linear regression, smaller improvements

1 files changed, 22 insertions, 0 deletions

diff --git a/linear.py b/linear.py
new file mode 100644
index 0000000..36ef411
--- /dev/null
+++ b/linear.py
@@ -0,0 +1,22 @@
+#!/usr/bin/python3
+from sklearn.datasets import *
+from sklearn.linear_model import LinearRegression
+from sklearn.cross_validation import train_test_split
+import numpy as np
+
+SEED = 42
+np.random.seed(SEED)
+
+data = load_boston()
+
+X_train, X_test, y_train, y_test = train_test_split(data.data, data.target, train_size=0.85)
+
+model = LinearRegression()
+model.fit(data.data, data.target)
+
+p=model.predict(X_test)
+squares=[]
+for n in range(len(p)):
+    squares.append((y_test[n] - p[n])**2)
+
+print(np.mean(squares))