{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "## Linear Regression for Diabetes dataset - Lesson 1" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Import needed libraries" ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "import matplotlib.pyplot as plt\n", "import numpy as np\n", "from sklearn import datasets, linear_model, model_selection\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Load the diabetes dataset, divided into `X` data and `y` features" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "(442, 10)\n", "[ 0.03807591 0.05068012 0.06169621 0.02187239 -0.0442235 -0.03482076\n", " -0.04340085 -0.00259226 0.01990749 -0.01764613]\n" ] } ], "source": [ "X, y = datasets.load_diabetes(return_X_y=True)\n", "print(X.shape)\n", "print(X[0])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Select just one feature to target for this exercise" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "(442,)\n" ] } ], "source": [ "# Selecting the 3rd feature\n", "X = X[:, 2]\n", "print(X.shape)\n" ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "(442, 1)\n", "[[ 0.06169621]\n", " [-0.05147406]\n", " [ 0.04445121]\n", " [-0.01159501]\n", " [-0.03638469]\n", " [-0.04069594]\n", " [-0.04716281]\n", " [-0.00189471]\n", " [ 0.06169621]\n", " [ 0.03906215]\n", " [-0.08380842]\n", " [ 0.01750591]\n", " [-0.02884001]\n", " [-0.00189471]\n", " [-0.02560657]\n", " [-0.01806189]\n", " [ 0.04229559]\n", " [ 0.01211685]\n", " [-0.0105172 ]\n", " [-0.01806189]\n", " [-0.05686312]\n", " [-0.02237314]\n", " [-0.00405033]\n", " [ 0.06061839]\n", " [ 0.03582872]\n", " [-0.01267283]\n", " [-0.07734155]\n", " [ 0.05954058]\n", " [-0.02129532]\n", " [-0.00620595]\n", " [ 0.04445121]\n", " [-0.06548562]\n", " [ 0.12528712]\n", " [-0.05039625]\n", " [-0.06332999]\n", " [-0.03099563]\n", " [ 0.02289497]\n", " [ 0.01103904]\n", " [ 0.07139652]\n", " [ 0.01427248]\n", " [-0.00836158]\n", " [-0.06764124]\n", " [-0.0105172 ]\n", " [-0.02345095]\n", " [ 0.06816308]\n", " [-0.03530688]\n", " [-0.01159501]\n", " [-0.0730303 ]\n", " [-0.04177375]\n", " [ 0.01427248]\n", " [-0.00728377]\n", " [ 0.0164281 ]\n", " [-0.00943939]\n", " [-0.01590626]\n", " [ 0.0250506 ]\n", " [-0.04931844]\n", " [ 0.04121778]\n", " [-0.06332999]\n", " [-0.06440781]\n", " [-0.02560657]\n", " [-0.00405033]\n", " [ 0.00457217]\n", " [-0.00728377]\n", " [-0.0374625 ]\n", " [-0.02560657]\n", " [-0.02452876]\n", " [-0.01806189]\n", " [-0.01482845]\n", " [-0.02991782]\n", " [-0.046085 ]\n", " [-0.06979687]\n", " [ 0.03367309]\n", " [-0.00405033]\n", " [-0.02021751]\n", " [ 0.00241654]\n", " [-0.03099563]\n", " [ 0.02828403]\n", " [-0.03638469]\n", " [-0.05794093]\n", " [-0.0374625 ]\n", " [ 0.01211685]\n", " [-0.02237314]\n", " [-0.03530688]\n", " [ 0.00996123]\n", " [-0.03961813]\n", " [ 0.07139652]\n", " [-0.07518593]\n", " [-0.00620595]\n", " [-0.04069594]\n", " [-0.04824063]\n", " [-0.02560657]\n", " [ 0.0519959 ]\n", " [ 0.00457217]\n", " [-0.06440781]\n", " [-0.01698407]\n", " [-0.05794093]\n", " [ 0.00996123]\n", " [ 0.08864151]\n", " [-0.00512814]\n", " [-0.06440781]\n", " [ 0.01750591]\n", " [-0.04500719]\n", " [ 0.02828403]\n", " [ 0.04121778]\n", " [ 0.06492964]\n", " [-0.03207344]\n", " [-0.07626374]\n", " [ 0.04984027]\n", " [ 0.04552903]\n", " [-0.00943939]\n", " [-0.03207344]\n", " [ 0.00457217]\n", " [ 0.02073935]\n", " [ 0.01427248]\n", " [ 0.11019775]\n", " [ 0.00133873]\n", " [ 0.05846277]\n", " [-0.02129532]\n", " [-0.0105172 ]\n", " [-0.04716281]\n", " [ 0.00457217]\n", " [ 0.01750591]\n", " [ 0.08109682]\n", " [ 0.0347509 ]\n", " [ 0.02397278]\n", " [-0.00836158]\n", " [-0.06117437]\n", " [-0.00189471]\n", " [-0.06225218]\n", " [ 0.0164281 ]\n", " [ 0.09618619]\n", " [-0.06979687]\n", " [-0.02129532]\n", " [-0.05362969]\n", " [ 0.0433734 ]\n", " [ 0.05630715]\n", " [-0.0816528 ]\n", " [ 0.04984027]\n", " [ 0.11127556]\n", " [ 0.06169621]\n", " [ 0.01427248]\n", " [ 0.04768465]\n", " [ 0.01211685]\n", " [ 0.00564998]\n", " [ 0.04660684]\n", " [ 0.12852056]\n", " [ 0.05954058]\n", " [ 0.09295276]\n", " [ 0.01535029]\n", " [-0.00512814]\n", " [ 0.0703187 ]\n", " [-0.00405033]\n", " [-0.00081689]\n", " [-0.04392938]\n", " [ 0.02073935]\n", " [ 0.06061839]\n", " [-0.0105172 ]\n", " [-0.03315126]\n", " [-0.06548562]\n", " [ 0.0433734 ]\n", " [-0.06225218]\n", " [ 0.06385183]\n", " [ 0.03043966]\n", " [ 0.07247433]\n", " [-0.0191397 ]\n", " [-0.06656343]\n", " [-0.06009656]\n", " [ 0.06924089]\n", " [ 0.05954058]\n", " [-0.02668438]\n", " [-0.02021751]\n", " [-0.046085 ]\n", " [ 0.07139652]\n", " [-0.07949718]\n", " [ 0.00996123]\n", " [-0.03854032]\n", " [ 0.01966154]\n", " [ 0.02720622]\n", " [-0.00836158]\n", " [-0.01590626]\n", " [ 0.00457217]\n", " [-0.04285156]\n", " [ 0.00564998]\n", " [-0.03530688]\n", " [ 0.02397278]\n", " [-0.01806189]\n", " [ 0.04229559]\n", " [-0.0547075 ]\n", " [-0.00297252]\n", " [-0.06656343]\n", " [-0.01267283]\n", " [-0.04177375]\n", " [-0.03099563]\n", " [-0.00512814]\n", " [-0.05901875]\n", " [ 0.0250506 ]\n", " [-0.046085 ]\n", " [ 0.00349435]\n", " [ 0.05415152]\n", " [-0.04500719]\n", " [-0.05794093]\n", " [-0.05578531]\n", " [ 0.00133873]\n", " [ 0.03043966]\n", " [ 0.00672779]\n", " [ 0.04660684]\n", " [ 0.02612841]\n", " [ 0.04552903]\n", " [ 0.04013997]\n", " [-0.01806189]\n", " [ 0.01427248]\n", " [ 0.03690653]\n", " [ 0.00349435]\n", " [-0.07087468]\n", " [-0.03315126]\n", " [ 0.09403057]\n", " [ 0.03582872]\n", " [ 0.03151747]\n", " [-0.06548562]\n", " [-0.04177375]\n", " [-0.03961813]\n", " [-0.03854032]\n", " [-0.02560657]\n", " [-0.02345095]\n", " [-0.06656343]\n", " [ 0.03259528]\n", " [-0.046085 ]\n", " [-0.02991782]\n", " [-0.01267283]\n", " [-0.01590626]\n", " [ 0.07139652]\n", " [-0.03099563]\n", " [ 0.00026092]\n", " [ 0.03690653]\n", " [ 0.03906215]\n", " [-0.01482845]\n", " [ 0.00672779]\n", " [-0.06871905]\n", " [-0.00943939]\n", " [ 0.01966154]\n", " [ 0.07462995]\n", " [-0.00836158]\n", " [-0.02345095]\n", " [-0.046085 ]\n", " [ 0.05415152]\n", " [-0.03530688]\n", " [-0.03207344]\n", " [-0.0816528 ]\n", " [ 0.04768465]\n", " [ 0.06061839]\n", " [ 0.05630715]\n", " [ 0.09834182]\n", " [ 0.05954058]\n", " [ 0.03367309]\n", " [ 0.05630715]\n", " [-0.06548562]\n", " [ 0.16085492]\n", " [-0.05578531]\n", " [-0.02452876]\n", " [-0.03638469]\n", " [-0.00836158]\n", " [-0.04177375]\n", " [ 0.12744274]\n", " [-0.07734155]\n", " [ 0.02828403]\n", " [-0.02560657]\n", " [-0.06225218]\n", " [-0.00081689]\n", " [ 0.08864151]\n", " [-0.03207344]\n", " [ 0.03043966]\n", " [ 0.00888341]\n", " [ 0.00672779]\n", " [-0.02021751]\n", " [-0.02452876]\n", " [-0.01159501]\n", " [ 0.02612841]\n", " [-0.05901875]\n", " [-0.03638469]\n", " [-0.02452876]\n", " [ 0.01858372]\n", " [-0.0902753 ]\n", " [-0.00512814]\n", " [-0.05255187]\n", " [-0.02237314]\n", " [-0.02021751]\n", " [-0.0547075 ]\n", " [-0.00620595]\n", " [-0.01698407]\n", " [ 0.05522933]\n", " [ 0.07678558]\n", " [ 0.01858372]\n", " [-0.02237314]\n", " [ 0.09295276]\n", " [-0.03099563]\n", " [ 0.03906215]\n", " [-0.06117437]\n", " [-0.00836158]\n", " [-0.0374625 ]\n", " [-0.01375064]\n", " [ 0.07355214]\n", " [-0.02452876]\n", " [ 0.03367309]\n", " [ 0.0347509 ]\n", " [-0.03854032]\n", " [-0.03961813]\n", " [-0.00189471]\n", " [-0.03099563]\n", " [-0.046085 ]\n", " [ 0.00133873]\n", " [ 0.06492964]\n", " [ 0.04013997]\n", " [-0.02345095]\n", " [ 0.05307371]\n", " [ 0.04013997]\n", " [-0.02021751]\n", " [ 0.01427248]\n", " [-0.03422907]\n", " [ 0.00672779]\n", " [ 0.00457217]\n", " [ 0.03043966]\n", " [ 0.0519959 ]\n", " [ 0.06169621]\n", " [-0.00728377]\n", " [ 0.00564998]\n", " [ 0.05415152]\n", " [-0.00836158]\n", " [ 0.114509 ]\n", " [ 0.06708527]\n", " [-0.05578531]\n", " [ 0.03043966]\n", " [-0.02560657]\n", " [ 0.10480869]\n", " [-0.00620595]\n", " [-0.04716281]\n", " [-0.04824063]\n", " [ 0.08540807]\n", " [-0.01267283]\n", " [-0.03315126]\n", " [-0.00728377]\n", " [-0.01375064]\n", " [ 0.05954058]\n", " [ 0.02181716]\n", " [ 0.01858372]\n", " [-0.01159501]\n", " [-0.00297252]\n", " [ 0.01750591]\n", " [-0.02991782]\n", " [-0.02021751]\n", " [-0.05794093]\n", " [ 0.06061839]\n", " [-0.04069594]\n", " [-0.07195249]\n", " [-0.05578531]\n", " [ 0.04552903]\n", " [-0.00943939]\n", " [-0.03315126]\n", " [ 0.04984027]\n", " [-0.08488624]\n", " [ 0.00564998]\n", " [ 0.02073935]\n", " [-0.00728377]\n", " [ 0.10480869]\n", " [-0.02452876]\n", " [-0.00620595]\n", " [-0.03854032]\n", " [ 0.13714305]\n", " [ 0.17055523]\n", " [ 0.00241654]\n", " [ 0.03798434]\n", " [-0.05794093]\n", " [-0.00943939]\n", " [-0.02345095]\n", " [-0.0105172 ]\n", " [-0.03422907]\n", " [-0.00297252]\n", " [ 0.06816308]\n", " [ 0.00996123]\n", " [ 0.00241654]\n", " [-0.03854032]\n", " [ 0.02612841]\n", " [-0.08919748]\n", " [ 0.06061839]\n", " [-0.02884001]\n", " [-0.02991782]\n", " [-0.0191397 ]\n", " [-0.04069594]\n", " [ 0.01535029]\n", " [-0.02452876]\n", " [ 0.00133873]\n", " [ 0.06924089]\n", " [-0.06979687]\n", " [-0.02991782]\n", " [-0.046085 ]\n", " [ 0.01858372]\n", " [ 0.00133873]\n", " [-0.03099563]\n", " [-0.00405033]\n", " [ 0.01535029]\n", " [ 0.02289497]\n", " [ 0.04552903]\n", " [-0.04500719]\n", " [-0.03315126]\n", " [ 0.097264 ]\n", " [ 0.05415152]\n", " [ 0.12313149]\n", " [-0.08057499]\n", " [ 0.09295276]\n", " [-0.05039625]\n", " [-0.01159501]\n", " [-0.0277622 ]\n", " [ 0.05846277]\n", " [ 0.08540807]\n", " [-0.00081689]\n", " [ 0.00672779]\n", " [ 0.00888341]\n", " [ 0.08001901]\n", " [ 0.07139652]\n", " [-0.02452876]\n", " [-0.0547075 ]\n", " [-0.03638469]\n", " [ 0.0164281 ]\n", " [ 0.07786339]\n", " [-0.03961813]\n", " [ 0.01103904]\n", " [-0.04069594]\n", " [-0.03422907]\n", " [ 0.00564998]\n", " [ 0.08864151]\n", " [-0.03315126]\n", " [-0.05686312]\n", " [-0.03099563]\n", " [ 0.05522933]\n", " [-0.06009656]\n", " [ 0.00133873]\n", " [-0.02345095]\n", " [-0.07410811]\n", " [ 0.01966154]\n", " [-0.01590626]\n", " [-0.01590626]\n", " [ 0.03906215]\n", " [-0.0730303 ]]\n" ] } ], "source": [ "#Reshaping to get a 2D array\n", "X = X.reshape(-1, 1)\n", "print(X.shape)\n", "print(X)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Split the training and test data for both `X` and `y`" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [], "source": [ "X_train, X_test, y_train, y_test = model_selection.train_test_split(X, y, test_size=0.33)\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Select the model and fit it with the training data" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
LinearRegression()
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.
" ], "text/plain": [ "LinearRegression()" ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" } ], "source": [ "model = linear_model.LinearRegression()\n", "model.fit(X_train, y_train)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Use test data to predict a line" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [], "source": [ "y_pred = model.predict(X_test)\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Display the results in a plot" ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "plt.scatter(X_test, y_test, color='black')\n", "plt.plot(X_test, y_pred, color='blue', linewidth=3)\n", "plt.show()" ] } ], "metadata": { "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.11.1" }, "metadata": { "interpreter": { "hash": "70b38d7a306a849643e446cd70466270a13445e5987dfa1344ef2b127438fa4d" } }, "orig_nbformat": 2 }, "nbformat": 4, "nbformat_minor": 2 }