Skip to content

Instantly share code, notes, and snippets.

@KhyatiMahendru
Last active August 12, 2020 16:15
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save KhyatiMahendru/6efeefd604afddd9ccedc41aad76a945 to your computer and use it in GitHub Desktop.
Save KhyatiMahendru/6efeefd604afddd9ccedc41aad76a945 to your computer and use it in GitHub Desktop.
For Jason Kabi.ipynb
Display the source blob
Display the rendered blob
Raw
{
"nbformat": 4,
"nbformat_minor": 0,
"metadata": {
"colab": {
"name": "For Jason Kabi.ipynb",
"provenance": [],
"authorship_tag": "ABX9TyOPbBIX1b+SAiZmEx+aJqwm",
"include_colab_link": true
},
"kernelspec": {
"name": "python3",
"display_name": "Python 3"
}
},
"cells": [
{
"cell_type": "markdown",
"metadata": {
"id": "view-in-github",
"colab_type": "text"
},
"source": [
"<a href=\"https://colab.research.google.com/gist/KhyatiMahendru/6efeefd604afddd9ccedc41aad76a945/for-jason-kabi.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
]
},
{
"cell_type": "code",
"metadata": {
"id": "Tn-n9nsoVx7r",
"colab_type": "code",
"colab": {}
},
"source": [
"from sklearn.datasets import make_blobs\n",
"\n",
"# Create dataset with 3 random cluster centers and 1000 datapoints\n",
"x, y = make_blobs(n_samples = 1000, centers = 3, n_features=2, shuffle=True, random_state=31)"
],
"execution_count": 1,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "1jhh9LHYWAk9",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 500
},
"outputId": "ac548a88-ee08-4baa-fdb4-af8dbd75c165"
},
"source": [
"import matplotlib.pyplot as plt\n",
"plt.figure(figsize = (14, 8))\n",
"plt.scatter(x[:, 0], x[:, 1])"
],
"execution_count": 4,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": [
"<matplotlib.collections.PathCollection at 0x7f9601b94390>"
]
},
"metadata": {
"tags": []
},
"execution_count": 4
},
{
"output_type": "display_data",
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 1008x576 with 1 Axes>"
]
},
"metadata": {
"tags": [],
"needs_background": "light"
}
}
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "6fWJSfbLXaSR",
"colab_type": "text"
},
"source": [
"# Optimal K has the largest Silhouette Score. So the task is simply to identify the K for which Silhouette score is maximum. In case of a tie between 2 K values, it might be better to choose the smaller one (but depends on your appliction)."
]
},
{
"cell_type": "code",
"metadata": {
"id": "pfyG30fZYDfu",
"colab_type": "code",
"colab": {}
},
"source": [
"from sklearn.metrics import silhouette_score\n",
"from sklearn.cluster import KMeans\n",
"\n",
"\n",
"# you need to input a value for kmax. I have added default value as 10\n",
"def optimal_k(max_sil = 0, kmax = 10):\n",
" for k in range(2, kmax+1):\n",
" kmeans = KMeans(n_clusters = k).fit(x)\n",
" labels = kmeans.labels_\n",
" curr_sil = silhouette_score(x, labels, metric = 'euclidean')\n",
" # max_sil variable is initialised to zero. It will store max silhouette score identified till the current value of k\n",
" if max_sil < curr_sil:\n",
" max_sil = curr_sil\n",
" k_optimal = k\n",
" return k_optimal"
],
"execution_count": 22,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "UIlmNJwsaM3V",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 282
},
"outputId": "e53b2e94-5445-4a26-8794-5dd08a3a9dc0"
},
"source": [
"kmeans_final = KMeans(n_clusters = optimal_k()).fit(x)\n",
"pred = kmeans_final.predict(x)\n",
"plt.scatter(x[:, 0], x[:, 1], c = pred)\n",
"plt.scatter(kmeans_final.cluster_centers_[:, 0], kmeans_final.cluster_centers_[:, 1], c = 'r', s = 200)"
],
"execution_count": 23,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": [
"<matplotlib.collections.PathCollection at 0x7f95fa7ce4a8>"
]
},
"metadata": {
"tags": []
},
"execution_count": 23
},
{
"output_type": "display_data",
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {
"tags": [],
"needs_background": "light"
}
}
]
}
]
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment