Python/ml/principal_component_analysis.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "kernelspec": {
      "display_name": "Python 3",
      "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": "3.6.8"
    },
    "colab": {
      "name": "PCA_example.ipynb",
      "provenance": []
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ijPhIQgyuE5V",
        "colab_type": "text"
      },
      "source": [
        "<h2 align=center> Principal Component Analysis</h2>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5jRnJxNBuE5Y",
        "colab_type": "text"
      },
      "source": [
        "## IMPORTS and LOADING"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vZ5WqEs8uE5Z",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "%matplotlib inline\n",
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "import seaborn as sns"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "k9YifIuNuE5c",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "plt.style.use(\"ggplot\")\n",
        "plt.rcParams[\"figure.figsize\"] = (12,8)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "P7CvCFf1uE5e",
        "colab_type": "code",
        "colab": {},
        "outputId": "d7bf5da5-3caf-4fe2-9895-debda083029c"
      },
      "source": [
        "iris = pd.read_csv(\"https://archive.ics.uci.edu/ml/machine-learning-databases/iris/iris.data\",\n",
        "                 header = None)\n",
        "iris.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>0</th>\n",
              "      <th>1</th>\n",
              "      <th>2</th>\n",
              "      <th>3</th>\n",
              "      <th>4</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>5.1</td>\n",
              "      <td>3.5</td>\n",
              "      <td>1.4</td>\n",
              "      <td>0.2</td>\n",
              "      <td>Iris-setosa</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>4.9</td>\n",
              "      <td>3.0</td>\n",
              "      <td>1.4</td>\n",
              "      <td>0.2</td>\n",
              "      <td>Iris-setosa</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>4.7</td>\n",
              "      <td>3.2</td>\n",
              "      <td>1.3</td>\n",
              "      <td>0.2</td>\n",
              "      <td>Iris-setosa</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>4.6</td>\n",
              "      <td>3.1</td>\n",
              "      <td>1.5</td>\n",
              "      <td>0.2</td>\n",
              "      <td>Iris-setosa</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>5.0</td>\n",
              "      <td>3.6</td>\n",
              "      <td>1.4</td>\n",
              "      <td>0.2</td>\n",
              "      <td>Iris-setosa</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "     0    1    2    3            4\n",
              "0  5.1  3.5  1.4  0.2  Iris-setosa\n",
              "1  4.9  3.0  1.4  0.2  Iris-setosa\n",
              "2  4.7  3.2  1.3  0.2  Iris-setosa\n",
              "3  4.6  3.1  1.5  0.2  Iris-setosa\n",
              "4  5.0  3.6  1.4  0.2  Iris-setosa"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 3
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2PXlrvRsuE5h",
        "colab_type": "code",
        "colab": {},
        "outputId": "54a28890-c999-433d-ef8f-4ebe16ebbec8"
      },
      "source": [
        "iris.columns=['sepal_length', 'sepal_width', 'petal_lenth', 'petal_width', 'species']\n",
        "iris.dropna(how = \"all\", inplace = True)\n",
        "iris.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>sepal_length</th>\n",
              "      <th>sepal_width</th>\n",
              "      <th>petal_lenth</th>\n",
              "      <th>petal_width</th>\n",
              "      <th>species</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>5.1</td>\n",
              "      <td>3.5</td>\n",
              "      <td>1.4</td>\n",
              "      <td>0.2</td>\n",
              "      <td>Iris-setosa</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>4.9</td>\n",
              "      <td>3.0</td>\n",
              "      <td>1.4</td>\n",
              "      <td>0.2</td>\n",
              "      <td>Iris-setosa</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>4.7</td>\n",
              "      <td>3.2</td>\n",
              "      <td>1.3</td>\n",
              "      <td>0.2</td>\n",
              "      <td>Iris-setosa</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>4.6</td>\n",
              "      <td>3.1</td>\n",
              "      <td>1.5</td>\n",
              "      <td>0.2</td>\n",
              "      <td>Iris-setosa</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>5.0</td>\n",
              "      <td>3.6</td>\n",
              "      <td>1.4</td>\n",
              "      <td>0.2</td>\n",
              "      <td>Iris-setosa</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   sepal_length  sepal_width  petal_lenth  petal_width      species\n",
              "0           5.1          3.5          1.4          0.2  Iris-setosa\n",
              "1           4.9          3.0          1.4          0.2  Iris-setosa\n",
              "2           4.7          3.2          1.3          0.2  Iris-setosa\n",
              "3           4.6          3.1          1.5          0.2  Iris-setosa\n",
              "4           5.0          3.6          1.4          0.2  Iris-setosa"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7_yYmEnWuE5j",
        "colab_type": "code",
        "colab": {},
        "outputId": "f2f50ac7-2244-4436-efa9-c6beb976d95e"
      },
      "source": [
        "iris.info()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "Int64Index: 150 entries, 0 to 149\n",
            "Data columns (total 5 columns):\n",
            "sepal_length    150 non-null float64\n",
            "sepal_width     150 non-null float64\n",
            "petal_lenth     150 non-null float64\n",
            "petal_width     150 non-null float64\n",
            "species         150 non-null object\n",
            "dtypes: float64(4), object(1)\n",
            "memory usage: 7.0+ KB\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4_oFpX2quE5n",
        "colab_type": "text"
      },
      "source": [
        "## Visualize"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "OI1vszRsuE5o",
        "colab_type": "code",
        "colab": {},
        "outputId": "b92f3bd2-1781-48f4-89ee-bf3ed0ba7e8b"
      },
      "source": [
        "sns.scatterplot(x = iris.sepal_length, y = iris.sepal_width,\n",
        "                hue = iris.species, style = iris.species);"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x576 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ziCE-gdduE5t",
        "colab_type": "text"
      },
      "source": [
        " "
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Ss6VHSSnuE5u",
        "colab_type": "text"
      },
      "source": [
        "## Standardize"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yKg6YDWEuE5v",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from sklearn.preprocessing import StandardScaler\n",
        "X = iris.iloc[:, 0:4].values\n",
        "y = iris.species.values\n",
        "\n",
        "X = StandardScaler().fit_transform(X)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9XafovdKuE5z",
        "colab_type": "text"
      },
      "source": [
        " "
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xTyYUXWMuE5z",
        "colab_type": "text"
      },
      "source": [
        "## Eigenvectors and Eigenvalues"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "F_CT6PlcuE50",
        "colab_type": "code",
        "colab": {},
        "outputId": "07f30133-01a4-480d-c097-0efe717c5de5"
      },
      "source": [
        "covariance_matrix = np.cov(X.T)\n",
        "print(\"Covariance matrix:\\n\", covariance_matrix)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Covariance matrix:\n",
            " [[ 1.00671141 -0.11010327  0.87760486  0.82344326]\n",
            " [-0.11010327  1.00671141 -0.42333835 -0.358937  ]\n",
            " [ 0.87760486 -0.42333835  1.00671141  0.96921855]\n",
            " [ 0.82344326 -0.358937    0.96921855  1.00671141]]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Vnhc3KVeuE52",
        "colab_type": "code",
        "colab": {},
        "outputId": "9740e77d-a595-400f-9bbe-41627db4319f"
      },
      "source": [
        "eigen_values, eigen_vectors = np.linalg.eig(covariance_matrix)\n",
        "print(\"Eigenvectors:\\n\", eigen_vectors, \"\\n\")\n",
        "print(\"Eigenvalues:\\n\", eigen_values)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Eigenvectors:\n",
            " [[ 0.52237162 -0.37231836 -0.72101681  0.26199559]\n",
            " [-0.26335492 -0.92555649  0.24203288 -0.12413481]\n",
            " [ 0.58125401 -0.02109478  0.14089226 -0.80115427]\n",
            " [ 0.56561105 -0.06541577  0.6338014   0.52354627]] \n",
            "\n",
            "Eigenvalues:\n",
            " [2.93035378 0.92740362 0.14834223 0.02074601]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "K-lbZ031uE5-",
        "colab_type": "text"
      },
      "source": [
        "## Picking Principal Components"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MT4uPchWuE5-",
        "colab_type": "code",
        "colab": {},
        "outputId": "d9953791-b903-4a1c-c0f2-f8f861d9eaee"
      },
      "source": [
        "for val in eigen_values:\n",
        "    print(val)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "2.930353775589317\n",
            "0.9274036215173419\n",
            "0.14834222648163944\n",
            "0.02074601399559593\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GetECYxPuE6A",
        "colab_type": "code",
        "colab": {},
        "outputId": "d407af2a-80f2-47ad-9707-3ae82ba89f02"
      },
      "source": [
        "variance_explained = [(i / sum(eigen_values)) * 100 for i in eigen_values]\n",
        "variance_explained"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[72.77045209380135, 23.03052326768065, 3.683831957627379, 0.5151926808906321]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Zy32lN10uE6C",
        "colab_type": "code",
        "colab": {},
        "outputId": "5f4f62ad-e2de-486f-dd97-94c48354ba4d"
      },
      "source": [
        "cumulative_variance_explained = np.cumsum(variance_explained)\n",
        "cumulative_variance_explained"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([ 72.77045209,  95.80097536,  99.48480732, 100.        ])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "HTfqM0OuuE6E",
        "colab_type": "code",
        "colab": {},
        "outputId": "f3a8c9c6-5810-4021-fa21-9e85c3659482"
      },
      "source": [
        "sns.lineplot(x = [1,2,3,4], y = cumulative_variance_explained);\n",
        "plt.xlabel(\"Number of components\")\n",
        "plt.ylabel(\"Cumulative explained variance\")\n",
        "plt.title(\"Explained variance vs Number of components\")\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x576 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "QzT8DPjZuE6I",
        "colab_type": "text"
      },
      "source": [
        " "
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "RhH9BhgouE6J",
        "colab_type": "text"
      },
      "source": [
        "## Lower-Dimensional"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xgitvn2puE6J",
        "colab_type": "code",
        "colab": {},
        "outputId": "5e2168ce-9b0d-473f-e061-6f1c448f97a6"
      },
      "source": [
        "projection_matrix = (eigen_vectors.T[:][:])[:2].T\n",
        "\n",
        "print(\"Projection matrix: \\n\", projection_matrix)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Projection matrix: \n",
            " [[ 0.52237162 -0.37231836]\n",
            " [-0.26335492 -0.92555649]\n",
            " [ 0.58125401 -0.02109478]\n",
            " [ 0.56561105 -0.06541577]]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qHCSwYpAuE6L",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "X_pca = X.dot(projection_matrix)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "V__vsYaYuE6M",
        "colab_type": "code",
        "colab": {},
        "outputId": "3193cede-664d-4f54-ed11-d2a5dbde5c3a"
      },
      "source": [
        "for species in ('Iris-setosa', 'Iris-versicolor', 'Iris-virginica'):\n",
        "    sns.scatterplot(X_pca[y==species, 0],\n",
        "                   X_pca[y==species, 1])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x576 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2usM_9pnuE6O",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}