From f107d004937100d1d49d5c19e3a3f9230604e582 Mon Sep 17 00:00:00 2001
From: nonunicorn <inonunicorn@gmail.com>
Date: Fri, 16 Dec 2022 16:18:41 +0000
Subject: [PATCH] Using OneClassSVM for outliers detection example #1890

---
 ...classsvm-for-outliers-detection-example.md | 34 +++++++++++++++++++
 1 file changed, 34 insertions(+)
 create mode 100644 python-scikit-learn/using-oneclasssvm-for-outliers-detection-example.md

diff --git a/python-scikit-learn/using-oneclasssvm-for-outliers-detection-example.md b/python-scikit-learn/using-oneclasssvm-for-outliers-detection-example.md
new file mode 100644
index 000000000..35648247c
--- /dev/null
+++ b/python-scikit-learn/using-oneclasssvm-for-outliers-detection-example.md
@@ -0,0 +1,34 @@
+# Using OneClassSVM for outliers detection example
+
+```python
+from sklearn import svm
+
+X = [[0.32], [0.31], [0], [0.32], [0.31], [1], [0.32], [0.31]]
+
+m = svm.OneClassSVM(gamma='auto').fit(X)
+detected = m.predict(X)
+```
+
+- `from sklearn import` - import module from [lib:scikit-learn](https://onelinerhub.com/python-scikit-learn/how-to-install-scikit-learn-using-pip)
+- `X = ` - example values set to detect outliers for
+- `.OneClassSVM(` - creates unsupervised outlier detection model
+- `.fit(` - train transformation model
+- `.predict(` - predict target variable based on given features dataset
+- `detected` - will contain a list of attributed outliers (`-1`) and values that are ok (`1`)
+
+## Example: 
+```python
+from sklearn import svm
+
+X = [[0.32], [0.31], [0], [0.32], [0.31], [1], [0.32], [0.31]]
+
+m = svm.OneClassSVM(gamma='auto').fit(X)
+detected = m.predict(X)
+
+print(detected)
+```
+```
+[ 1  1 -1  1  1 -1  1  1]
+
+```
+