From 5ee42fc999747dfbdd953200a2fe8f69bd37da3b Mon Sep 17 00:00:00 2001 From: Damian Bregier Date: Mon, 28 Jun 2021 10:25:18 +0200 Subject: [PATCH] ADD: Polynomial regression --- Polynomial Regression.ipynb | 17 ++++++++++++++++- 1 file changed, 16 insertions(+), 1 deletion(-) diff --git a/Polynomial Regression.ipynb b/Polynomial Regression.ipynb index c6933b5..49a4d21 100644 --- a/Polynomial Regression.ipynb +++ b/Polynomial Regression.ipynb @@ -105,6 +105,21 @@ "## 2. Metody regresji wielomianowej" ] }, + { + "attachments": { + "image.png": { + "image/png": "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" + } + }, + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Regresja wielomianowa służy do obliczania zależności pomiędzy zmienną zależną a jedną lub więcej zmiennymi niezależnymi, które mogą występować w wyższych potęgach. Wykorzystywana w przypadku zbiorów cechujacych się nieliniowymi zależnościami.\n", + "\n", + "Równanie regresji wielomianowej:\n", + "![image.png](attachment:image.png)" + ] + }, { "cell_type": "code", "execution_count": 40, @@ -118,7 +133,7 @@ " print(f\"{x}x^{i}\", end=\" + \")\n", " print(f\"{theta.tolist()[-1][0]}^{len(theta) - 1}\")\n", "\n", - "# Implementacja wzrosu na regresję wielomianową\n", + "# Implementacja wzoru na regresję wielomianową\n", "def polynomial_regression(theta, x):\n", " x = x/data[\"Height\"].max()\n", " return sum(theta * np.power(x, i) for i, theta in enumerate(theta.tolist()))\n",