cuetessa.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "YsJgpVbgHrAE"
      },
      "source": [
        "# Song mood prediction using shallow models"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4F2lGH3lHs_V"
      },
      "source": [
        "# Objective: Predicting valence of songs"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "KBtmDn8mINSn"
      },
      "source": [
        "### What is valence?"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "oFo6ysKJIj2x"
      },
      "source": [
        "Score used to objectively define the musical positiveness of a song, including lyrics, instruments and the audio itself."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "a0oy3B8WIW5p"
      },
      "source": [
        "### What are shallow models?"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "g4uRYTSFI7Xg"
      },
      "source": [
        "Contrary to the commonly mentioned deep learning models, shallow models use lower ammount of features created from the raw data."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6axfjR72IZFG"
      },
      "source": [
        "### Why is this study relevant?"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "PlaPOPI4JL1H"
      },
      "source": [
        "Valence prediciton can be used on music platforms to create playlists, suggest songs based on moods (e.g. for different moments of the day) and help in identifying similar songs.\n",
        "\n",
        "Data from different sources were used in this study, so this could be used as a4 comparative study about data quality for training of shallow models."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "TC0JDG4pKK6I"
      },
      "source": [
        "# Methodology"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-pvYWNHrKa8f"
      },
      "source": [
        "### Decide what kind of data will be used"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gpioa-CLKrup"
      },
      "source": [
        "We opted to extract features from raw audio data, so valence scores can be predicted even for instrumental songs."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "58dmscKiKMiO"
      },
      "source": [
        "### Find data on the internet"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cS2BnGauKUp5"
      },
      "source": [
        "We used both the Spotify API for developers and the Dataset for Emotional Analysis of Music (DEAM).\n",
        "\n",
        "Spotify API is widely available, it has valence and audio previews for any song on the platform.\n",
        "\n",
        "DEAM has been used in scientific articles as a dataset for model training and it is a reliable source. Songs may not be commercial or famous."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "pSbV9_afKi_M"
      },
      "source": [
        "### Extract features"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8L9xP5KFKeEs"
      },
      "source": [
        "Feature extraction methods were found in the literature for raw audio data using the Librosa module. We tried to replicate methods by using:\n",
        "\n",
        "- Chroma\n",
        "- RMS Energy\n",
        "- Spectral Centroid\n",
        "- Rolloff\n",
        "- Zero Cross Ratings\n",
        "- Mel-frequency Ceptrum Coefficient\n",
        "\n",
        "We took the means and variances of both these signals and their first order differentiation signals.\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "PNosr3RxKnMk"
      },
      "source": [
        "### Analyze the data (EDA) and Train Models"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "EkT5HTToGQ6Z",
        "outputId": "5474ba27-91d4-45e5-f907-354ae147db81"
      },
      "outputs": [],
      "source": [
        "from google.colab import drive\n",
        "# mounting google drive to access data file\n",
        "drive.mount('/content/drive')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "noxkes2iXeSc",
        "outputId": "40591e49-d063-47b8-f013-be2d8deeb4fb"
      },
      "outputs": [],
      "source": [
        "# Import spotipy for easy access to the spotify API\n",
        "!pip install spotipy --quiet\n",
        "import spotipy\n",
        "import spotipy.util as util\n",
        "from spotipy.oauth2 import SpotifyClientCredentials\n",
        "from spotipy.oauth2 import SpotifyOAuth\n",
        "\n",
        "# Import python stuff\n",
        "import requests\n",
        "import shutil\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "import os\n",
        "import os.path as path\n",
        "from IPython.display import display, Audio\n",
        "import matplotlib.pyplot as plt\n",
        "import warnings\n",
        "import joblib\n",
        "import yaml\n",
        "from tqdm import tqdm\n",
        "import seaborn as sns\n",
        "from sklearn.preprocessing import minmax_scale\n",
        "\n",
        "# Import librosa for audio processing\n",
        "import librosa\n",
        "import librosa.display"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XS2i5BoTsxJb",
        "outputId": "188f3034-1558-48e5-f789-02a4c0d19e92"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Feature Data\n",
            "x train shape:  (1196, 104)\n",
            "x test shape (298, 104)\n",
            "x valid shape (298, 104)\n",
            "\n",
            "Target Data\n",
            "y train shape:  (1196, 1)\n",
            "y test shape (298, 1)\n",
            "y valid shape (298, 1)\n"
          ]
        }
      ],
      "source": [
        "# importing features and train files\n",
        "x_train = pd.read_csv('/content/drive/MyDrive/data_project_cuetessa/X_train.csv', index_col = 'Unnamed: 0')\n",
        "x_test =  pd.read_csv('/content/drive/MyDrive/data_project_cuetessa/X_test.csv', index_col = 'Unnamed: 0')\n",
        "x_valid =  pd.read_csv('/content/drive/MyDrive/data_project_cuetessa/X_valid.csv', index_col = 'Unnamed: 0')\n",
        "y_train = pd.read_csv('/content/drive/MyDrive/data_project_cuetessa/Y_train.csv', index_col = 'Unnamed: 0')\n",
        "y_test =  pd.read_csv('/content/drive/MyDrive/data_project_cuetessa/Y_test.csv', index_col = 'Unnamed: 0')\n",
        "y_valid =  pd.read_csv('/content/drive/MyDrive/data_project_cuetessa/Y_valid.csv', index_col = 'Unnamed: 0')\n",
        "\n",
        "print('Feature Data')\n",
        "print('x train shape: ', x_train.shape)\n",
        "print('x test shape', x_test.shape)\n",
        "print('x valid shape', x_valid.shape)\n",
        "print()\n",
        "print('Target Data')\n",
        "print('y train shape: ', y_train.shape)\n",
        "print('y test shape', y_test.shape)\n",
        "print('y valid shape', y_valid.shape)\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "Below is a least of every feature considered for model training."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "fFjpeMwlORyM",
        "outputId": "eee76956-69c1-426a-b292-c8c1b31cfaa2"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "array(['chroma_mean', 'chroma_var', 'chroma_meandif', 'chroma_vardif',\n",
              "       'rmse_mean', 'rmse_var', 'rmse_meandif', 'rmse_vardif',\n",
              "       'spec_cent_mean', 'spec_cent_var', 'spec_cent_meandif',\n",
              "       'spec_cent_vardif', 'spec_bw_mean', 'spec_bw_var',\n",
              "       'spec_bw_meandif', 'spec_bw_vardif', 'rolloff_mean', 'rolloff_var',\n",
              "       'rolloff_meandif', 'rolloff_vardif', 'zcr_mean', 'zcr_var',\n",
              "       'zcr_meandif', 'zcr_vardif', 'mfcc0_mean', 'mfcc0_var',\n",
              "       'mfcc0_meandif', 'mfcc0_vardif', 'mfcc1_mean', 'mfcc1_var',\n",
              "       'mfcc1_meandif', 'mfcc1_vardif', 'mfcc2_mean', 'mfcc2_var',\n",
              "       'mfcc2_meandif', 'mfcc2_vardif', 'mfcc3_mean', 'mfcc3_var',\n",
              "       'mfcc3_meandif', 'mfcc3_vardif', 'mfcc4_mean', 'mfcc4_var',\n",
              "       'mfcc4_meandif', 'mfcc4_vardif', 'mfcc5_mean', 'mfcc5_var',\n",
              "       'mfcc5_meandif', 'mfcc5_vardif', 'mfcc6_mean', 'mfcc6_var',\n",
              "       'mfcc6_meandif', 'mfcc6_vardif', 'mfcc7_mean', 'mfcc7_var',\n",
              "       'mfcc7_meandif', 'mfcc7_vardif', 'mfcc8_mean', 'mfcc8_var',\n",
              "       'mfcc8_meandif', 'mfcc8_vardif', 'mfcc9_mean', 'mfcc9_var',\n",
              "       'mfcc9_meandif', 'mfcc9_vardif', 'mfcc10_mean', 'mfcc10_var',\n",
              "       'mfcc10_meandif', 'mfcc10_vardif', 'mfcc11_mean', 'mfcc11_var',\n",
              "       'mfcc11_meandif', 'mfcc11_vardif', 'mfcc12_mean', 'mfcc12_var',\n",
              "       'mfcc12_meandif', 'mfcc12_vardif', 'mfcc13_mean', 'mfcc13_var',\n",
              "       'mfcc13_meandif', 'mfcc13_vardif', 'mfcc14_mean', 'mfcc14_var',\n",
              "       'mfcc14_meandif', 'mfcc14_vardif', 'mfcc15_mean', 'mfcc15_var',\n",
              "       'mfcc15_meandif', 'mfcc15_vardif', 'mfcc16_mean', 'mfcc16_var',\n",
              "       'mfcc16_meandif', 'mfcc16_vardif', 'mfcc17_mean', 'mfcc17_var',\n",
              "       'mfcc17_meandif', 'mfcc17_vardif', 'mfcc18_mean', 'mfcc18_var',\n",
              "       'mfcc18_meandif', 'mfcc18_vardif', 'mfcc19_mean', 'mfcc19_var',\n",
              "       'mfcc19_meandif', 'mfcc19_vardif'], dtype=object)"
            ]
          },
          "execution_count": 4,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "x_train.columns.values"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "These are the Pearon correlation values for every feature."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1kzl_cQwT0AB",
        "outputId": "9078c53a-6b65-42e1-ac95-45f06cfbc80f"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "mfcc0_mean        0.564614\n",
              "rolloff_mean      0.547272\n",
              "spec_cent_mean    0.524815\n",
              "spec_bw_mean      0.509111\n",
              "chroma_vardif     0.411423\n",
              "zcr_mean          0.384039\n",
              "mfcc13_mean       0.363900\n",
              "mfcc7_mean        0.355339\n",
              "mfcc9_mean        0.349365\n",
              "mfcc16_vardif     0.343538\n",
              "mfcc15_mean       0.334761\n",
              "mfcc17_vardif     0.332201\n",
              "mfcc14_vardif     0.330344\n",
              "mfcc17_mean       0.327830\n",
              "mfcc15_vardif     0.326495\n",
              "mfcc11_mean       0.321137\n",
              "mfcc18_vardif     0.320691\n",
              "mfcc13_vardif     0.312322\n",
              "mfcc8_vardif      0.304462\n",
              "mfcc12_vardif     0.302653\n",
              "Name: valence_mean, dtype: float64"
            ]
          },
          "execution_count": 5,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "x_train.join(y_train).corr()['valence_mean'].sort_values(ascending=False)[1:].head(20)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "The graph below shows that data has been correctly distributed between training, testing and validation datasets."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 446
        },
        "id": "gh-ifkVIQbZ0",
        "outputId": "3f52c28b-ab62-4bde-9d30-8ef9efd2f9ff"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 684x684 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "plt.figure(figsize=(9.5,9.5))\n",
        "sns.violinplot(data=[y_train, y_valid, y_test])\n",
        "ax = plt.gca()\n",
        "ax.set_xticklabels(labels=['train', 'validation', 'test'])\n",
        "ax.set_title('Violin plots for valence scores of DEAM')\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "DvFlqRLeO70D",
        "outputId": "4f6bbf63-022f-4e9d-903d-ab59c7f01f1f"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Feature Data\n",
            "x train shape:  (1084, 156)\n",
            "x test shape (272, 156)\n",
            "x valid shape (270, 156)\n",
            "\n",
            "Target Data\n",
            "y train shape:  (1084, 1)\n",
            "y test shape (272, 1)\n",
            "y valid shape (270, 1)\n"
          ]
        }
      ],
      "source": [
        "# importing features and train files\n",
        "spotify_x_train = pd.read_csv('/content/drive/MyDrive/data_project_cuetessa/spotify_X_train.csv', index_col = 'Unnamed: 0')\n",
        "spotify_x_test =  pd.read_csv('/content/drive/MyDrive/data_project_cuetessa/spotify_X_test.csv', index_col = 'Unnamed: 0')\n",
        "spotify_x_valid =  pd.read_csv('/content/drive/MyDrive/data_project_cuetessa/spotify_X_valid.csv', index_col = 'Unnamed: 0')\n",
        "spotify_y_train = pd.read_csv('/content/drive/MyDrive/data_project_cuetessa/spotify_Y_train.csv', index_col = 'Unnamed: 0')\n",
        "spotify_y_test =  pd.read_csv('/content/drive/MyDrive/data_project_cuetessa/spotify_Y_test.csv', index_col = 'Unnamed: 0')\n",
        "spotify_y_valid =  pd.read_csv('/content/drive/MyDrive/data_project_cuetessa/spotify_Y_valid.csv', index_col = 'Unnamed: 0')\n",
        "\n",
        "print('Feature Data')\n",
        "print('x train shape: ', spotify_x_train.shape)\n",
        "print('x test shape', spotify_x_test.shape)\n",
        "print('x valid shape', spotify_x_valid.shape)\n",
        "print()\n",
        "print('Target Data')\n",
        "print('y train shape: ', spotify_y_train.shape)\n",
        "print('y test shape', spotify_y_test.shape)\n",
        "print('y valid shape', spotify_y_valid.shape)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "Since valence is prone to subjective error, it is valid to analyze distributions of both datasets.\n",
        "The probability distributions for both datasets are very different from each other.\n",
        "Spotify is reportedly considered less reliable."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 446
        },
        "id": "BwldasyzQaMa",
        "outputId": "93b49643-e51c-4f4a-a6f2-fda9e12172e0"
      },
      "outputs": [
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjcAAAIzCAYAAAD8uTi4AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdd3hc1YH+8e+RRmXUZUnuNsbgguk1lU3CJptOssnu/jZkE1rwJhhI6IaYBEgghCyQZQmkmVBMJwnNDhgbGzAuuOIq2ZaLLFu2ZPU69fz+mBkjZMuS1e7cmffzPHoeaWY0885cSfPq3HPPNdZaRERERBJFitMBRERERAaSyo2IiIgkFJUbERERSSgqNyIiIpJQVG5EREQkoajciIiISEJRuRFxkDHmu8aY+UP8mLuMMZ8fysdMVMaYTcaYz0Y/N8aYvxhj6o0x7zuYaYIxxhpjPE5lEHGayo0kBGPMRcaYVcaYFmNMlTHmH8aYTzudqyfW2qestf/idI7eckMxMsZ8wxizzhjTZIw5aIx5yxhz/ADc72PGmF92vsxae7K1dnH0y08DXwDGWmvP6+/jDZboNmw3xjQbYxqMMUuNMT80xqR0us1jxhh/9Pcp9vFBl/vJiV7+j24ew2+MKe5y+dpo8ZowWM9PBFRuJAEYY64DfgvcDYwAxgMPA99wMldP9J/1wDPGnAg8AVwP5APHA78DQkPw8McBu6y1rUPwWP31dWttLpHM9wA3A7O73OZea21Op4/Tu1z/bcAHfMEYM/IIj7ET+E7sC2PMqUDWgD0DkaNQuRFXM8bkA3cCM6y1f7PWtlprA9baV621N0Zvk2GM+a0xZl/047fGmIzodZ81xlQaY24yxlRHR32+aYz5ijFmqzGmzhhza6fHu90Y86Ix5rnof75rjDGnd7p+pjGmPHrdZmPMv3a67hJjzHvGmAeMMbXA7dHLlkSvN9HrqqOjDhuMMafEnqcx5gljTI0xZrcxZlbsP+3YfRhj/ie6S2SnMebLPbx050bz1Ud3pWR2yvm16MhH7L/606KXP0mkOL4a/Y/9JmPM48aY66PXj4n+Vz4j+vUJ0dcv5Wj3G71utDHmr9Hnt9MYc02X1/z56PNvNpFdQed087zOAHZaaxfaiGZr7V+ttRW93H4nGWMWRzNuMsZcGL18OvBd4Kboc381evkuY8znjTGXA38GPhG9/g/GmDZjTFGn+z4r+vzSuoY2xpxnjFkWfdwqY8xDxpj0TtdbExld2Ra9ze+MMSZ6XWp02x80xuwAvtrDtj/EWttorX0F+H/AxbGft166GPg9sB74ryNc/yTw/S63f+IY7l+k76y1+tCHaz+ALwFBwHOU29wJLAeGAyXAUuAX0es+G/3+nwFpwBVADfA0kAucDLQDx0dvfzsQAP4tevsbiPyHmha9/t+B0UT+cfh/QCswKnrdJdHHuhrwAN7oZUui138RWA0UAAY4qdP3PgG8HM00AdgKXN7pfgPR7KnAj4B9gOnm9dgFbATGAcOA94BfRq87E6gGPha9r4ujt8/o9L2f73RflwGvRj+/CCgHnut03cs93W/0tVod3QbpwERgB/DFTq95B/CV6Pf+CljezXObGL3tA8DngJwu13e7/aIf24FbozkuAJqBKdHvfSz2OnV5LT/faTss6XTdPOBHnb5+APi/bnKfDXw8+nMxAdgC/KTT9RZ4LfqzMZ7Iz+iXotf9ECjttD0XRW9/xN+Jrtuw0+UVsbxHeq5dbnscEAamERklW3+kxwDKiPwcpwKV0e+zwASn/3boI7E/NHIjblcEHLTWBo9ym+8Cd1prq621NcAdwPc6XR8A7rLWBoBngWLgf23kv/5NwGag85D8amvti9Hb3w9kEnljwlr7grV2n7U2bK19DtgGdJ5/sc9a+3/W2qC1tr1LzgCR8jKVSDHZYq2tMsakAv8J3BLNtAu4r8tz2G2t/ZO1NgQ8DowisouuOw9Za/dYa+uAu/hw98F04A/W2hXW2pC19nEiux4+3s39vA18Ojo680/AvcCnotd9Jnp9T/d7LlBirb3TWuu31u4A/hR9zjFLrLXzos/vST66PQ6Jfu9ngTHA88BBE5k/ktPpZt1tv48DOcA90RxvESkU36FvHic6ohHdht+JZj9S7tXW2uXRn4tdwB+IvH6d3WOtbbCRUahFREapAP4D+G2n7fmrPubdR6QcxdwQHSWKfTze6brvESk0m4n8zpxsjDnzCPcZG735ApHCtreP2USOicqNuF0tUGyOPn9lNLC709e7o5cduo/omyZERmkADnS6vp3Im17Mntgn1towkf9IRwMYY77faddLA3AKkbJ02Pd2FX0zfYjIHJFqY8wfjTF50e9PO8JzGNPp6/2d7qct+mnnzF11ztH59TgOuL7zmxqREYHRXe8g+ljlREanzgDOJ1IG9hljpvDRcnO0+z0OGN3lulv5aDnb3+nzNiCzu20eLQn/Ya0tiWb6J+CnR3ruXbbfaGBP9LLOr03n1/lYvAxMM5HJzF8AGq21RzyKyhgz2RjzmjFmvzGmicj8seIuN+v6GsS272gO3559MQao6/T1/1hrCzp9XNzpuu8DTwFYa/cS2c6dr495ksiI3iVol5QMIZUbcbtlREYAvnmU2+wj8gYaMz56WV+Ni30SHbEYS+QN/TgiIw5XAUXW2gIiu39Mp++1R7tja+2D1tqziQz3TwZuBA4SGdXp+hz681/wuE6fd3499hAZxer8ppZlrX3mKPnfJrKbJ73LG10hsK4X97uHyDyZztflWmu/0o/nFwlr7Urgb0RK5mHPvfP2i36MM52OGuKjr/NRt90RHruDyOjRfxEZ6TjiqE3UI0R2LU2y1uYRKXfmKLfvrIrDt+cxMcacS6TcLOnFbT8JTAJuiZax/UR2N17UtXBaa3cT2e33FSLbQWRIqNyIq1lrG4nM1fidiUwEzjLGpBljvmyMuTd6s2eAWcaYEhM5NPVnwJx+POzZxphvRf+Q/4RIuVoOZBN5A6wBMMZcykffVI/KGHOuMeZj0QmnrUTmjoSjo0rPA3cZY3KjJeq6fj6HGcaYscaYYURGNZ6LXv4n4IfRHMYYk22M+aoxJjd6/QEi81o6e5tIoXsn+vXi6NdLOo2IHe1+3weajTE3G2O80Qmyp0TfcI+JMebTxpgrjDHDo19PBS4ksn1iutt+K4iMiNwU/Rn6LPB1IrtdunvuPXmCyKjFhRy93OQCTUBLNPOPjuExngeuiW7PQmBmb7/RGJNnjPkakec4x1q7oRffdjHwJpECfkb04xQic8iONJH9cuAC646jyCRBqNyI61lr7yPyZj+LSLHYQ+TN9aXoTX4JrCJyVMcGYE30sr56mchk4Xoi/5F/y0aO0NpMZC7MMiJvhKcSmazbW3lESkA9kV0LtcBvotddTaTw7CDy3/XTwKP9eA5PA/Oj91dO9PWw1q4iMjH5oWiO7UTenGN+RaQoNhhjbohe9jaRN+dYuVlC5JDf2NdHvd9oAfoa0SOdiIxU/ZnIodzHqoFIkdhgjGkBXgf+TmQuUEx3289PpMx8OZrhYeD71trS6PfNJrKbqcEY8xK9YK19j8jE2zXRUYzu3EBk900zkZ+B545y267+BLwBfEDkZ7s3IySvGmOaifyu/JTI3KNLu9wmdmRY7OOgiRxV9x9EJkbv7/Sxk0h5O2zXlLW2PLr9RYaMsfaYRlpFkpox5nbgRGvtkQ59lTjnxPYzxrwFPG2t/fNQPaZIstMiYiIigyS6a+0s4nxBSZFEo91SIiKDIHro9AIi69U0O51HJJlot5SIiIgkFI3ciIiISEJxbM5NcXGxnTBhglMPLyIiIi62evXqg9HFOg/jWLmZMGECq1bp6EARERE5dsaYbpdX0G4pERERSSgqNyIiIpJQVG5EREQkoajciIiISEJRuREREZGEonIjIiIiCUXlRkRERBKKyo2IiIgkFJUbERERSSgqNyIiIpJQVG5EREQkoajciIiISEJRuREREZGEonIjIiIiCUXlRkRERBKKyo2IiIgkFJUbERERSSgqNyIiIpJQVG5EREQkoajciIiISEJRuREREZGEonIjcWH16tUsWrQIa63TUURExOU8TgcQCQaDXH/99QCMGTOGyZMnO5xIRETcTCM34riOjo5Dn7e0tDiYREREEoHKjTiuvb39iJ+LiIj0hcqNOK5zoek8iiMiItIXKjfiuM6FRiM3IiLSXyo34rjW1tYjfi4iItIXKjfiuM6TiDWhWERE+kvlRhzX3Nx8xM9FRET6QuVGHHdotCYtUyM3IiLSbyo34rjm5mYwhnB6jkZuRESk31RuxHENDQ2YtEzCqRnU1dc7HUdERFxO5UYcV1dXh03zYtO81NXVOR1HRERcTuVGHFdXV0cwNZNwmpeGhgadPFNERPpF5UYcd7C2DuuJjNyEgkFNKhYRkX5RuRFHWWtpaKjHpmVi07wA2jUlIiL9onIjjmpsbCTg9xNOz8GmZwNw4MABh1OJiIibqdyIo/bv3w+AzcghnJ4DqNyIiEj/qNyIo2JFJpyRg03PApOiciMiIv2iciOOio3chNNzwKRgMrIPXSYiItIXKjfiqKqqKownA1LTAQik5bB3716HU4mIiJup3IijKioqCGXmgTEAhDPz2V2xR2vdiIhIn6nciKN27tpFKCP/0Ndhbz5trS3U6zQMIiLSRyo34pjW1lbq6+oIezuVm8wCIDKiIyIi0hcqN+KYPXv2AB8WmsjnkaKze/duRzKJiIj7qdyIY8rLywEIeT8sNzY9G+NJZ+fOnU7FEhERl1O5Ecds27YN40nHZuR9eKExBL3DKCsrcy6YiIi4msqNOKasrIygd9ihI6ViQllFbC8vJxgMOpRMRETcTOVGHBEMBtleXk4oq+iw60JZRQT8/kNzckRERI6Fyo04oqKigoDfTyhr2GHXhbMjhWfr1q1DHUtERBKAyo04YtOmTQCEcoYfdl04Mx/jyWDjxo1DHUtERBJAj+XGGPOoMabaGHPEdxoT8aAxZrsxZr0x5qyBjymJZsOGDZh070cnE8eYFALZJXywfv3QBxMREdfrzcjNY8CXjnL9l4FJ0Y/pwCP9jyWJbt0H6/FnDz9sMnFMKGcEFbt309TUNMTJRETE7XosN9bad4C6o9zkG8ATNmI5UGCMGTVQASXx1NTUUH1gP6Gckd3eJpQ7AkC7pkRE5JgNxJybMUDnw1oqo5cdxhgz3RizyhizqqamZgAeWtxofXR3U6zAHEkouxhSUlm3bt1QxRIRkQQxpBOKrbV/tNaeY609p6SkZCgfWuLIqlWrMGkZhI9wpNQhKR5COcNZuWrV0AUTEZGEMBDlZi8wrtPXY6OXiRzGWsuK99/HnzMKzNF//IJ5o9m5Ywe1tbVDlE5ERBLBQJSbV4DvR4+a+jjQaK2tGoD7lQS0a9cu6mprCeUfcc/lRwTzIrdZvXr1YMcSEZEE0ptDwZ8BlgFTjDGVxpjLjTE/NMb8MHqTecAOYDvwJ+DKQUsrrrcqupspVlyOJpxVhEnzsnLlysGOJSIiCcTT0w2std/p4XoLzBiwRJLQli1bDt4CbEZOzzc2Bn/uKJYvX0EoFCI1NXXwA4qIiOtphWIZMs3Nzaz7YB2+/HE93zgqWDie5uamQysai4iI9ETlRobMihUrCIdCBAuP6/X3BPPHQkoKS5YsGcRkIiKSSFRuZMgsWbIEk55FOPsYlgFITSeYO5p33n2XyB5QERGRo1O5kSHh9/tZtnw5vryx3Z5yoTvBgvHsr6pi586dg5ROREQSicqNDImVK1fi6+g4pl1SMcGC8QC8/fbbAx1LREQSkMqNDIm33noLk5ZJqJtDwDMqlpNRsfyI19n0LEK5o1j41lvaNSUiIj1SuZFB19HRwZL33sNXMB5Sjvwjl9JWR0pb9+dnDQw7nso9e9ixY8dgxRQRkQShciODbvny5ZFdUsMm9vk+goXHgTG89dZbA5hMREQSkcqNDLpFixZh0r2Eckf2+T5smpdQ7igWLFyoXVMiInJUKjcyqJqbm3nvvaX4Co7v8USZPfEXncCB/fu1oJ+IiByVyo0MqsWLFxMMBggUn9jv+woWTsCkpvHGG28MQDIREUlUKjcyqF5//Q1sViHhrKL+31lqGv6C8Sx86y18Pl//709ERBKSyo0Mmr1797Jp00b8wyYe88J93QkUnUhbayvLli0bkPsTEZHEo3Ijg2b+/PkABIadMGD3GcobBRnZ/OP11wfsPkVEJLGo3MigCIVCvDZ3HsG8MdiMnIG7Y5OCr/AE3l+xgoMHDw7c/YqISMJQuZFBsXbtWmoP1hAomTTg9x0omYS1VhOLRUTkiFRuZFDMmzcPk5ZJsODYzyXVE5uZTyh3JK/Nnac1b0RE5DAqNzLgmpqaeOedd/EVToSU1EF5DH/xJKr27WXDhg2Dcv8iIuJeKjcy4N58883I2jaDsEsqJlg4AeNJ57XXXhu0xxAREXdSuZEBZa3llVdfJZxTMjBr23QnNQ1f4fEsWryY5ubmwXscERFxHZUbGVCbN29m965d+IsnD/pjBUqmEPD7WbBgwaA/loiIuIfKjQyoV199FZOaRqAfZwDvrXB2MTa7iFdefVUTi0VE5BCVGxkwLS0tvPXWInyFx0Nq2pA8pq94Mjt37KC0tHRIHk9EROKfyo0MmIULF+L3+wiUTBmyxwwUnYBJ9WhisYiIHKJyIwPCWstLL7+MzS4inF08dA+cmo6vcCILFiyktbV16B5XRETilsqNDIjS0lJ27tiBr3jKgJ0ks7cCJVPw+To0sVhERACVGxkghyYSFw3+ROKuYhOLX37lFU0sFhERlRvpv9bWVhYsWBidSJw+9AGMwVc8mR3l5ZpYLCIiKjfSfwsWLBjyicRdRSYWp2lisYiIqNxI/1hrI7uDhnoicVep6fgKJ7BgwULa2tqcyyEiIo5TuZF+KSsrY0d5Ob7iyUM+kbir2MTihQsXOppDREScpXIj/fLaa69FVyQ+wekohLNLsFnDePmVV5yOIiIiDlK5kT5ra2vjzTcX4C+cAB4HJhJ3FZ1YvH3bNrZu3ep0GhERcYjKjfTZW2+9hc/Xgd/BicRdBYpOwKR4mDt3rtNRRETEISo30mevvvYaNquQcHaJ01E+5MnAX3gc8+e/SUdHh9NpRETEASo30ic7duygrLQUX9EkxycSdxUomUJ7exuLFy92OoqIiDhA5Ub6ZO7cuZCSQrDoRKejHCaUMwK8+bz6qta8ERFJRio3csz8fj+vvzGfQP54bFqm03EOZwy+okls2rSRiooKp9OIiMgQU7mRY7Z06VJaW5oJlEx2Okq3AkUngjG8/vrrTkcREZEhpnIjx2zu3HmQkUMob7TTUbpl07MI5o9l3j9eJxgMOh1HRESGkMqNHJPq6mpWrVqJb9gJYOL7xydQNImG+jpWrVrldBQRERlC8f3uJHFn/vz5WGsJFE9yOkqPggXjMGmZzJs3z+koIiIyhFRupNestfzj9dcJ5Y7EZuY5HadnKan4hk3kvaVLaWpqcjqNiIgMEZUb6bXS0lL2VlZGJuu6RKBoEqFgkEWLFjkdRUREhojKjfTaG2+8gUnxECic4HSUXgtnDcNmDeP1N95wOoqIiAwRlRvplUAgwJsLFuIvGBcfJ8nsLWPwDTuBLZs3U1lZ6XQaEREZAio30isrVqyIrG3jol1SMcGiE8AY5s+f73QUEREZAio30isLFizApHsJ5Y9xOsoxs+lZhHJH8eabC7DWOh1HREQGmcqN9KitrY333nsPX8GEuF/bpjv+YROpqtpHWVmZ01FERGSQufOdSobUkiVLCAQCBIpOcDpKnwULJ2BSUlmwYIHTUUREZJCp3EiP5r/5JmTmEs4ucTpK33nS8eePZcGChYRCIafTiIjIIFK5kaNqaGhgzerV+AqPB2OcjtMvwWETaWioZ926dU5HERGRQaRyI0f17rvvEg6HCQ473uko/RbMH4dJTWPx4sVORxERkUGkciNH9daiReDNJ+wd5nSU/kv14M8fy+LFb+tM4SIiCUzlRrpVX1/PurVro0dJuXuXVExw2PE0Nzdp15SISAJTuZFuvfPOO1hrE2KXVEwwf6x2TYmIJDiVG+nW22+/E90lVeh0lIGT4sGfP463335HR02JiCQolRs5oqamJtatW4uv4LiE2SUVEyw8jubmJjZs2OB0FBERGQQqN3JES5cujRwlVXCc01EGXDB/LKSk8u677zodRUREBoHKjRzRu+++Cxk5hLOLnY4y8FLTCOSNYfHb7+hcUyIiCUjlRg7T3t7O+++/jz9/fMLtkooJFh5H7cEanWtKRCQBqdzIYVatWkUgECBYON7pKIMmWDAOjOG9995zOoqIiAwwlRs5zNKlSzGeDEI5I52OMng8mYRyRrBE5UZEJOGo3MhHhEIhlry3FH/eGEhJ7B+PQP44du7Ywf79+52OIiI92LVrF6tXr3Y6hrhEYr97yTErLS2luakxstsmwQULIrvdli1b5nASEenJ9OnTuf7669mzZ4/TUcQFVG7kI5YuXQrGRA6XTnDWmw/efO2aEnEBv98PQF1dncNJxA1UbuQjli5bRihnBHgynI4yJPx5Y1m3bh3t7e1ORxGRbnQ+0W1LS4uDScQtVG7kkJqaGnbu2JEUozYxwfyxhIJB1q5d63QUEelGc3Pzoc8bGxsdTCJuoXIjh7z//vsASVVuQrkjMalph567iMSf2traI34u0h2VGzlkxYoVmIycxDpRZk9SUvHnjmLp0mVarVgkTqncyLFSuREgsk975cpV+PJGJ+yqxN0J5Y+luvoAFRUVTkcRkSOoqakBIDPVUl1d7XAacYNelRtjzJeMMWXGmO3GmJlHuH68MWaRMWatMWa9MeYrAx9VBtOWLVtob28jlJc8u6RigvljALSGhkic2rt3L6kpMLUgwL69lU7HERfosdwYY1KB3wFfBqYB3zHGTOtys1nA89baM4H/BB4e6KAyuFatWhU5BDxvlNNRhpzNyAVvPu+/v9LpKCJyBPv27aPEaxmRFaKqqopwOOx0JIlzvRm5OQ/Ybq3dYa31A88C3+hyGwvkRT/PB/YNXEQZCivef59wdknSHALelT93FGvXriUQCDgdRUS62LmjnFHeAKOzQvj8Aa0qLj3qTbkZA3ReErIyellntwP/ZYypBOYBVw9IOhkSzc3NlJWWEsgb7XQUx4TyxuDzdbBp0yano4hIJz6fj8rKvYzPCTI+NwRAeXm5w6kk3g3UhOLvAI9Za8cCXwGeNMYcdt/GmOnGmFXGmFWxCWLivHXr1mGtJZTE5SaYOwqMYc2aNU5HEZFOdu7cSdhaxuWEGJsdxBjYvn2707EkzvWm3OwFOp9oaGz0ss4uB54HsNYuAzKB4q53ZK39o7X2HGvtOSUlJX1LLANu7dq1mNQ0QtlJvE086YSzS1i9WuVGJJ7ERlNPyAuSkQpjc8Js2rjR4VQS73pTblYCk4wxxxtj0olMGH6ly20qgH8GMMacRKTcaGjGJVatXk0gZzikpDodxVGB3JFsKd1CW1ub01FEJGr9+vUUeaEoMzKJeHKen02bNn7klAwiXfVYbqy1QeAq4A1gC5GjojYZY+40xlwYvdn1wBXGmA+AZ4BLrFZEc4Xa2loqdu8mlJt8R0l1FcodRTgUYsOGDU5HEREgHA6z/oN1TM7zHbpscn6A9g6fdk3JUXl6cyNr7TwiE4U7X/azTp9vBj41sNFkKKxbtw6AYBLPt4kJ5YyAlFTWrFnDxz72MafjiCS9bdu2Ud/QyKkn+Q9ddvKwAIbIiupTp051LpzENa1QnOTWrVuH8aQTzhrmdBTnpXoIZRezNlr4RMRZK1aswACnFX24RENeumVifohly5Y6F0zinspNklu77gMC2cPh8IPbklIwZyTbt23TvBuROPDuO+8wMT9EXvpHZzmcMcxHWVmZTsUg3dI7WhKrr6+nck8FodwRTkeJG6HckYTDYa13I+Kw3bt3s237dj42vOOw6z42woe18NZbbzmQTNxA5SaJxSbOBnNGOpwkfoRyhoMxfPDBB05HEUlqCxYswBj4+HDfYdeNzAozMS/EgjfnO5BM3EDlJomtX78ek+ohnH3YkkTJKzWNcFaxyo2Ig4LBIP+YN5dTCgMUZBz5wNtPjexge/kOSktLhziduIHKTRLbsGEjwazipF/fpqtgznBKS0u1joaIQ5YsWcLB2jo+P7a929t8eqSPTA/87W9/G8Jk4hYqN0nK5/Oxbfs2gtnDnY4Sd0I5wwkEAlpHQ8QhL77wAsOzLKd3OkpqztYs5mzNOvS112P59IgOFr21kLq6OidiShxTuUlSZWVlhEMhQrkqN12FciKvyUYt8S4y5NauXcvGTZv4lzFtpJgPL69o8VDR8tGl2b4wrp1gKMRzzz03xCkl3qncJKnY0UBhjdwcxqZnQ2aOjpgSGWLWWv7yl0cpzITPjj78KKmuRmWF+eQIHy/9/W/U1tYOQUJxC5WbJLVp0ybw5mPTMp2OEpcCWSVs0MiNyJBatWoV69dv4OvjW0jv5VTAb05oIxAI8NRTTw1uOHEVlZsktXnzFgLeIqdjxK1QdjEHa2qor693OopIUggEAvzfg//L8CzLZ0Yffvh3d0Zkhfns6A5eeuklduzYMYgJxU1UbpJQbW0tdXW1hLJLnI4St8LR10aHmYoMjRdeeIGKPZV878Rm0o7xnenfJraR5QnzwP33oXM2C6jcJKWysjIArW9zFKGsIjDm0GslIoNn7969PP74Y5xV7Of04kDP39BFTprlPya2sGHjJubNm9fzN0jCU7lJQqWlpWAMIZ0ss3upaVhvgUZuRAZZMBjkrrt+SWrYz39Nbu3z/fzTKB8nFQZ56P8eZO/evQOYUNxI5SYJbd26FestgNQ0p6PEtaC3iNKyrU7HEEloTz/9NJs3b+HiSc0UZ4b7fD8pBq44qRkT8nHXXb/UIpxJTuUmCZVt3UbQq1GbnoSyhtFQX6dJxSKDZO3atTz22GN8YoSPT4z09/v+ijPDXDypmc2btzB79uwBSChupXKTZBoaGqivqyWkctOjcFbkaLLy8sjT2NEAACAASURBVHKHk4gknv3793P7z3/GSG+IS6b0fXdUV58Y6edzozt45plnWLRo0YDdr7iLyk2Sib1RhzXfpkexOUk6DYPIwOro6GDWrJ/ib2/mx6c24PUM7BFO35vcyqSCIPfc8yu2bds2oPct7qByk2RUbo6BJwOTkaNyIzKAgsEgd95xB+Xby/nRSU2Myur7PJvueFLg6pObyDJ+bpl5MwcOHBjwx5D4pnKTZHbt2oVJ92LTvE5HcYVAZgE7du50OoZIQrDWcv/997N02TK+N7mFM/pw2HdvFWRYrj+tgdamOm664XqampoG7bEk/qjcJJmdO3cSzChwOoZrhDML2FOxh1Ao5HQUEdebPXs28+bN48IJbXx+bO9XIe6r8TkhfnJKI3v3VjJz5s20t7cP+mNKfFC5SSLWWnbt2k3Im+90FNcIeQsIBPzs37/f6Sgirvbkk08yZ84cPjOqg28fP3Ql46TCID+a1syWLVu49ZZb6Ojo+YSc4n4qN0nk4MGDtLe3Ec7UyE1vhb2R12r37t0OJxFxr6effprZs2fzyRE+Lp3aijFD+/jnDvczfWoz69at46e33orPN/ijRuIslZskEnuDjr1hS89iRVDlRqRvnn32Wf74xz/y8RE+pk9rIWWIi03Mp0b5+cFJLaxZs4ZZs36qgpPgVG6SSGxJ8nCmdkv1micdk55FZWWl00lEXOeJJ57g97//PecN9/HfJzlXbGLOH+XjsqktrFq5SnNwEpzKTRLZu3cvJtWDTctyOoqrBNNzqKzUuWpEestay5///GceffRRPjXSx4+mtZAaJ+82nxntY/pJkV1UN95wA62tA7eAoMSPOPlxk6FQWVlJOCOPId/h7XLhjDwq9uxxOoaIK4TDYR566CHmzJnD50Z3cMVJ8VNsYj41ys+Mk5vZsnkT1137ExobG52OJAMszn7kZDBV7KkkmJ7jdAzXCWfmUV9Xq6MsRHoQDAa59957+etf/8oXx7VzyZRWx3dFdee84X6uObWJHdu3cc3VV1FTU+N0JBlAKjdJIhwOs39/FTYzz+korhPOyAVg3759DicRiV9+v587br+d119/nW8d38ZFJ7bF/SDxmcUBbji9kQP7Krnmqhn6HU8gKjdJor6+nmAgQFgjN8csVm60hLvIkbW1tXHLzJm8u2QJ/zWplW8e3x73xSbmpMIgM8+op7m+mqtmXMmOHTucjiQDQOUmScQWoYu9UUvv2WghVLkROVxjYyPXX3cta9eu4YqTWviXce7bfTsxL8RPz2zAtjdwzdVXsWnTJqcjST+p3CSJ2Buz1cjNMbNpXkhJ1SrFIl3U1NRwzdVXsX1rGdec0sT5o9y7dsyY7BCzzqwn27Zy3XXXsnLlSqcjST+o3CSJD0duVG6OmTGQkaORG5FOKisrufqqGRzYt4frT2/krJLBOwnmUCnxhpl1Vj0j0ju4ZeZMFi9e7HQk6SOVmyRRXV2NScuA1DSno7hSMC2LqiqN3IgAlJeXc/VVM2ipr2bmGQ1MKww6HWnA5KdbbjmjgeNzfdx5xx3MnTvX6UjSByo3SaKmpgabnu10DNeyadlU61BRETZt2sSPr7kaOhr56ZkNTMwLOR1pwGWnWW46vZFThvn5zW9+w3PPPed0JDlGKjdJ4kB1NUGP1+kYrhVOz6ahvo5QKPH+kIv01urVq7n+umvJsq3MOrOeMdmJ+/uQkQo/ObWJ84b7eOSRR3j00Uex1jodS3pJ5SZJ1NQcJJymkZu+sunZhMNh6uvrnY4i4oj33nuPmTffRHFaB7POrKfEG3Y60qDzpMCVJ7fwmVEdPPHEEzz00EOEw4n/vBOByk0SCAaDNDU2YNN1Tqm+Ckdfu+rqaoeTiAy9hQsXcttttzEuy8+tZzZQkJE8IxgpBi6b2soXx7Xz17/+lXvvvVcjuC7gcTqADL7a2lqstZpz0w+xk43W1tY6nERkaL366qvcf/99TMkPcu1pjXiT8F3DGLjoxDa8qZaXXn+d9vZ2Zs2aRVqaDtCIV0n4Y5p8Ym/IYZ0NvM9io14qN5JMnnvuOR555BFOKwpwzSlNpKc6ncg5xsC3Jrbj9VieefttOjraufPOX5CRkeF0NDkC7ZZKArE3ZKty02fWkwnGqNxIUrDW8pe//IVHHnmE84b7+MmpyV1sOvvy+A4undLC+yve56Ybb6S1tdXpSHIEKjdJ4ODBgwCac9MfJgWTnqVyIwkvHA7z4IMP8vjjj3P+qA6uPLkFj94pPuJzY3z897RmNmxYz3XX/oSGhganI0kX+pFNArW1tWAM1qPh0/4IebyHiqJIIgoGg9xzzz38/e9/54vj2rl8aispLjkB5lD75Eg/Pz61iR3l2/jx1VdRo3Ww4orKTRKora3FpGeBic/NnVGxnNS2WlLbavGWziOjYrnTkY4onOalRuVGEpTP5+NnP/sZ8+fP51vHt3HRiW0qNj04szjAjac1Ul1VyVUzrqSiosLpSBIVn+92MqBqa2sJxfECfiltdZhQABMK4GneT0pbndORjiicpt1Skpiampq47rprWbZ0Kd+b3Mo3j2/HqNj0ytTCIDef0UB7Yw1XzbiSzZs3Ox1JULlJCtU1BwlpMnG/2bQsmhobCQYT5zw6ItXV1Vx91QzKtmxmxinNfGFsh9ORXGdiXuSM4hmhZq699iesWLHC6UhJT+UmCRw8eBCbFr8jN24Rew3r6uJzZEnkWG3bto0rf/RDqqv2cOPpjZw33O90JNcamRXmtjPrGZnewS233MKrr77qdKSkpnKT4AKBAC3NTToMfADEVinWpGJJBO+99x5XXTUD21bHT89s4KQEOrO3UwoyLLee2cAphT7uu+8+Hn74Ya1m7BCVmwT34WHgWp24v2z03FwqN+Jm1lqef/55Zs36KWMy2vn5WXWMz9Eb8EDxeizXntrE58e28/zzz3PbbbfR1tbmdKyko3KT4GJvxGGtcdNvViM34nI+n49f/epXPPzww5xT7OOWJDtP1FBJTYHvT27je5NbWbZsKTOu/BGVlZVOx0oqKjcJ7tDIjXZL9VtkleIUrWchrlRVVcWMGVfy5vz5/Ovxbcw4pYUMrTo8qL4wtoMbT2+iZt9u/nv6FSxbtszpSElD5SbBxd6Iw9ot1X/GYDKyVW7EdVauXMn0K37Avt3lXHtaE/96fLvWsBkipwwLcMfZ9RSntnLrrbfw2GOPaR7OEFC5SXDV1dWYVA+kpjsdJSEE07Korq52OoZIrwSDQf70pz9x0003UkAzd5xdzxnFAadjJZ0Sb5jbzqrnkyM6eOyxx7jxxhu0ZtYgU7lJcNXV1diMHLQi18AIp2VTtf+A0zFEelRVVcU111zNU089xWdGdfCzs+sZkRV2OlbSSk+F6Se1cvnUFjZ+sJbLL71E6+EMIpWbBFdVVUXQo11SAyWckUNt7UENK0tcW7x4MVf84HJ2btvClSc3c9nUVs2viQPGwGdG+7jjnAZywo3cfPPNPPLII/j9Wl9ooKncJLj9B6oJZ+Q4HSNh2PQcwqGQhpQlLjU3N3PXXXdx++23Mzy1mV+cU8/HR+iNM96MyQ5x+9n1XDCmg+eee44f/vd0duzY4XSshKJyk8A6OjpobmrUGjcDKFYUDxzQrimJL2vWrOGySy9h4YI3+eaENmad1cBwr3ZDxav0VLhkSivXntZE7b5d/Pf0K3j22Wc1KjxAVG4SWFVVFQDhjFyHkySOcHrktYy9tiJO6+jo4KGHHuK6664jtf0gt53dyLcmtuPRX3dXOLM4wF3n1nFaYTu///3vufban7Bv3z6nY7mefvwT2N69ewEIZ+Q5nCRxxCZnx15bESdt3LiRyy+7lBdffJHPj+ngF+fUcUKeTqPgNnnplmtOaeaKk1rYtmUDl116CX//+98JhzXy1lcepwPI4DlUbjJVbgZMSiomI0flRhzl8/mYPXs2L7zwPEWZlplnNDFtmEqNmxkD54/yMa0wwOzSHP73f/+Xd95ezE03z2TUqFFOx3MdjdwksH379mHSMsCT4XSUhBJIz9FS6uKYTZs28YPLL+P555/nc6M7uOvcOhWbBFKUGebG05u4bGoLpRs/4LJLL+Hll1/GWp0m41ho5CaBVVZWEkrXfJuBFs7IY0+lRm5kaPl8Ph577DGee/ZZCjMtN53RxCnDtCBfIjIGPjvaxymFAf5clssDDzzAO2+/zU0338yIESOcjucKGrlJYDt37SaUme90jIQTzsyntaWZhoYGp6NIkigtLWX6FT/gmWee4fyR7dx9bp2KTRIo9oa5+fRGLpnSwsb1a7j0kouZN2+eRnF6QeUmQbW1tVFXe5BwZoHTURJO2Bt5TXfv3u1wEkl0wWCQJ554giuvvJKm6j1cf3oTl5/UitejN7dkYQxcMMbH3efWc1xmC/feey+33nIL9fX1TkeLayo3CaqiogKAsFcjNwMtVhhVbmQwVVVV8eNrruHRRx/lYyWR0ZrTizRak6xKvGFuPqOR705qZdX7y7n0kotZvny507HilspNgjpUbjRyM+BsejYm1aNyI4PCWssbb7zB5Zddys5tm/nRtGZ+dHIL2WkarUl2KQa+OK6DO85pIDfcwMyZM/ntb39LR0eH09HijiYUJ6idO3eCSdEaN4PBGEKZBZHXWGQA+Xw+HnjgAV5//XWmFAT57zOaKdYqw9LF2JwQPz+rnhd3ZPHSSy+x/oN1/OKXdzFmzBino8UNjdwkqPLyHdisQkjRJh4MIW8h28vLnY4hCWTfvn3MuPJHvP7663xzQhu3nNmoYiPdSk+Fiya1ccPpTRyo3MX0K37A0qVLnY4VN/TOl6C2bd9GULukBk3IO4ymxkbq6uqcjiIJYPny5Uy/4gfsq9jB9ac18a2J7aQYp1OJG5xWFODOc+op8bRy66238uc//1krG6Nyk5AaGhqor6sjlDXM6SgJKxx9bXUmX+mvV155hVtvuYWilBbuPKee04s1aViOTYk3zKwz6/nMqA7mzJnDnXfeid+f3GeDV7lJQLG5IGGvys1gCXsLAdi+fbvDScStrLX85S9/4f777+fUYX5mnVWvs3hLn6WnwmVTW/nPE1tZvHgxN990Iy0tLU7HcozKTQLatm0b8OHoggw8m5aJychRuZE+CYfD3HfffTz++OOcP6qDH5/aREaq06nE7YyBr4zv4IfTmln/wQdcc/VVSbvrvFflxhjzJWNMmTFmuzFmZje3+Q9jzGZjzCZjzNMDG1OOxdatWzEZOdg0r9NREprfO4wtpWVOxxCXsdby4IMP8tprr/H149r4wdRWPPo3UwbQJ0f6ue60JvZW7ObGG66nubnZ6UhDrsdfKWNMKvA74MvANOA7xphpXW4zCbgF+JS19mTgJ4OQVXppS2kpfu2SGnThrCL27a2kra3N6SjiIrNnz+all17iK+Pb+beJ7RhNHJZBcGpRgB+f2siuXTuZefPNtLe3Ox1pSPXm/4XzgO3W2h3WWj/wLPCNLre5AvidtbYewFpbPbAxpbfa2trYt3cv4awip6MkvFB2EdZa7ZqSXnvxxReZM2cOnx3dwf87oU3FRgbVKcMCXDmtmc1bNvOzn91GMJg8Z4/vTbkZA+zp9HVl9LLOJgOTjTHvGWOWG2O+NFAB5dhs374day2h7GKnoyS8cFbkNS4r064p6dmaNWt4+He/4+wSH5dMaVWxkSFx7nA/l0xuYeXKVcyePdvpOENmoPb0eoBJwGeB7wB/MsYctsiKMWa6MWaVMWZVTU3NAD20dFZaWgpAWOVm0Nn0LExGtsqN9Ki6upo7b/85I7PDTD+pRWvYyJD63BgfF4zp4JlnnuHtt992Os6Q6E252QuM6/T12OhlnVUCr1hrA9bancBWImXnI6y1f7TWnmOtPaekpKSvmeUotmzZgsnM1WTiIeL3FrNp8xanY0gcCwaD3H77z+loa+aakxvw6qQ34oDvTmrlhPwQ9/zqbiorK52OM+h6U25WApOMMccbY9KB/wRe6XKbl4iM2mCMKSaym0qrmzlg0+Yt+L2abzNUwjnFVO3bS1NTk9NRJE4988wzbN68hcumNDM6W+vYiDPSUuCqk5swIR+/uvtuQqGQ05EGVY/lxlobBK4C3gC2AM9bazcZY+40xlwYvdkbQK0xZjOwCLjRWls7WKHlyBoaGqg+sJ9QtkbFhkrstdauKTmSbdu28dhjf+Hjw318fERyrxgrzivKDPO9Sc1s2ryZ5557zuk4g6pXc26stfOstZOttSdYa++KXvYza+0r0c+ttfY6a+00a+2p1tpnBzO0HJnm2wy92MTtLVu0a0o+KhAI8Ku77iI3Lcz3p7Q6HUcEgE+O8HNuiZ+/PDo7oU8fo6WjEsjmzZvBGB0pNZRS0yGrMPLai3QyZ84cduzaxWWTm8hJs07HEQEiqxhfPKWFzJQQv/71PQl7eLjKTQLZvHkz1lsIqWlOR0kqgaxiNm3ajLV6A5OI7du3M2fOk3xqpI8zdCLMfpuzNYvdzansbk7l7jV5zNma5XQkV8tLt3xvUjNlZVt58cUXnY4zKFRuEkQ4HGbT5s0ENGoz5ELZJTQ3N7Fv3z6no0gcCAQC3POru8n2hLloknZHDYSKFg/toRTaQymUNqRR0aJDzvrrY8P9nF3sZ/bsP7Nr1y6n4ww4lZsEsWfPHtrb2ghlD3c6StIJ5UQmFWvejQA89thjbC/fwWWTm8jV7iiJU8bAJVNbyDRB7vrlLwgEEmuEUeUmQcTmfIRzdKTUUAt7CzGpHs27ETZu3MgzTz/NP43q4KySxHqzkMSTn265dEoz27aX88QTTzgdZ0Cp3CSILVu2YDzphDMPWxhaBptJIRiddyPJq7GxkTvvuJ0ib5jvTtLJVMUdzinxc/7IDp6aM4c1a9Y4HWfAqNwkiE2bNhPIKkInrHFGKLuE7eXb8fu1lkkyCofD3H33XdTVHmTGtEa8Hu2OEvf43uRWRmaFufOO2zl48KDTcQaEyk0C6OjoYOfOHZpv46BQdgmhYJBt27Y5HUUc8NRTT7FixftcdGILE/MSe+VXSTyZHrj6lEbaW5q44/afJ8Th4So3CWDr1q2Ew+FDE1tl6MVee827ST7vvvsujz46m0+M8PHPY3xOxxHpkzHZIS6d0syGjZt44IEHXL+0hcpNAogdpRPWaRccY9OzMRnZh1aJluSwdetWfvnLXzAxL8TlU1u0V1hc7ZMj/Xz9uDbmzp3L888/73ScflG5SQClpaU6E3gc8GfpDOHJpLq6mltm3kxOip+fnNpIeqrTiUT679sT2zm3xMfvf/8I7777rtNx+kzlJgFs3LQZv1eL9zktnF3C/qp9OkN4EmhqauKG66+jrbme605tID/d3UP4IjEpBqZPi8wd+8Wdd7J+/XqnI/WJyo3L1dfXU1N9gFCOyo3TYmcI166pxNbR0cHMmTdTtbeSa09pZFyOJhBLYslIhetObaQo3c+tt8x05Qk2VW5c7sMzgWu+jdNC2UWAVipOZMFgkNtv/zlbNm/hh9OamVro/qNKRI4kN91y42kNpIXauPGG66mqqnI60jFRuXG5srIyAEJZRQ4nkcgZwgsObRNJLOFwmF//+tcsX76Ci6e0cO5wrWkkia3YG+bG0xpob67nhuuvo76+3ulIvaZy43JlZWWQpTOBx4uAt4gt2i2VcKy1PPLII7z55pt8+/g2LtAh35IkxuaEuP60RmoOVHHTjTfQ2uqOk8Gq3LjcltIyAt5hTseQqFBWMfV1dQmzyqdEPPvss7zwwgv8y9h2LpzQ7nQckSE1KT/I1Sc3saO8nFmzfuqKk2yq3LjYwYMHaaivI5StycTxIhzdFlu3bnU4iQyUhQsX8oc//IGPDfdx0aQ2rWUjSen04gCXT21m7dp13HvvvXG/yJ/KjYvF3kDDWSo38SKUNQyM0bybBLFu3Tru+dXdTCkIMn1aCykqNpLEPj3Kz7ePb+PNN9/k0UcfdTrOUancuNj27duB6BuqxIfUNPDmH9o24l5VVVXcNuunlGQE+fGpTaTpr6UIF05o5zOjOnjyySdZuHCh03G6pV9XF9u2bRt48zWZOM4EMgsp024pV/P5fPz8Z7cR8rVx7akN5KTF9xC8yFAxBi6e0sqkgiD33vtrdu7c6XSkI1K5cbGyrVsJZGrUJt6Es4o4WFOjlYpd7MEHH2Trtu1Mn9rIiKyw03FE4oonBa4+uYlM/Nw266dxeQSVyo1LNTc3U33gAOFslZt4E1tzSLum3GnJkiXMnTuXrx/Xxlkl8X9UiIgTCjIsM6Y1snffPh5++GGn4xxG5calYkOBIR0GHnfC0TlQ5eXlDieRY9XS0sID99/H+Nww/3q8DvkWOZqphUG+PK6duXPnsnbtWqfjfITKjUvFzvUR1mTiuGPTvJh0b9zui5bu/eEPf6Cuvp7LpjTh0V9HkR796/FtDM+y/ObeX+Pzxc/ilvr1damdO3di0jKwaVlOR5EjCGQWUO7Ck80lsz179jD3tdf4l7HtTMzTyTBFeiMjFS6Z3My+qv289tprTsc5ROXGpcrLywlmFqAVxeJTOLOQXTt3EQ5rMqpbPPXUU3hS4OvHaXeUyLE4ZViAKQVBnnn6Kfz++DjnmsqNC1lr2bFzJ6HMQqejSDfCWYX4fB2uO5Nusjpw4ABvvjmfz45qIy9dh32LHKtvTGjjYG0d8+fPdzoKoHLjSnV1dbS1thL2FjgdRboRyoxsm4qKCoeTSG8sWrSIUCjMF8d1OB1FxJVOLgwwJifM/PlvOB0FULlxpT179gAQzsx3OIl0J+yNbJvYtpL49s47bzMhL0yJV7sRRfrCGDinuIONGzbS0NDgdByVGzeKjQao3MQxTyYmLVPlxgXq6+vZsmULZxdp1EakP84u8RO2lmXLljkdReXGjfbs2YNJ9WDTs52OIkcRzMhjt3ZLxb3y8nKshckFWrBPpD+OywmR4YmPBUxVblyooqIiMmqjI6XiWjgzn927VW7iXWwkdFSWDv8W6Q9jYFRWOC7mGqrcuFDl3n0E03OcjiE9CGfk0thQT0eHdnfEs71795LpMeTrKCmRfhuZGaByj8qNHCNrLdUHDmDTc52OIj0IZ0S20f79+x1OIkfj8/nI9GggVGQgZKTauFipWOXGZerq6ggE/IQzNHIT71Ru3CEcDpNiNGojMhBSDHGxeKnKjcvE3ihjb5wSv2x016EW8ot/1mrYRmQgxMu/CSo3LhMrN1YjN3HPpnkhJVUjN3GusLCQJp8lHC9/lUVcrNGfwrBC51fPV7lxmbq6OgDCiXTCzJCfzMxM/u3f/o3MzEwIxce5SfrNGEx61qFtJvGpqKiIkIXmgEZv4lV70Hzkb0R7UNsqXjX4UykePsLpGCo3blNbWwspqZCa7nSUAWOCfr72ta9x1VVX8dWvfhUTTJByA4Q83sg2k7g1YkTkD3F1e6rDSaQ7bUHzkb8RbSo3cclaqOnwMHz4cKej4HE6gByburo6THpWQh3aYT3pvPbaa1hrmTt3LtaTOKNSIY+Xgyo3cW3ixIkA7GlJZVJ+0OE0ciRZHvuRvxEjPNqHGI/q/Sm0+O2h3yknaeTGZWprawl5Mp2OMbBS0+no6OCvf/1rZE2YBBqVsmkauYl3I0eOJDvLS0WL/teLV16P/cjfCK/KTVyqaI6Mfp544okOJ1G5cZ26+nrCiVZuEphNy6S1pYVQSKvfxitjDJOnTKW8KXFKtYgTtjd5SElJUbmRY9fa2opNoJGNRBfbVm1tbQ4nkaM5/fTTqWhOoVWTikX6rKwhnUmTTiQry/mpBSo3LqNy4y6xbdXa2upwEjma008/HQtsbdSuKZG+8IegvMnD6aef4XQUQOXGVay1tLe3Y1PTnI4ivaVy4wrTpk0jIz2NDbX6x0GkL0ob0giG4ayzznI6CqBy4yo+n49wKJRQE24TXayIqtzEt4yMDM46+2w+qMvAaq6qyDH7oDaNjPQ0zjzzTKejACo3ruL3R9Z/sSlaj8M1otsqtu0kfn3845+gpt2wr02/XyLHwlpYV5fJmWedTUZGhtNxAJUbV/nwZGSa9OgW1kR+xXS0VPz71Kc+hTGwslojoyLHYndLKjVthk9/+tNORzlE5cZFDr1BGm0294gU0Xg4S64cXXFxMaeecgorarTUgsixWHEgg9TUFM4//3ynoxyid0kX+bDcaOTGNaLbSiM37vC5C/6ZvS0phxYjE5GjC1tYUePl7LPPJj8/3+k4h6jcuEjsv3+r3VLuYTRy4yYXXHABaWkeFu/T6I1Ib2ysS+NgO3z5y19xOspHqNy4iMcTWYPDWL1Rukb00JvYtpP4lp+fz2c+81neq87Ep8E2kR4t2ptJQV5uXM23AZUbV0lPj050DOuvrluYcOREjIe2ncS9Cy+8kPYAvFsVH0d9iMSr6vYU1tam8+Wvfo20tPhaf03lxkVib5DGqty4RrSIqty4x6mnnsq0aScxtyKHoAZJRbr16i4vqR4P3/72t52OchiVGxfRyI0LWZUbtzHGcPHFl1DbodEbke4c7EhhyYFMvvrVr1FcXOx0nMOo3LhISkoKnrS0Q7s6JP6ZUGRbZWZqgqqbnHfeeUydOoWXdufQrl83kcO8UJ5FSkoqF110kdNRjkjlxmVyc3Ih6HM6hvSSCUW2VV5ensNJ5FgYY7jmmh9T3wEv7XT+DMci8WRzvYdlBzK46Lv/xfDhw52Oc0QqNy6Tl5+PCXY4HUN6yQQi20rlxn2mTZvGV7/6Vd6o9LKnReveiAAEw/DktlxGjhget6M2oHLjOoUFBaSENHLjFibow5uVpUPBXeqKK64gLzeX32/Jw6+pbiK8UJ7F3pYUfvyTa+PmPFJHonLjMgUF+aSq3LiGCXaQlxc/q3bKsSkoKODmW25lT3MKz5dr95Qktw21afxjj5cLL7yQT3ziE07HOSqVG5cpKCjABNqdjiG9y5B/OQAAIABJREFUZALtFA0rdDqG9MMnPvEJvv3tbzO/0suamvhay0NkqDT4DH8szeO48eOYMWOG03F6pHLjMsOHD8cGfBAKOB1FesETbGPEiBFOx5B+mj59OpNOPJE/bMmnUvNvJMn4Q/DbDfl0kM7Pb78jrndHxajcuExsZnqKv9XhJNIjazG+VkpKSpxOIv2UkZHBXXffjTc3n99uzKfZr/O7SXKwFmaX5rCjKZVZs25j4sSJTkfqFZUbl4mVG+NvcTiJ9MQEO7DhoEZuEsTw4cP55V13Ux9I48GNmmAsyeHlXV6WHcjgBz/4Aeeff77TcXpN5cZlNHLjHia6jeJ1HQg5dtOmTeOWW26lrMHDHzbnErZOJxIZPIv3ZfC3nVl84Qtf4Lvf/a7TcY6Jyo3LFBcXk5qaiulodjqK9CDFF9lGI0eOdDiJDKQLLriAGTNmsLImnSe3ZsdO/C6SUNbUpPGXshzOO/dcbr75Zoxx165YlRuX8Xg8jBw1ihRfo9NRpAcpHZFtNGbMGIeTyED793//d77zne+wcG8mL+/yOh1HZECVNXj43eY8Jk+ezO133OHKdbpUblxo/LhxeHwauYl3KR1NFA4bRlaW1kdJRNOnT+dLX/oSf9uZxYLK+D96RKQ3KppTeWBDPiNHjeHXv77XtX+/VG5caNy4caT4mtB4eHxL8TUxftx4p2PIIDHGcMMNN/DJT36SJ7fmsGy/zvwu7nagLYXfrC8gO7+I/7nvfgoKCpyO1GcqNy40ZswYbCh4aMKqxCePv5lx48Y6HUMGkcfj4ec//zmnnXYaf9ySy8Y6LfIn7tToN/xmfQE2PYf/ue9+1x/lqXLjQscddxwAKR0NDieR7phAO9bffmhbSeKKrYFz3HETeHBjHruatcifuEtHEO5fn09jKIN7fn1vQvzd6lW5McZ8yRhTZozZboyZeZTbfdsYY40x5wxcROlqwoQJAKS0q9zEq9i2iW0rSWw5OTn8+je/Ia+wmPvWF1DTrv8bxR2CYXhoUx67W9K4/fY7mDZtmtORBkSPv4HGmFTgd8CXgWnAd4wxhz17Y0wu8GNgxUCHlI8qKCggLz9f5SaOxUbVVG6SR0lJCb/5n/sIe7K5f0M+7UF3HTobz8bnBPGmhvGmhplaEGB8TtDpSAnjqW3ZrK9N47r/396dx8dR3dkCP7+q6m61dlm2ZFmyLWu35EXyIsm2LC/ybmMMYxjM8kJCIBMCmRDWAcwQMBAgQAIEjCFhS2ayz2dIhgl5wySBJI/FxA7EC8Z437Uvrb37vj9aMvKGZau7b3f1+X4+fNzqLlUdYUt9VPdW3W9+M+wXwzwXg/n1ohzATqXULqVUN4CfALjwNNvdD+BhAJ0BzEdnMC57HCwOS4Uto6MRbncshg8frjsKhdDYsWNx73334XC7hWe3xvMmfwFyZUE7xiZ4MTbBizuntODKgnbdkWzhzQMuvHkwBpdddhlWrFihO05ADabcZALYP+DjA33PHSciUwCMVkr91+ftSESuE5GNIrKxtrb2nMPSZ8aNy4bZ2cQrpsKU2dGE7HHZEXfjKxq6qVOn4sYbb8TmOid+sSsyL6Ml+9vWaOHVT+JRWVmBa6+9VnecgBvywLCIGAAeB3Dz2bZVSm1QSk1TSk3jYoJDk5ubC9XbzTWmwpFSsDobkZebqzsJabJq1SqsWLECv9nrxqY6XkFF4aWpS/DM1iRkZWVh7dp7YJr2mwQ/mHJzEMDoAR9n9T3XLwHABAB/EJE9ACoBvMZJxcGV2/fGabY3aE5CJ5NuD1RPF/Ly8nRHIU1EBDfeeCPycnPw/PZE1HdygjGFB58C1m9LRKdy4N5v3Ye4uDjdkYJiMN9x7wPIF5FxIuIEcBmA1/pfVEo1K6WGK6WylVLZAN4BsFIptTEoiQmAf6KqiMDoaNQdhU7S/3eSk5OjOQnp5HK5cO+37oPXcOEZLrJJYeI3e93Y2mDh6//8DVv/jDpruVFK9QK4AcAbALYB+JlSaouI3CciK4MdkE4vNjYW6SMzYPDMTdjpP5tm5x8cNDhZWVn45s234JMmC/+1N0Z3HIpyu1tM/MfuWMyfPw/Lli3THSeoBrUallLqdQCvn/TcPWfYdu7QY9FgFOTn4cj7f+PlaWHGaG9AWnq6bU/30rlZsGAB/vSnt/Grt9/C5OE9GBPv1R2JolC3F9iwPRHJySn4xjdusv3FDhwIjmD5+flARzPg7dYdhQZwdDagsKBAdwwKEyKCm276JhITk7BhWyJ6fboTUTT61e5YHGwzcNsddyAxMVF3nKBjuYlg/RNWOak4jHh7gI5mf/Ek6pOcnIxbbr0N+1oN/HqvW3ccijI7my389343li9fjoqKCt1xQoLlJoL1v4Fy3k346C+avFKKTjZr1iwsWLAAr+2JxT6uP0Uh0u0FXtieiOGpqbj++ut1xwkZlpsIlpqaisSkJJjt9bqjUB+j7++CZ27odG688UYkJiVhw/ZE9HB4ikLgl7tjccgjuPW226NqHiDLTQQTERQWFMBiuQkbZns9EhKTuOwCnVZSUhJuve127Gs18CvevZiCbFujhd/uc+OCCy5AeXm57jghxXIT4QoLCyEdjYCPC8mFA6u9HuOLCm1/JQKdv5kzZ2LFihV4fZ8b2xsHdcEq0Tnz9Ag2bE/CqFEZUTUc1Y/lJsIVFBQASnHeTTjw9UI6Gv1/J0Sf42tf+xpGjcrAs9uS0NzNIkyBpRTwwvZ4NHYZuHvtPXC7o28SO8tNhOt/I+W8G/2M9gZAKZYbOiu3241v3Xc/2n0OfH8LLw+nwPrN3hh8UOvEV7/6VYwfP153HC1YbiJceno64hMSYXjqdEeJev0Fk+WGBiMvLw8333Irtjda+OmnnH9DgfFRvQO/2B2H+fPnYfXq1brjaMNyE+FEBMXjx8PBMzfamW21SExKQnp6uu4oFCEWLVqEiy++GG/sd+ON/VyegYZmT6uJp7ckInvsWNx6621RPfeP5cYGxo8v8k8q9vbojhLVHB31KCkujuofKHTurr/+esyuqsKPP4nDn484dcehCHWk3cB3PkxGQspwPPzIo1E5z2YglhsbKCoqApTivBudvN1Ae2PUjm/T+bMsC3evXYuy0lK8sC0Bm+ocuiNRhKnvNPDo35IhrgQ89vgTSEtL0x1JO5YbGygqKgIAGJ5azUmil+nxF8v+vwuic+FyubDugQeQl5+PJz9KxDtHeQaHBudwu4F1m1LggRuPPPodjB49WneksMByYwMpKSkYkZYOs43lRhez7RgAlhs6f3FxcXjs8SdQMnEint2SgN8fdOmORGFuX6uJBzalwOtIwPeefAqFhYW6I4UNlhubmDxpIpzttf4bHFDImW3HkJk1OipW26XgiY+Px6OPfgflFeV48eN4vLbHzW9pOq1tjRYe3JwMV0Iqnnz6+1zy5SQsNzYxYcIEqC4PpLtNd5TooxQc7bWYPGmi7iRkAy6XC+vWPYCFCxfgF7ti8eyWeHR5daeicPI/B1x4eHMSRozMwlNPfx9jxozRHSnssNzYRElJCYDPhkcodIzOZqieTkyYMEF3FLIJh8OBO++8C9dddx3erXXhwU3JaOjkj+to1+sDXvo4Dq/siEd5RQWeWf8cRo4cqTtWWOJ3i02MGzcOrpgYmG1HdUeJOkZfoewvmESBICK4/PLL8cADD+JoTxzu+SAFf2/glVTRqqHTwLc3J+F/D8Yc/3cRTat8nyuWG5uwLAsTSkrg4JmbkLNajyAhIZGnhikoZs6ciWeeXY+U9NF4dHMifvGpG14u1xBVNtU5cPfGFOzvjMPatWtx3XXXwTRN3bHCGsuNjZSVlUHaGyA9nbqjRBWH5yjKykp58z4KmuzsbKx/bgOWLluG1/bG4sHNSajnMJXt9fiAH38Siyc+TMTIMbl4/oUfoKamRnesiMDvDhspLS0FAJitRzQniR7S1Qp0th7/f08ULG63G7fddhvWrl2LA51xuOv9FPzliJNXU9nUvjYT3/ogBW/sd+Piiy/GM888i6ysLN2xIoalOwAFTlFREZwuF7pbD6N3WLbuOFGhv0iWlZVpTkLRoqamBoWFhXjowQexfutWbKztxtWFbUh0suXYgdcHvL4vBr/aE4fExCQ8+ODtmDlzpu5YEYdnbmzEsixMmjgRjjaeuQkVq+UwEhISkZ2drTsKRZGsrCw8+dRT+MpXvoK/Nbpx5/vDsLGWdzWOdIc9Bh7YlIyf74pD1exqvPjSyyw254nlxmamTJkCaW+EdLfrjmJ/SsHZehhTp07hfBsKOdM0sWbNGjy34Xmkjc7Fkx8l4Om/x6Opi/8WI02vD/j1nhjc/X4KjnoTsXbtWtx777eQnJysO1rEYrmxmenTpwMAzJaDmpPYn9HRCNXtQXl5ue4oFMVycnKwfv1zuOaaa7CpIRb/8t4wvHXIxbk4EWJXi4l7P0jBz3fFYUbVbLz08iuoqanhL0xDxDk3NpObm4uk5GT0NB9A7/DIuB23L3YYVN+K5t7YVPhih2lONDhms79ATps2TXMSinaWZeGqq65CdXU1HvvOo3jho7/jL8di8MWCVqTH8rrxcNTlBX61Kxa/PeDGsORk3H//zZg9e7buWLbBMzc2YxgGKsrL4Ww9HDHrTHWNqYQ3NhXe2FR0FC1D15hK3ZEGxdFyEGPGjkVaWpruKEQAgLFjx+K733sSN910E/Z0xOPO91Lwn7vd6GG/CSub6hy4471U/Pd+N1asuAAvvfIqi02AsdzY0PTp06F6OmF46nRHsS9vD6y2o6isqNCdhOgEhmHgwgsvxCuv/gizqufil7tjcff7w7C1kSfqdavvNPC9DxPwxIeJSEwfi6eeego333wzEhISdEezHf5rt6Hy8nKICKzm/eiOH6E7ji1ZLYegfF5UVkbGWSaKPsOHD8e9996Ld999F9994nF8e5Ng1sguXJbnQRIvGw+pXh/wfw/E4Fd74gHDwle+8iVccsklsCy+BQcLz9zYUFJSEkomTICzeb/uKLZlNu2DOzYWkyZN0h2F6HNVVFTgpZdfwZVXXol3a2Nx+7vD8D8HXPCx34TEx00W7tmYgn/fGYcp0/x/F2vWrGGxCTKWG5uqmjUL4qmHdLXpjmI/SsHVcgAzKiv5A4oigsvlwpe//GX84Ic/RNHEMryyIx7f+iAZn7bw32+wtHQLntsahwf+moRudxrWrVuHBx96CBkZGbqjRQWWG5uaMWMGAMDi2ZuAMzy1UN0dvLkWRZyxY8fisccex9q1a9FipuK+jUl4cXscPD287DhQfAp484ALt707DO/WxuGKK67Ay6+8iqqqKl7eHUKs7TY1ZswYZIzKxIHGvehJG687jq1YjftgmCYqOJmYIpCIoKamBpWVlXjxxRfxq1/+Eh/Ux+CynFbMGtkNvv+evz2tJl76OAG7WkyUlZbiGzfdhLFjx+qOFZV45samRARz51T71z7q7dIdxz6Ugqt5L0pLS3mFA0W0uLg43HDDDXhuwwZk5RRhw7YEPLQpCQc9pu5oEae9V/DKjlj868ZkNBrDcPfdd+PxJ55gsdGI5cbGqqurAeWD1bRPdxTbMDqagI5mzKmu1h2FKCDy8/Px9PefwS233IKDvUm4+71k/PxTN7q9upOFP6WA9445cce7w/DmQTdWrboIr7z6IyxYsIBDUJpxWMrGioqKkJo6HEca90bM3YrDndW4ByKCqqoq3VGIAsYwDKxYsQJVVVV45pln8Ovf/Q7v1brxxYIWFA/r1R0vLNV1GnhlRxw21zmRn5eHb99yC4qKinTHoj48c2NjIoK5c+fA0XIQ8PbojmMLzqa9KCkpQWpqqu4oRAGXnJyMO++8E4899hjMpAx8e3MSnt8ah1ZOOD7Op4A39sfgX95LwbbWeFx//fV4dv16Fpsww3Jjc9XV1YDPC6uJV00NlXQ0Q9obMGfOHN1RiIJq6tSpePGll3HFFVfgL7X+xTg/qHXojqXdYY+BdX9Nxo8/icPksul46aWXcemll/KWEGGI5cbmJkyYgOSUYbAaduuOEvEcjf7/hyw3FA1cLheuvfZabNjwPNJG5+J7HyVi/ZZ4tEXhWRyfAn67LwZ3b0zBUW8i7rrrLjz8yCO8Z00YY7mxOdM0MX/eXDg5NDVkzsY9KCmZwIUyKark5uZi/frncPXVV+PdOjfufH8YNtdFz1mco+0GHtyUhH/bGYdp0yvx0suvYOHChZwwHOZYbqLA3LlzoXy9vGpqCIyOJkh7A+bPn6c7ClHIWZaFq6++Gs8+ux7DRmbj8Q8T8aMdsbZfbfwvR5xYuzEFB7sTcMcdd+DBhx7ifLsIwXITBSZMmICUYRyaGgqrcQ+AvjlMRFGqoKAA6zdswOrVq/G7A27c/9cUHG2339tIlxf4wbY4rN+agNyCYvzwxZewZMkSnq2JIPb7V0mnMAwD8+fN49DUEPQPSY0YwVXWKbo5nU7ccMMNWLduHep9Cbhn4zC8c9SpO1bAHPSYuPeDFLx1JAZXXHEFvvfkk0hPT9cdi84Ry02U8A9NeTk0dR76h6RqaubrjkIUNqqqqvDCD36InILxeGZLAn65yx3xK41/WO/AfX9NhsdIwiOPPIprr72WV0JFKJabKFFSUoJhqakcmjoPVsNuiAiHpIhOkp6eju9+73tYunQp/nNPLJ7ZEo+uCL2z8f8ccOHxDxORkZWN555/AdOnT9cdiYaA5SZK9A9N+W/o1607TkRxNu3BhIkTMXz4cN1RiMKOw+HAbbfdhn/6p3/C+7UuPLQpGS3dkTM3xaeAH+2IxSs74lFZOQNPf/8ZDkPZAMtNFJk3b57/hn6NHJoaLP+QVCPmzZ2rOwpR2BIRXHbZZVi37gEc6IjBtzdHRsHxKeDF7XH43QE3Vq9ejfvXrUNsbKzuWBQALDdRZPz48f6hqb4rf+js+teS4pAU0dnNmjUL3374YdR2u/DQ5mQ0h3HB8SngB9vj8MfDMbjqqqvwta99DabJFdHtguUmihiGgblz5sDZcohXTQ2Ss2kviotLOCRFNEhTpkzBtx9+BHU9MXhoc0pY3tFY9Z2xeftwDL7whS/gS1/6Ei/zthmWmyhTXV3tv6FfM9eaOhvpbIF46jF3LpdbIDoXZWVlePjhR1DbaeHpvyeiN8xu9vf6vhj88XAMrrzySnzxi19ksbEhlpsoM3HiRCQmJcNq2Ks7Sthz9A3fzZ49W28QoghUWlqKW269DVsbLby6Iw4qTC4T/6DWgZ99Goe5c+fimmuu0R2HgoTlJsqYpok51bPhbDkA+Hp1xwlrjqZ9yM/Px8iRI3VHIYpIixcvxpo1a/D7QzH4wyGX7jg43G5g/bZEFBTk44477uAZGxtjuYlCs2bNgvL2wGw9ojtK2JKedhhtx3jWhmiIrr32WkyZUoZ/+zQBxzr0veX4FPD8tkQ4XLG4f90DiImJ0ZaFgo/lJgqVlZXB5YrhJeGfw2ryz0maNWuW5iREkc0wDNxxx7/AdMTg+W0J2u5i/F97Y7Cz2cQ3bvom0tLS9ISgkGG5iUIulwvl5dPhbNmPsBkIDzNW0z6MSEtHTk6O7ihEES8tLQ1f/+d/xsdNFn5/MPTDU0fbDfzHnjjMmVONmpqakB+fQo/lJkrNmjUL6PLAaK/XHSX8eHvhaDmM6tlVHJMnCpDFixejdPJk/MfeeHT0hvb76mefxsFyuPD1r/8zv6ejBMtNlKqsrISIwGo+oDtK2DFbD0P5ejFjxgzdUYhsQ0Tw1euvR0sX8Ou97pAd9+MmC+/XOnH5FVcgNTU1ZMclvVhuolRycjLy8vPhYLk5hdV8EA6nExMnTtQdhchWCgsLsXDhQrxxwB2y5Rl+uSsOqcNScMkll4TkeBQeWG6iWGVFBQxPLdDbpTtKWHG2HsSUsjK4XPovXSWym6uuugq9PuCN/cG/WumTZgvbmyxctuZyuN2hO1tE+rHcRLHp06cDSsFqOaw7StiQzhagoxnl5eW6oxDZ0pgxYzB7djXePBQb9Lk3v9nrRkJ8HJYvXx7U41D4YbmJYsXFxXDHxsLk0NRxVsshAH3Fj4iC4vLLL0d7D/D24eCdHT3abmBTnRMXXfwPXOk7CrHcRDHLsjClrAzONt7Mr5/Zcgipw0dg9OjRuqMQ2VZRURGKigrx5qHYoN2N4s2DMTBNAytXrgzOASissdxEudLSUqCzBdLVpjuKfkrB6TmKqVPKeLkoUZBddNHFOOwRbGl0BHzf3V7g7SNuVFXNxvDhwwO+fwp/LDdRrrS0FAC4FAMAo7MJqrvj+P8TIgqeuXPnIiE+LihrTr1/zAlPD7Bq1aqA75siA8tNlMvNzUVcXDzMVk4qNvsmVrPcEAWfy+XCosVL8EGdC60Bviz8D4fdyByVwe/lKMZyE+UMw0Bp6WQ4247qjqKd2XoEw1KHIyMjQ3cUoqiwfPlyeH3An48E7uzN4XYDHzdZWL7iAg4vRzGWG/LfrK6zBdLToTuKVs72OpROnsQfiEQhkpOTg/FFhfjjEXfAJha/fTgGhmFg8eLFgdkhRSSWG0JJSQkAwGw7pjmJPtLVBtXVhgkTJuiOQhRVli1fgYNtBna3mkPel9cH/OmoG5UVFVxqIcqx3BAKCgpgmiaMKC43psf/tRcXF2tOQhRd5s+fD5fTgT8eGvodiz9scKCpE1jGm/ZFPZYbgsvlQn5+PixPFJebtlo4nE7k5eXpjkIUVeLi4jBn7jy8UxuDLu/Q9vXWYReSkxJRWVkZmHAUsVhuCIB/aMry1APKpzuKFpanFkWFhbAsS3cUoqizbNkydPQAG485z3sfzd2CzXUuLF6ylN/HxHJDfkVFRVC+XhgdTbqjhJ7PB7O9AePHj9edhCgqTZ48GZmjMvDHw+e/uOWfj7jgVf6iRDSociMiS0TkYxHZKSJ3nOb1b4rIVhH5UETeFJGxgY9KwVRUVAQAMD11mpOEntHZCOXrRWFhoe4oRFFJRLB02XJsb7JwtP3cf+dWCnjrcCyKi8dj7Fi+/dAgyo2ImAC+D2ApgGIAa0Tk5FmXmwBMU0pNAvALAI8EOigFV2ZmJtzuWBhRWG76Cx3LDZE+ixcvhiGCt85jMc2dLRYOeQTLl68IQjKKRIOpyOUAdiqldimlugH8BMCFAzdQSv1eKdXe9+E7ALICG5OCzTAMFBUVwmqPvnJjeOoQGxeHzMxM3VGIotaIESNQXlGOt4/EwnuOU//eOuRCjMuFefPmBSccRZzBlJtMAPsHfHyg77kzuQbAf5/uBRG5TkQ2isjG2trawaekkCgsLITR0Qj4omtSsdVej8KCAt68j0iz5ctXoKkL+HvD4BfT7OwF3q11Y35NDWJjY4OYjiJJQCcUi8iVAKYBePR0ryulNiilpimlpo0YMSKQh6YAyM/PB3xeGJ2NuqOEjs8Ho6MRBQUFupMQRb3KykokJybgj+cwNPV+rQudvQpLly4NYjKKNIMpNwcBjB7wcVbfcycQkQUA7gKwUinVFZh4FEr5+fkAAMNTrzlJ6BidTYDPy3JDFAYcDgcWLl6CTfWDX0zzrcMxyByVwbuL0wkGU27eB5AvIuNExAngMgCvDdxARMoAPAd/sYneO8FFuKysLLhcMTDbG3RHCRmj3V/kePM+ovCwePFieH3AO8fOfvamtsO/SOaSpcs4rEwnOGu5UUr1ArgBwBsAtgH4mVJqi4jcJyIr+zZ7FEA8gJ+LyGYRee0Mu6MwZhgG8vLyYHZEz5kb01MPp8uFrCzOgScKB3l5ecjNGYc/HTlxOYYx8b0YE997wnN/PuKCCLBo0aJQRqQIMKjbOCqlXgfw+knP3TPg8YIA5yJNCgrysXX7x/4bR0TBb0JmR72/0JlDX7SPiAJj0eIlePbZZ3Gk3cDIWP8FDlcWtJ+wjVLAX465MXnSZKSnp+uISWGMdyimE+Tn50N5eyBdLbqjBJ9SsNobUNA314iIwsO8efMgArxz9MxDU3vbTBzxCGoW8HdrOhXLDZ2gf1KxGQWTiqWrBcrbc/xrJqLwkJaWhokTJ+KdY2dejuGdoy6YpoHq6uoQJqNIwXJDJ8jOzoZpmscn2tqZ2fc1stwQhZ958+bjkEdwyHPq25RSwMa6GEydOhVJSUka0lG4Y7mhEzgcDmRnjzv+xm9nhqcBhmkiOztbdxQiOsmsWbMAAB/UnrpS+AGPiWPtgtmzedaGTo/lhk5RVFQIR0eD/9cjGzPb65AzLgdO56k/PIlIr7S0NBQWFuCv9TGnvPbXWidEgJkzZ2pIRpGA5YZOUVhYCNXTCelu0x0leJSCo6MeRUVcLJMoXM2YMRO7mk209Zx45eZHjU4UFBQgNTVVUzIKdyw3dIr+1bFNG68QLl2tUD1dXAmcKIxNmzYNCsDWxs/WmuroFexstjB9erm+YBT2WG7oFOPGjfNPKrZxuemfU8RyQxS+ioqKEOuOOWEhze1NFnwKmDp1qsZkFO5YbugUTqcTObm5sOxcbjy1sCwHxo0bpzsKEZ2BZVmYOGkSPmn5bF7cjiYHLNNEcXGxxmQU7lhu6LQmlJTAaq8DlE93lKAwPbUoKCiAw+E4+8ZEpE1JyQQcbDPg6Zt380mLAwUF+XC5Br9yOEUflhs6reLiYihvD4yORt1RAs/nheWpx4QJJbqTENFZ9K/2vavFgtcH7G61UFzCFcDp87Hc0GmVlPjf+M22Ws1JAs9ob4Dy9fK0NlEEyM3NBQDsazNxpMNEj5c33qSzY7mh08rIyEBCYhLMtmO6owSc6fEXNpYbovCXlJSE4akp2N9mYX+bf4Hb/sJDdCYsN3RaIoLJkybC4bFhuWk9gtThI5CWlqY7ChENwricPBxsd+CQx4QhgjFjxuiORGGO5YbOaPLkyUBnC6TLRjfzUwpOz1FMnVKmOwkRDVJmZiaOdZo42mEibcRw3lWczorlhs6otLQUgP9Mh10Ync1Q3R1YY6YbAAAQu0lEQVT+4kZEESEzMxMdPQqftjiQkZmlOw5FAJYbOqOcnBzExsXZqtyYrYcBfFbciCj8jRw5EgBwrMNARkaG5jQUCVhu6IxM00Tp5Mlwttmo3LQcxrDUVIwaNUp3FCIapIFrSHE9KRoMlhv6XNOmTfPPu+ls0R1l6JQPzrbDKJ8+HSJy9u2JKCyw3NC5Yrmhz1Ve7l+czmo5qDnJ0BmeOqieruNfExFFhuTk5NM+JjoTlhv6XJmZmUgfORJW8wHdUYbMaj4AEeGCe0QRZuBSC/Hx8RqTUKRguaHPJSKorKiAo/UI4PPqjjMkjpZDKCwsRFJSku4oRHSe+P1Lg8FyQ2dVXl4O5e2J6KumpKcDhqcWFRUVuqMQ0Xl46KGHcMMNN/DuxDQolu4AFP6mTp0Kh9MJq2kvvEmZuuOcF6tpP6AUqqqqdEchovMwY8YM3REogvDMDZ1VTEwMKisq4Gr2F4RIZDXuRVpaOvLy8nRHISKiIGO5oUGZPXs2VJcHhicCVwn39sBqPYQ5c6p5CTgRURRguaFBmTFjBgzThNW4V3eUc2Y17Qd8XsyePVt3FCIiCgGWGxqUhIQETJkyBa6mvRE3NGU17kFScjJKSkp0RyEiohBguaFBW1BTA3S2RNbQVG83nM37saCmBqZp6k5DREQhwHJDgzZ79mw4HA446j/VHWXQrMY9UD4vFixYoDsKERGFCMsNDVpcXBxmzpwJV9MewOfTHWdQnA27MDIjA0VFRbqjEBFRiLDc0DlZuHAhVHcHzAhYa0q622G2HMLiRYt4lRQRURRhuaFzUl5ejviEBDjqPtEd5awc9f6MCxcu1JyEiIhCieWGzonT6cSSxYvhaN4H6enUHefMlIKrficmTpqErKws3WmIiCiEWG7onC1btgzw+WCF8cRis+0o0NGM5cuW6Y5CREQhxnJD5ywnJwcFhYVw1X8Stve8cdTuQIzbjTlz5uiOQkREIcZyQ+dlxfLlkPaG8LznTW8XnE17sHDBArjdbt1piIgoxFhu6LzU1NTAFRMD57HtuqOcwlG3E8rbiwsuuEB3FCIi0oDlhs5LXFwcli5ZAmfj7vCaWKwUYuq2o2j8eBQUFOhOQ0REGrDc0HlbuXIllM8LK4wuCzdbDwMdzbj4oot0RyEiIk1Ybui85eTkYOKkSYip2x42E4sdx7YhPiGBE4mJiKIYyw0NyUWrVgGdrTCb9+uOAulqg6NpH1YsXw6Xy6U7DhERacJyQ0NSXV2NlGGpcB3dqjsKHMe2QQCsWrVKdxQiItKI5YaGxLIs/MPFF8FsOQSjo1FfEG8vYup3oKqqCiNHjtSXg4iItGO5oSFbsWIFHA4HHBrP3jjqd0L1dGH16tXaMhARUXhguaEhS05OxsKFC+Fq+BTo7Qp9AKXgqt2GnNxcTJo0KfTHJyKisMJyQwGxevVqKG8vnLUfh/zYZsshSHsjLlm9GiIS8uMTEVF4YbmhgMjJyUFZ2RS4arcBPl9Ij+08ugVJycmoqakJ6XGJiCg8sdxQwFx66SVAlwdW4+6QHdPoaILVfAAXX3QRnE5nyI5LREThi+WGAqaiogKZWVlwHdsaspv6OY5ugWU5sHLlypAcj4iIwh/LDQWMYRi49JJLYLTVwmw7FvwD9nbC1fApFi1aiJSUlOAfj4iIIgLLDQXUokWLEBsXB8fRLUE/lvPYx1DeXl7+TUREJ2C5oYByu924cOVKOJr2Qrragncgnw+uuu2YMmUqcnJygnccIiKKOCw3FHCrVq2CIQLnsW1BO4bVuAfo8uCSS3jWhoiITsRyQwGXnp6O6upquOp2AN6eoBzDdWwrRo3KREVFRVD2T0REkYvlhoJi9erVUL1dcNTvDPi+jbZjMNqOYfXqf4Bh8J8wERGdiO8MFBQlJSXIy8uHq3Z7wC8Ldx7bhhi3G0uWLAnofomIyB5YbigoRAQXX3wRpL0RZuuRwO23pwOOxj1YsngxYmNjA7ZfIiKyD5YbCpr58+cjLi4ejgBOLHbU7QB8XqxatSpg+yQiInthuaGgiYmJwfLly+Bo2gfpbh/6DpUPrrodKC0tRXZ29tD3R0REtsRyQ0F14YUXAsrnP+MyRGbzQaCzFRdddFEAkhERkV2x3FBQZWZm+lcLr/9kyBOLHbU7kJiUjFmzZgUoHRER2RHLDQXdihXLgc5WmC2HzriNL3YYfLHDzvi69HTA0bwfS5cshmVZwYhJREQ2wXcJCrqqqirExcWjp24HvEmZp92ma0zl5+7DqtsJKB+WLVsWjIhERGQjPHNDQedyubBo0UI4mvYBvZ3nvgOlEFP/CYqLSzB27NjAByQiIlthuaGQWLZsGeDzwtGw55w/1/DUAR1NWLZsaeCDERGR7bDcUEjk5eUhMysLjoZd5/y5joZdME0Tc+bMCUIyIiKyG5YbCgkRwcIFC2C2HoF0ewb/iUrB2bQHFRUVSEhICF5AIiKyDZYbCpmamhoAgNWwe9CfY7YeAbo8xz+XiIjobFhuKGRGjx6NvLx8OBsHX26sht1wOl2YOXNmEJMREZGdsNxQSM2bNxdGW+3ghqaUgqt5H2bMqITb7Q5+OCIisgWWGwqpGTNmAACspv1n3dZor4fqbudZGyIiOieDKjciskREPhaRnSJyx2led4nIT/tef1dEsgMdlOxh3LhxGJGWPqhyYzXtg4igoqIiBMmIiMguzlpuRMQE8H0ASwEUA1gjIsUnbXYNgEalVB6AJwA8HOigZA8igtlVs+BoPQx4ez93W2fzARQXlyA5OTlE6YiIyA4Gc+amHMBOpdQupVQ3gJ8AuPCkbS4E8HLf418AqBERCVxMspOZM2dC+Xphth4+4zbS3Q7x1GHmzBkhTEZERHYwmHKTCWDgGMKBvudOu41SqhdAM4DUk3ckIteJyEYR2VhbW3t+iSniTZw4EaZlwfqcctNffKZNmxaqWEREZBMhnVCslNqglJqmlJo2YsSIUB6awojL5UJxcTGs1iNn3MZsPQJ3bCzy8vJCmIyIiOxgMOXmIIDRAz7O6nvutNuIiAUgCUB9IAKSPU0pK4PRXg/0dp32dWfbEZROLoVpmiFORkREkW4w5eZ9APkiMk5EnAAuA/DaSdu8BuALfY9XA/hfpZQKXEyym9LSUkApmG1HT3lNutuBjmaUlZVqSEZERJHurOWmbw7NDQDeALANwM+UUltE5D4RWdm32Q8ApIrITgDfBHDK5eJEAxUXF8M0TZitp5ab/sIzadKkUMciIiIbsAazkVLqdQCvn/TcPQMedwK4JLDRyM5cLhfG5eRg+9G6U14zPHUwLQu5ubkakhERUaTjHYpJm/FFRXB0NAAnjWBa7XXIycmBw+HQlIyIiCIZyw1pU1BQANXTBelq/exJpWC1N6CosFBfMCIiimgsN6RNYV+BMT2fDU1JVytUb9fx14iIiM4Vyw1pM27cOBiGAaOj8fhzZkcDAPD+NkREdN5Ybkgbh8OB9JEZMDqbjz9ndPgfjxkzRlcsIiKKcCw3pNW47LGwugaUm85mpAxLRWxsrMZUREQUyVhuSKsxY8ZAOlsA5QMAmF3NGJc9VnMqIiKKZCw3pNWYMWMAnxfS1ea/Y3FnC4ekiIhoSFhuSKuMjAwAgNHVCni7oXq7MGrUKM2piIgokg3qDsVEwZKeng4AkG4PDIcbAJCWlqYzEhERRTiWG9Jq+PDhEBEY3R4oRwyAzwoPERHR+eCwFGnldDqRmJTsP3PT7QHAMzdERDQ0LDekXXp6GozuNkiXB6ZpIiUlRXckIiKKYCw3pN3w1FSY3i4YvR1ITEqGYfCfJRERnT++i5B2ycnJMHo7IT2dSElJ1h2HiIgiHMsNaZeUlATV3QHD24lhHJIiIqIhYrkh7ZKTkwHlg9HV6n9MREQ0BCw3pF1SUpL/QU/nZ4+JiIjOE8sNaRcXF3fax0REROeD5Ya0i4+PP/6Y5YaIiIaK5Ya045kbIiIKJJYb0i42Nvb4Y5YbIiIaKpYb0m5guXG73RqTEBGRHbDckHYxMTGnfUxERHQ+WG5IO6fTefyxy+XSmISIiOyA5Ya0syzr+GOWGyIiGiqWGworHJYiIqKhss6+CVHw3X777aitrUVGRobuKEREFOFYbigsLF26VHcEIiKyCQ5LERERka2w3BAREZGtsNwQERGRrbDcEBERka2w3BAREZGtsNwQERGRrbDcEBERka2w3BAREZGtsNwQERGRrbDcEBERka2w3BAREZGtsNwQERGRrbDcEBERka2w3BAREZGtsNwQERGRrbDcEBERka2w3BAREZGtsNwQERGRrbDcEBERka2w3BAREZGtsNwQERGRrYhSSs+BRWoB7NVycApXwwHU6Q5BRGGLPyNooLFKqRGne0FbuSE6mYhsVEpN052DiMITf0bQYHFYioiIiGyF5YaIiIhsheWGwskG3QGIKKzxZwQNCufcEBERka3wzA0RERHZCssNERER2QrLDQWMiNwlIltE5EMR2SwiFeexj1UiUjzg4/tEZEHf49l9+98sIu5AZiei0BERb9/38RYR+ZuI3CwiRt9rc0Wkue/1/v8WDPjcVSKiRKRowHPZfc+tG/DccBHpEZGnQ/vVUTiwdAcgexCRGQBWAJiilOoSkeEAnOexq1UAfgNgKwAope4Z8NoVAB5SSv1oqHmJSKsOpVQpAIhIGoB/A5AI4F/7Xn9bKbXiDJ+7BsCf+v781wHP7wawHMDdfR9fAmBLgHNThOCZGwqUDAB1SqkuAFBK1SmlDonIHhF5REQ+EpH3RCQPOP6b1v/2neV5U0TGiMhMACsBPNr321quiLwkIqtF5MsALgVwv4j8WETeEpHS/oOLyJ9EZLKGr5uIhkApdQzAdQBuEBH5vG1FJB5AFYBrAFx20svtALaJSP9N/v4RwM8CHJciBMsNBcrvAIwWkR0i8oyIzBnwWrNSaiKApwF8t++5pwC8rJSaBODHAJ5USv0FwGsAblVKlSqlPu3fgVLqhQGvXQHgBwCuBgARKQAQo5T6W3C/RCIKBqXULgAmgLS+p2afNCyV2/f8hQB+q5TaAaBeRKaetKufALhMREYD8AI4FIr8FH5YbigglFJtAKbC/xtYLYCfisjVfS//+4A/Z/Q9ngH/qWgAeBX+38bOxc8BrBARB4AvAXjpvIITUTh6u+8XnNKTftFZA3+BQd+fa076vN8CWAj/WZ2fhiYqhSPOuaGAUUp5AfwBwB9E5CMAX+h/aeBmATpWu4j8X/h/k7sU/mJFRBFIRHLgP9NyDMD4M2wzDMB8ABNFRMF/pkeJyK392yilukXkAwA3AyiGf5ibohDP3FBAiEihiOQPeKoUn636/o8D/vx/fY//gs/GzK8A8Hbf41YACYM87AsAngTwvlKq8XxyE5FeIjICwHoAT6vPv6vsagCvKqXGKqWylVKj4Z9EPPuk7R4DcLtSqiE4iSkS8MwNBUo8gKdEJBlAL4Cd8A9RrQCQIiIfAujCZ6eRbwTwYt9vXbUAvtj3/E8APC8iX4f/h9kZKaU+EJEWAC8G+oshoqByi8hmAA74f168CuDxAa/P7nu93zr4f3Y8fNJ+fnny80qpLeBVUlGPyy9QUInIHgDTlFJ1Qdj3KPiHwYqUUr5A75+IiCITh6UoIonI/wHwLoC7WGyIiGggnrkhIiIiW+GZGyIiIrIVlhsiIiKyFZYbIiIishWWGyIiIrIVlhsiIiKylf8P0EwTwKy/7c4AAAAASUVORK5CYII=",
            "text/plain": [
              "<Figure size 684x684 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "plt.figure(figsize=(9.5,9.5))\n",
        "sns.violinplot(data=[spotify_y_train, y_train])\n",
        "ax = plt.gca()\n",
        "ax.set_xticklabels(['Spotify', 'DEAM'])\n",
        "ax.set_title('Comparison between Spotify and DEAM')\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "We will try several models for machine learning training."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "heWt3ZYJIzfJ",
        "outputId": "aacfebee-9033-4949-e13d-92d4f4889b4f"
      },
      "outputs": [],
      "source": [
        "#libraries used for training\n",
        "from sklearn.linear_model import LinearRegression as lr\n",
        "from sklearn.model_selection import train_test_split, GridSearchCV, RandomizedSearchCV\n",
        "from sklearn.metrics import mean_squared_error, mean_absolute_error\n",
        "import math\n",
        "from sklearn.svm import SVR\n",
        "from sklearn.ensemble import RandomForestRegressor\n",
        "import xgboost as xgb\n",
        "from hyperopt import STATUS_OK, Trials, fmin, hp, tpe\n",
        "from sklearn.neighbors import KNeighborsRegressor\n",
        "from sklearn.preprocessing import StandardScaler\n",
        "!pip install catboost\n",
        "import catboost"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "63kxIgjINLIx"
      },
      "source": [
        "Model 0: Linear Regression Model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "1qxoT8asNTBo"
      },
      "outputs": [],
      "source": [
        "# defining feature, target, validation variables for model\n",
        "features_train_0 = x_train\n",
        "features_valid_0 = x_valid\n",
        "features_test_0 = x_test\n",
        "\n",
        "target_train_0 = y_train\n",
        "target_valid_0 = y_valid\n",
        "target_test_0 = y_test"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "xnmG036zNGO8",
        "outputId": "abb19303-116e-4315-f70f-110670c34131"
      },
      "outputs": [],
      "source": [
        "model_0 = lr()\n",
        "model_0.fit(features_train_0, target_train_0)\n",
        "predicted_0 = model_0.predict(features_valid_0)\n",
        "\n",
        "\n",
        "mse_0 = mean_squared_error(predicted_0, target_valid_0)\n",
        "rmse_0 = math.sqrt(mse_0)\n",
        "mae_0 = mean_absolute_error(predicted_0, target_valid_0)\n",
        "print('mae_0 :', mae_0)\n",
        "print('rmse_0 :', rmse_0)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qiKsbaLNOak4"
      },
      "source": [
        "Model_0b: Linear Regression Model, Normalized Features\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5ldG9ZydK6ld",
        "outputId": "341ab688-c4fb-4102-894c-5609fa842f2a"
      },
      "outputs": [],
      "source": [
        "all_features = pd.concat([x_train, x_valid, x_test])\n",
        "print('all features shape',all_features.shape)\n",
        "\n",
        "#norm\n",
        "scaler_0b = StandardScaler()\n",
        "scaler_0b.fit(all_features)\n",
        "\n",
        "features_train_0b = scaler_0b.transform(features_train_0)\n",
        "features_valid_0b = scaler_0b.transform(features_valid_0)\n",
        "features_test_0b = scaler_0b.transform(features_test_0)\n",
        "\n",
        "target_train_0b = target_train_0\n",
        "target_valid_0b = target_valid_0\n",
        "target_test_0 = target_test_0"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "WDunBJg3QEb2",
        "outputId": "ce495da6-ca67-487a-8ae3-bdef1005edf5"
      },
      "outputs": [],
      "source": [
        "model_0b = lr()\n",
        "model_0b.fit(features_train_0b, target_train_0b)\n",
        "predicted_0b = model_0b.predict(features_valid_0b)\n",
        "\n",
        "\n",
        "mse_0b = mean_squared_error(predicted_0b, target_valid_0b)\n",
        "rmse_0b = math.sqrt(mse_0b)\n",
        "mae_0b = mean_absolute_error(predicted_0b, target_valid_0b)\n",
        "print('mae_0b :', mae_0b)\n",
        "print('rmse_0b :', rmse_0b)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "WAlgxeXUR3xf"
      },
      "source": [
        "Model 1: XGBoost Regressor"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "tb9iOwyLSuFO"
      },
      "outputs": [],
      "source": [
        "# defining feature, target, validation variables for model\n",
        "features_train_1 = x_train\n",
        "features_valid_1 = x_valid\n",
        "features_test_1 = x_test\n",
        "\n",
        "target_train_1 = y_train\n",
        "target_valid_1 = y_valid\n",
        "target_test_1 = y_test"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "dgSJD_pCS7Ax"
      },
      "outputs": [],
      "source": [
        "#using RandomizedSearchCV for hyper parameter tuning\n",
        "\n",
        "params_1 = {'booster' : ['gbtree', 'gblinear'],\n",
        "         'eta' : list(np.arange(0.01, 0.3, 0.01)),\n",
        "         'max_depth': list(range(3,11)),\n",
        "         'gamma': list(range(1,11)),\n",
        "         'min_child_weight' : list(range(1,11))}"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "IgqrWpcXTB3J"
      },
      "outputs": [],
      "source": [
        "model_1 = xgb.XGBRegressor(seed=12345)\n",
        "\n",
        "rscv_1 = RandomizedSearchCV(estimator=model_1, param_distributions = params_1, cv=4, verbose=2, scoring = 'neg_root_mean_squared_error')\n",
        "rscv_1.fit(features_train_1, target_train_1)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "suDDxdqzS-Ok",
        "outputId": "a3b538d3-5368-47e0-efb4-419f0fc48fdc"
      },
      "outputs": [],
      "source": [
        "rscv_1.best_params_"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "9zTy4ONOTQzM",
        "outputId": "16e23d00-7f7d-4fe4-886c-d6a877dc994d"
      },
      "outputs": [],
      "source": [
        "model_1v = xgb.XGBRegressor(seed=12345,\n",
        "                            booster = rscv_1.best_params_['booster'],\n",
        "                            eta = rscv_1.best_params_['eta'],\n",
        "                            max_depth = rscv_1.best_params_['max_depth'],\n",
        "                            gamma = rscv_1.best_params_['gamma'],\n",
        "                            min_child_weight = rscv_1.best_params_['min_child_weight'])\n",
        "\n",
        "model_1v.fit(features_train_1, target_train_1)\n",
        "predicted_1 = model_1v.predict(features_valid_1)\n",
        "rmse_1 = math.sqrt(mean_squared_error(predicted_1, target_valid_1))\n",
        "mae_1 = mean_absolute_error(predicted_1, target_valid_1)\n",
        "\n",
        "print('mae_1 :', mae_1)\n",
        "print('rmse_1 :', rmse_1)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xzt4GPFuGqzy"
      },
      "source": [
        "Model 1b: XGboost Regressor, Normalized Features"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Vp8IpDneUOUY"
      },
      "outputs": [],
      "source": [
        "# defining training, validation, and testing varaibles\n",
        "\n",
        "features_train_1b = scaler_0b.transform(features_train_0)\n",
        "features_valid_1b = scaler_0b.transform(features_valid_0)\n",
        "features_test_1b = scaler_0b.transform(features_test_0)\n",
        "\n",
        "target_train_1b = target_train_0\n",
        "target_valid_1b = target_valid_0\n",
        "target_test_1b = target_test_0"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "NWuZcRuOUZwz"
      },
      "outputs": [],
      "source": [
        "#using RandomizedSearchCV for hyper parameter tuning\n",
        "\n",
        "params_1b = {'booster' : ['gbtree', 'gblinear'],\n",
        "         'eta' : list(np.arange(0.01, 0.3, 0.01)),\n",
        "         'max_depth': list(range(3,11)),\n",
        "         'gamma': list(range(1,11)),\n",
        "         'min_child_weight' : list(range(1,11))}"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "9JRBdhDoUgdQ"
      },
      "outputs": [],
      "source": [
        "model_1b = xgb.XGBRegressor(seed=12345)\n",
        "\n",
        "rscv_1b = RandomizedSearchCV(estimator=model_1b, param_distributions = params_1b, cv=4, verbose=2, scoring = 'neg_root_mean_squared_error')\n",
        "rscv_1b.fit(features_train_1b, target_train_1b)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "mY8Srm7OUseT",
        "outputId": "57c508ea-9203-46b2-9a28-cb94433e942a"
      },
      "outputs": [],
      "source": [
        "rscv_1b.best_params_"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "AyD1rs1ZUte_",
        "outputId": "0dd5fee3-3725-48cc-9d3f-10e09e10c50c"
      },
      "outputs": [],
      "source": [
        "model_1vb = xgb.XGBRegressor(seed=12345,\n",
        "                            booster = rscv_1b.best_params_['booster'],\n",
        "                            eta = rscv_1b.best_params_['eta'],\n",
        "                            max_depth = rscv_1b.best_params_['max_depth'],\n",
        "                            gamma = rscv_1b.best_params_['gamma'],\n",
        "                            min_child_weight = rscv_1b.best_params_['min_child_weight'])\n",
        "\n",
        "model_1vb.fit(features_train_1b, target_train_1b)\n",
        "predicted_1b = model_1vb.predict(features_valid_1b)\n",
        "rmse_1b = math.sqrt(mean_squared_error(predicted_1b, target_valid_1b))\n",
        "mae_1b = mean_absolute_error(predicted_1b, target_valid_1b)\n",
        "\n",
        "print('mae_1b :', mae_1b)\n",
        "print('rmse_1b :', rmse_1b)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "PvblPvLoVaSw"
      },
      "source": [
        "Model 2: Support Vector Regressor"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "w2_HP-iLVZNC"
      },
      "outputs": [],
      "source": [
        "# defining feature, target, validation variables for model\n",
        "features_train_2 = x_train\n",
        "features_valid_2 = x_valid\n",
        "features_test_2 = x_test\n",
        "\n",
        "target_train_2 = y_train\n",
        "target_valid_2 = y_valid\n",
        "target_test_2 = y_test"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "hCnKf5PGWjCs"
      },
      "outputs": [],
      "source": [
        "#using RandomizedSearchCV for hyper parameter tuning\n",
        "\n",
        "params_2 = {'kernel' : ['rbf'],\n",
        "            'gamma' : list(np.arange(0.001, 10.1, 0.1)),\n",
        "            'C' : list(np.arange(0.1, 100.1, 0.1)),\n",
        "            'epsilon' : list(np.arange(0.1, 0.5, 0.1))}\n",
        "\n",
        "params_2['gamma'].append('scale')\n",
        "params_2['gamma'].append('auto')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "sSAQ6Lw0WoDc"
      },
      "outputs": [],
      "source": [
        "model_2 = SVR()\n",
        "\n",
        "rscv_2 = RandomizedSearchCV(estimator=model_2, param_distributions = params_2, cv=3, verbose=2, scoring = 'neg_root_mean_squared_error')\n",
        "rscv_2.fit(features_train_2, target_train_2)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "M9Ee-4QrWxVf",
        "outputId": "41ea6dbc-63f7-4508-c868-5d595b1822c1"
      },
      "outputs": [],
      "source": [
        "rscv_2.best_params_"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "oDmGTP7dW1Z0",
        "outputId": "601058e8-3473-4736-a882-7344cd0ee067"
      },
      "outputs": [],
      "source": [
        "model_2v = SVR(kernel = rscv_2.best_params_['kernel'],\n",
        "               gamma = rscv_2.best_params_['gamma'],\n",
        "               epsilon = rscv_2.best_params_['epsilon'],\n",
        "               C = rscv_2.best_params_['C'])\n",
        "\n",
        "model_2v.fit(features_train_2, target_train_2)\n",
        "predicted_2 = model_2v.predict(features_valid_2)\n",
        "rmse_2 = math.sqrt(mean_squared_error(predicted_2, target_valid_2))\n",
        "mae_2 = mean_absolute_error(predicted_2, target_valid_2)\n",
        "\n",
        "print('mae_2 :', mae_2)\n",
        "print('rmse_2 :', rmse_2)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "drXocgaDXSpo"
      },
      "source": [
        "Model 2b: Support Vector Regressor, Normalized Data"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "6EBx462cXEmi"
      },
      "outputs": [],
      "source": [
        "# defining training, validation, and testing varaibles\n",
        "\n",
        "features_train_2b = scaler_0b.transform(features_train_0)\n",
        "features_valid_2b = scaler_0b.transform(features_valid_0)\n",
        "features_test_2b = scaler_0b.transform(features_test_0)\n",
        "\n",
        "target_train_2b = target_train_0\n",
        "target_valid_2b = target_valid_0\n",
        "target_test_2b = target_test_0"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "3DaLHOHZYtxj"
      },
      "outputs": [],
      "source": [
        "#using RandomizedSearchCV for hyper parameter tuning\n",
        "\n",
        "params_2b = {'kernel' : ['rbf'],\n",
        "            'gamma' : list(np.arange(0.001, 10.1, 0.1)),\n",
        "            'C' : list(np.arange(0.1, 100.1, 0.1)),\n",
        "            'epsilon' : list(np.arange(0.1, 0.5, 0.1))}\n",
        "\n",
        "params_2b['gamma'].append('scale')\n",
        "params_2b['gamma'].append('auto')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "WqWPanSzY4tm"
      },
      "outputs": [],
      "source": [
        "model_2b = SVR()\n",
        "\n",
        "rscv_2b = RandomizedSearchCV(estimator=model_2b, param_distributions = params_2b, cv=3, verbose=2, scoring = 'neg_root_mean_squared_error')\n",
        "rscv_2b.fit(features_train_2b, target_train_2b)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "TSAiH0RnZA90",
        "outputId": "fbd51cf6-05b1-42cf-99a9-5a5ad9a3fdfe"
      },
      "outputs": [],
      "source": [
        "model_2vb = SVR(kernel = rscv_2b.best_params_['kernel'],\n",
        "               gamma = rscv_2b.best_params_['gamma'],\n",
        "               epsilon = rscv_2b.best_params_['epsilon'],\n",
        "               C = rscv_2b.best_params_['C'])\n",
        "\n",
        "model_2vb.fit(features_train_2b, target_train_2b)\n",
        "predicted_2b = model_2vb.predict(features_valid_2b)\n",
        "rmse_2b = math.sqrt(mean_squared_error(predicted_2b, target_valid_2b))\n",
        "mae_2b = mean_absolute_error(predicted_2b, target_valid_2b)\n",
        "\n",
        "print('mae_2b :', mae_2b)\n",
        "print('rmse_2b :', rmse_2b)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "KqzkMMR0ajuI"
      },
      "source": [
        "Model 3: K-Nearest Neighbors Regressor"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "XHuNA-PPZT2j"
      },
      "outputs": [],
      "source": [
        "# defining feature, target, validation variables for model\n",
        "features_train_3 = x_train\n",
        "features_valid_3 = x_valid\n",
        "features_test_3 = x_test\n",
        "\n",
        "target_train_3 = y_train\n",
        "target_valid_3 = y_valid\n",
        "target_test_3 = y_test"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "bHYtA4y8bhVo"
      },
      "outputs": [],
      "source": [
        "#using RandomizedSearchCV for hyper parameter tuning\n",
        "\n",
        "params_3 = {'n_neighbors' : list(range(1,50)),\n",
        "            'weights' : ['uniform', 'distance'],\n",
        "            'p' : [1,2],\n",
        "            'metric' : ['l1','l2']}"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "D_UA3lqgbva7"
      },
      "outputs": [],
      "source": [
        "model_3 = KNeighborsRegressor()\n",
        "\n",
        "rscv_3 = RandomizedSearchCV(estimator=model_3, param_distributions = params_3, cv=4, verbose=2, scoring = 'neg_root_mean_squared_error')\n",
        "rscv_3.fit(features_train_3, target_train_3)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "CioaRm3Lb21d",
        "outputId": "d742bd6e-3b92-4e41-c231-6fdc3d72ad48"
      },
      "outputs": [],
      "source": [
        "rscv_3.best_params_"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "iSFFOJ-Rb84_",
        "outputId": "dec0f4df-7182-4548-febe-02acbaac8520"
      },
      "outputs": [],
      "source": [
        "model_3v = KNeighborsRegressor(weights = rscv_3.best_params_['weights'],\n",
        "                               p= rscv_3.best_params_['p'],\n",
        "                               n_neighbors= rscv_3.best_params_['n_neighbors'],\n",
        "                               metric = rscv_3.best_params_['metric'])\n",
        "\n",
        "model_3v.fit(features_train_3, target_train_3)\n",
        "predicted_3 = model_3v.predict(features_valid_3)\n",
        "rmse_3 = math.sqrt(mean_squared_error(predicted_3, target_valid_3))\n",
        "mae_3 = mean_absolute_error(predicted_3, target_valid_3)\n",
        "\n",
        "print('mae_3 :', mae_3)\n",
        "print('rmse_3 :', rmse_3)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "CTedEPCYcvDD"
      },
      "source": [
        "Model 3b: K-Nearest Neighbors Regressor, Normalized Data\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "xoGZNLqNcmyJ"
      },
      "outputs": [],
      "source": [
        "# defining training, validation, and testing varaibles\n",
        "\n",
        "features_train_3b = scaler_0b.transform(features_train_0)\n",
        "features_valid_3b = scaler_0b.transform(features_valid_0)\n",
        "features_test_3b = scaler_0b.transform(features_test_0)\n",
        "\n",
        "target_train_3b = target_train_0\n",
        "target_valid_3b = target_valid_0\n",
        "target_test_3b = target_test_0"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "foSr0st3c900"
      },
      "outputs": [],
      "source": [
        "#using RandomizedSearchCV for hyper parameter tuning\n",
        "\n",
        "params_3b = {'n_neighbors' : list(range(1,50)),\n",
        "            'weights' : ['uniform', 'distance'],\n",
        "            'p' : [1,2],\n",
        "            'metric' : ['l1','l2']}"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Hj5i280hdCYE"
      },
      "outputs": [],
      "source": [
        "model_3b = KNeighborsRegressor()\n",
        "\n",
        "rscv_3b = RandomizedSearchCV(estimator=model_3b, param_distributions = params_3b, cv=4, verbose=2, scoring = 'neg_root_mean_squared_error')\n",
        "rscv_3b.fit(features_train_3b, target_train_3b)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "uY2iiUT7dNR5",
        "outputId": "524a6946-236c-4d79-b17c-f3991fceb5d8"
      },
      "outputs": [],
      "source": [
        "rscv_3b.best_params_"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PLfxHkh9dOlm",
        "outputId": "d645970d-528f-4bc5-cffd-8be3f9a834e2"
      },
      "outputs": [],
      "source": [
        "model_3vb = KNeighborsRegressor(weights = rscv_3b.best_params_['weights'],\n",
        "                               p= rscv_3b.best_params_['p'],\n",
        "                               n_neighbors= rscv_3b.best_params_['n_neighbors'],\n",
        "                               metric = rscv_3b.best_params_['metric'])\n",
        "\n",
        "model_3vb.fit(features_train_3b, target_train_3b)\n",
        "predicted_3b = model_3vb.predict(features_valid_3b)\n",
        "rmse_3b = math.sqrt(mean_squared_error(predicted_3b, target_valid_3b))\n",
        "mae_3b = mean_absolute_error(predicted_3b, target_valid_3b)\n",
        "\n",
        "print('mae_3b :', mae_3b)\n",
        "print('rmse_3b :', rmse_3b)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HQyPk4A9dwDG"
      },
      "source": [
        "Model 4: CatBoost Regressor"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "2UFZ1ShUgUz0"
      },
      "outputs": [],
      "source": [
        "# defining feature, target, validation variables for model\n",
        "features_train_4 = x_train\n",
        "features_valid_4 = x_valid\n",
        "features_test_4 = x_test\n",
        "\n",
        "target_train_4 = y_train\n",
        "target_valid_4 = y_valid\n",
        "target_test_4 = y_test"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "vdmxktLxdzhB"
      },
      "outputs": [],
      "source": [
        "model_4 = catboost.CatBoostRegressor(random_state=1995, loss_function='RMSE')\n",
        "params_4 = {\n",
        "    'depth': list(range(3,11,2)),\n",
        "    'learning_rate': [0.003,0.01],\n",
        "    'iterations': [\n",
        "         #500,\n",
        "         700,\n",
        "         #1000,\n",
        "         #1500,\n",
        "        #2000\n",
        "        ]\n",
        "}"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "EFwC-kuEgt9n"
      },
      "outputs": [],
      "source": [
        "rscv_4 = RandomizedSearchCV(estimator=model_4, param_distributions = params_4, cv=3, verbose=2, scoring = 'neg_root_mean_squared_error')\n",
        "rscv_4.fit(features_train_4, target_train_4)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "calq_aR9g3m3",
        "outputId": "4d98d664-e9ad-40e0-b4cd-0232471185d4"
      },
      "outputs": [],
      "source": [
        "rscv_4.best_params_"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "c2Wh65gFtvpc"
      },
      "outputs": [],
      "source": [
        "model_4v = catboost.CatBoostRegressor(learning_rate = rscv_4.best_params_['learning_rate'],\n",
        "                               iterations = rscv_4.best_params_['iterations'],\n",
        "                               depth= rscv_4.best_params_['depth'])\n",
        "\n",
        "model_4v.fit(features_train_4, target_train_4)\n",
        "predicted_4 = model_4v.predict(features_valid_4)\n",
        "rmse_4 = math.sqrt(mean_squared_error(predicted_4, target_valid_4))\n",
        "mae_4 = mean_absolute_error(predicted_4, target_valid_4)\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Xg5MK9osD2qN",
        "outputId": "b3baa1e8-d9dd-44a7-98fa-ade33a9c15fc"
      },
      "outputs": [],
      "source": [
        "print('mae_4 :', mae_4)\n",
        "print('rmse_4 :', rmse_4)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qYr-i6q2EYyE"
      },
      "source": [
        "Model 4b: CatboostRegressor, Normalized Data"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "WY1ujNrlElc4"
      },
      "outputs": [],
      "source": [
        "# defining training, validation, and testing varaibles\n",
        "\n",
        "features_train_4b = scaler_0b.transform(features_train_0)\n",
        "features_valid_4b = scaler_0b.transform(features_valid_0)\n",
        "features_test_4b = scaler_0b.transform(features_test_0)\n",
        "\n",
        "target_train_4b = target_train_0\n",
        "target_valid_4b = target_valid_0\n",
        "target_test_4b = target_test_0"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "At2LBB_REo2l"
      },
      "outputs": [],
      "source": [
        "model_4b = catboost.CatBoostRegressor(random_state=1995, loss_function='RMSE')\n",
        "params_4b = {\n",
        "    'depth': list(range(3,11,2)),\n",
        "    'learning_rate': list(np.arange(0.001, 1, 0.01)),\n",
        "    'iterations': [\n",
        "         #500,\n",
        "         700,\n",
        "         #1000,\n",
        "         #1500,\n",
        "        #2000\n",
        "        ]\n",
        "}"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "nqL2WnqVE6Xd"
      },
      "outputs": [],
      "source": [
        "rscv_4b = RandomizedSearchCV(estimator=model_4b, param_distributions = params_4b, cv=3, verbose=2, scoring = 'neg_root_mean_squared_error')\n",
        "rscv_4b.fit(features_train_4b, target_train_4b)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "h7glkjebFInJ",
        "outputId": "7933cc45-47ff-491d-9ecb-20ab7022023f"
      },
      "outputs": [],
      "source": [
        "rscv_4b.best_params_"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "GcMI5G4KFNfe"
      },
      "outputs": [],
      "source": [
        "model_4vb = catboost.CatBoostRegressor(learning_rate = rscv_4b.best_params_['learning_rate'],\n",
        "                               iterations = rscv_4b.best_params_['iterations'],\n",
        "                               depth= rscv_4b.best_params_['depth'])\n",
        "\n",
        "model_4vb.fit(features_train_4b, target_train_4b)\n",
        "predicted_4b = model_4vb.predict(features_valid_4b)\n",
        "rmse_4b = math.sqrt(mean_squared_error(predicted_4b, target_valid_4b))\n",
        "mae_4b = mean_absolute_error(predicted_4b, target_valid_4b)\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rEtEQfLZERxl"
      },
      "source": [
        "### Performance Results and Selection of best Model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 600
        },
        "id": "yWxCy9AyEQq0",
        "outputId": "f40b567d-9ead-42fa-ff92-8131a56b42b1"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-b3ff6578-a495-417c-b750-564a94d040e0\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>model_description</th>\n",
              "      <th>model</th>\n",
              "      <th>rmse</th>\n",
              "      <th>mae</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>catboost regressor</td>\n",
              "      <td>&lt;catboost.core.CatBoostRegressor object at 0x7...</td>\n",
              "      <td>0.083495</td>\n",
              "      <td>0.068551</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>catboost regressor, normalized features</td>\n",
              "      <td>&lt;catboost.core.CatBoostRegressor object at 0x7...</td>\n",
              "      <td>0.083874</td>\n",
              "      <td>0.067704</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>linear regression</td>\n",
              "      <td>LinearRegression()</td>\n",
              "      <td>0.087565</td>\n",
              "      <td>0.070914</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>linear regression, normalized features</td>\n",
              "      <td>LinearRegression()</td>\n",
              "      <td>0.087565</td>\n",
              "      <td>0.070914</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>XGBoost regressor</td>\n",
              "      <td>XGBRegressor(seed=12345)</td>\n",
              "      <td>0.088296</td>\n",
              "      <td>0.07207</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>XGBoost regressor, normalized features</td>\n",
              "      <td>XGBRegressor(seed=12345)</td>\n",
              "      <td>0.089524</td>\n",
              "      <td>0.07306</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>k-nearest neighbors regressor, normalized feat...</td>\n",
              "      <td>KNeighborsRegressor()</td>\n",
              "      <td>0.091061</td>\n",
              "      <td>0.072954</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>k-nearest neighbors regressor</td>\n",
              "      <td>KNeighborsRegressor()</td>\n",
              "      <td>0.107722</td>\n",
              "      <td>0.089935</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>support vector regressor</td>\n",
              "      <td>SVR()</td>\n",
              "      <td>0.122508</td>\n",
              "      <td>0.10229</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>support vector regressor, normalized features</td>\n",
              "      <td>SVR()</td>\n",
              "      <td>0.122508</td>\n",
              "      <td>0.10229</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-b3ff6578-a495-417c-b750-564a94d040e0')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-b3ff6578-a495-417c-b750-564a94d040e0 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-b3ff6578-a495-417c-b750-564a94d040e0');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "                                   model_description  \\\n",
              "8                                 catboost regressor   \n",
              "9            catboost regressor, normalized features   \n",
              "0                                  linear regression   \n",
              "1             linear regression, normalized features   \n",
              "2                                  XGBoost regressor   \n",
              "3             XGBoost regressor, normalized features   \n",
              "7  k-nearest neighbors regressor, normalized feat...   \n",
              "6                      k-nearest neighbors regressor   \n",
              "4                           support vector regressor   \n",
              "5      support vector regressor, normalized features   \n",
              "\n",
              "                                               model      rmse       mae  \n",
              "8  <catboost.core.CatBoostRegressor object at 0x7...  0.083495  0.068551  \n",
              "9  <catboost.core.CatBoostRegressor object at 0x7...  0.083874  0.067704  \n",
              "0                                 LinearRegression()  0.087565  0.070914  \n",
              "1                                 LinearRegression()  0.087565  0.070914  \n",
              "2                           XGBRegressor(seed=12345)  0.088296   0.07207  \n",
              "3                           XGBRegressor(seed=12345)  0.089524   0.07306  \n",
              "7                              KNeighborsRegressor()  0.091061  0.072954  \n",
              "6                              KNeighborsRegressor()  0.107722  0.089935  \n",
              "4                                              SVR()  0.122508   0.10229  \n",
              "5                                              SVR()  0.122508   0.10229  "
            ]
          },
          "execution_count": 54,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# printing results with validation data to select best model for testing\n",
        "\n",
        "val_results = { 'model_description' : ['linear regression', 'linear regression, normalized features',\n",
        "                                       'XGBoost regressor', ' XGBoost regressor, normalized features',\n",
        "                                       'support vector regressor', 'support vector regressor, normalized features',\n",
        "                                       'k-nearest neighbors regressor', 'k-nearest neighbors regressor, normalized features',\n",
        "                                       'catboost regressor', 'catboost regressor, normalized features'],\n",
        "\n",
        "              'model': [model_0, model_0b, model_1, model_1b, model_2, model_2b, model_3, model_3b, model_4, model_4b],\n",
        "\n",
        "               'rmse' : [rmse_0, rmse_0b, rmse_1, rmse_1b, rmse_2, rmse_2b,\n",
        "                         rmse_3, rmse_3b, rmse_4, rmse_4b],\n",
        "\n",
        "               'mae'  : [mae_0, mae_0b, mae_1, mae_1b, mae_2, mae_2b,\n",
        "                         mae_3, mae_3b, mae_4, mae_4b]}\n",
        "\n",
        "v_results = pd.DataFrame.from_dict(val_results,orient='index').transpose()\n",
        "v_results.sort_values('rmse', ascending=True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "-_F6WwF1U_oU"
      },
      "outputs": [],
      "source": [
        "deam = pd.read_csv('/content/drive/MyDrive/data_project_cuetessa/deam.csv', index_col = 'Unnamed: 0')\n",
        "\n",
        "lim_up = deam.iloc[[i for i in y_test.index.values]]['valence_mean'] + 0.5*deam.iloc[[i for i in y_test.index.values]]['valence_std']\n",
        "lim_down = deam.iloc[[i for i in y_test.index.values]]['valence_mean'] - 0.5*deam.iloc[[i for i in y_test.index.values]]['valence_std']"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "BKchD18dVZrf"
      },
      "outputs": [],
      "source": [
        "def plot_ordered(data, keys=None, figsize=(12,12)):\n",
        "    if keys is None:\n",
        "        keys = ['data '+str(i) for i in range(len(data))]\n",
        "\n",
        "    sorted = pd.DataFrame({keys[i]: data[i] for i in range(len(data))}).sort_values(keys[-1], ascending=True).reset_index(drop=True)\n",
        "    plt.figure(figsize=figsize)\n",
        "    for i in keys:\n",
        "        plt.plot(sorted[i])"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8ZXFkkafJiEG"
      },
      "source": [
        "Prediction made by our final model:"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "QIfLaMZSVeNI"
      },
      "outputs": [],
      "source": [
        "predicted_test = pd.read_csv('/content/drive/MyDrive/data_project_cuetessa/predicted_test.csv', index_col = 'Unnamed: 0')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "nX_KZmxXXm3h"
      },
      "outputs": [],
      "source": [
        "predicted_test.index = y_test.index"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "LgWp7DrGVbks"
      },
      "outputs": [],
      "source": [
        "def plot_ordered(data, keys=None, figsize=(12,8)):\n",
        "    if keys is None:\n",
        "        keys = ['data '+str(i) for i in range(len(data))]\n",
        "\n",
        "    sorted = pd.DataFrame({keys[i]: data[i] for i in range(len(data))}).sort_values(keys[-1], ascending=True).reset_index(drop=True)\n",
        "    plt.figure(figsize=figsize)\n",
        "    for i in keys:\n",
        "        plt.plot(sorted[i])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "CPn66O6oZ1PG",
        "outputId": "688f5d9d-eca5-460f-a5a4-2ca33250e004"
      },
      "outputs": [],
      "source": [
        "predicted_test['0']"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 483
        },
        "id": "_xcY_lumZ5fr",
        "outputId": "75415b2f-df22-4bfd-9a20-eda135ada402"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x7fee6d2e1640>"
            ]
          },
          "execution_count": 62,
          "metadata": {},
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 864x576 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "plot_ordered( [\n",
        "    lim_up,\n",
        "    lim_down,\n",
        "     predicted_test['0']*10,\n",
        "     y_test['valence_mean']*10] )\n",
        "\n",
        "plt.legend(['mean + 0.5 std', 'mean - 0.5 std', 'predicted', 'target'])"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "Although the final predicted result of valence is not satisfactory when compared against the target, it is important to further consider the circumstances of this test. In fact, the audio files for the annotated dataset were not readily available, and spotify data was considered instead. Unfortunately, since the spotify data does not have a reported method of defining valence values, it is possible that they were using a model to assign valence values to their songs.\n",
        "\n",
        "Unfortunately, considering the flat distribution of valence values of spotify when compared with the annotated dataset, this suggests that our training data is not reliable as a source of information. I.E. if the training dataset has random values assigned to each song, then the model is more likely to behave randomly.\n",
        "\n",
        "Therefore, the main findings in this project are:\n",
        "\n",
        "- Spotify data was not a reliable source of information for valence reference.\n",
        "- Human annotated datasets are extremely important for audio related emotion trainig.\n",
        "- Valence values tend to behave as a normal distribution."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": []
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": []
    }
  ],
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}