iris-svm

sklearnv1.0.0

A beginner-friendly multi-class classification model using Support Vector Machine with RBF kernel on the classic Iris dataset. Uses scikit-learn Pipeline with StandardScaler preprocessing and cross-validation for evaluation. Reports per-class precision, recall, and F1 scores.

Classificationtabularmulti-classsvmbeginner

Install

1openmodelstudio install iris-svm

SDK Usage

1import openmodelstudio as oms
2
3model = oms.use_model("iris-svm")
4handle = oms.register_model("my-iris-svm", model=model)
5job = oms.start_training(handle.model_id, wait=True)

Source Preview(76 lines)

View full source

1"""Iris flower classification with SVM."""
2
3import numpy as np
4from sklearn.svm import SVC
5from sklearn.model_selection import cross_val_score
6from sklearn.preprocessing import StandardScaler
7from sklearn.pipeline import Pipeline
8
9def train(ctx):
10    hp = ctx.hyperparameters
11    C = float(hp.get("C", 1.0))
12    kernel = hp.get("kernel", "rbf")
13
14    model = Pipeline([
15        ("scaler", StandardScaler()),
16        ("svc", SVC(C=C, kernel=kernel, probability=True)),
17    ])
18
19    from sklearn.datasets import load_iris
20    data = load_iris()
21    X, y = data.data, data.target
22    # ... cross-validation, metrics logging

Details

Author: openmodelstudio
License: MIT
Source: 76 lines
Version: 1.0.0

Dependencies

scikit-learn>=1.3

numpy>=1.24

Hyperparameters

C1.0

Regularization parameter

kernelrbf

Kernel type (rbf, linear, poly)

gammascale

Kernel coefficient

View on GitHub