From 14f7bc3c03b37dbd9c1462d01a0efc1d660925b6 Mon Sep 17 00:00:00 2001 From: harkirat-data Date: Mon, 6 Apr 2026 23:05:29 +0530 Subject: [PATCH] Added cross-validation tutorial notebook with explanation and visualization --- tutorials/cross_validation.ipynb | 160 +++++++++++++++++++++++++++++++ 1 file changed, 160 insertions(+) create mode 100644 tutorials/cross_validation.ipynb diff --git a/tutorials/cross_validation.ipynb b/tutorials/cross_validation.ipynb new file mode 100644 index 0000000..5cbebb0 --- /dev/null +++ b/tutorials/cross_validation.ipynb @@ -0,0 +1,160 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "id": "ab57a3bf", + "metadata": {}, + "source": [ + "# Cross Validation in Machine Learning\n", + "\n", + "## What is Cross Validation?\n", + "\n", + "Cross-validation is a technique used to evaluate how well a machine learning model performs on unseen data.\n", + "\n", + "Instead of splitting the data once into training and testing sets, cross-validation splits the data multiple times and evaluates the model repeatedly." + ] + }, + { + "cell_type": "markdown", + "id": "6d78907c", + "metadata": {}, + "source": [ + "## Why do we need Cross Validation?\n", + "\n", + "When we use a simple train-test split:\n", + "- The result may depend heavily on how the data is split\n", + "- Model may overfit or underfit\n", + "\n", + "Cross-validation solves this by:\n", + "- Using multiple splits\n", + "- Giving a more reliable performance estimate" + ] + }, + { + "cell_type": "markdown", + "id": "c3184883", + "metadata": {}, + "source": [ + "## Dataset Used\n", + "\n", + "We are using the Diabetes dataset from sklearn, which is commonly used for regression problems." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "0964706b", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Scores: [0.42955615 0.52259939 0.48268054 0.42649776 0.55024834]\n", + "Mean Score: 0.48231643590864215\n" + ] + } + ], + "source": [ + "import numpy as np\n", + "import pandas as pd\n", + "from sklearn.model_selection import cross_val_score\n", + "from sklearn.linear_model import LinearRegression\n", + "from sklearn.datasets import load_diabetes\n", + "\n", + "#load dataset\n", + "data = load_diabetes()\n", + "#features and targets\n", + "X = data.data\n", + "y = data.target\n", + "\n", + "model = LinearRegression()\n", + "\n", + "scores = cross_val_score(model, X, y, cv=5) #Here, cv=5 means the dataset is split into 5 equal parts. The model is trained and tested 5 times, each time using a different part as test data. \n", + "\n", + "print(\"Scores:\", scores)\n", + "print(\"Mean Score:\", scores.mean())" + ] + }, + { + "cell_type": "markdown", + "id": "0279aaaa", + "metadata": {}, + "source": [ + "## Understanding the Results\n", + "\n", + "- Each value in the scores array represents the model's performance on one fold\n", + "- The mean score gives an overall performance estimate\n", + "\n", + "A higher score indicates better model performance.\n", + "Conclusion: Cross-validation provides a more reliable evaluation of model performance and helps reduce overfitting." + ] + }, + { + "cell_type": "markdown", + "id": "4d7fa9a0", + "metadata": {}, + "source": [ + "## Visualization of Cross-Validation Scores\n", + "\n", + "The plot above shows the model performance for each fold in cross-validation.\n", + "\n", + "- Each point represents the score obtained in one fold\n", + "- Variation in scores indicates how stable the model is across different data splits\n", + "- If the scores are similar, the model is consistent\n", + "- Large variation may indicate that the model is sensitive to the data\n", + "\n", + "This visualization helps us better understand the reliability of our model." + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "id": "5ce55a1e", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "import matplotlib.pyplot as plt\n", + "\n", + "plt.plot(scores, marker='o')\n", + "plt.title(\"Cross Validation Scores Across Folds\")\n", + "plt.xlabel(\"Fold Number\")\n", + "plt.ylabel(\"Score\")\n", + "plt.grid()\n", + "plt.show()" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "base", + "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": "3.13.5" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +}