Search Results
https://dev.mysql.com/doc/heatwave/en/mys-hw-debugging-queries.html
This topic describes how to debug queries that fail to offload to MySQL HeatWave for execution. To learn how to download the sample database, see AirportDB Analytics Quickstart. Setting Secondary Engine Definition By default, SELECT queries are ...
https://dev.mysql.com/doc/heatwave/en/mys-hw-genai-byo-embedding-tables.html
As of MySQL 9.2.1, MySQL HeatWave GenAI lets you use tables containing your own vector embedding to run retrieval-augemented generation (RAG) with vector search. The ML_RAG and ML_RAG_TABLE routines let you specify the table column names to use as ...
https://dev.mysql.com/doc/heatwave/en/mys-hw-genai-chat-details.html
Chat history: which includes both your queries and responses generated by MySQL HeatWave Chat. This topic describes how to view a chat session details. It contains the following sections: Before You Begin View Details What's Next Before You Begin ...
https://dev.mysql.com/doc/heatwave/en/mys-hw-genai-rag.html
Retrieve Context and Generate Content Using Hybrid Search As of MySQL 9.5.0, you can use both keyword-based search with vector-based semantic search, which helps improve the quality of search results and accuracy of the response generated for your ... The ML_RAG routine runs retrieval-augmented generation which aims to generate more accurate responses for your ...
https://dev.mysql.com/doc/heatwave/en/mys-hw-genai-supported-models.html
This topic provides the list of languages that MySQL HeatWave GenAI supports and the embedding models as well as large language models (LLMs) that are available in MySQL HeatWave. This topic contains the following sections: Before You Begin View ...
https://dev.mysql.com/doc/heatwave/en/mys-hw-genai-vector-search.html
As of MySQL 9.2.1, you can use both inbuilt vector store tables and tables containing your own vector embeddings for running RAG with vector search. When you enter a query, MySQL HeatWave GenAI performs a vector-based similarity search to retrieve ...
https://dev.mysql.com/doc/heatwave/en/mys-hw-genai-vector-store-load.html
The specified LLM must be capable of processing both textual and visual inputs. This section describes how to generate vector embeddings for files or folders stored in Object Storage, and load the embeddings into a vector store table. If not ...
https://dev.mysql.com/doc/heatwave/en/mys-hw-lakehouse-auto-parallel-load-overview.html
If you use an ambiguous format to specify temporal data types (for example, date_format as %m:%d:%y and time_format as %h:%i:%s), and the data matches both formats, the following order of precedence is used for inferring the data type: DATE > TIME > ... Lakehouse Auto Parallel Load, which extends the Auto Parallel Load feature of MySQL HeatWave, facilitates the process of loading data from Object Storage into MySQL HeatWave by automating many of the steps involved, including: Excluding schemas, tables, and columns that Auto Parallel Load cannot ...
https://dev.mysql.com/doc/heatwave/en/mys-hw-lakehouse-error-handling.html
The console Total Warnings will include both the displayed and the filtered warnings count. See MySQL 9.5 Error Message Reference for Lakehouse error messages. As of MySQL 9.2.2, a detailed summary of warnings is available when loading data to ...
https://dev.mysql.com/doc/heatwave/en/mys-hw-lakehouse-export-query-results-csv-non-json.html
Therefore, instead of having queries compute results by scanning all the rows for a given range of dates, you can run queries against this aggregate table, saving both time and cost. This topic shows the different ways to export query results as a ...