AlloyDB AI は、AlloyDB for PostgreSQL と AlloyDB Omni に含まれる機能スイートで、機械学習(ML)モデルのセマンティック機能と予測機能をデータに適用できます。このページでは、AlloyDB で利用可能な ML を活用した 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 2025-03-06 UTC。"],[[["AlloyDB AI enhances AlloyDB for PostgreSQL and AlloyDB Omni with machine learning capabilities, allowing users to leverage ML models for data analysis and predictions."],["AlloyDB offers two extensions for vector operations: a customized `vector` extension, which is an enhanced version of `pgvector`, and `alloydb_scann`, a highly efficient nearest-neighbor index powered by the ScaNN algorithm."],["The `google_ml_integration` extension provides SQL functions to interact with models, enabling users to generate embeddings from text and invoke model predictions within database transactions."],["AlloyDB can be configured to work with Vertex AI, providing access to models in the Vertex AI Model Garden and the ability to use `textembedding-gecko` English models for embedding generation."],["The `alloydb_scann` extension, which was previously known as `postgres_ann` requires users to drop indexes made with `postgres_ann`, upgrade to AlloyDB Omni version 15.5.5, and recreate them with `alloydb_scann` before use."]]],[]]