"<[target_column_specs][google.cloud.automl.v1beta1.TablesModelMetadata.target_column_spec]

[display_name][google.cloud.automl.v1beta1.ColumnSpec.display_name]>__score"where M is the number of distinct target values, i.e. number of distinctvalues in the target column of the table used to train the model.Subsequent lines will contain the respective values of successfullypredicted rows, with the last, i.e. the target, columns having thecorresponding predictionscores. ForREGRESSION and FORECASTING

[prediction_type-s][google.cloud.automl.v1beta1.TablesModelMetadata.prediction_type]:Each .csv file will contain a header, listing all columns'[display_name-s][google.cloud.automl.v1beta1.display_name] given oninput followed by the predicted target column with name in the format of

"predicted_<[target_column_specs][google.cloud.automl.v1beta1.TablesModelMetadata.target_column_spec]

[display_name][google.cloud.automl.v1beta1.ColumnSpec.display_name]>"Subsequent lines will contain the respective values of successfullypredicted rows, with the last, i.e. the target, column having thepredicted target value. If prediction for any rows failed, then anadditionalerrors_1.csv,errors_2.csv,...,errors_N.csv willbe created (N depends on total number of failed rows). These files willhave analogous format astables_*.csv, but always with a singletarget column having

`google.rpc.Status <https:%20//github.com/googleapis/googleapis/blob/master/google/rpc/status.proto>__represented as a JSON string, and containing onlycodeandmessage`. BigQuery case:

[bigquery_destination][google.cloud.automl.v1beta1.OutputConfig.bigquery_destination]pointing to a BigQuery project must be set. In the given project a newdataset will be created with nameprediction_<model-display-name>_<timestamp-of-prediction-call> wherewill be made BigQuery-dataset-name compatible (e.g. most specialcharacters will become underscores), and timestamp will be inYYYY_MM_DDThh_mm_ss_sssZ "based on ISO-8601" format. In the datasettwo tables will be created,predictions, anderrors. Thepredictions table's column names will be the input columns'

[display_name-s][google.cloud.automl.v1beta1.ColumnSpec.display_name]followed by the target column with name in the format of

"predicted_<[target_column_specs][google.cloud.automl.v1beta1.TablesModelMetadata.target_column_spec]

[display_name][google.cloud.automl.v1beta1.ColumnSpec.display_name]>"The input feature columns will contain the respective values ofsuccessfully predicted rows, with the target column having an ARRAY of

AnnotationPayloads,represented as STRUCT-s, containingTablesAnnotation. Theerrors table contains rows for which the prediction has failed, ithas analogous input columns while the target column name is in theformat of

"errors_<[target_column_specs][google.cloud.automl.v1beta1.TablesModelMetadata.target_column_spec]

[display_name][google.cloud.automl.v1beta1.ColumnSpec.display_name]>",and as a value has

`google.rpc.Status <https:%20//github.com/googleapis/googleapis/blob/master/google/rpc/status.proto>__represented as a STRUCT, and containing onlycodeandmessage`.

The Google Cloud Storage location of the directory where the output is to be written to.

Request message forPredictionService.BatchPredict.

Required. The input configuration for batch prediction.

Additional domain-specific parameters for the predictions, any string must be up to 25000 characters long. - For Text Classification:score_threshold - (float) A value from 0.0 to 1.0. When the model makes predictions for a text snippet, it will only produce results that have at least this confidence score. The default is 0.5. - For Image Classification:score_threshold - (float) A value from 0.0 to 1.0. When the model makes predictions for an image, it will only produce results that have at least this confidence score. The default is 0.5. - For Image Object Detection:score_threshold - (float) When Model detects objects on the image, it will only produce bounding boxes which have at least this confidence score. Value in 0 to 1 range, default is 0.5.max_bounding_box_count - (int64) No more than this number of bounding boxes will be produced per image. Default is 100, the requested value may be limited by server. - For Video Classification :score_threshold - (float) A value from 0.0 to 1.0. When the model makes predictions for a video, it will only produce results that have at least this confidence score. The default is 0.5.segment_classification - (boolean) Set to true to request segment-level classification. AutoML Video Intelligence returns labels and their confidence scores for the entire segment of the video that user specified in the request configuration. The default is "true".shot_classification - (boolean) Set to true to request shot-level classification. AutoML Video Intelligence determines the boundaries for each camera shot in the entire segment of the video that user specified in the request configuration. AutoML Video Intelligence then returns labels and their confidence scores for each detected shot, along with the start and end time of the shot. WARNING: Model evaluation is not done for this classification type, the quality of it depends on training data, but there are no metrics provided to describe that quality. The default is "false".1s_interval_classification - (boolean) Set to true to request classification for a video at one-second intervals. AutoML Video Intelligence returns labels and their confidence scores for each second of the entire segment of the video that user specified in the request configuration. WARNING: Model evaluation is not done for this classification type, the quality of it depends on training data, but there are no metrics provided to describe that quality. The default is "false". - For Video Object Tracking:score_threshold - (float) When Model detects objects on video frames, it will only produce bounding boxes which have at least this confidence score. Value in 0 to 1 range, default is 0.5.max_bounding_box_count - (int64) No more than this number of bounding boxes will be returned per frame. Default is 100, the requested value may be limited by server.min_bounding_box_size - (float) Only bounding boxes with shortest edge at least that long as a relative value of video frame size will be returned. Value in 0 to 1 range. Default is 0.

Result of the Batch Predict. This message is returned inresponse][google.longrunning.Operation.response] of the operationreturned by thePredictionService.BatchPredict.

The BigQuery location for the output content.

The BigQuery location for the input content.

Bounding box matching model metrics for a singleintersection-over-union threshold and multiple label match confidencethresholds.

Output only. The mean average precision, most often close to au_prc.

A bounding polygon of a detected object on a plane. Onoutput both vertices and normalized_vertices are provided. The polygonis formed by connecting vertices in the order they are listed.

API documentation forautoml_v1beta1.types.CancelOperationRequest class.

The data statistics of a series of CATEGORY values.

Contains annotation details specific to classification.

Model evaluation metrics for classification problems.Note: For Video Classification this metrics only describe quality of theVideo Classification predictions of "segment_classification" type.

Output only. The Area Under Precision-Recall Curve metric based on priors. Micro-averaged for the overall evaluation. Deprecated.

Output only. The Log Loss metric.

Output only. Confusion matrix of the evaluation. Only set for MULTICLASS classification problems where number of labels is no more than 10. Only set for model level evaluation, not for evaluation per label.

A representation of a column in a relational table. Whenlisting them, column specs are returned in the same order in which theywere given on import . Used by: * Tables

The data type of elements stored in the column.

Output only. Stats of the series of values in the column. This field may be stale, see the ancestor's Dataset.tables_dataset_metadata.stats_update_time field for the timestamp at which these stats were last updated.

Used to perform consistent read-modify-write updates. If not set, a blind "overwrite" update happens.

A correlation statistics between two series of DataTypevalues. The series may have differing DataType-s, but within a singleseries the DataType must be the same.

Request message forAutoMl.CreateDataset.

The dataset to create.

Details of CreateModel operation.

Request message forAutoMl.CreateModel.

The model to create.

The data statistics of a series of values that share thesame DataType.

The statistics for FLOAT64 DataType.

The statistics for TIMESTAMP DataType.

The statistics for STRUCT DataType.

The number of distinct values.

The number of values that are valid.

Indicated the type of data that can be stored in astructured data entity (e.g. a table).

If [type_code][google.cloud.automl.v1beta1.DataType.type_code] ==ARRAY, thenlist_element_type is the type of the elements.

If [type_code][google.cloud.automl.v1beta1.DataType.type_code] ==TIMESTAMP thentime_format provides the format in which that time field is expressed. The time_format must either be one of: *UNIX_SECONDS *UNIX_MILLISECONDS *UNIX_MICROSECONDS *UNIX_NANOSECONDS (for respectively number of seconds, milliseconds, microseconds and nanoseconds since start of the Unix epoch); or be written instrftime syntax. If time_format is not set, then the default format as described on the type_code is used.

If true, this DataType can also beNULL. In .CSV filesNULL value is expressed as an empty string.

A workspace for solving a single, particular machinelearning (ML) problem. A workspace contains examples that may beannotated.

Metadata for a dataset used for translation.

Metadata for a dataset used for text classification.

Metadata for a dataset used for video classification.

Metadata for a dataset used for text extraction.

Metadata for a dataset used for Tables.

Required. The name of the dataset to show in the interface. The name can be up to 32 characters long and can consist only of ASCII Latin letters A-Z and a-z, underscores (_), and ASCII digits 0-9.

Output only. The number of examples in the dataset.

Used to perform consistent read-modify-write updates. If not set, a blind "overwrite" update happens.

Request message forAutoMl.DeleteDataset.

Request message forAutoMl.DeleteModel.

Details of operations that perform deletes of anyentities.

API documentation forautoml_v1beta1.types.DeleteOperationRequest class.

Details of DeployModel operation.

Request message forAutoMl.DeployModel.

Model deployment metadata specific to Image Object Detection.

Resource name of the model to deploy.

A structured text document e.g. a PDF.

The plain text version of this document.

The dimensions of the page in the document.

Message that describes dimension of a document.

Width value of the document, works together with the unit.

Input configuration of aDocument.

A range between two double numbers.

End of the range, exclusive.

API documentation forautoml_v1beta1.types.Duration class.

Example data used for training or prediction.

Example image.

Example document.

Details of ExportData operation.

Request message forAutoMl.ExportData.

Required. The desired output location.

Details of EvaluatedExamples operation.

Output configuration for ExportEvaluatedExamples Action.Note that this call is available only for 30 days since the moment themodel was evaluated. The output depends on the domain, as follows (notethat only examples from the TEST set are exported):

• For Tables:

[bigquery_destination][google.cloud.automl.v1beta1.OutputConfig.bigquery_destination]pointing to a BigQuery project must be set. In the given project a newdataset will be created with name

export_evaluated_examples_<model-display-name>_<timestamp-of-export-call>where will be made BigQuery-dataset-name compatible (e.g. most specialcharacters will become underscores), and timestamp will be inYYYY_MM_DDThh_mm_ss_sssZ "based on ISO-8601" format. In the datasetanevaluated_examples table will be created. It will have all thesame columns as the

[primary_table][google.cloud.automl.v1beta1.TablesDatasetMetadata.primary_table_spec_id]of the [dataset][google.cloud.automl.v1beta1.Model.dataset_id] fromwhich the model was created, as they were at the moment of model'sevaluation (this includes the target column with its ground truth),followed by a column called "predicted_". That last column will containthe model's prediction result for each respective row, given as ARRAY ofAnnotationPayloads,represented as STRUCT-s, containingTablesAnnotation.

The BigQuery location where the output is to be written to.

Request message forAutoMl.ExportEvaluatedExamples.

Required. The desired output location and configuration.

Details of ExportModel operation.

Request message forAutoMl.ExportModel.Models need to be enabled for exporting, otherwise an error code will bereturned.

Required. The desired output location and configuration.

API documentation forautoml_v1beta1.types.FieldMask class.

The data statistics of a series of FLOAT64 values.

The standard deviation of the series.

Histogram buckets of the data series. Sorted by the min value of the bucket, ascendingly, and the number of the buckets is dynamically generated. The buckets are non-overlapping and completely cover whole FLOAT64 range with min of first bucket being"-Infinity", and max of the last one being"Infinity".

The GCR location where the image must be pushed to.

The Google Cloud Storage location where the output is tobe written to.

The Google Cloud Storage location for the input content.

Request message forAutoMl.GetAnnotationSpec.

Request message forAutoMl.GetColumnSpec.

Mask specifying which fields to read.

Request message forAutoMl.GetDataset.

Request message forAutoMl.GetModelEvaluation.

Request message forAutoMl.GetModel.

API documentation forautoml_v1beta1.types.GetOperationRequest class.

Request message forAutoMl.GetTableSpec.

Mask specifying which fields to read.

A representation of an image. Only images up to 30MB insize are supported.

Image content represented as a stream of bytes. Note: As with allbytes fields, protobuffers use a pure binary representation, whereas JSON representations use base64.

Output only. HTTP URI to the thumbnail image.

Dataset metadata that is specific to image classification.

Model deployment metadata specific to ImageClassification.

Model metadata for image classification.

Required. The train budget of creating this model, expressed in hours. The actualtrain_cost will be equal or less than this value.

Output only. The reason that this create model operation stopped, e.g.BUDGET_REACHED,MODEL_CONVERGED.

Output only. An approximate number of online prediction QPS that can be supported by this model per each node on which it is deployed.

Annotation details for image object detection.

Output only. The confidence that this annotation is positive for the parent example, value in [0, 1], higher means higher positivity confidence.

Dataset metadata specific to image object detection.

Model evaluation metrics for image object detectionproblems. Evaluates prediction quality of labeled bounding boxes.

Output only. The bounding boxes match metrics for each Intersection-over-union threshold 0.05,0.10,...,0.95,0.96,0.97,0.98,0.99 and each label confidence threshold 0.05,0.10,...,0.95,0.96,0.97,0.98,0.99 pair.

Model deployment metadata specific to Image ObjectDetection.

Model metadata specific to image object detection.

Output only. The number of nodes this model is deployed on. A node is an abstraction of a machine resource, which can handle online prediction QPS as given in the qps_per_node field.

Output only. The reason that this create model operation stopped, e.g.BUDGET_REACHED,MODEL_CONVERGED.

Output only. The actual train cost of creating this model, expressed in milli node hours, i.e. 1,000 value in this field means 1 node hour. Guaranteed to not exceed the train budget.

Details of ImportData operation.

Request message forAutoMl.ImportData.

Required. The desired input location and its domain specific semantics, if any.

Input configuration for ImportData Action.

The format of input depends on dataset_metadata the Dataset into whichthe import is happening has. As input source the[gcs_source][google.cloud.automl.v1beta1.InputConfig.gcs_source] isexpected, unless specified otherwise. Additionally any input .CSV fileby itself must be 100MB or smaller, unless specified otherwise. If an"example" file (that is, image, video etc.) with identical content (evenif it had different GCS_FILE_PATH) is mentioned multiple times, thenits label, bounding boxes etc. are appended. The same file should bealways provided with the same ML_USE and GCS_FILE_PATH, if it is not,then these values are nondeterministically selected from the given ones.

The formats are represented in EBNF with commas being literal and withnon-terminal symbols defined near the end of this comment. The formatsare:

• For Image Classification: CSV file(s) with each line in format:ML_USE,GCS_FILE_PATH,LABEL,LABEL,... GCS_FILE_PATH leads toimage of up to 30MB in size. Supported extensions: .JPEG, .GIF, .PNG,.WEBP, .BMP, .TIFF, .ICO For MULTICLASS classification type, at mostone LABEL is allowed per image. If an image has not yet been labeled,then it should be mentioned just once with no LABEL. Some samplerows: TRAIN,gs://folder/image1.jpg,daisyTEST,gs://folder/image2.jpg,dandelion,tulip,roseUNASSIGNED,gs://folder/image3.jpg,daisyUNASSIGNED,gs://folder/image4.jpg

• For Image Object Detection: CSV file(s) with each line in format:ML_USE,GCS_FILE_PATH,(LABEL,BOUNDING_BOX | ,,,,,,,)GCS_FILE_PATH leads to image of up to 30MB in size. Supportedextensions: .JPEG, .GIF, .PNG. Each image is assumed to beexhaustively labeled. The minimum allowed BOUNDING_BOX edge lengthis 0.01, and no more than 500 BOUNDING_BOX-es per image are allowed(one BOUNDING_BOX is defined per line). If an image has not yet beenlabeled, then it should be mentioned just once with no LABEL and the",,,,,,," in place of the BOUNDING_BOX. For images which are knownto not contain any bounding boxes, they should be labelled explictlyas "NEGATIVE_IMAGE", followed by ",,,,,,," in place of theBOUNDING_BOX. Sample rows:TRAIN,gs://folder/image1.png,car,0.1,0.1,,,0.3,0.3,,TRAIN,gs://folder/image1.png,bike,.7,.6,,,.8,.9,,UNASSIGNED,gs://folder/im2.png,car,0.1,0.1,0.2,0.1,0.2,0.3,0.1,0.3TEST,gs://folder/im3.png,,,,,,,,,TRAIN,gs://folder/im4.png,NEGATIVE_IMAGE,,,,,,,,,

• For Video Classification: CSV file(s) with each line in format:ML_USE,GCS_FILE_PATH where ML_USE VALIDATE value should not beused. The GCS_FILE_PATH should lead to another .csv file whichdescribes examples that have given ML_USE, using the following rowformat:GCS_FILE_PATH,(LABEL,TIME_SEGMENT_START,TIME_SEGMENT_END | ,,)Here GCS_FILE_PATH leads to a video of up to 50GB in size and up to3h duration. Supported extensions: .MOV, .MPEG4, .MP4, .AVI.TIME_SEGMENT_START and TIME_SEGMENT_END must be within the lengthof the video, and end has to be after the start. Any segment of avideo which has one or more labels on it, is considered a hardnegative for all other labels. Any segment with no labels on it isconsidered to be unknown. If a whole video is unknown, then it shuoldbe mentioned just once with ",," in place of LABEL,TIME_SEGMENT_START,TIME_SEGMENT_END. Sample top level CSV file:TRAIN,gs://folder/train_videos.csv TEST,gs://folder/test_videos.csvUNASSIGNED,gs://folder/other_videos.csv Sample rows of a CSV filefor a particular ML_USE: gs://folder/video1.avi,car,120,180.000021gs://folder/video1.avi,bike,150,180.000021gs://folder/vid2.avi,car,0,60.5 gs://folder/vid3.avi,,,

• For Video Object Tracking: CSV file(s) with each line in format:ML_USE,GCS_FILE_PATH where ML_USE VALIDATE value should not beused. The GCS_FILE_PATH should lead to another .csv file whichdescribes examples that have given ML_USE, using one of thefollowing row format:GCS_FILE_PATH,LABEL,[INSTANCE_ID],TIMESTAMP,BOUNDING_BOX orGCS_FILE_PATH,,,,,,,,,, Here GCS_FILE_PATH leads to a video of upto 50GB in size and up to 3h duration. Supported extensions: .MOV,.MPEG4, .MP4, .AVI. Providing INSTANCE_IDs can help to obtain abetter model. When a specific labeled entity leaves the video frame,and shows up afterwards it is not required, albeit preferable, thatthe same INSTANCE_ID is given to it. TIMESTAMP must be within thelength of the video, the BOUNDING_BOX is assumed to be drawn on theclosest video's frame to the TIMESTAMP. Any mentioned by theTIMESTAMP frame is expected to be exhaustively labeled and no morethan 500 BOUNDING_BOX-es per frame are allowed. If a whole video isunknown, then it should be mentioned just once with ",,,,,,,,,," inplace of LABEL, [INSTANCE_ID],TIMESTAMP,BOUNDING_BOX. Sample toplevel CSV file: TRAIN,gs://folder/train_videos.csvTEST,gs://folder/test_videos.csvUNASSIGNED,gs://folder/other_videos.csv Seven sample rows of a CSVfile for a particular ML_USE:gs://folder/video1.avi,car,1,12.10,0.8,0.8,0.9,0.8,0.9,0.9,0.8,0.9gs://folder/video1.avi,car,1,12.90,0.4,0.8,0.5,0.8,0.5,0.9,0.4,0.9gs://folder/video1.avi,car,2,12.10,.4,.2,.5,.2,.5,.3,.4,.3gs://folder/video1.avi,car,2,12.90,.8,.2,,,.9,.3,,gs://folder/video1.avi,bike,,12.50,.45,.45,,,.55,.55,,gs://folder/video2.avi,car,1,0,.1,.9,,,.9,.1,,gs://folder/video2.avi,,,,,,,,,,,

• For Text Extraction: CSV file(s) with each line in format:ML_USE,GCS_FILE_PATH GCS_FILE_PATH leads to a .JSONL (that is,JSON Lines) file which either imports text in-line or as documents.Any given .JSONL file must be 100MB or smaller. The in-line .JSONLfile contains, per line, a proto that wraps a TextSnippet proto (injson representation) followed by one or more AnnotationPayload protos(called annotations), which have display_name and text_extractiondetail populated. The given text is expected to be annotatedexhaustively, for example, if you look for animals and text contains"dolphin" that is not labeled, then "dolphin" is assumed to not be ananimal. Any given text snippet content must be 10KB or smaller, andalso be UTF-8 NFC encoded (ASCII already is). The document .JSONLfile contains, per line, a proto that wraps a Document proto. TheDocument proto must have either document_text or input_config set.In document_text case, the Document proto may also contain thespatial information of the document, including layout, documentdimension and page number. In input_config case, only PDF documentsare supported now, and each document may be up to 2MB large.Currently, annotations on documents cannot be specified at import.Three sample CSV rows: TRAIN,gs://folder/file1.jsonlVALIDATE,gs://folder/file2.jsonl TEST,gs://folder/file3.jsonl

• For Text Classification: CSV file(s) with each line in format:ML_USE,(TEXT_SNIPPET | GCS_FILE_PATH),LABEL,LABEL,...TEXT_SNIPPET and GCS_FILE_PATH are distinguished by a pattern. Ifthe column content is a valid gcs file path, i.e. prefixed by"gs://", it will be treated as a GCS_FILE_PATH, else if the contentis enclosed within double quotes (""), it is treated as aTEXT_SNIPPET. In the GCS_FILE_PATH case, the path must lead to a.txt file with UTF-8 encoding, for example,"gs://folder/content.txt", and the content in it is extracted as atext snippet. In TEXT_SNIPPET case, the column content excludingquotes is treated as to be imported text snippet. In both cases, thetext snippet/file size must be within 128kB. Maximum 100 uniquelabels are allowed per CSV row. Sample rows: TRAIN,"They have badfood and very rude",RudeService,BadFoodTRAIN,gs://folder/content.txt,SlowService TEST,"Typically always badservice there.",RudeService VALIDATE,"Stomach ache to go.",BadFood

• For Text Sentiment: CSV file(s) with each line in format:ML_USE,(TEXT_SNIPPET | GCS_FILE_PATH),SENTIMENT TEXT_SNIPPETand GCS_FILE_PATH are distinguished by a pattern. If the columncontent is a valid gcs file path, that is, prefixed by "gs://", it istreated as a GCS_FILE_PATH, otherwise it is treated as aTEXT_SNIPPET. In the GCS_FILE_PATH case, the path must lead to a.txt file with UTF-8 encoding, for example,"gs://folder/content.txt", and the content in it is extracted as atext snippet. In TEXT_SNIPPET case, the column content itself istreated as to be imported text snippet. In both cases, the textsnippet must be up to 500 characters long. Sample rows:TRAIN,"@freewrytin this is way too good for your product",2 TRAIN,"Ineed this product so bad",3 TEST,"Thank you for this product.",4VALIDATE,gs://folder/content.txt,2

• For Tables: Either[gcs_source][google.cloud.automl.v1beta1.InputConfig.gcs_source] or

[bigquery_source][google.cloud.automl.v1beta1.InputConfig.bigquery_source]can be used. All inputs is concatenated into a single

[primary_table][google.cloud.automl.v1beta1.TablesDatasetMetadata.primary_table_name]For gcs_source: CSV file(s), where the first row of the first file isthe header, containing unique column names. If the first row of asubsequent file is the same as the header, then it is also treated as aheader. All other rows contain values for the corresponding columns.Each .CSV file by itself must be 10GB or smaller, and their total sizemust be 100GB or smaller. First three sample rows of a CSV file:"Id","First Name","Last Name","Dob","Addresses"

"1","John","Doe","1968-01-22","[{"status":"current","address":"123_First_Avenue","city":"Seattle","state":"WA","zip":"11111","numberOfYears":"1"},{"status":"previous","address":"456_Main_Street","city":"Portland","state":"OR","zip":"22222","numberOfYears":"5"}]"

"2","Jane","Doe","1980-10-16","[{"status":"current","address":"789_Any_Avenue","city":"Albany","state":"NY","zip":"33333","numberOfYears":"2"},{"status":"previous","address":"321_Main_Street","city":"Hoboken","state":"NJ","zip":"44444","numberOfYears":"3"}]}For bigquery_source: An URI of a BigQuery table. The user data size ofthe BigQuery table must be 100GB or smaller. An imported table must havebetween 2 and 1,000 columns, inclusive, and between 1000 and 100,000,000rows, inclusive. There are at most 5 import data running in parallel.Definitions: ML_USE = "TRAIN" | "VALIDATE" | "TEST" | "UNASSIGNED"Describes how the given example (file) should be used for modeltraining. "UNASSIGNED" can be used when user has no preference.GCS_FILE_PATH = A path to file on GCS, e.g. "gs://folder/image1.png".LABEL = A display name of an object on an image, video etc., e.g. "dog".Must be up to 32 characters long and can consist only of ASCII Latinletters A-Z and a-z, underscores(_), and ASCII digits 0-9. For eachlabel an AnnotationSpec is created which display_name becomes thelabel; AnnotationSpecs are given back in predictions. INSTANCE_ID = Apositive integer that identifies a specific instance of a labeled entityon an example. Used e.g. to track two cars on a video while being ableto tell apart which one is which. BOUNDING_BOX =VERTEX,VERTEX,VERTEX,VERTEX | VERTEX,,,VERTEX,, A rectangle parallel tothe frame of the example (image, video). If 4 vertices are given theyare connected by edges in the order provided, if 2 are given they arerecognized as diagonally opposite vertices of the rectangle. VERTEX =COORDINATE,COORDINATE First coordinate is horizontal (x), the second isvertical (y). COORDINATE = A float in 0 to 1 range, relative to totallength of image or video in given dimension. For fractions the leadingnon-decimal 0 can be omitted (i.e. 0.3 = .3). Point 0,0 is in top left.TIME_SEGMENT_START = TIME_OFFSET Expresses a beginning, inclusive, ofa time segment within an example that has a time dimension (e.g. video).TIME_SEGMENT_END = TIME_OFFSET Expresses an end, exclusive, of a timesegment within an example that has a time dimension (e.g. video).TIME_OFFSET = A number of seconds as measured from the start of anexample (e.g. video). Fractions are allowed, up to a microsecondprecision. "inf" is allowed, and it means the end of the example.TEXT_SNIPPET = A content of a text snippet, UTF-8 encoded, enclosedwithin double quotes (""). SENTIMENT = An integer between 0 andDataset.text_sentiment_dataset_metadata.sentiment_max (inclusive).Describes the ordinal of the sentiment - higher value means a morepositive sentiment. All the values are completely relative, i.e. neither0 needs to mean a negative or neutral sentiment nor sentiment_max needsto mean a positive one - it is just required that 0 is the leastpositive sentiment in the data, and sentiment_max is the most positiveone. The SENTIMENT shouldn't be confused with "score" or "magnitude"from the previous Natural Language Sentiment Analysis API. All SENTIMENTvalues between 0 and sentiment_max must be represented in the importeddata. On prediction the same 0 to sentiment_max range will be used. Thedifference between neighboring sentiment values needs not to be uniform,e.g. 1 and 2 may be similar whereas the difference between 2 and 3 maybe huge.

Errors: If any of the provided CSV files can't be parsed or if more thancertain percent of CSV rows cannot be processed then the operation failsand nothing is imported. Regardless of overall success or failure theper-row failures, up to a certain count cap, is listed inOperation.metadata.partial_failures.

The Google Cloud Storage location for the input content. In ImportData, the gcs_source points to a csv with structure described in the comment.

Additional domain-specific parameters describing the semantic of the imported data, any string must be up to 25000 characters long. - For Tables:schema_inference_version

• (integer) Required. The version of the algorithm thatshould be used for the initial inference of the schema(columns' DataTypes) of the table the data is beingimported into. Allowed values: "1".

Request message forAutoMl.ListColumnSpecs.

Mask specifying which fields to read.

Requested page size. The server can return fewer results than requested. If unspecified, the server will pick a default size.

Response message forAutoMl.ListColumnSpecs.

A token to retrieve next page of results. Pass to [ListColumnS pecsRequest.page_token][google.cloud.automl.v1beta1.ListColum nSpecsRequest.page_token] to obtain that page.

Request message forAutoMl.ListDatasets.

An expression for filtering the results of the request. -dataset_metadata - for existence of the case (e.g. image_classification_dataset_metadata:*). Some examples of using the filter are: -translation_dataset_metadata:* --> The dataset has translation_dataset_metadata.

A token identifying a page of results for the server to return Typically obtained via [ListDatasetsResponse.next_page_token ][google.cloud.automl.v1beta1.ListDatasetsResponse.next_page _token] of the previous [AutoMl.ListDatasets][google.cloud.aut oml.v1beta1.AutoMl.ListDatasets] call.

Response message forAutoMl.ListDatasets.

A token to retrieve next page of results. Pass to [ListDataset sRequest.page_token][google.cloud.automl.v1beta1.ListDatasets Request.page_token] to obtain that page.

Request message forAutoMl.ListModelEvaluations.

An expression for filtering the results of the request. -annotation_spec_id - for =, != or existence. See example below for the last. Some examples of using the filter are:

• annotation_spec_id!=4 --> The model evaluation was donefor annotation spec with ID different than 4. -NOTannotation_spec_id:* --> The model evaluation was done foraggregate of all annotation specs.

  A token identifying a page of results for the server toreturn. Typically obtained via [ListModelEvaluationsResponse.next_page_token][google.cloud.automl.v1beta1.ListModelEvaluationsResponse.next_page_token] of the previous [AutoMl.ListModelEvaluations][google.cloud.automl.v1beta1.AutoMl.ListModelEvaluations] call.

Response message forAutoMl.ListModelEvaluations.

A token to retrieve next page of results. Pass to the [ListMod elEvaluationsRequest.page_token][google.cloud.automl.v1beta1. ListModelEvaluationsRequest.page_token] field of a new [AutoM l.ListModelEvaluations][google.cloud.automl.v1beta1.AutoMl.Lis tModelEvaluations] request to obtain that page.

Request message forAutoMl.ListModels.

An expression for filtering the results of the request. -model_metadata - for existence of the case (e.g. video_classification_model_metadata:*). -dataset_id

• for = or !=. Some examples of using the filter are: -image_classification_model_metadata:* --> The model hasimage_classification_model_metadata. -dataset_id=5--> The model was created from a dataset with ID 5.

  A token identifying a page of results for the server to returnTypically obtained via [ListModelsResponse.next_page_token][google.cloud.automl.v1beta1.ListModelsResponse.next_page_token] of the previous [AutoMl.ListModels][google.cloud.automl.v1beta1.AutoMl.ListModels] call.

Response message forAutoMl.ListModels.

A token to retrieve next page of results. Pass to [ListModelsR equest.page_token][google.cloud.automl.v1beta1.ListModelsRequ est.page_token] to obtain that page.

API documentation forautoml_v1beta1.types.ListOperationsRequest class.

API documentation forautoml_v1beta1.types.ListOperationsResponse class.

Request message forAutoMl.ListTableSpecs.

Mask specifying which fields to read.

Requested page size. The server can return fewer results than requested. If unspecified, the server will pick a default size.

Response message forAutoMl.ListTableSpecs.

A token to retrieve next page of results. Pass to [ListTableSp ecsRequest.page_token][google.cloud.automl.v1beta1.ListTableS pecsRequest.page_token] to obtain that page.

API documentation forautoml_v1beta1.types.ListValue class.

API proto representing a trained machine learning model.

Metadata for translation models.

Metadata for text classification models.

Metadata for video classification models.

Metadata for text extraction models.

Metadata for text sentiment models.

Required. The name of the model to show in the interface. The name can be up to 32 characters long and can consist only of ASCII Latin letters A-Z and a-z, underscores (_), and ASCII digits 0-9. It must start with a letter.

Output only. Timestamp when the model training finished and can be used for prediction.

Output only. Deployment state of the model. A model can only serve prediction requests after it gets deployed.

Evaluation results of a model.

Model evaluation metrics for image, text, video and tables classification. Tables problem is considered a classification when the target column is CATEGORY DataType.

Model evaluation metrics for translation.

Model evaluation metrics for video object tracking.

Evaluation metrics for text extraction models.

Output only. The ID of the annotation spec that the model evaluation applies to. The The ID is empty for the overall model evaluation. For Tables annotation specs in the dataset do not exist and this ID is always not set, but for CLASSIFICATION [prediction_type-s][google.cloud.automl.v1bet a1.TablesModelMetadata.prediction_type] the [display_name][g oogle.cloud.automl.v1beta1.ModelEvaluation.display_name] field is used.

Output only. Timestamp when this model evaluation was created.

Output configuration for ModelExport Action.

The Google Cloud Storage location where the model is to be written to. This location may only be set for the following model formats: "tflite", "edgetpu_tflite", "tf_saved_model", "tf_js", "core_ml". Under the directory given as the destination a new one with name "model-export--", where timestamp is in YYYY-MM-DDThh:mm:ss.sssZ ISO-8601 format, will be created. Inside the model and any of its supporting files will be written.

The format in which the model must be exported. The available, and default, formats depend on the problem and model type (if given problem and type combination doesn't have a format listed, it means its models are not exportable): - For Image Classification mobile-low-latency-1, mobile-versatile-1, mobile-high-accuracy-1: "tflite" (default), "edgetpu_tflite", "tf_saved_model", "tf_js", "docker". - For Image Classification mobile-core-ml-low-latency-1, mobile-core- ml-versatile-1, mobile-core-ml-high-accuracy-1: "core_ml" (default). Formats description: - tflite - Used for Android mobile devices. - edgetpu_tflite - Used forEdge TPU <https://cloud.google.com/edge-tpu/> devices. - tf_saved_model - A tensorflow model in SavedModel format. - tf_js - ATensorFlow.js <https://www.tensorflow.org/js> model that can be used in the browser and in Node.js using JavaScript. - docker - Used for Docker containers. Use the params field to customize the container. The container is verified to work correctly on ubuntu 16.04 operating system. See more at [containers quickstart](https: //cloud.google.com/vision/automl/docs/containers-gcs- quickstart) * core_ml - Used for iOS mobile devices.

Required. Horizontal coordinate.

API documentation forautoml_v1beta1.types.Operation class.

API documentation forautoml_v1beta1.types.OperationInfo class.

Metadata used across all long running operations returnedby AutoML API.

Details of a Delete operation.

Details of an UndeployModel operation.

Details of ImportData operation.

Details of ExportData operation.

Details of ExportEvaluatedExamples operation.

Output only. Partial failures encountered. E.g. single files that couldn't be read. This field should never exceed 20 entries. Status details field will contain standard GCP error details.

Output only. Time when the operation was updated for the last time.

* For Translation: CSV filetranslation.csv, witheach line in format: ML_USE,GCS_FILE_PATH GCS_FILE_PATH leads to a.TSV file which describes examples that have given ML_USE, using thefollowing row format per line: TEXT_SNIPPET (in source language)\tTEXT_SNIPPET (in target language)

• For Tables: Output depends on whether the dataset was imported fromGCS or BigQuery. GCS case:

[gcs_destination][google.cloud.automl.v1beta1.OutputConfig.gcs_destination]must be set. Exported are CSV file(s)tables_1.csv,tables_2.csv,...,\tables_N.csv with each having as header linethe table's column names, and all other lines contain values for theheader columns. BigQuery case:

[bigquery_destination][google.cloud.automl.v1beta1.OutputConfig.bigquery_destination]pointing to a BigQuery project must be set. In the given project a newdataset will be created with name

export_data_<automl-dataset-display-name>_<timestamp-of-export-call>where will be made BigQuery-dataset-name compatible (e.g. most specialcharacters will become underscores), and timestamp will be inYYYY_MM_DDThh_mm_ss_sssZ "based on ISO-8601" format. In thatdataset a new table calledprimary_table will be created, and filledwith precisely the same data as this obtained on import.

The Google Cloud Storage location where the output is to be written to. For Image Object Detection, Text Extraction, Video Classification and Tables, in the given directory a new directory will be created with name: export_data-- where timestamp is in YYYY-MM-DDThh:mm:ss.sssZ ISO-8601 format. All export output will be written into that directory.

Request message forPredictionService.Predict.

Required. Payload to perform a prediction on. The payload must match the problem type that the model was trained to solve.

Response message forPredictionService.Predict.

The preprocessed example that AutoML actually makes prediction on. Empty if AutoML does not preprocess the input example. * For Text Extraction: If the input is a .pdf file, the OCR'ed text will be provided in [document_text][google.cloud.automl. v1beta1.Document.document_text].

Metrics for regression problems.

Output only. Mean Absolute Error (MAE).

Output only. R squared.

A representation of a row in a relational table.

Required. The values of the row cells, given in the same order as the column_spec_ids, or, if not set, then in the same order as input feature [column_specs][google.cloud.automl.v1 beta1.TablesModelMetadata.input_feature_column_specs] of the Model this row is being passed to.

API documentation forautoml_v1beta1.types.Status class.

The data statistics of a series of STRING values.

API documentation forautoml_v1beta1.types.Struct class.

The data statistics of a series of STRUCT values.

\StructType defines the DataType-s of aSTRUCT type.

A specification of a relational table. The table's schemais represented via its child column specs. It is pre-populated as partof ImportData by schema inference algorithm, the version of which is arequired parameter of ImportData InputConfig. Note: While working with atable, at times the schema may be inconsistent with the data in thetable (e.g. string in a FLOAT64 column). The consistency validation isdone upon creation of a model. Used by: * Tables

column_spec_id of the time column. Only used if the parent dataset's ml_use_column_spec_id is not set. Used to split rows into TRAIN, VALIDATE and TEST sets such that oldest rows go to TRAIN set, newest to TEST, and those in between to VALIDATE. Required type: TIMESTAMP. If both this column and ml_use_column are not set, then ML use of all rows will be assigned by AutoML. NOTE: Updates of this field will instantly affect any other users concurrently working with the dataset.

Output only. The number of valid rows (i.e. without values that don't match DataType-s of their columns).

Output only. Input configs via which data currently residing in the table had been imported.

Contains annotation details specific to Tables.

Output only. Only populated when [target_column_spec][googl e.cloud.automl.v1beta1.TablesModelMetadata.target_column_spe c] has FLOAT64 data type. An interval in which the exactly correct target value has 95% chance to be in.

Output only. Auxiliary information for each of the model's [i nput_feature_column_specs][google.cloud.automl.v1beta1.Tabl esModelMetadata.input_feature_column_specs] with respect to this particular prediction. If no other fields than [column_ spec_name][google.cloud.automl.v1beta1.TablesModelColumnInfo. column_spec_name] and [column_display_name][google.cloud. automl.v1beta1.TablesModelColumnInfo.column_display_name] would be populated, then this whole field is not.

Metadata for a dataset used for AutoML Tables.

column_spec_id of the primary table's column that should be used as the training & prediction target. This column must be non-nullable and have one of following data types (otherwise model creation will error): - CATEGORY - FLOAT64 If the type is CATEGORY , only up to 100 unique values may exist in that column across all rows. NOTE: Updates of this field will instantly affect any other users concurrently working with the dataset.

column_spec_id of the primary table column which specifies a possible ML use of the row, i.e. the column will be used to split the rows into TRAIN, VALIDATE and TEST sets. Required type: STRING. This column, if set, must either have all ofTRAIN,VALIDATE,TEST among its values, or only haveTEST,UNASSIGNED values. In the latter case the rows withUNASSIGNED value will be assigned by AutoML. Note that if a given ml use distribution makes it impossible to create a "good" model, that call will error describing the issue. If both this column_spec_id and primary table's time_column_spec_id are not set, then all rows are treated asUNASSIGNED. NOTE: Updates of this field will instantly affect any other users concurrently working with the dataset.

Output only. The most recent timestamp when target_column_correlations field and all descendant ColumnSpec.data_stats and ColumnSpec.top_correlated_columns fields were last (re-)generated. Any changes that happened to the dataset afterwards are not reflected in these fields values. The regeneration happens in the background on a best effort basis.

An information specific to given column and Tables Model,in context of the Model and the predictions created by it.

Output only. The display name of the column (same as the display_name of its ColumnSpec).

Model metadata specific to AutoML Tables.

Required when optimization_objective is "MAXIMIZE_PRECISION_AT_RECALL". Must be between 0 and 1, inclusive.

Column spec of the dataset's primary table's column the model is predicting. Snapshotted when model creation started. Only 3 fields are used: name - May be set on CreateModel, if it's not then the ColumnSpec corresponding to the current target_column_spec_id of the dataset the model is trained from is used. If neither is set, CreateModel will error. display_name - Output only. data_type - Output only.

Objective function the model is optimizing towards. The training process creates a model that maximizes/minimizes the value of the objective function over the validation set. The supported optimization objectives depend on the prediction type. If the field is not set, a default objective function is used. CLASSIFICATION_BINARY: "MAXIMIZE_AU_ROC" (default) - Maximize the area under the receiver operating characteristic (ROC) curve. "MINIMIZE_LOG_LOSS" - Minimize log loss. "MAXIMIZE_AU_PRC" - Maximize the area under the precision- recall curve. "MAXIMIZE_PRECISION_AT_RECALL" - Maximize precision for a specified recall value. "MAXIMIZE_RECALL_AT_PRECISION" - Maximize recall for a specified precision value. CLASSIFICATION_MULTI_CLASS : "MINIMIZE_LOG_LOSS" (default) - Minimize log loss. REGRESSION: "MINIMIZE_RMSE" (default) - Minimize root-mean- squared error (RMSE). "MINIMIZE_MAE" - Minimize mean-absolute error (MAE). "MINIMIZE_RMSLE" - Minimize root-mean-squared log error (RMSLE).

Required. The train budget of creating this model, expressed in milli node hours i.e. 1,000 value in this field means 1 node hour. The training cost of the model will not exceed this budget. The final cost will be attempted to be close to the budget, though may end up being (even) noticeably smaller

• at the backend's discretion. This especially may happen whenfurther model training ceases to provide any improvements. Ifthe budget is set to a value known to be insufficient to traina model for the given dataset, the training won't be attemptedand will error. The train budget must be between 1,000 and72,000 milli node hours, inclusive.

  Use the entire training budget. This disables the earlystopping feature. By default, the early stopping feature isenabled, which means that AutoML Tables might stop trainingbefore the entire training budget has been used.

Dataset metadata for classification.

Model metadata that is specific to text classification.

Annotation for identifying spans of text.

An entity annotation will set this, which is the part of the original text to which the annotation pertains.

Dataset metadata that is specific to text extraction

Model evaluation metrics for text extraction problems.

Output only. Metrics that have confidence thresholds. Precision-recall curve can be derived from it.

Model metadata that is specific to text extraction.

A contiguous part of a text (string), assuming it has anUTF-8 NFC encoding.

Required. Zero-based character index of the first character of the text segment (counting characters from the beginning of the text).

Contains annotation details specific to text sentiment.

Dataset metadata for text sentiment.

Model evaluation metrics for text sentiment problems.

Output only. Recall.

Output only. Mean absolute error. Only set for the overall model evaluation, not for evaluation of a single annotation spec.

Output only. Linear weighted kappa. Only set for the overall model evaluation, not for evaluation of a single annotation spec.

Output only. Confusion matrix of the evaluation. Only set for the overall model evaluation, not for evaluation of a single annotation spec.

Model metadata that is specific to text sentiment.

A representation of a text snippet.

Optional. The format ofcontent. Currently the only two allowed values are "text/html" and "text/plain". If left blank, the format is automatically determined from the type of the uploadedcontent.

A time period inside of an example that has a timedimension (e.g. video).

End of the time segment (exclusive), represented as the duration since the example start.

API documentation forautoml_v1beta1.types.Timestamp class.

The data statistics of a series of TIMESTAMP values.

Annotation details specific to translation.

Dataset metadata that is specific to translation.

Required. The BCP-47 language code of the target language.

Evaluation metrics for the dataset.

Output only. BLEU score for base model.

Model metadata that is specific to translation.

Output only. Inferred from the dataset. The source languge (The BCP-47 language code) that is used for training.

Details of UndeployModel operation.

Request message forAutoMl.UndeployModel.

Request message forAutoMl.UpdateColumnSpec

The update mask applies to the resource.

Request message forAutoMl.UpdateDataset

The update mask applies to the resource.

Request message forAutoMl.UpdateTableSpec

The update mask applies to the resource.

API documentation forautoml_v1beta1.types.Value class.

Contains annotation details specific to videoclassification.

Output only . The classification details of this annotation.

Dataset metadata specific to video classification. AllVideo Classification datasets are treated as multi label.

Model metadata specific to video classification.

Annotation details for video object tracking.

Required. A time (frame) of a video to which this annotation pertains. Represented as the duration since the video's start.

Output only. The confidence that this annotation is positive for the video at the time_offset, value in [0, 1], higher means higher positivity confidence. For annotations created by the user the score is 1. When user approves an annotation, the original float score is kept (and not changed to 1).

Dataset metadata specific to video object tracking.

Model evaluation metrics for video object trackingproblems. Evaluates prediction quality of both labeled bounding boxesand labeled tracks (i.e. series of bounding boxes sharing same label andinstance ID).

Output only. The total number of bounding boxes (i.e. summed over all frames) the ground truth used to create this evaluation had.

Output only. The single metric for bounding boxes evaluation: the mean_average_precision averaged over all bounding_box_metrics_entries.

Model metadata specific to video object tracking.

API documentation forautoml_v1beta1.types.WaitOperationRequest class.