02-Naive-Gaussian.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "352982e0",
   "metadata": {},
   "source": [
    "# Naive Bayes example\n",
    "\n",
    "## Gaussian"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "1b9d5215",
   "metadata": {
    "vscode": {
     "languageId": "rust"
    }
   },
   "outputs": [],
   "source": [
    "// cargo install the newest version\n",
    "// this may take a while, needs to download and install the library\n",
    ":dep smartcore = {git = \"https://github.com/m7142yosuke/smartcore.git\", branch = \"feature/improve-display-for-naive-bayes\",  features = [\"datasets\"]}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "c62e1893",
   "metadata": {
    "vscode": {
     "languageId": "rust"
    }
   },
   "outputs": [],
   "source": [
    "use smartcore::dataset::iris::load_dataset;\n",
    "use smartcore::naive_bayes::gaussian::{GaussianNB, GaussianNBParameters};\n",
    "use smartcore::linalg::basic::matrix::DenseMatrix;\n",
    "// Model performance\n",
    "use smartcore::metrics::accuracy;"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "0acb0e13",
   "metadata": {
    "vscode": {
     "languageId": "rust"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "GaussianNB:\n",
      " inner: BaseNaiveBayes: {\n",
      " distribution: GaussianNBDistribution: {\n",
      "    class_labels: [0, 1, 2],\n",
      "    class_count: [50, 50, 50],\n",
      "    class_priors: [0.3333333333333333, 0.3333333333333333, 0.3333333333333333],\n",
      "    var: [[0.12176398698426993, 0.1422760001411465, 0.0295040034446723, 0.01126400058841702], [0.26110400788880384, 0.09650000076294418, 0.21639999752045114, 0.03832399889564586], [0.3962559386291673, 0.10192399745559833, 0.2984959836425922, 0.07392400431251556]],\n",
      "    theta: [[5.006000003814697, 3.41800000667572, 1.4639999961853027, 0.24400000482797624], [5.935999975204468, 2.770000009536743, 4.259999980926514, 1.3259999918937684], [6.588000001907349, 2.9739999914169313, 5.551999988555909, 2.0259999775886537]]\n",
      "}}\n",
      "\n",
      "\n",
      "\n",
      "accuracy: 0.96\n"
     ]
    }
   ],
   "source": [
    "// Load Iris dataset\n",
    "let iris_dataset = load_dataset();\n",
    "let iris_data = load_dataset().data;\n",
    "let iris_targets = load_dataset().target;\n",
    "\n",
    "// Input data\n",
    "let x: DenseMatrix<f32> = DenseMatrix::new(\n",
    "    iris_dataset.num_samples,      // num rows\n",
    "    iris_dataset.num_features,     // num columns\n",
    "    iris_data,                     // data as Vec\n",
    "    false,                         // column_major\n",
    ");\n",
    "// Ground truth\n",
    "let y: Vec<u32> = iris_targets;\n",
    "\n",
    "// Run Gaussian Naive Bayes without priors\n",
    "let parameters = GaussianNBParameters::default();\n",
    "let gnb = GaussianNB::fit(&x, &y, parameters).unwrap();\n",
    "\n",
    "// with priors would be:\n",
    "// let priors = vec![...];\n",
    "// let parameters = GaussianNBParameters::default().with_priors(priors.clone());\n",
    "\n",
    "println!(\"{}\", &gnb);\n",
    "\n",
    "let y_hat = gnb.predict(&x).unwrap(); // Predict class labels\n",
    "// Calculate training error\n",
    "println!(\"\\naccuracy: {}\", accuracy(&y, &y_hat)); // Prints 0.96"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9b9545b8",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Rust",
   "language": "rust",
   "name": "rust"
  },
  "language_info": {
   "codemirror_mode": "rust",
   "file_extension": ".rs",
   "mimetype": "text/rust",
   "name": "Rust",
   "pygment_lexer": "rust",
   "version": ""
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}