Create one_class_svm.py

projects/piguardian/ai_ml/one_class_svm.py

@@ -0,0 +1,28 @@
+import numpy as np
+from sklearn.svm import OneClassSVM
+from sklearn.preprocessing import StandardScaler
+
+class OneClassSVMAnomalyDetector:
+    def __init__(self, kernel='rbf', gamma=0.1, nu=0.1):
+        self.kernel = kernel
+        self.gamma = gamma
+        self.nu = nu
+        self.svm = OneClassSVM(kernel=kernel, gamma=gamma, nu=nu)
+        self.scaler = StandardScaler()
+
+    def fit(self, X):
+        self.scaler.fit(X)
+        X_scaled = self.scaler.transform(X)
+        self.svm.fit(X_scaled)
+
+    def predict(self, X):
+        X_scaled = self.scaler.transform(X)
+        return self.svm.predict(X_scaled)
+
+    def decision_function(self, X):
+        X_scaled = self.scaler.transform(X)
+        return self.svm.decision_function(X_scaled)
+
+    def score_samples(self, X):
+        X_scaled = self.scaler.transform(X)
+        return self.svm.score_samples(X_scaled)