notebooks/02_numerical_pipeline_sol_00.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# \ud83d\udcc3 Solution for Exercise M1.02\n",
    "\n",
    "The goal of this exercise is to fit a similar model as in the previous\n",
    "notebook to get familiar with manipulating scikit-learn objects and in\n",
    "particular the `.fit/.predict/.score` API."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's load the adult census dataset with only numerical variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "adult_census = pd.read_csv(\"../datasets/adult-census-numeric.csv\")\n",
    "data = adult_census.drop(columns=\"class\")\n",
    "target = adult_census[\"class\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In the previous notebook we used `model = KNeighborsClassifier()`. All\n",
    "scikit-learn models can be created without arguments. This is convenient\n",
    "because it means that you don't need to understand the full details of a\n",
    "model before starting to use it.\n",
    "\n",
    "One of the `KNeighborsClassifier` parameters is `n_neighbors`. It controls\n",
    "the number of neighbors we are going to use to make a prediction for a new\n",
    "data point.\n",
    "\n",
    "What is the default value of the `n_neighbors` parameter? Hint: Look at the\n",
    "documentation on the [scikit-learn\n",
    "website](https://scikit-learn.org/stable/modules/generated/sklearn.neighbors.KNeighborsClassifier.html)\n",
    "or directly access the description inside your notebook by running the\n",
    "following cell. This will open a pager pointing to the documentation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "\n",
    "KNeighborsClassifier?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": [
     "solution"
    ]
   },
   "source": [
    "We can see that the default value for `n_neighbors` is 5."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Create a `KNeighborsClassifier` model with `n_neighbors=50`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# solution\n",
    "model = KNeighborsClassifier(n_neighbors=50)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Fit this model on the data and target loaded above"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# solution\n",
    "model.fit(data, target)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Use your model to make predictions on the first 10 data points inside the\n",
    "data. Do they match the actual target values?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# solution\n",
    "first_data_values = data.iloc[:10]\n",
    "first_predictions = model.predict(first_data_values)\n",
    "first_predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "tags": [
     "solution"
    ]
   },
   "outputs": [],
   "source": [
    "first_target_values = target.iloc[:10]\n",
    "first_target_values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "tags": [
     "solution"
    ]
   },
   "outputs": [],
   "source": [
    "number_of_correct_predictions = (\n",
    "    first_predictions == first_target_values).sum()\n",
    "number_of_predictions = len(first_predictions)\n",
    "print(\n",
    "    f\"{number_of_correct_predictions}/{number_of_predictions} \"\n",
    "    \"of predictions are correct\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Compute the accuracy on the training data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# solution\n",
    "model.score(data, target)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now load the test data from `\"../datasets/adult-census-numeric-test.csv\"` and\n",
    "compute the accuracy on the test data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# solution\n",
    "adult_census_test = pd.read_csv(\"../datasets/adult-census-numeric-test.csv\")\n",
    "\n",
    "data_test = adult_census_test.drop(columns=\"class\")\n",
    "target_test = adult_census_test[\"class\"]\n",
    "\n",
    "model.score(data_test, target_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": [
     "solution"
    ]
   },
   "source": [
    "Looking at the previous notebook, the accuracy seems slightly higher with\n",
    "`n_neighbors=50` than with `n_neighbors=5` (the default value)."
   ]
  }
 ],
 "metadata": {
  "jupytext": {
   "main_language": "python"
  },
  "kernelspec": {
   "display_name": "Python 3",
   "name": "python3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}