- Input
- Train and Test Set
- Example 1: Linear Model
- Example 2: Polynomial Model
- Example 3: Radial Basis Model (RBF) Model
- Example 4: Sigmoid Model
In all of these examples, the DenseSVMTrainer function creates the model and the DenseSVMPredictor function uses that model on a test set to make a prediction. The Polynomial, RBF and sigmoid models generally obtain better prediction accuracy with higher values of the hashbits and subspacedimension arguments. The value of subspacedimension argument cannot be greater than the number of rows input. You can tune the model using the cost and bias arguments. For details on model-specific tuning parameters, refer to the arguments section.