Train Regressor (SVM) Node

AI/ML/Regression

Fit/Train a Support Vector Regressor. Learns non-linear targets through a kernel, with epsilon-SVR or nu-SVR.

fit_svm_regressionml

Inputs9

Outputs3

Security exposure6/10

Packageml

Ratings

Scores range from 0 to 10. Higher values mean more impact, exposure, or operational weight.

SecurityAttack surface and exposure impact.

6/10Medium

PrivacyPotential sensitivity of processed data.

6/10Medium

PerformanceRuntime or resource pressure.

4/10Medium

GovernancePolicy, audit, or compliance impact.

5/10Medium

ReliabilityOperational stability considerations.

7/10High

CostExternal or compute cost impact.

5/10Medium

Input Pins

Input

Execution

exec_in

Execution trigger that begins SVR training

Data Source

String

source

Choose which backend supplies the training data

Default Database

Database

Mode

String

mode

Epsilon-SVR penalises deviations larger than Epsilon. Nu-SVR replaces Epsilon with Nu, the target fraction of support vectors.

Default Epsilon-SVR

Epsilon-SVRNu-SVR

Kernel

String

kernel

Feature-space mapping. Gaussian for smooth non-linear targets, Linear for the plain SVR, Polynomial for interaction terms.

Default Gaussian

GaussianLinearPolynomial

Kernel Parameter

Float

kernel_param

Gaussian: the eps in exp(-||x - x'||^2 / eps), larger means smoother. Polynomial: the degree of (<x, x'> + 1)^degree. Ignored for Linear.

Default 30

Range 0.0001 to 1000

C

Float

Penalty for deviations outside the tolerated margin. Higher values fit the training data harder and risk overfitting. Used by both modes.

Default 1

Range 0.0001 to 100000

Epsilon

Float

epsilon

Width of the insensitive tube: errors smaller than this are not penalised. Epsilon-SVR only.

Default 0.1

Range 0.0001 to 100

Nu

Float

Upper bound on the fraction of training errors and lower bound on the fraction of support vectors, in (0, 1]. Nu-SVR only.

Default 0.5

Range 0.0001 to 1

Solver Tolerance

Float

tolerance

Stopping threshold of the SMO solver. Smaller values train longer for a more precise solution.

Default 0.001

Range 1e-7 to 1

Output Pins

Done

Execution

exec_out

Activated once training completes

Model

Struct

model

Thread-safe handle to the trained support vector regressor

Schema enforced

Support Vectors

Integer

support_vectors

Number of training rows that ended up contributing to the regression

Node Info

Internal name: fit_svm_regression
Category: AI/ML/Regression