VideoContentSearchDescriptionSpanInfo

GoogleApi.ContentWarehouse.V1.Model.VideoContentSearchDescriptionSpanInfo

Table of Contents ▼

Jump to a specific part of the page:

Metadata about the span when the anchor source is description span.

This module has the following attributes (case-insensitive ascending order):

contextTokenCount (type: integer(), default: nil)
- The number of tokens in the context (sentence) where the description span is extracted from.
dolphinScores (type: GoogleApi.ContentWarehouse.V1.Model.VideoContentSearchSpanDolphinScores, default: nil)
- The inference result from the Dolphin span model if the anchor's source is description span.
spanDolphinScoreStats (type: GoogleApi.ContentWarehouse.V1.Model.VideoContentSearchMetricStats, default: nil)
- A summary of the Dolphin span token scores for the selected candidate. Currently, dolphin_scores in DescriptionSpanInfo would only contain a single span candidate so this field is essentially the copy of the score_stats for that span candidate.
spanTokenCount (type: integer(), default: nil)
- The number of tokens in the description span. The description span is formed from non-contiguous segment spans of a sentence (context), where each segment span's score satifsies the min span thresholds.
spanTokenCountRatio (type: number(), default: nil)
- The ratio of span_token_count / context_token_count.

@type t() :: %GoogleApi.ContentWarehouse.V1.Model.VideoContentSearchDescriptionSpanInfo{
contextTokenCount: integer() | nil,
dolphinScores: GoogleApi.ContentWarehouse.V1.Model.VideoContentSearchSpanDolphinScores.t() | nil,
spanDolphinScoreStats: GoogleApi.ContentWarehouse.V1.Model.VideoContentSearchMetricStats.t() | nil,
spanTokenCount: integer() | nil,
spanTokenCountRatio: number() | nil
}

@spec decode(struct(), keyword()) :: struct()