SynapCores AutoML Guide
Build powerful machine learning models directly in SQL without writing any Python code.
Overview
SynapCores AutoML provides comprehensive options for creating machine learning experiments through SQL syntax. Train, tune, and deploy production-ready models using familiar database commands.
Task Types
| Task Type | Description | Default Metric |
|---|---|---|
regression |
Continuous value prediction | R-squared |
binary_classification |
Two-class classification | AUC |
classification/multiclass |
Multi-class classification | Accuracy |
clustering |
Unsupervised grouping | Silhouette Score |
anomaly |
Anomaly detection | F1 Score |
time_series |
Time series forecasting | MAPE |
Creating AutoML Experiments
Basic Syntax
Option 1: AS Syntax
CREATE EXPERIMENT <experiment_name> AS
<SELECT_query>
WITH (<options>)
Option 2: USING Syntax
CREATE EXPERIMENT <experiment_name>
USING (<SELECT_query>)
TARGET <target_column>
OPTIONS (<options>)
Configuration Options
General Options
| Option | Type | Default | Description |
|---|---|---|---|
task_type |
string | 'binary_classification' |
Type of ML task |
target_column |
string | Required | Column to predict |
max_trials |
integer | 100 | Maximum training trials |
time_budget_minutes |
integer | 60 | Maximum time budget |
validation_split |
float | 0.2 | Validation data proportion |
cv_folds |
integer | 5 | Cross-validation folds |
optimization_metric |
string | Task-dependent | Metric to optimize |
ensemble |
boolean | true | Create ensemble models |
early_stopping_patience |
integer | 10 | Trials without improvement |
random_seed |
integer | 42 | Random seed for reproducibility |
###Available Algorithms
'linear_regression'- Linear Regression'logistic_regression'- Logistic Regression'decision_tree'- Decision Tree'random_forest'- Random Forest'gradient_boosting'- Gradient Boosting'xgboost'- XGBoost'neural_network'- Neural Network'knn'- K-Nearest Neighbors'naive_bayes'- Naive Bayes'svm'- Support Vector Machine
Algorithm Selection Strategies
'all'- Try all available algorithms'fast'- Only fast algorithms (linear models, decision trees, naive bayes, knn)'accurate'- Only highly accurate algorithms (random forest, gradient boosting, xgboost, neural networks)'interpretable'- Only interpretable algorithms (linear regression, logistic regression, decision trees)
Algorithm-Specific Options
Random Forest
| Hyperparameter | Type | Default | Description |
|---|---|---|---|
n_estimators |
integer | 100 | Number of trees |
max_depth |
integer | None | Maximum tree depth |
min_samples_split |
integer | 2 | Minimum samples to split |
max_features |
string/float | 'sqrt' | Features to consider |
WITH (
task_type='classification',
algorithms=['random_forest'],
n_estimators=200,
max_depth=10
)
Neural Network
| Hyperparameter | Type | Default | Description |
|---|---|---|---|
hidden_layers |
array | [100] | Hidden layer sizes |
learning_rate |
float | 0.001 | Initial learning rate |
batch_size |
integer | 32 | Mini-batch size |
n_epochs |
integer | 100 | Maximum epochs |
activation |
string | 'relu' | Activation function |
dropout_rate |
float | 0.0 | Dropout rate |
WITH (
task_type='classification',
algorithms=['neural_network'],
hidden_layers=[128, 64, 32],
dropout_rate=0.2
)
Gradient Boosting / XGBoost
| Hyperparameter | Type | Default | Description |
|---|---|---|---|
n_estimators |
integer | 100 | Number of boosting stages |
learning_rate |
float | 0.1 | Learning rate |
max_depth |
integer | 3 | Maximum tree depth |
subsample |
float | 1.0 | Fraction of samples |