AlloyDB AI は、AlloyDB for PostgreSQL と AlloyDB Omni に含まれる機能スイートであり、機械学習(ML)モデルのセマンティック機能と予測機能で AlloyDB のデータを利用できます。このページでは、AlloyDB で利用可能な ML AI 関数の概要について説明します。
[[["わかりやすい","easyToUnderstand","thumb-up"],["問題の解決に役立った","solvedMyProblem","thumb-up"],["その他","otherUp","thumb-up"]],[["わかりにくい","hardToUnderstand","thumb-down"],["情報またはサンプルコードが不正確","incorrectInformationOrSampleCode","thumb-down"],["必要な情報 / サンプルがない","missingTheInformationSamplesINeed","thumb-down"],["翻訳に関する問題","translationIssue","thumb-down"],["その他","otherDown","thumb-down"]],["最終更新日 2025-04-21 UTC。"],[[["AlloyDB AI integrates machine learning capabilities into AlloyDB for PostgreSQL and AlloyDB Omni, allowing users to apply ML models to their data for semantic and predictive insights."],["AlloyDB offers two extensions for managing vectors: a customized `vector` extension based on `pgvector`, and `alloydb_scann`, which utilizes the ScaNN algorithm for efficient nearest-neighbor indexing."],["Users can fine-tune vector query performance by adjusting indexes to optimize for query-per-second (QPS) and recall."],["AlloyDB AI extends SQL syntax with functions like `Invoke predictions` and `Generate embeddings`, enabling direct model querying and text-to-vector conversion within transactions."],["AlloyDB Omni can be configured to work with Vertex AI, allowing applications to invoke predictions using models in the Vertex AI Model Garden and generate embeddings using `text-embedding-005` English models."]]],[]]