Search Results
https://dev.mysql.com/doc/heatwave/en/mys-hw-string-column-encoding-reference.html
It applies to string columns by default when tables are loaded with the exception of string columns defined explicitly as dictionary-encoded columns. Dictionary encoding: This is suited for string columns with a low number of distinct values ...
https://dev.mysql.com/doc/heatwave/en/mys-hw-uniform-resource-identifiers.html
What's Next Learn how to do the following: Create External Tables Using Auto Parallel Load. As of MySQL 9.3.1, (supported in MySQL HeatWave on OCI only) MySQL HeatWave Lakehouse lets you use Uniform Resource Identifiers (URI) in a unified way to ...
https://dev.mysql.com/doc/heatwave/en/mys-hw-workload-optimization-olap.html
Unload Advisor Recommends tables to unload, reducing MySQL HeatWave memory usage. The recommendations are based upon when the tables were last queried. Workload optimization for online analytical processing, OLAP, includes using dictionary encoding ...
https://dev.mysql.com/doc/heatwave/en/mys-hwaml-automl-workflow.html
The data can originate from either DB System tables or external Lakehouse tables. A typical MySQL HeatWave AutoML workflow is described below: When you run the ML_TRAIN routine, MySQL HeatWave AutoML retrieves the data to use for training. The ...
https://dev.mysql.com/doc/heatwave/en/mys-hwaml-classification-train.html
Train Model Train the model with the ML_TRAIN routine and use the training_data table previously created. mysql> CALL sys.ML_TRAIN('table_name', 'target_column_name', JSON_OBJECT('task', 'task_name'), model_handle); Replace table_name, ... After ...
https://dev.mysql.com/doc/heatwave/en/mys-hwaml-model-quality.html
After training and loading the model, prepare a table of labeled data to score that has a different set of data from the trained model. mysql> CALL sys.ML_SCORE(table_name, target_column_name, model_handle, metric, score, [options]); The following ... ML_SCORE scores a model by generating predictions using the feature columns in a labeled dataset as input and comparing the predictions to ground truth values in the target column of the labeled ...
https://dev.mysql.com/doc/heatwave/en/mys-hwaml-regression-train.html
Train Model Train the model with the ML_TRAIN routine and use the house_price_training table previously created. mysql> CALL sys.ML_TRAIN('table_name', 'target_column_name', JSON_OBJECT('task', 'task_name'), @variable); Replace table_name, ... After ...
https://dev.mysql.com/doc/heatwave/en/mys-hwaml-topic-modeling-train.html
Therefore, you cannot use the following options for topic modeling: model_list optimization_metric exclude_model_list exclude_column_list include_column_list Unsupported Routines You cannot run the following routines for topic modeling: ML_EXPLAIN ... After preparing the data for topic modeling, you can train the ...
https://dev.mysql.com/doc/heatwave/en/mys-hwgenai-ml-generate.html
See Also Section 7.6, “Generate Text-Based Content” Section 10.3.2, “ML_GENERATE_TABLE” . The ML_GENERATE routine uses the specified large language model (LLM) to generate text-based content as a response for the given natural-language ...
https://dev.mysql.com/doc/relnotes/heatwave/en/news-8-0-26-u2.html
The number of megabytes scanned by an individual MySQL HeatWave query can be obtained by querying the performance_schema.rpd_query_stats table. An estimated number of megabytes scanned by an individual query can be obtained by running the query with ... Functionality Added or Changed The following function support was added: YEARWEEK(date), YEARWEEK(date,mode) The mode argument for the two-argument form of the WEEK() function: WEEK(date[,mode]) MAKEDATE() “Zero” handling for dates such as '2001-11-00' was implemented for WEEK(), YEARWEEK(), and MAKEDATE() ...