> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.nvidia.com/sdgm/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.nvidia.com/sdgm/_mcp/server.

# Model Plan

> **Prerelease documentation**
>
> This Kumo Studio information is prerelease and is not yet supported. Stay tuned for the Docker and Kubernetes releases.

Kumo simplifies the process of building high-performing Graph Neural Networks (GNNs) with two key tools:

* **AutoML:** Automatically selects the best GNN architecture, column encodings, and training table generation strategy based on your dataset and predictive query.
* **Model Planner:** Provides fine-grained control over GNN architecture, column encoding, and training configurations for experienced users who want to optimize performance.

## Model Plan

Whenever you write a predictive query, Kumo generates a modeling plan covering:

### Column Encoding

Kumo automates column encoding, transforming raw tabular data into model-ready inputs.
The AutoML algorithm analyzes data types, column semantics, and statistical properties to determine the best encoding strategy.
Supported encodings include:

* **Hash Encoding:** For high-cardinality identifiers (for example, `product_code`).
* **Datetime Encoding:** For timestamps.
* **Numerical Encoding:** For quantities (for example, `num_visits`).
* **Index Encoding:** For boolean values.

### Training Table Generation

Kumo automatically generates training tables with properly ordered train, validation, and holdout splits, preventing data leakage in temporal queries.
Kumo optimally samples data to ensure balanced splits, even for complex predictive tasks involving time-based aggregations.

### Architecture Search

Kumo supports GNN architectures including GraphSAGE, GIN, ID-GNN, GCN, PNA, and GAT.
Kumo selects the best architecture and hyperparameters for each predictive query, optimizing aspects such as:

* Neighborhood sampling method
* Layer connectivity
* Embedding size
* Aggregation methods

Kumo runs multiple experiments (typically 2–8) to find the best configuration and displays the final architecture and hyperparameters in the UI.

## Fine-Grained Control with Model Plan

The Model Plan lets you override AutoML defaults and control model configurations directly.
Common use cases:

### **Data Split Strategy**

Specify exact holdout datasets using `TimeRangeSplit`, ensuring compatibility with external models or enforcing organizational constraints.

### **Faster Job Execution**

Skip the full AutoML search by specifying a known architecture, reducing the number of experiments and improving iteration speed.

### **Performance Optimization**

Adjust hyperparameters beyond AutoML defaults to maximize accuracy, such as increasing channel limits or enabling `refit` for full dataset training.

### **Custom Data Encoding**

Override AutoML's encoding choices to:

* Treat missing numerical values as `0`.
* Use advanced NLP encoding for specific text columns.

### **Optimization Method Customization**

Modify Kumo's tuning metric (e.g., AUROC, MAE, Loss) or optimize for recommendation objectives like diversity vs. recall.

### **Embedding Export**

Ensure stable embeddings for downstream use in KNN lookups or feature engineering, or disable embedding generation to prioritize accuracy with advanced GNN architectures.

> The [Column Preprocessing reference](/kumo-engine/1.0.0/kumo-studio-guide-coming-soon/preliminary-documentation/connect-data/column-preprocessing) details the available options.