VideoContentSearchDolphinScoringConfig
VideoSearch InfrastructureGoogleApi.ContentWarehouse.V1.Model.VideoContentSearchDolphinScoringConfig
SEO Analysis
AI GeneratedRelated to video content processing and YouTube integration. Video content appears in universal search results, video carousels, and YouTube search. This model processes video signals that determine how video content is ranked and displayed in both web search and YouTube search results.
Actionable Insights for SEOs
- Understanding this model helps SEOs grasp Google's internal data architecture
- Consider how this system might interact with other ranking signals
Attributes
11descriptivenessOutputKeystringnilFull type: String.tThe output put keys for Dolphin PredictResponse
ensembleModelNamesstringnilFull type: list(String.tIf the dolphin model is an ensemble model (e.g. Video QnA model which consists of 4 teacher models), stores each individual model name.
inferenceBatchSizeinteger(nilThe inference batch size to use for inference methods that handle batching.
inferenceMethodstringnilFull type: String.tThe method to use for inference. This must be set or inference will fail.
maxRpcRetriesinteger(nilHolds value of flag --max_rpc_retries.
modelNamestringnilFull type: String.tModel name used for ModelSpec in PredictRequest used in the PredictionService API.
modelPathstringnilFull type: String.tOnly used when using the bulk_inference API. See go/dolphin-models to learn about the different dolphin models.
outputKeystringnilFull type: String.tTODO(alexiaxu) To deprecate this field in the future Output key for Dolphin PredictResponse.
rpcDeadlineSecondsfloat(nilHolds value of flag --rpc_deadline (converted to seconds).
serviceBnsstringnilFull type: String.tTensorflow inference BNS address when using PredictionService API.
usefulnessOutputKeystringnilFull type: String.t