teradata-onnx-model-scoring

name	Teradata ONNX Model Scoring
description	Score data using ONNX models (Bring Your Own Model)

Teradata ONNX Model Scoring

Property	Value
Skill Name	Teradata ONNX Model Scoring
Description	Score data using ONNX models (Bring Your Own Model)
Category	BYOM Scoring
Primary Function	ONNXPredict
Framework	BYOM

Core Capabilities

• Automated table structure analysis via DBC.ColumnsV
• Dynamic SQL generation for ONNXPredict
• Complete workflow from data preparation to results interpretation
• Data quality validation and preprocessing guidance
• Parameter optimization and tuning

Key Parameters

• InputTable: Data to score
• ModelTable: ONNX model stored as BLOB (model_id, model columns)
• Accumulate: Columns to pass through
• Overwrite: 'true'/'false' for output column handling

Use Cases

1. Score data with externally trained ONNX models
2. Deploy PyTorch/TensorFlow models in-database
3. Real-time scoring at scale
4. Model portability across platforms

Example Usage

-- ONNXPredict execution
SELECT * FROM ONNXPredict (
    ON {USER_DATABASE}.{USER_TABLE} AS InputTable
    ON (SELECT model_id, model FROM {USER_DATABASE}.{MODEL_TABLE}
        WHERE model_id = '{MODEL_NAME}') AS ModelTable DIMENSION
    USING
    Accumulate ('{ID_COLUMN}', '{OTHER_COLUMNS}')
    Overwrite ('true')
) AS dt;

Scripts Included

Core Analytics Scripts

• table_analysis.sql: Automatic table structure discovery
• preprocessing.sql: Data preparation and feature engineering
• model_training.sql: ONNXPredict execution
• evaluation.sql: Results analysis and metrics
• complete_workflow_template.sql: End-to-end workflow

Utility Scripts

• data_quality_checks.sql: Comprehensive data validation
• parameter_tuning.sql: Parameter optimization
• diagnostic_queries.sql: Results diagnostics and interpretation

Best Practices

• Always run table_analysis.sql first to understand your data structure
• Validate data quality before executing the analytical function
• Use parameter_tuning.sql to find optimal configuration
• Review diagnostic_queries.sql output for model/results validation

Limitations

• Requires Teradata Vantage 17.20+ with ClearScape Analytics
• Input data must meet function-specific requirements
• Results depend on data quality and parameter configuration

Teradata ONNX Model Scoring - ClearScape Analytics skill for Teradata Vantage