{ "cells": [ { "cell_type": "markdown", "id": "41a9f174", "metadata": {}, "source": [ "
\n", " \n", "# ๐Ÿ“K-NN Hyperparamaters\n", "\n", " ๐Ÿ‘จโ€๐Ÿซ **Vikesh K** \n", " ๐Ÿ““ **Lab 07** \n", " \n", "\n", "
" ] }, { "cell_type": "markdown", "id": "86fe86a1", "metadata": {}, "source": [ " ๐Ÿ’ก \n", " **\"There is no substitute for hard work\"** \n", " ๐Ÿ’ก " ] }, { "cell_type": "markdown", "id": "43a1c13c", "metadata": {}, "source": [ "\n", "## ๐Ÿ“Lab Agenda\n", " \n", "We will cover:\n", "\n", "* sklearn datasets\n", "* train-test split with stratification\n", "* Model Building with different values of k in k-nn\n", "* Visualize the accuracy score for different k" ] }, { "cell_type": "markdown", "id": "0976373e", "metadata": {}, "source": [ "## Importing Modules and data" ] }, { "cell_type": "code", "execution_count": 24, "id": "db04c530", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Modules ran\n" ] } ], "source": [ "import numpy as np\n", "import pandas as pd\n", "\n", "from sklearn.neighbors import KNeighborsClassifier\n", "from sklearn.model_selection import train_test_split\n", "import matplotlib.pyplot as plt\n", "\n", "from pprint import pprint\n", "import warnings\n", "warnings.filterwarnings(\"ignore\")\n", "print(\"Modules ran\")" ] }, { "cell_type": "markdown", "id": "b84c897c", "metadata": {}, "source": [ "
\n", " \n", "**๐Ÿ’ก We will be using the Wisconsin Cancer Dataset to build a `k-nn` model** \n", " \n", "**๐Ÿ“ Please read more about the Wisconsin Breast cancer dataset [here](https://scikit-learn.org/stable/datasets/toy_dataset.html)**.[*scroll down in the page*]\n", " \n", "
" ] }, { "cell_type": "code", "execution_count": 25, "id": "341167e9", "metadata": {}, "outputs": [], "source": [ "url = 'https://raw.githubusercontent.com/vkoul/data/main/misc/wisconsin.csv'\n", "df = pd.read_csv(url)" ] }, { "cell_type": "code", "execution_count": 26, "id": "d688ef49", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
diagnosisradius_meantexture_meanperimeter_meanarea_meansmoothness_meancompactness_meanconcavity_meanconcave points_meansymmetry_mean...radius_worsttexture_worstperimeter_worstarea_worstsmoothness_worstcompactness_worstconcavity_worstconcave points_worstsymmetry_worstfractal_dimension_worst
0117.9910.38122.801001.00.118400.277600.30010.147100.2419...25.3817.33184.602019.00.16220.66560.71190.26540.46010.11890
1120.5717.77132.901326.00.084740.078640.08690.070170.1812...24.9923.41158.801956.00.12380.18660.24160.18600.27500.08902
2119.6921.25130.001203.00.109600.159900.19740.127900.2069...23.5725.53152.501709.00.14440.42450.45040.24300.36130.08758
3111.4220.3877.58386.10.142500.283900.24140.105200.2597...14.9126.5098.87567.70.20980.86630.68690.25750.66380.17300
4120.2914.34135.101297.00.100300.132800.19800.104300.1809...22.5416.67152.201575.00.13740.20500.40000.16250.23640.07678
\n", "

5 rows ร— 31 columns

\n", "
" ], "text/plain": [ " diagnosis radius_mean texture_mean perimeter_mean area_mean \\\n", "0 1 17.99 10.38 122.80 1001.0 \n", "1 1 20.57 17.77 132.90 1326.0 \n", "2 1 19.69 21.25 130.00 1203.0 \n", "3 1 11.42 20.38 77.58 386.1 \n", "4 1 20.29 14.34 135.10 1297.0 \n", "\n", " smoothness_mean compactness_mean concavity_mean concave points_mean \\\n", "0 0.11840 0.27760 0.3001 0.14710 \n", "1 0.08474 0.07864 0.0869 0.07017 \n", "2 0.10960 0.15990 0.1974 0.12790 \n", "3 0.14250 0.28390 0.2414 0.10520 \n", "4 0.10030 0.13280 0.1980 0.10430 \n", "\n", " symmetry_mean ... radius_worst texture_worst perimeter_worst \\\n", "0 0.2419 ... 25.38 17.33 184.60 \n", "1 0.1812 ... 24.99 23.41 158.80 \n", "2 0.2069 ... 23.57 25.53 152.50 \n", "3 0.2597 ... 14.91 26.50 98.87 \n", "4 0.1809 ... 22.54 16.67 152.20 \n", "\n", " area_worst smoothness_worst compactness_worst concavity_worst \\\n", "0 2019.0 0.1622 0.6656 0.7119 \n", "1 1956.0 0.1238 0.1866 0.2416 \n", "2 1709.0 0.1444 0.4245 0.4504 \n", "3 567.7 0.2098 0.8663 0.6869 \n", "4 1575.0 0.1374 0.2050 0.4000 \n", "\n", " concave points_worst symmetry_worst fractal_dimension_worst \n", "0 0.2654 0.4601 0.11890 \n", "1 0.1860 0.2750 0.08902 \n", "2 0.2430 0.3613 0.08758 \n", "3 0.2575 0.6638 0.17300 \n", "4 0.1625 0.2364 0.07678 \n", "\n", "[5 rows x 31 columns]" ] }, "execution_count": 26, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.head()" ] }, { "cell_type": "markdown", "id": "9d64bf3a", "metadata": {}, "source": [ "## Exploring the data " ] }, { "cell_type": "code", "execution_count": 27, "id": "a39c1101", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(569, 31)" ] }, "execution_count": 27, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.shape" ] }, { "cell_type": "code", "execution_count": 28, "id": "adf0f7a1", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "diagnosis 0\n", "radius_mean 0\n", "texture_mean 0\n", "perimeter_mean 0\n", "area_mean 0\n", "smoothness_mean 0\n", "compactness_mean 0\n", "concavity_mean 0\n", "concave points_mean 0\n", "symmetry_mean 0\n", "fractal_dimension_mean 0\n", "radius_se 0\n", "texture_se 0\n", "perimeter_se 0\n", "area_se 0\n", "smoothness_se 0\n", "compactness_se 0\n", "concavity_se 0\n", "concave points_se 0\n", "symmetry_se 0\n", "fractal_dimension_se 0\n", "radius_worst 0\n", "texture_worst 0\n", "perimeter_worst 0\n", "area_worst 0\n", "smoothness_worst 0\n", "compactness_worst 0\n", "concavity_worst 0\n", "concave points_worst 0\n", "symmetry_worst 0\n", "fractal_dimension_worst 0\n", "dtype: int64" ] }, "execution_count": 28, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.isna().sum()" ] }, { "cell_type": "code", "execution_count": 29, "id": "5eae22f2", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\n", "RangeIndex: 569 entries, 0 to 568\n", "Data columns (total 31 columns):\n", " # Column Non-Null Count Dtype \n", "--- ------ -------------- ----- \n", " 0 diagnosis 569 non-null int64 \n", " 1 radius_mean 569 non-null float64\n", " 2 texture_mean 569 non-null float64\n", " 3 perimeter_mean 569 non-null float64\n", " 4 area_mean 569 non-null float64\n", " 5 smoothness_mean 569 non-null float64\n", " 6 compactness_mean 569 non-null float64\n", " 7 concavity_mean 569 non-null float64\n", " 8 concave points_mean 569 non-null float64\n", " 9 symmetry_mean 569 non-null float64\n", " 10 fractal_dimension_mean 569 non-null float64\n", " 11 radius_se 569 non-null float64\n", " 12 texture_se 569 non-null float64\n", " 13 perimeter_se 569 non-null float64\n", " 14 area_se 569 non-null float64\n", " 15 smoothness_se 569 non-null float64\n", " 16 compactness_se 569 non-null float64\n", " 17 concavity_se 569 non-null float64\n", " 18 concave points_se 569 non-null float64\n", " 19 symmetry_se 569 non-null float64\n", " 20 fractal_dimension_se 569 non-null float64\n", " 21 radius_worst 569 non-null float64\n", " 22 texture_worst 569 non-null float64\n", " 23 perimeter_worst 569 non-null float64\n", " 24 area_worst 569 non-null float64\n", " 25 smoothness_worst 569 non-null float64\n", " 26 compactness_worst 569 non-null float64\n", " 27 concavity_worst 569 non-null float64\n", " 28 concave points_worst 569 non-null float64\n", " 29 symmetry_worst 569 non-null float64\n", " 30 fractal_dimension_worst 569 non-null float64\n", "dtypes: float64(30), int64(1)\n", "memory usage: 137.9 KB\n" ] } ], "source": [ "df.info()" ] }, { "cell_type": "markdown", "id": "d738fb80", "metadata": {}, "source": [ "**Split the dataset into Train and Test** " ] }, { "cell_type": "code", "execution_count": 30, "id": "7a060ea5", "metadata": {}, "outputs": [], "source": [ "X = df.drop(columns = 'diagnosis')\n", "y = df['diagnosis']" ] }, { "cell_type": "markdown", "id": "55b3a3c4", "metadata": {}, "source": [ "
\n", "\n", "**๐Ÿ’กStratified Train-Test Splits** \n", "One final consideration is for classification problems only.\n", "\n", "Some classification problems do not have a balanced number of examples for each class label. As such, it is desirable to split the dataset into train and test sets in a way that preserves the same proportions of examples in each class as observed in the original dataset.\n", "\n", "This is called a stratified train-test split.\n", "\n", "We can achieve this by setting the `stratify` argument to the y component of the original dataset. This will be used by the `train_test_split()` function to ensure that both the train and test sets have the proportion of examples in each class that is present in the provided โ€œyโ€ array.\n", "\n", "Source: [MachineLearning Mastery](https://machinelearningmastery.com/train-test-split-for-evaluating-machine-learning-algorithms/#:~:text=Stratified%20Train%2DTest%20Splits,-One%20final%20consideration&text=As%20such%2C%20it%20is%20desirable,a%20stratified%20train%2Dtest%20split.)\n", "\n", "
" ] }, { "cell_type": "markdown", "id": "a667d64c", "metadata": {}, "source": [ "## Train Test Split" ] }, { "cell_type": "code", "execution_count": 31, "id": "251ab277", "metadata": {}, "outputs": [], "source": [ "X_train, X_test, y_train, y_test = train_test_split(X, y \n", " ,stratify = y # stratification \n", " ,random_state = 66 # for reproducibility\n", " ,test_size = 0.20)" ] }, { "cell_type": "code", "execution_count": 32, "id": "27fb9b44", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "(455, 30)\n", "(114, 30)\n" ] } ], "source": [ "print(X_train.shape)\n", "print(X_test.shape)" ] }, { "cell_type": "markdown", "id": "40c11453", "metadata": {}, "source": [ "### Checking the effect of Stratifcation Effect" ] }, { "cell_type": "markdown", "id": "148bc7f2", "metadata": {}, "source": [ "**Overall Distribution**" ] }, { "cell_type": "code", "execution_count": 33, "id": "4883e2fb", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
proportion
diagnosis
00.63
10.37
\n", "
" ], "text/plain": [ " proportion\n", "diagnosis \n", "0 0.63\n", "1 0.37" ] }, "execution_count": 33, "metadata": {}, "output_type": "execute_result" } ], "source": [ "y.value_counts(normalize = True).round(2).to_frame()" ] }, { "cell_type": "markdown", "id": "b24a0310", "metadata": {}, "source": [ "**Train Data Distribution**" ] }, { "cell_type": "code", "execution_count": 34, "id": "f3d5edad", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
proportion
diagnosis
00.63
10.37
\n", "
" ], "text/plain": [ " proportion\n", "diagnosis \n", "0 0.63\n", "1 0.37" ] }, "execution_count": 34, "metadata": {}, "output_type": "execute_result" } ], "source": [ "y_train.value_counts(normalize = True).round(2).to_frame()" ] }, { "cell_type": "markdown", "id": "af6b0f35", "metadata": {}, "source": [ "**Test Data Distribution**" ] }, { "cell_type": "code", "execution_count": 35, "id": "52f071dd", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
proportion
diagnosis
00.63
10.37
\n", "
" ], "text/plain": [ " proportion\n", "diagnosis \n", "0 0.63\n", "1 0.37" ] }, "execution_count": 35, "metadata": {}, "output_type": "execute_result" } ], "source": [ "y_test.value_counts(normalize = True).round(2).to_frame()" ] }, { "cell_type": "markdown", "id": "5bf509de", "metadata": {}, "source": [ "
\n", " \n", "๐Ÿ‘จ๐Ÿผโ€๐Ÿ’ป**Run the above ratio check after commenting out the `stratify` argument in `train_test_split`. Note down your findings**" ] }, { "cell_type": "markdown", "id": "d63448ed", "metadata": {}, "source": [ "## Building the model with 3 neighbours" ] }, { "cell_type": "markdown", "id": "40dfc93c", "metadata": {}, "source": [ "**You may get errors with the code below if your are using Sklearn version 1.3.0. In that case, you would need to install the version 1.2.2 of sklearn and may also need to install `threadpoolctl==3.1.0` to remove the `split` error**" ] }, { "cell_type": "code", "execution_count": 36, "id": "fde3d59c", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "'1.2.2'" ] }, "execution_count": 36, "metadata": {}, "output_type": "execute_result" } ], "source": [ "import sklearn\n", "\n", "sklearn.__version__" ] }, { "cell_type": "code", "execution_count": 37, "id": "2a8f101d", "metadata": {}, "outputs": [], "source": [ "# !pip install threadpoolctl==3.1.0" ] }, { "cell_type": "markdown", "id": "4ca1fc99", "metadata": {}, "source": [ "**[Sklearn Error](https://github.com/scikit-learn/scikit-learn/issues/26768)**" ] }, { "cell_type": "code", "execution_count": 38, "id": "1657f034", "metadata": {}, "outputs": [], "source": [ "# !pip uninstall scikit-learn" ] }, { "cell_type": "code", "execution_count": 39, "id": "95d3de10", "metadata": {}, "outputs": [], "source": [ "# !pip install scikit-learn==1.2.2" ] }, { "cell_type": "code", "execution_count": 40, "id": "a09da78d", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "The train data accuracy is 95.82%\n", "The test data accuracy is 91.23%\n" ] } ], "source": [ "# build the model\n", "clf = KNeighborsClassifier(n_neighbors= 3)\n", "\n", "# fit the model \n", "clf.fit(X_train, y_train)\n", "\n", "# # record training set accuracy\n", "train_acc = clf.score(X_train, y_train)\n", "print(f\"The train data accuracy is {train_acc :.2%}\")\n", "\n", "# # record test accuracy\n", "test_acc = clf.score(X_test, y_test)\n", "print(f\"The test data accuracy is {test_acc :.2%}\")" ] }, { "cell_type": "markdown", "id": "42d408c3", "metadata": {}, "source": [ "## Accuracy scores across values of k" ] }, { "cell_type": "markdown", "id": "f08083c9", "metadata": {}, "source": [ "**We will make use of the for- loop to run k-nn models, varying the `n_neighbors` argument of the function**" ] }, { "cell_type": "code", "execution_count": 41, "id": "2f4ca449", "metadata": {}, "outputs": [], "source": [ "# empty list that will be populated\n", "training_accuracy = [] \n", "test_accuracy = []\n", "\n", "# try n_neighbors from 1 to 51\n", "neighbors_settings = range(1, 51)\n", "\n", "for n_neighbors in neighbors_settings:\n", " \n", " # build the model\n", " clf = KNeighborsClassifier(n_neighbors= n_neighbors)\n", " clf.fit(X_train, y_train)\n", " \n", " # record training set accuracy\n", " training_accuracy.append(clf.score(X_train, y_train))\n", " \n", " # record generalization accuracy\n", " test_accuracy.append(clf.score(X_test, y_test))\n", " " ] }, { "cell_type": "markdown", "id": "830ebe2a", "metadata": {}, "source": [ "**[`pprint`](https://docs.python.org/3/library/pprint.html) provides a capability to โ€œpretty-printโ€ arbitrary Python data structures**" ] }, { "cell_type": "code", "execution_count": 42, "id": "c1cbb394", "metadata": {}, "outputs": [], "source": [ "# pprint(test_accuracy)\n", "# # print(\"-------\")\n", "# pprint(training_accuracy)" ] }, { "cell_type": "code", "execution_count": 43, "id": "6c80ae3f", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "50" ] }, "execution_count": 43, "metadata": {}, "output_type": "execute_result" } ], "source": [ "len(test_accuracy)" ] }, { "cell_type": "code", "execution_count": 44, "id": "ee92824b", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "50" ] }, "execution_count": 44, "metadata": {}, "output_type": "execute_result" } ], "source": [ "len(training_accuracy)" ] }, { "cell_type": "markdown", "id": "d6c3af9a", "metadata": {}, "source": [ "## Visualise the results" ] }, { "cell_type": "code", "execution_count": 45, "id": "55dc1cdd", "metadata": { "scrolled": false }, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "plt.figure(figsize= (10, 8))\n", "\n", "plt.plot(neighbors_settings, training_accuracy, marker = \"o\", label = \"train data accuracy\")\n", "plt.plot(neighbors_settings, test_accuracy, marker = \"o\", label = \"test data accuracy\")\n", "\n", "# highlight the max and min accuracy values\n", "plt.axvline(x = test_accuracy.index(max(test_accuracy))+1, linewidth = 20, alpha = 0.2, color = \"green\")\n", "plt.axvline(x = test_accuracy.index(min(test_accuracy))+1, linewidth = 20, alpha = 0.2, color = 'red')\n", "\n", "\n", "plt.ylabel(\"Accuracy\")\n", "plt.xlabel(\"n_neighbors\")\n", "plt.title(\"Accuracy scores for different values of `k` in train and test data\")\n", "\n", "plt.legend();" ] } ], "metadata": { "kernelspec": { "display_name": "Python 3 (ipykernel)", "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": "3.9.13" }, "toc": { "base_numbering": 1, "nav_menu": {}, "number_sections": true, "sideBar": true, "skip_h1_title": false, "title_cell": "Table of Contents", "title_sidebar": "Contents", "toc_cell": false, "toc_position": {}, "toc_section_display": true, "toc_window_display": false } }, "nbformat": 4, "nbformat_minor": 5 }