Colab_Python_Data_Science.ipynb

{"nbformat":4,"nbformat_minor":0,"metadata":{"kernelspec":{"display_name":"Python 3","language":"python","name":"python3"},"language_info":{"codemirror_mode":{"name":"ipython","version":3},"file_extension":".py","mimetype":"text/x-python","name":"python","nbconvert_exporter":"python","pygments_lexer":"ipython3","version":"3.7.2"},"colab":{"name":"Python_Data_Science.ipynb","provenance":[],"toc_visible":true}},"cells":[{"cell_type":"code","metadata":{"id":"s74V1WgvmSGn","executionInfo":{"status":"ok","timestamp":1616171575650,"user_tz":300,"elapsed":8291,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}}},"source":["# !pip3 install scikit-learn==0.24.1"],"execution_count":1,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"NfaOZoqjl2XO"},"source":["# Python Data Science Tools Overview\n","\n","## Analysis goals \n","\n","We're going to build a model that classifies plants by species based on their flower measurements. We're hoping to find a set of flower measurements that are good at discriminating between species.\n","\n","--- \n","\n","# Table of contents\n","\n","## [0. Load packages](#pkg)\n","\n","## [1. Load Data](#load)\n","\n","### [1.1 Load Data From a Google Drive Mount](#gdrive)\n","\n","### [1.2 Load Data From a Google Sheet](#gsheet)\n","\n","### [1.2 Load Data From Your Computer](#local)\n","\n","### [1.3 Load Data From GitHub](#github)\n","\n","## [2. Describe Data](#describe)\n","\n","### [2.1 Describe Data Structure](#structure)\n","\n","### [2.2 Describe Data Types](#types)\n","\n","### [2.3 Visual Descriptions](#graphs)\n","\n","### [2.4 Data Iterables](#iter)\n","    \n","## [3. Clean Data](#clean)\n","\n","### [3.1 Clean Messy Data](#mess)\n","    \n","### [3.2 Describe Missing Data](#miss)\n","    \n","#### [3.2.1 Missing Data Can Have Meaning](#mean)\n","\n","#### [3.2.2 Missing Data vs Zero](#zero)\n","\n","#### [3.2.3 Find Missing Data by Row](#row)\n","\n","#### [3.3.4 Describe Distribution of Missing Data](#distr)\n","\n","### [3.3 Handle Missing Data](#handle)\n","\n","#### [3.3.1 Ignore Missing Data](#ignore)\n","\n","#### [3.3.2 Interpolate Missing Data](#interp)\n","\n","#### [3.3.3 Remove Missing Data](#remove)\n","\n","## [4. Classification](#class)\n","\n","### [4.1 Class Balance](#balance)\n","\n","### [4.2 Measure Standardization](#std)\n","\n","### [4.3 Multinomial Logistic Regression](#multi)\n","\n","#### [4.3.1 Fit the Model](#fit1)\n","\n","#### [4.3.2 Evaluate the Model: Accuracy](#acc)\n","\n","### [4.4 One-vs-Rest Regression](#ovr)\n","\n","#### [4.4.1 Fit the Model](#fit2)\n","\n","#### [4.4.2 Evaluate the Model: Reciever-Operator Criterion](#roc)\n"]},{"cell_type":"markdown","metadata":{"id":"qm_TMMfr_8AI"},"source":["# 0. Load packages into notebook namespace <a name=\"pkg\"></a>\n","\n","This workbook relies on data manipulation, plotting, and machine learning packages. These packages are pre-installed on Google Colab virtual machines, but must be imported before use in the notebook.\n","\n"]},{"cell_type":"code","metadata":{"id":"5e0-fO-Fl2XH","executionInfo":{"status":"ok","timestamp":1616171585352,"user_tz":300,"elapsed":1075,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}}},"source":["from platform import python_version\n","import pandas as pd\n","import numpy as np\n","import matplotlib\n","import matplotlib.pyplot as plt\n","import seaborn as sns\n","import sklearn as sk\n","from sklearn.datasets import load_iris"],"execution_count":1,"outputs":[]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"jHxFTroPl2XM","executionInfo":{"status":"ok","timestamp":1616171586659,"user_tz":300,"elapsed":235,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"dc3709c3-f6e7-4432-b87b-c40527c5c96d"},"source":["print('The current Python version is {}'.format(python_version()))\n","print('The current pandas version is {}'.format(pd.__version__))\n","print('The current numpy version is {}'.format(np.__version__))\n","print('The current matplotlib version is {}'.format(matplotlib.__version__))\n","print('The current sklearn version is {}'.format(sk.__version__))"],"execution_count":2,"outputs":[{"output_type":"stream","text":["The current Python version is 3.7.10\n","The current pandas version is 1.1.5\n","The current numpy version is 1.19.5\n","The current matplotlib version is 3.2.2\n","The current sklearn version is 0.22.2.post1\n"],"name":"stdout"}]},{"cell_type":"markdown","metadata":{"id":"Dy2yJrLUKQNb"},"source":["# 1. Load data <a name=\"load\"></a>\n","\n","We will work with an example dataset that describes flowers from several different plant species. The \"iris\" dataset is commonly used for data science and machine learning training. This dataset is approximately 3.8kb and will fit in memory.\n","\n","There are multiple ways to load data depending on (1) where the data are stored and (2) what format the data are in. We will explore a couple of different techniques for loading \"iris\" data."]},{"cell_type":"markdown","metadata":{"id":"D4YVuD9PAEzs"},"source":["## 1.1 Load text data from Google Drive mount <a name=\"gdrive\"></a>\n","\n","Google Colab can access files saved in your Google Drive, which makes loading some file types very straightforward. You must mount your Google Drive to the virtual machine running the Colab Notebook; after this is complete you will be able to browse files and load supported file types with Pandas.\n","\n","Supported file types include `.csv` (comma separated values) using the `pd.read_csv()` method, or `.txt` files that have standard character separation for columns and rows using the `pd.read_table()` method."]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"6ckrqqCuAZu9","executionInfo":{"status":"ok","timestamp":1616171590602,"user_tz":300,"elapsed":341,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"7efced25-6d2d-4f32-ec9a-0a167dc5205b"},"source":["from google.colab import drive\n","drive.mount('/content/drive')"],"execution_count":3,"outputs":[{"output_type":"stream","text":["Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n"],"name":"stdout"}]},{"cell_type":"markdown","metadata":{"id":"IsjDe9KpNs5H"},"source":["Data in the `.csv` format can be loaded with the `pd.read_csv()` function, which expects a comma (`\",\"`) to separate columns and a newline character to separate rows."]},{"cell_type":"code","metadata":{"id":"RDlLfKNGAlG4","executionInfo":{"status":"ok","timestamp":1616171593270,"user_tz":300,"elapsed":227,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}}},"source":["iris_drive_csv = pd.read_csv('/content/drive/MyDrive/Colab Notebooks/iris.csv')"],"execution_count":4,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"JGkdpLcqNxjO"},"source":["Data in the `.txt` format can be loaded with the `pd.read_table()` function.  The default column separating character is a tab (`\"\\t\"`), but if your data had a different column separating character you could indicate that with the `sep` argument. Newline characters separate rows."]},{"cell_type":"code","metadata":{"id":"ICtWo6QYMKON","executionInfo":{"status":"ok","timestamp":1616171597194,"user_tz":300,"elapsed":224,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}}},"source":["iris_drive_table = pd.read_table('/content/drive/MyDrive/Colab Notebooks/iris.txt', sep='\\t')"],"execution_count":5,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"hAqDAt4lN6uK"},"source":["## 1.2 Load data from a Google Sheet file with the Google Sheets API <a name=\"*gsheet*\"></a>\n","\n","If your data are saved in Google Sheets, there are a few extra steps. Pandas does not support the `.gsheet` format, so even if your Google Drive is mounted to your notebook virtual machine, you won't be able to access Google Sheets files directly (that I've found so far...).\n","\n","Instead, you'll need to authorize Google Colab to access your Google Drive through an API (rather than the mounted directories), and then use the Google Sheets API via the package `gspread` to load your data. This is very similar to what you would do if you were running a local Jupyter notebook (instead of running the notebook in Colab).\n","\n","After you've authenticated Google Colab's access to your Google Drive files, you will open the `.gsheet` file you want and convert it to a Pandas dataframe."]},{"cell_type":"code","metadata":{"id":"MdHLCycNAVB3","executionInfo":{"status":"ok","timestamp":1616171600404,"user_tz":300,"elapsed":779,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}}},"source":["from google.colab import auth\n","auth.authenticate_user()\n","\n","import gspread\n","from oauth2client.client import GoogleCredentials\n","\n","gc = gspread.authorize(GoogleCredentials.get_application_default())"],"execution_count":6,"outputs":[]},{"cell_type":"code","metadata":{"id":"c1hcka7BBCvE","executionInfo":{"status":"ok","timestamp":1616171603539,"user_tz":300,"elapsed":1297,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}}},"source":["iris_gsheet = gc.open_by_url('https://docs.google.com/spreadsheets/d/1dxgYstbQt0HRrPu7J4sZD60e_beybGths6XVzSpoTlw/edit?usp=sharing')\n","\n","iris_gsheet_data = iris_gsheet.worksheet('iris')\n","\n","iris_gsheet_df = pd.DataFrame(iris_gsheet_data.get_all_records()).astype(\n","    {\n","        'sepal_length': float, 'sepal_width': float, 'petal_length': float, 'petal_width': float, 'species': str\n","    }\n",")"],"execution_count":7,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"eHer7AJ3AJQR"},"source":["## 1.3 Load data from your computer <a name=\"local\"></a>\n","\n","If you have a file saved on your computer that you'd like to work with in Colab, you can upload it to Google Drive directly from your Colab Notebook and then convert to a Pandas dataframe."]},{"cell_type":"code","metadata":{"colab":{"resources":{"http://localhost:8080/nbextensions/google.colab/files.js":{"data":"Ly8gQ29weXJpZ2h0IDIwMTcgR29vZ2xlIExMQwovLwovLyBMaWNlbnNlZCB1bmRlciB0aGUgQXBhY2hlIExpY2Vuc2UsIFZlcnNpb24gMi4wICh0aGUgIkxpY2Vuc2UiKTsKLy8geW91IG1heSBub3QgdXNlIHRoaXMgZmlsZSBleGNlcHQgaW4gY29tcGxpYW5jZSB3aXRoIHRoZSBMaWNlbnNlLgovLyBZb3UgbWF5IG9idGFpbiBhIGNvcHkgb2YgdGhlIExpY2Vuc2UgYXQKLy8KLy8gICAgICBodHRwOi8vd3d3LmFwYWNoZS5vcmcvbGljZW5zZXMvTElDRU5TRS0yLjAKLy8KLy8gVW5sZXNzIHJlcXVpcmVkIGJ5IGFwcGxpY2FibGUgbGF3IG9yIGFncmVlZCB0byBpbiB3cml0aW5nLCBzb2Z0d2FyZQovLyBkaXN0cmlidXRlZCB1bmRlciB0aGUgTGljZW5zZSBpcyBkaXN0cmlidXRlZCBvbiBhbiAiQVMgSVMiIEJBU0lTLAovLyBXSVRIT1VUIFdBUlJBTlRJRVMgT1IgQ09ORElUSU9OUyBPRiBBTlkgS0lORCwgZWl0aGVyIGV4cHJlc3Mgb3IgaW1wbGllZC4KLy8gU2VlIHRoZSBMaWNlbnNlIGZvciB0aGUgc3BlY2lmaWMgbGFuZ3VhZ2UgZ292ZXJuaW5nIHBlcm1pc3Npb25zIGFuZAovLyBsaW1pdGF0aW9ucyB1bmRlciB0aGUgTGljZW5zZS4KCi8qKgogKiBAZmlsZW92ZXJ2aWV3IEhlbHBlcnMgZm9yIGdvb2dsZS5jb2xhYiBQeXRob24gbW9kdWxlLgogKi8KKGZ1bmN0aW9uKHNjb3BlKSB7CmZ1bmN0aW9uIHNwYW4odGV4dCwgc3R5bGVBdHRyaWJ1dGVzID0ge30pIHsKICBjb25zdCBlbGVtZW50ID0gZG9jdW1lbnQuY3JlYXRlRWxlbWVudCgnc3BhbicpOwogIGVsZW1lbnQudGV4dENvbnRlbnQgPSB0ZXh0OwogIGZvciAoY29uc3Qga2V5IG9mIE9iamVjdC5rZXlzKHN0eWxlQXR0cmlidXRlcykpIHsKICAgIGVsZW1lbnQuc3R5bGVba2V5XSA9IHN0eWxlQXR0cmlidXRlc1trZXldOwogIH0KICByZXR1cm4gZWxlbWVudDsKfQoKLy8gTWF4IG51bWJlciBvZiBieXRlcyB3aGljaCB3aWxsIGJlIHVwbG9hZGVkIGF0IGEgdGltZS4KY29uc3QgTUFYX1BBWUxPQURfU0laRSA9IDEwMCAqIDEwMjQ7CgpmdW5jdGlvbiBfdXBsb2FkRmlsZXMoaW5wdXRJZCwgb3V0cHV0SWQpIHsKICBjb25zdCBzdGVwcyA9IHVwbG9hZEZpbGVzU3RlcChpbnB1dElkLCBvdXRwdXRJZCk7CiAgY29uc3Qgb3V0cHV0RWxlbWVudCA9IGRvY3VtZW50LmdldEVsZW1lbnRCeUlkKG91dHB1dElkKTsKICAvLyBDYWNoZSBzdGVwcyBvbiB0aGUgb3V0cHV0RWxlbWVudCB0byBtYWtlIGl0IGF2YWlsYWJsZSBmb3IgdGhlIG5leHQgY2FsbAogIC8vIHRvIHVwbG9hZEZpbGVzQ29udGludWUgZnJvbSBQeXRob24uCiAgb3V0cHV0RWxlbWVudC5zdGVwcyA9IHN0ZXBzOwoKICByZXR1cm4gX3VwbG9hZEZpbGVzQ29udGludWUob3V0cHV0SWQpOwp9CgovLyBUaGlzIGlzIHJvdWdobHkgYW4gYXN5bmMgZ2VuZXJhdG9yIChub3Qgc3VwcG9ydGVkIGluIHRoZSBicm93c2VyIHlldCksCi8vIHdoZXJlIHRoZXJlIGFyZSBtdWx0aXBsZSBhc3luY2hyb25vdXMgc3RlcHMgYW5kIHRoZSBQeXRob24gc2lkZSBpcyBnb2luZwovLyB0byBwb2xsIGZvciBjb21wbGV0aW9uIG9mIGVhY2ggc3RlcC4KLy8gVGhpcyB1c2VzIGEgUHJvbWlzZSB0byBibG9jayB0aGUgcHl0aG9uIHNpZGUgb24gY29tcGxldGlvbiBvZiBlYWNoIHN0ZXAsCi8vIHRoZW4gcGFzc2VzIHRoZSByZXN1bHQgb2YgdGhlIHByZXZpb3VzIHN0ZXAgYXMgdGhlIGlucHV0IHRvIHRoZSBuZXh0IHN0ZXAuCmZ1bmN0aW9uIF91cGxvYWRGaWxlc0NvbnRpbnVlKG91dHB1dElkKSB7CiAgY29uc3Qgb3V0cHV0RWxlbWVudCA9IGRvY3VtZW50LmdldEVsZW1lbnRCeUlkKG91dHB1dElkKTsKICBjb25zdCBzdGVwcyA9IG91dHB1dEVsZW1lbnQuc3RlcHM7CgogIGNvbnN0IG5leHQgPSBzdGVwcy5uZXh0KG91dHB1dEVsZW1lbnQubGFzdFByb21pc2VWYWx1ZSk7CiAgcmV0dXJuIFByb21pc2UucmVzb2x2ZShuZXh0LnZhbHVlLnByb21pc2UpLnRoZW4oKHZhbHVlKSA9PiB7CiAgICAvLyBDYWNoZSB0aGUgbGFzdCBwcm9taXNlIHZhbHVlIHRvIG1ha2UgaXQgYXZhaWxhYmxlIHRvIHRoZSBuZXh0CiAgICAvLyBzdGVwIG9mIHRoZSBnZW5lcmF0b3IuCiAgICBvdXRwdXRFbGVtZW50Lmxhc3RQcm9taXNlVmFsdWUgPSB2YWx1ZTsKICAgIHJldHVybiBuZXh0LnZhbHVlLnJlc3BvbnNlOwogIH0pOwp9CgovKioKICogR2VuZXJhdG9yIGZ1bmN0aW9uIHdoaWNoIGlzIGNhbGxlZCBiZXR3ZWVuIGVhY2ggYXN5bmMgc3RlcCBvZiB0aGUgdXBsb2FkCiAqIHByb2Nlc3MuCiAqIEBwYXJhbSB7c3RyaW5nfSBpbnB1dElkIEVsZW1lbnQgSUQgb2YgdGhlIGlucHV0IGZpbGUgcGlja2VyIGVsZW1lbnQuCiAqIEBwYXJhbSB7c3RyaW5nfSBvdXRwdXRJZCBFbGVtZW50IElEIG9mIHRoZSBvdXRwdXQgZGlzcGxheS4KICogQHJldHVybiB7IUl0ZXJhYmxlPCFPYmplY3Q+fSBJdGVyYWJsZSBvZiBuZXh0IHN0ZXBzLgogKi8KZnVuY3Rpb24qIHVwbG9hZEZpbGVzU3RlcChpbnB1dElkLCBvdXRwdXRJZCkgewogIGNvbnN0IGlucHV0RWxlbWVudCA9IGRvY3VtZW50LmdldEVsZW1lbnRCeUlkKGlucHV0SWQpOwogIGlucHV0RWxlbWVudC5kaXNhYmxlZCA9IGZhbHNlOwoKICBjb25zdCBvdXRwdXRFbGVtZW50ID0gZG9jdW1lbnQuZ2V0RWxlbWVudEJ5SWQob3V0cHV0SWQpOwogIG91dHB1dEVsZW1lbnQuaW5uZXJIVE1MID0gJyc7CgogIGNvbnN0IHBpY2tlZFByb21pc2UgPSBuZXcgUHJvbWlzZSgocmVzb2x2ZSkgPT4gewogICAgaW5wdXRFbGVtZW50LmFkZEV2ZW50TGlzdGVuZXIoJ2NoYW5nZScsIChlKSA9PiB7CiAgICAgIHJlc29sdmUoZS50YXJnZXQuZmlsZXMpOwogICAgfSk7CiAgfSk7CgogIGNvbnN0IGNhbmNlbCA9IGRvY3VtZW50LmNyZWF0ZUVsZW1lbnQoJ2J1dHRvbicpOwogIGlucHV0RWxlbWVudC5wYXJlbnRFbGVtZW50LmFwcGVuZENoaWxkKGNhbmNlbCk7CiAgY2FuY2VsLnRleHRDb250ZW50ID0gJ0NhbmNlbCB1cGxvYWQnOwogIGNvbnN0IGNhbmNlbFByb21pc2UgPSBuZXcgUHJvbWlzZSgocmVzb2x2ZSkgPT4gewogICAgY2FuY2VsLm9uY2xpY2sgPSAoKSA9PiB7CiAgICAgIHJlc29sdmUobnVsbCk7CiAgICB9OwogIH0pOwoKICAvLyBXYWl0IGZvciB0aGUgdXNlciB0byBwaWNrIHRoZSBmaWxlcy4KICBjb25zdCBmaWxlcyA9IHlpZWxkIHsKICAgIHByb21pc2U6IFByb21pc2UucmFjZShbcGlja2VkUHJvbWlzZSwgY2FuY2VsUHJvbWlzZV0pLAogICAgcmVzcG9uc2U6IHsKICAgICAgYWN0aW9uOiAnc3RhcnRpbmcnLAogICAgfQogIH07CgogIGNhbmNlbC5yZW1vdmUoKTsKCiAgLy8gRGlzYWJsZSB0aGUgaW5wdXQgZWxlbWVudCBzaW5jZSBmdXJ0aGVyIHBpY2tzIGFyZSBub3QgYWxsb3dlZC4KICBpbnB1dEVsZW1lbnQuZGlzYWJsZWQgPSB0cnVlOwoKICBpZiAoIWZpbGVzKSB7CiAgICByZXR1cm4gewogICAgICByZXNwb25zZTogewogICAgICAgIGFjdGlvbjogJ2NvbXBsZXRlJywKICAgICAgfQogICAgfTsKICB9CgogIGZvciAoY29uc3QgZmlsZSBvZiBmaWxlcykgewogICAgY29uc3QgbGkgPSBkb2N1bWVudC5jcmVhdGVFbGVtZW50KCdsaScpOwogICAgbGkuYXBwZW5kKHNwYW4oZmlsZS5uYW1lLCB7Zm9udFdlaWdodDogJ2JvbGQnfSkpOwogICAgbGkuYXBwZW5kKHNwYW4oCiAgICAgICAgYCgke2ZpbGUudHlwZSB8fCAnbi9hJ30pIC0gJHtmaWxlLnNpemV9IGJ5dGVzLCBgICsKICAgICAgICBgbGFzdCBtb2RpZmllZDogJHsKICAgICAgICAgICAgZmlsZS5sYXN0TW9kaWZpZWREYXRlID8gZmlsZS5sYXN0TW9kaWZpZWREYXRlLnRvTG9jYWxlRGF0ZVN0cmluZygpIDoKICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgJ24vYSd9IC0gYCkpOwogICAgY29uc3QgcGVyY2VudCA9IHNwYW4oJzAlIGRvbmUnKTsKICAgIGxpLmFwcGVuZENoaWxkKHBlcmNlbnQpOwoKICAgIG91dHB1dEVsZW1lbnQuYXBwZW5kQ2hpbGQobGkpOwoKICAgIGNvbnN0IGZpbGVEYXRhUHJvbWlzZSA9IG5ldyBQcm9taXNlKChyZXNvbHZlKSA9PiB7CiAgICAgIGNvbnN0IHJlYWRlciA9IG5ldyBGaWxlUmVhZGVyKCk7CiAgICAgIHJlYWRlci5vbmxvYWQgPSAoZSkgPT4gewogICAgICAgIHJlc29sdmUoZS50YXJnZXQucmVzdWx0KTsKICAgICAgfTsKICAgICAgcmVhZGVyLnJlYWRBc0FycmF5QnVmZmVyKGZpbGUpOwogICAgfSk7CiAgICAvLyBXYWl0IGZvciB0aGUgZGF0YSB0byBiZSByZWFkeS4KICAgIGxldCBmaWxlRGF0YSA9IHlpZWxkIHsKICAgICAgcHJvbWlzZTogZmlsZURhdGFQcm9taXNlLAogICAgICByZXNwb25zZTogewogICAgICAgIGFjdGlvbjogJ2NvbnRpbnVlJywKICAgICAgfQogICAgfTsKCiAgICAvLyBVc2UgYSBjaHVua2VkIHNlbmRpbmcgdG8gYXZvaWQgbWVzc2FnZSBzaXplIGxpbWl0cy4gU2VlIGIvNjIxMTU2NjAuCiAgICBsZXQgcG9zaXRpb24gPSAwOwogICAgd2hpbGUgKHBvc2l0aW9uIDwgZmlsZURhdGEuYnl0ZUxlbmd0aCkgewogICAgICBjb25zdCBsZW5ndGggPSBNYXRoLm1pbihmaWxlRGF0YS5ieXRlTGVuZ3RoIC0gcG9zaXRpb24sIE1BWF9QQVlMT0FEX1NJWkUpOwogICAgICBjb25zdCBjaHVuayA9IG5ldyBVaW50OEFycmF5KGZpbGVEYXRhLCBwb3NpdGlvbiwgbGVuZ3RoKTsKICAgICAgcG9zaXRpb24gKz0gbGVuZ3RoOwoKICAgICAgY29uc3QgYmFzZTY0ID0gYnRvYShTdHJpbmcuZnJvbUNoYXJDb2RlLmFwcGx5KG51bGwsIGNodW5rKSk7CiAgICAgIHlpZWxkIHsKICAgICAgICByZXNwb25zZTogewogICAgICAgICAgYWN0aW9uOiAnYXBwZW5kJywKICAgICAgICAgIGZpbGU6IGZpbGUubmFtZSwKICAgICAgICAgIGRhdGE6IGJhc2U2NCwKICAgICAgICB9LAogICAgICB9OwogICAgICBwZXJjZW50LnRleHRDb250ZW50ID0KICAgICAgICAgIGAke01hdGgucm91bmQoKHBvc2l0aW9uIC8gZmlsZURhdGEuYnl0ZUxlbmd0aCkgKiAxMDApfSUgZG9uZWA7CiAgICB9CiAgfQoKICAvLyBBbGwgZG9uZS4KICB5aWVsZCB7CiAgICByZXNwb25zZTogewogICAgICBhY3Rpb246ICdjb21wbGV0ZScsCiAgICB9CiAgfTsKfQoKc2NvcGUuZ29vZ2xlID0gc2NvcGUuZ29vZ2xlIHx8IHt9OwpzY29wZS5nb29nbGUuY29sYWIgPSBzY29wZS5nb29nbGUuY29sYWIgfHwge307CnNjb3BlLmdvb2dsZS5jb2xhYi5fZmlsZXMgPSB7CiAgX3VwbG9hZEZpbGVzLAogIF91cGxvYWRGaWxlc0NvbnRpbnVlLAp9Owp9KShzZWxmKTsK","ok":true,"headers":[["content-type","application/javascript"]],"status":200,"status_text":""}},"base_uri":"https://localhost:8080/","height":72},"id":"BzB_GitBBMhH","executionInfo":{"status":"ok","timestamp":1616171623183,"user_tz":300,"elapsed":5381,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"a7a689b3-2e45-48c9-8cd3-541878c0d339"},"source":["from google.colab import files\n","uploaded = files.upload()"],"execution_count":9,"outputs":[{"output_type":"display_data","data":{"text/html":["\n","     <input type=\"file\" id=\"files-0a294339-b7c0-47c1-bc07-f6b5eb9cc2dd\" name=\"files[]\" multiple disabled\n","        style=\"border:none\" />\n","     <output id=\"result-0a294339-b7c0-47c1-bc07-f6b5eb9cc2dd\">\n","      Upload widget is only available when the cell has been executed in the\n","      current browser session. Please rerun this cell to enable.\n","      </output>\n","      <script src=\"/nbextensions/google.colab/files.js\"></script> "],"text/plain":["<IPython.core.display.HTML object>"]},"metadata":{"tags":[]}},{"output_type":"stream","text":["Saving iris.csv to iris (1).csv\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"41pasYUeBhEL","executionInfo":{"status":"ok","timestamp":1616171627239,"user_tz":300,"elapsed":236,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}}},"source":["import io\n","iris_upload = pd.read_csv(io.StringIO(uploaded['iris.csv'].decode('utf-8')))"],"execution_count":10,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"Co-7BzmeJ7no"},"source":["### 1.4 Load data from GitHub <a name=\"github\"></a>\n","\n","Collaborative software projects typically use version control to track changes in software; `git` is a popular version control tool and GitHub is an online service for hosting remote `git` repositories. Code and small datasets are often made publicly available through GitHub, and we can load data directly from a GitHub URL using Pandas:"]},{"cell_type":"code","metadata":{"id":"rCidi13_ASZY","executionInfo":{"status":"ok","timestamp":1616171632768,"user_tz":300,"elapsed":466,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}}},"source":["iris_github = pd.read_csv('https://raw.githubusercontent.com/kellypierce/python_data_science/master/iris.csv')"],"execution_count":11,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"gP-4EMB9l2XQ"},"source":["## 2. Describe data <a name=\"describe\"></a>\n","\n","We will proceed with two files loaded from GitHub.\n","\n","As demonstrated earlier, we can load a `.csv` file directly from GitHub with the appropriate URL:"]},{"cell_type":"code","metadata":{"id":"MnBK-9sTUnol","executionInfo":{"status":"ok","timestamp":1616171655815,"user_tz":300,"elapsed":236,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}}},"source":["iris = pd.read_csv('https://raw.githubusercontent.com/kellypierce/python_data_science/master/iris.csv')"],"execution_count":13,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"ROXxC7m3U2vY"},"source":["If you had a larger dataset that wouldn't fit in memory, you could read it in chunks. The `chunksize=` argument to `pd.read_csv()` determines the number of rows read at a time. When you use the `chunksize=` argument, `pd.read_csv()` returns an iterable. (This is more useful for local data or data in Google Drive because GitHub doesn't support storage of really large data files, but we're using a small dataset simply as an example.)"]},{"cell_type":"code","metadata":{"id":"RYqx3M7AUplN","executionInfo":{"status":"ok","timestamp":1616171657269,"user_tz":300,"elapsed":331,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}}},"source":["iris_chunks = pd.read_csv('https://raw.githubusercontent.com/kellypierce/python_data_science/master/iris.csv', chunksize=10)"],"execution_count":14,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"FyLQhb56l2XQ"},"source":["## 2.1 Describe data structure <a name=\"structure\"></a>\n","\n","We now have two example data structures loaded:\n","\n","- `iris`, a pandas dataframe\n","- `iris_chunks`, an iterable object comprised of small dataframe subsets\n","\n","First we'll expore the `iris` dataframe."]},{"cell_type":"markdown","metadata":{"id":"QrmNNxwWl2XQ"},"source":["Inspect the first several rows in the dataframe:"]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":204},"id":"O-jcN0uyl2XQ","executionInfo":{"status":"ok","timestamp":1616171659436,"user_tz":300,"elapsed":482,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"dcee3822-e9e0-4477-e93f-2675ea383aba"},"source":["iris.head()"],"execution_count":15,"outputs":[{"output_type":"execute_result","data":{"text/html":["<div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>sepal_length</th>\n","      <th>sepal_width</th>\n","      <th>petal_length</th>\n","      <th>petal_width</th>\n","      <th>species</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>0</th>\n","      <td>5.1</td>\n","      <td>3.5</td>\n","      <td>1.4</td>\n","      <td>0.2</td>\n","      <td>setosa</td>\n","    </tr>\n","    <tr>\n","      <th>1</th>\n","      <td>4.9</td>\n","      <td>3.0</td>\n","      <td>1.4</td>\n","      <td>0.2</td>\n","      <td>setosa</td>\n","    </tr>\n","    <tr>\n","      <th>2</th>\n","      <td>4.7</td>\n","      <td>3.2</td>\n","      <td>1.3</td>\n","      <td>0.2</td>\n","      <td>setosa</td>\n","    </tr>\n","    <tr>\n","      <th>3</th>\n","      <td>4.6</td>\n","      <td>3.1</td>\n","      <td>1.5</td>\n","      <td>0.2</td>\n","      <td>setosa</td>\n","    </tr>\n","    <tr>\n","      <th>4</th>\n","      <td>5.0</td>\n","      <td>3.6</td>\n","      <td>1.4</td>\n","      <td>0.2</td>\n","      <td>setosa</td>\n","    </tr>\n","  </tbody>\n","</table>\n","</div>"],"text/plain":["   sepal_length  sepal_width  petal_length  petal_width species\n","0           5.1          3.5           1.4          0.2  setosa\n","1           4.9          3.0           1.4          0.2  setosa\n","2           4.7          3.2           1.3          0.2  setosa\n","3           4.6          3.1           1.5          0.2  setosa\n","4           5.0          3.6           1.4          0.2  setosa"]},"metadata":{"tags":[]},"execution_count":15}]},{"cell_type":"markdown","metadata":{"id":"eqls49k9l2XR"},"source":["The `.head()` method displays the first 5 rows of the dataframe.\n","\n","The column on the left, without a header, represents the \"index\" of the dataframe. Note that indexing starts at 0 -- in other words, the first row is row 0.\n","\n","There are 5 columns in this dataframe. You can also access the column name attribute of the dataframe directly:"]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"yPhTVHKGl2XR","executionInfo":{"status":"ok","timestamp":1616171665137,"user_tz":300,"elapsed":421,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"1f4cdbc0-0525-4d3b-b3cb-7b5e188919c2"},"source":["iris.columns"],"execution_count":16,"outputs":[{"output_type":"execute_result","data":{"text/plain":["Index(['sepal_length', 'sepal_width', 'petal_length', 'petal_width',\n","       'species'],\n","      dtype='object')"]},"metadata":{"tags":[]},"execution_count":16}]},{"cell_type":"markdown","metadata":{"id":"GXRa3pI3l2XS"},"source":["We can access individual columns in the dataframe by their names. For example, we can look only at the data in the `sepal_length` column:"]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"IC_GiVGBl2XS","executionInfo":{"status":"ok","timestamp":1616171672170,"user_tz":300,"elapsed":247,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"2c115d6d-cb44-47d0-ae9b-14101d1c9a95"},"source":["iris['sepal_length']"],"execution_count":17,"outputs":[{"output_type":"execute_result","data":{"text/plain":["0      5.1\n","1      4.9\n","2      4.7\n","3      4.6\n","4      5.0\n","      ... \n","145    6.7\n","146    6.3\n","147    6.5\n","148    6.2\n","149    5.9\n","Name: sepal_length, Length: 150, dtype: float64"]},"metadata":{"tags":[]},"execution_count":17}]},{"cell_type":"markdown","metadata":{"id":"TizShGaPl2XS"},"source":["Referencing a single column returns a pandas series, not a dataframe.\n","\n","We can also look at two columns at a time by referencing a list of names rather than a single name. This will return a dataframe:"]},{"cell_type":"code","metadata":{"scrolled":true,"colab":{"base_uri":"https://localhost:8080/","height":419},"id":"LpfrVCZHl2XS","executionInfo":{"status":"ok","timestamp":1616171674764,"user_tz":300,"elapsed":230,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"b31e0d43-0902-4758-d15b-219fee786aad"},"source":["iris[['sepal_length', 'petal_length']]"],"execution_count":18,"outputs":[{"output_type":"execute_result","data":{"text/html":["<div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>sepal_length</th>\n","      <th>petal_length</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>0</th>\n","      <td>5.1</td>\n","      <td>1.4</td>\n","    </tr>\n","    <tr>\n","      <th>1</th>\n","      <td>4.9</td>\n","      <td>1.4</td>\n","    </tr>\n","    <tr>\n","      <th>2</th>\n","      <td>4.7</td>\n","      <td>1.3</td>\n","    </tr>\n","    <tr>\n","      <th>3</th>\n","      <td>4.6</td>\n","      <td>1.5</td>\n","    </tr>\n","    <tr>\n","      <th>4</th>\n","      <td>5.0</td>\n","      <td>1.4</td>\n","    </tr>\n","    <tr>\n","      <th>...</th>\n","      <td>...</td>\n","      <td>...</td>\n","    </tr>\n","    <tr>\n","      <th>145</th>\n","      <td>6.7</td>\n","      <td>5.2</td>\n","    </tr>\n","    <tr>\n","      <th>146</th>\n","      <td>6.3</td>\n","      <td>5.0</td>\n","    </tr>\n","    <tr>\n","      <th>147</th>\n","      <td>6.5</td>\n","      <td>5.2</td>\n","    </tr>\n","    <tr>\n","      <th>148</th>\n","      <td>6.2</td>\n","      <td>5.4</td>\n","    </tr>\n","    <tr>\n","      <th>149</th>\n","      <td>5.9</td>\n","      <td>5.1</td>\n","    </tr>\n","  </tbody>\n","</table>\n","<p>150 rows × 2 columns</p>\n","</div>"],"text/plain":["     sepal_length  petal_length\n","0             5.1           1.4\n","1             4.9           1.4\n","2             4.7           1.3\n","3             4.6           1.5\n","4             5.0           1.4\n","..            ...           ...\n","145           6.7           5.2\n","146           6.3           5.0\n","147           6.5           5.2\n","148           6.2           5.4\n","149           5.9           5.1\n","\n","[150 rows x 2 columns]"]},"metadata":{"tags":[]},"execution_count":18}]},{"cell_type":"markdown","metadata":{"id":"oZWMG-4_l2XS"},"source":["We can also learn how many rows and columns are in the dataframe with the `shape` method:"]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"mjAUbmGxl2XT","executionInfo":{"status":"ok","timestamp":1616171676732,"user_tz":300,"elapsed":342,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"eaae7343-4f81-40f6-b707-f46185ded3a1"},"source":["iris.shape"],"execution_count":19,"outputs":[{"output_type":"execute_result","data":{"text/plain":["(150, 5)"]},"metadata":{"tags":[]},"execution_count":19}]},{"cell_type":"markdown","metadata":{"id":"8WdD657_l2XT"},"source":["`shape` returns a tuple of (n rows, n columns)."]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"nEDiJ3f9l2XT","executionInfo":{"status":"ok","timestamp":1616171679355,"user_tz":300,"elapsed":231,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"f7a20d2a-f7bf-4958-f5a2-71a89cff1ee0"},"source":["print('There are {} rows in the iris dataframe'.format(iris.shape[0]))\n","print('There are {} columns in the iris dataframe'.format(iris.shape[1]))"],"execution_count":20,"outputs":[{"output_type":"stream","text":["There are 150 rows in the iris dataframe\n","There are 5 columns in the iris dataframe\n"],"name":"stdout"}]},{"cell_type":"markdown","metadata":{"id":"OgKmWn4Ml2XT"},"source":["### 2.2 Describe data types <a name=\"types\"></a>\n","\n","The `dtypes` method will report the data type in each column.\n","\n","If you expect numbers in certain columns, you can verify that they contain either `float64` or `integer` data types. Pandas dataframe columns cannot have mixed data types, so Pandas will interpret any column that contains both numbers and character strings as containing character strings."]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"-M4DWfUwl2XU","executionInfo":{"status":"ok","timestamp":1616171683283,"user_tz":300,"elapsed":243,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"56aa7f75-93b3-4843-edd2-041e7b822013"},"source":["iris.dtypes"],"execution_count":21,"outputs":[{"output_type":"execute_result","data":{"text/plain":["sepal_length    float64\n","sepal_width     float64\n","petal_length    float64\n","petal_width     float64\n","species          object\n","dtype: object"]},"metadata":{"tags":[]},"execution_count":21}]},{"cell_type":"markdown","metadata":{"id":"wMTYNKksl2XU"},"source":["The `describe()` method will provide summary statistics for columns containing numeric data:"]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":297},"id":"sHBO-Ce-l2XU","executionInfo":{"status":"ok","timestamp":1616171685283,"user_tz":300,"elapsed":239,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"459b618e-bc16-4826-d661-3dfe0fe04961"},"source":["iris.describe()"],"execution_count":22,"outputs":[{"output_type":"execute_result","data":{"text/html":["<div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>sepal_length</th>\n","      <th>sepal_width</th>\n","      <th>petal_length</th>\n","      <th>petal_width</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>count</th>\n","      <td>150.000000</td>\n","      <td>150.000000</td>\n","      <td>150.000000</td>\n","      <td>150.000000</td>\n","    </tr>\n","    <tr>\n","      <th>mean</th>\n","      <td>5.843333</td>\n","      <td>3.057333</td>\n","      <td>3.758000</td>\n","      <td>1.199333</td>\n","    </tr>\n","    <tr>\n","      <th>std</th>\n","      <td>0.828066</td>\n","      <td>0.435866</td>\n","      <td>1.765298</td>\n","      <td>0.762238</td>\n","    </tr>\n","    <tr>\n","      <th>min</th>\n","      <td>4.300000</td>\n","      <td>2.000000</td>\n","      <td>1.000000</td>\n","      <td>0.100000</td>\n","    </tr>\n","    <tr>\n","      <th>25%</th>\n","      <td>5.100000</td>\n","      <td>2.800000</td>\n","      <td>1.600000</td>\n","      <td>0.300000</td>\n","    </tr>\n","    <tr>\n","      <th>50%</th>\n","      <td>5.800000</td>\n","      <td>3.000000</td>\n","      <td>4.350000</td>\n","      <td>1.300000</td>\n","    </tr>\n","    <tr>\n","      <th>75%</th>\n","      <td>6.400000</td>\n","      <td>3.300000</td>\n","      <td>5.100000</td>\n","      <td>1.800000</td>\n","    </tr>\n","    <tr>\n","      <th>max</th>\n","      <td>7.900000</td>\n","      <td>4.400000</td>\n","      <td>6.900000</td>\n","      <td>2.500000</td>\n","    </tr>\n","  </tbody>\n","</table>\n","</div>"],"text/plain":["       sepal_length  sepal_width  petal_length  petal_width\n","count    150.000000   150.000000    150.000000   150.000000\n","mean       5.843333     3.057333      3.758000     1.199333\n","std        0.828066     0.435866      1.765298     0.762238\n","min        4.300000     2.000000      1.000000     0.100000\n","25%        5.100000     2.800000      1.600000     0.300000\n","50%        5.800000     3.000000      4.350000     1.300000\n","75%        6.400000     3.300000      5.100000     1.800000\n","max        7.900000     4.400000      6.900000     2.500000"]},"metadata":{"tags":[]},"execution_count":22}]},{"cell_type":"markdown","metadata":{"id":"M47dgqJYl2XU"},"source":["Note that one column, `species`, is not included in the `describe()` output. The `species` column contains the species names, which are character strings. We can learn about which species are present using the `unique` method:"]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"ZL79mL5al2XU","executionInfo":{"status":"ok","timestamp":1616171688261,"user_tz":300,"elapsed":249,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"3dad4e47-95af-4f31-c051-e59366331c23"},"source":["iris['species'].unique()"],"execution_count":23,"outputs":[{"output_type":"execute_result","data":{"text/plain":["array(['setosa', 'versicolor', 'virginica'], dtype=object)"]},"metadata":{"tags":[]},"execution_count":23}]},{"cell_type":"markdown","metadata":{"id":"BU84k6oil2XV"},"source":["We can see that there are three plant species represented in this dataframe.\n","\n","The `value_counts()` method will count the number of observations of each unique value in a column:"]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"TtzTGB7Ql2XV","executionInfo":{"status":"ok","timestamp":1616171690288,"user_tz":300,"elapsed":230,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"ee69f61b-da02-4e6d-ed0c-c72f1d27d401"},"source":["iris['species'].value_counts()"],"execution_count":24,"outputs":[{"output_type":"execute_result","data":{"text/plain":["versicolor    50\n","virginica     50\n","setosa        50\n","Name: species, dtype: int64"]},"metadata":{"tags":[]},"execution_count":24}]},{"cell_type":"markdown","metadata":{"id":"SPJuJYbZl2XV"},"source":["You can see that there are 50 observations of each species. As we saw above, there are 150 rows in this data frame. 50 observations x 3 species = 150 observations total, which happens to be the same number of rows we have in the dataframe. This is a good sanity check that we understand the contents of the dataframe."]},{"cell_type":"markdown","metadata":{"id":"K3X57BP_l2XV"},"source":["### 2.3 Visual description <a name=\"graphs\"></a>\n","\n","We can make some simple descriptive graphs to get a better understanding of our data. Because we know we have four columns of numeric data (sepal width, sepal length, petal width, petal length), we can make scatterplots of each pairwise combination of variables.\n","\n","The easiest way to do this is with the `pairplot()` method in the graphing library `seaborn`:"]},{"cell_type":"code","metadata":{"scrolled":false,"colab":{"base_uri":"https://localhost:8080/","height":743},"id":"jsydQLTAl2XV","executionInfo":{"status":"ok","timestamp":1616171700336,"user_tz":300,"elapsed":7926,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"8870d9dc-e5dd-4471-ddc7-4ede3a6e0ab2"},"source":["sns.pairplot(iris, hue='species')"],"execution_count":25,"outputs":[{"output_type":"execute_result","data":{"text/plain":["<seaborn.axisgrid.PairGrid at 0x7f2c02b90750>"]},"metadata":{"tags":[]},"execution_count":25},{"output_type":"display_data","data":{"image/png":"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\n","text/plain":["<Figure size 804.75x720 with 20 Axes>"]},"metadata":{"tags":[],"needs_background":"light"}}]},{"cell_type":"markdown","metadata":{"id":"-QLecERKl2XW"},"source":["Which variable looks to be the most variable between species?"]},{"cell_type":"markdown","metadata":{"id":"f-NmQ6F3l2XW"},"source":["### 2.4 Apply description tools to a data iterable <a name=\"iter\"></a>\n","\n","Now let's look at `iris_chunks`. If you try to look at the `head()` of this object you'll get an error because it isn't a dataframe, it's an iterable object of class `TextFileReader`:"]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":164},"id":"F88h2LiYl2XW","executionInfo":{"status":"error","timestamp":1616171707132,"user_tz":300,"elapsed":265,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"45398af4-e744-4b8d-90f5-0d6c396281e4"},"source":["iris_chunks.head()"],"execution_count":26,"outputs":[{"output_type":"error","ename":"AttributeError","evalue":"ignored","traceback":["\u001b[0;31m---------------------------------------------------------------------------\u001b[0m","\u001b[0;31mAttributeError\u001b[0m                            Traceback (most recent call last)","\u001b[0;32m<ipython-input-26-ff61412b3e88>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0miris_chunks\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mhead\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m","\u001b[0;31mAttributeError\u001b[0m: 'TextFileReader' object has no attribute 'head'"]}]},{"cell_type":"markdown","metadata":{"id":"UHtXec6zl2XW"},"source":["You can interact with `iris_chunks` by iterating over the chunks of data in this object. Each of those chunks is a dataframe and can be accessed using the methods described above."]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"CFGjtZVbl2XW","executionInfo":{"status":"ok","timestamp":1616171716439,"user_tz":300,"elapsed":227,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"b51b9f00-606b-44b3-d1e4-a95ab743e4f9"},"source":["for i, chunk in enumerate(iris_chunks):\n","    print('Chunk {} has shape {}'.format(i, chunk.shape))"],"execution_count":27,"outputs":[{"output_type":"stream","text":["Chunk 0 has shape (10, 5)\n","Chunk 1 has shape (10, 5)\n","Chunk 2 has shape (10, 5)\n","Chunk 3 has shape (10, 5)\n","Chunk 4 has shape (10, 5)\n","Chunk 5 has shape (10, 5)\n","Chunk 6 has shape (10, 5)\n","Chunk 7 has shape (10, 5)\n","Chunk 8 has shape (10, 5)\n","Chunk 9 has shape (10, 5)\n","Chunk 10 has shape (10, 5)\n","Chunk 11 has shape (10, 5)\n","Chunk 12 has shape (10, 5)\n","Chunk 13 has shape (10, 5)\n","Chunk 14 has shape (10, 5)\n"],"name":"stdout"}]},{"cell_type":"markdown","metadata":{"id":"s9zWoR_Yl2XX"},"source":["If you try to iterate through the list a second time, nothing happens:"]},{"cell_type":"code","metadata":{"id":"UN9lZM_ml2XX","executionInfo":{"status":"ok","timestamp":1616171719686,"user_tz":300,"elapsed":242,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}}},"source":["for i, chunk in enumerate(iris_chunks):\n","    print('Chunk {} has shape {}'.format(i, chunk.shape))"],"execution_count":28,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"BuLEPvJgl2XX"},"source":["This is because the `iris_chunk` iterable is mutable and behaves much like a queue. Once elements from it have been accessed they are dropped from memory.\n","\n","If you have a large data file that doesn't fit in memory, you can access each chunk, summarize in a meaningful way, and assign the summaries to new variables such that everything fits in memory.\n","\n","We'll close our discussion of iterable dataframes and delete the `iris_chunks` variable because the tutorial will use only small data structures. You can delete variables you no longer need using the built in function `del`."]},{"cell_type":"code","metadata":{"id":"a8kCsV4Wl2XX","executionInfo":{"status":"ok","timestamp":1616171724060,"user_tz":300,"elapsed":234,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}}},"source":["del iris_chunks"],"execution_count":29,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"BkdkrkA_l2XX"},"source":["## 3. Clean data <a name=\"clean\"></a>\n","\n","Good models require good data.\n","\n","Let's load in a new dataset, `iris_missing_data.csv`. This is a copy of `iris.csv` with a few values intentionally removed or altered so you can practice finding different types of missing or messy data."]},{"cell_type":"code","metadata":{"id":"x3-jdxuul2XX","executionInfo":{"status":"ok","timestamp":1616171728103,"user_tz":300,"elapsed":367,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}}},"source":["iris_missing = pd.read_csv('https://raw.githubusercontent.com/kellypierce/python_data_science/master/iris_missing_data.csv')"],"execution_count":30,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"vbVBaSWfl2XY"},"source":["**On your own:**\n","\n","- Use the `head()` method to look at the first few rows of data.\n","- Use the `shape` method to determine the number of rows and columns.\n","- Use the `value_counts()` method to see how many observations of each species are present.\n","- Use the `dtypes` method to inspect the column datatypes.\n","- Use the `describe()` method to look at quantitative summaries of numerical columns.\n","\n","Based on the results of this exploration, where do you think the missing data are?"]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":204},"id":"oEoVjT3el2XY","executionInfo":{"status":"ok","timestamp":1616171733066,"user_tz":300,"elapsed":241,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"e8d733cc-e5bd-498b-9c07-3f4525e2e4af"},"source":["iris_missing.head()"],"execution_count":31,"outputs":[{"output_type":"execute_result","data":{"text/html":["<div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>sepal_length</th>\n","      <th>sepal_width</th>\n","      <th>petal_length</th>\n","      <th>petal_width</th>\n","      <th>species</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>0</th>\n","      <td>4.9</td>\n","      <td>3.1</td>\n","      <td>1.5</td>\n","      <td>0.1</td>\n","      <td>setosa</td>\n","    </tr>\n","    <tr>\n","      <th>1</th>\n","      <td>4.8</td>\n","      <td>3.0</td>\n","      <td>1.4</td>\n","      <td>0.1</td>\n","      <td>setosa</td>\n","    </tr>\n","    <tr>\n","      <th>2</th>\n","      <td>4.3</td>\n","      <td>3.0</td>\n","      <td>1.1</td>\n","      <td>NaN</td>\n","      <td>setosa</td>\n","    </tr>\n","    <tr>\n","      <th>3</th>\n","      <td>5.2</td>\n","      <td>4.1</td>\n","      <td>1.5</td>\n","      <td>0.1</td>\n","      <td>setosa</td>\n","    </tr>\n","    <tr>\n","      <th>4</th>\n","      <td>4.9</td>\n","      <td>3.6</td>\n","      <td>1.4</td>\n","      <td>0.1</td>\n","      <td>setosa</td>\n","    </tr>\n","  </tbody>\n","</table>\n","</div>"],"text/plain":["   sepal_length  sepal_width  petal_length petal_width species\n","0           4.9          3.1           1.5         0.1  setosa\n","1           4.8          3.0           1.4         0.1  setosa\n","2           4.3          3.0           1.1         NaN  setosa\n","3           5.2          4.1           1.5         0.1  setosa\n","4           4.9          3.6           1.4         0.1  setosa"]},"metadata":{"tags":[]},"execution_count":31}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"f4g6q9jpl2XY","executionInfo":{"status":"ok","timestamp":1616171735433,"user_tz":300,"elapsed":497,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"09774876-81b9-4275-a89c-d85f64fab501"},"source":["iris_missing.shape"],"execution_count":32,"outputs":[{"output_type":"execute_result","data":{"text/plain":["(150, 5)"]},"metadata":{"tags":[]},"execution_count":32}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"A9T6e9H4l2XY","executionInfo":{"status":"ok","timestamp":1616171737325,"user_tz":300,"elapsed":229,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"235073d0-6bfd-4749-80bb-d88dd2ba9f51"},"source":["iris_missing['species'].value_counts()"],"execution_count":33,"outputs":[{"output_type":"execute_result","data":{"text/plain":["versicolor    50\n","virginica     50\n","setosa        50\n","Name: species, dtype: int64"]},"metadata":{"tags":[]},"execution_count":33}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"yHPxDnqwl2XY","executionInfo":{"status":"ok","timestamp":1616171738691,"user_tz":300,"elapsed":235,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"b2ae872a-7332-47a4-d078-8be96e3a842c"},"source":["iris_missing.dtypes"],"execution_count":34,"outputs":[{"output_type":"execute_result","data":{"text/plain":["sepal_length    float64\n","sepal_width     float64\n","petal_length    float64\n","petal_width      object\n","species          object\n","dtype: object"]},"metadata":{"tags":[]},"execution_count":34}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":297},"id":"fB9vRO6jl2XZ","executionInfo":{"status":"ok","timestamp":1616171740230,"user_tz":300,"elapsed":410,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"795ad189-274f-4186-fdf9-dd2f1ce71bcf"},"source":["iris_missing.describe()"],"execution_count":35,"outputs":[{"output_type":"execute_result","data":{"text/html":["<div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>sepal_length</th>\n","      <th>sepal_width</th>\n","      <th>petal_length</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>count</th>\n","      <td>150.000000</td>\n","      <td>150.000000</td>\n","      <td>150.000000</td>\n","    </tr>\n","    <tr>\n","      <th>mean</th>\n","      <td>5.843333</td>\n","      <td>3.057333</td>\n","      <td>3.758000</td>\n","    </tr>\n","    <tr>\n","      <th>std</th>\n","      <td>0.828066</td>\n","      <td>0.435866</td>\n","      <td>1.765298</td>\n","    </tr>\n","    <tr>\n","      <th>min</th>\n","      <td>4.300000</td>\n","      <td>2.000000</td>\n","      <td>1.000000</td>\n","    </tr>\n","    <tr>\n","      <th>25%</th>\n","      <td>5.100000</td>\n","      <td>2.800000</td>\n","      <td>1.600000</td>\n","    </tr>\n","    <tr>\n","      <th>50%</th>\n","      <td>5.800000</td>\n","      <td>3.000000</td>\n","      <td>4.350000</td>\n","    </tr>\n","    <tr>\n","      <th>75%</th>\n","      <td>6.400000</td>\n","      <td>3.300000</td>\n","      <td>5.100000</td>\n","    </tr>\n","    <tr>\n","      <th>max</th>\n","      <td>7.900000</td>\n","      <td>4.400000</td>\n","      <td>6.900000</td>\n","    </tr>\n","  </tbody>\n","</table>\n","</div>"],"text/plain":["       sepal_length  sepal_width  petal_length\n","count    150.000000   150.000000    150.000000\n","mean       5.843333     3.057333      3.758000\n","std        0.828066     0.435866      1.765298\n","min        4.300000     2.000000      1.000000\n","25%        5.100000     2.800000      1.600000\n","50%        5.800000     3.000000      4.350000\n","75%        6.400000     3.300000      5.100000\n","max        7.900000     4.400000      6.900000"]},"metadata":{"tags":[]},"execution_count":35}]},{"cell_type":"markdown","metadata":{"id":"QRKOAx-hl2XZ"},"source":["### 3.1 Clean messy data <a name=\"mess\"></a>\n","\n","You'll notice that the data types in `iris_missing` are not quite what we expect: the `petal_width` column data type is `object`, not `float64`. This indicates that some of the data in this column are not numeric. But when we look at the `head()` of the dataframe we see numbers (`NaN` is a numeric datatype in Python).\n","\n","Before we proceed with our exploration we need to identify the non-numeric data.\n","\n","The first thing we can try is to convert the values in column `petal_width` to floats:"]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":215},"id":"z4gcvdJwl2XZ","executionInfo":{"status":"error","timestamp":1616171743824,"user_tz":300,"elapsed":246,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"515c68ee-41b4-45c0-e1bc-b3d976f403c3"},"source":["iris_missing['petal_width'] = [float(i) for i in iris_missing['petal_width']]"],"execution_count":36,"outputs":[{"output_type":"error","ename":"ValueError","evalue":"ignored","traceback":["\u001b[0;31m---------------------------------------------------------------------------\u001b[0m","\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)","\u001b[0;32m<ipython-input-36-f18659f388e0>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0miris_missing\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'petal_width'\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mfloat\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mi\u001b[0m \u001b[0;32min\u001b[0m \u001b[0miris_missing\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'petal_width'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m","\u001b[0;32m<ipython-input-36-f18659f388e0>\u001b[0m in \u001b[0;36m<listcomp>\u001b[0;34m(.0)\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0miris_missing\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'petal_width'\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mfloat\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mi\u001b[0m \u001b[0;32min\u001b[0m \u001b[0miris_missing\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'petal_width'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m","\u001b[0;31mValueError\u001b[0m: could not convert string to float: 'O.6'"]}]},{"cell_type":"markdown","metadata":{"id":"FWuWB6Zcl2XZ"},"source":["While this first attempt didn't work, the error helps us to locate the problem. At least one number has the letter 'o' in place of a zero. We can use string replacement to fix this, then convert the string to a float."]},{"cell_type":"code","metadata":{"id":"xl5pr6pcl2Xa","executionInfo":{"status":"ok","timestamp":1616171747250,"user_tz":300,"elapsed":241,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}}},"source":["iris_missing['petal_width'] = [float(i.replace('O', '0')) if isinstance(i, str) else i for i in iris_missing['petal_width']]\n"],"execution_count":37,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"wyEjHGiol2Xa"},"source":["No errors -- that worked! \n","\n","Let's break this line down to understand how it works:\n","\n","This is a list comprehension, which is a for-loop flattened out into a single line. This code will iterate over all the values in `iris_missing['petal_width']` to perform the type conversion.\n","\n","    [float(i.replace('O', 0)) if isinstance(i, str) else i for i in iris_missing['petal_width']]\n","    \n","This is the same as `for i in iris_missing['petal_width']: ...`\n","\n","The \"O\" is removed with `i.replace('O', 0)`, which will take the string variable `i` and replace it with the number 0.\n","\n","`i.replace()` won't work if `i` is actually a numeric type, because numeric types don't have a `replace()` method. Not all of the values in `iris_missing` are strings -- some are already floats. We handle this with an `if else` clause: `isinstance(i, str)` we will use the `replace()` method, `else i` means if `i` isn't a string we will just keep it as is.\n","\n","The whole list comprehension returns a list, which is then assigned to be the new value of the column `iris_missing['petal_width']`."]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"hN0e_5gMl2Xa","executionInfo":{"status":"ok","timestamp":1616171753326,"user_tz":300,"elapsed":560,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"1fd2e90e-227e-4c93-e5d7-6d112183eb01"},"source":["iris_missing.dtypes"],"execution_count":38,"outputs":[{"output_type":"execute_result","data":{"text/plain":["sepal_length    float64\n","sepal_width     float64\n","petal_length    float64\n","petal_width     float64\n","species          object\n","dtype: object"]},"metadata":{"tags":[]},"execution_count":38}]},{"cell_type":"markdown","metadata":{"id":"62JcQYkTl2Xa"},"source":["Now the `petal_width` column contains `float64` data, as expected."]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":297},"id":"CLg0eiWXl2Xb","executionInfo":{"status":"ok","timestamp":1616171755596,"user_tz":300,"elapsed":297,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"1caa84ac-86d8-48ef-bedf-dd648d085bb1"},"source":["iris_missing.describe()"],"execution_count":39,"outputs":[{"output_type":"execute_result","data":{"text/html":["<div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>sepal_length</th>\n","      <th>sepal_width</th>\n","      <th>petal_length</th>\n","      <th>petal_width</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>count</th>\n","      <td>150.000000</td>\n","      <td>150.000000</td>\n","      <td>150.000000</td>\n","      <td>140.000000</td>\n","    </tr>\n","    <tr>\n","      <th>mean</th>\n","      <td>5.843333</td>\n","      <td>3.057333</td>\n","      <td>3.758000</td>\n","      <td>1.268571</td>\n","    </tr>\n","    <tr>\n","      <th>std</th>\n","      <td>0.828066</td>\n","      <td>0.435866</td>\n","      <td>1.765298</td>\n","      <td>0.741580</td>\n","    </tr>\n","    <tr>\n","      <th>min</th>\n","      <td>4.300000</td>\n","      <td>2.000000</td>\n","      <td>1.000000</td>\n","      <td>0.100000</td>\n","    </tr>\n","    <tr>\n","      <th>25%</th>\n","      <td>5.100000</td>\n","      <td>2.800000</td>\n","      <td>1.600000</td>\n","      <td>0.400000</td>\n","    </tr>\n","    <tr>\n","      <th>50%</th>\n","      <td>5.800000</td>\n","      <td>3.000000</td>\n","      <td>4.350000</td>\n","      <td>1.400000</td>\n","    </tr>\n","    <tr>\n","      <th>75%</th>\n","      <td>6.400000</td>\n","      <td>3.300000</td>\n","      <td>5.100000</td>\n","      <td>1.800000</td>\n","    </tr>\n","    <tr>\n","      <th>max</th>\n","      <td>7.900000</td>\n","      <td>4.400000</td>\n","      <td>6.900000</td>\n","      <td>2.500000</td>\n","    </tr>\n","  </tbody>\n","</table>\n","</div>"],"text/plain":["       sepal_length  sepal_width  petal_length  petal_width\n","count    150.000000   150.000000    150.000000   140.000000\n","mean       5.843333     3.057333      3.758000     1.268571\n","std        0.828066     0.435866      1.765298     0.741580\n","min        4.300000     2.000000      1.000000     0.100000\n","25%        5.100000     2.800000      1.600000     0.400000\n","50%        5.800000     3.000000      4.350000     1.400000\n","75%        6.400000     3.300000      5.100000     1.800000\n","max        7.900000     4.400000      6.900000     2.500000"]},"metadata":{"tags":[]},"execution_count":39}]},{"cell_type":"markdown","metadata":{"id":"nXwAZdqol2Xb"},"source":["### 3.2 Describe missing data <a name=\"miss\"></a>\n","\n","Most real world data are missing some values. Finding holes in datasets and learning how to handle them is a critical part of doing data science. Pandas offers a number of tools to help you handling missing data.\n","\n","#### 3.2.1 Missing data may have meaning <a name=\"mean\"></a>\n","\n","It's possible that some types of measurements are more likely to be missed than others. If that's the case, seeing missing measurements in your dataset can tell you something useful about your data and you probably will want to retain the missing data in your analysis.\n","\n","Alternatively, if the data are missing at random then they likely just represent noise and lack meaning. In this case it's appropriate to handle them using one of the techniques below.\n","\n","#### 3.2.2 Missing data vs zero <a name=\"zero\"></a>\n","\n","Missing data is data that were not measured or not recorded.\n","\n","Zero is a measurement of, well, zero.\n","\n","Sometimes it can be hard to tell if empty cells in your dataframe are missing data (not measured) or measured as zero. You can use the context of the data to try to decide: would a measure of zero make sense?\n","\n","Even if zeros would make sense for your data, be careful about assuming that any missing entry should be zero.\n","\n","If you're confident that your missing data really should be zeros, you can use the `df.fillna(value=0)` method to substitute missing data with 0.\n","\n","#### 3.2.3 Find rows with missing data <a name=\"row\"></a>\n","\n","We can also use the `isnull()` method, in cojunction with dataframe subsetting (also called slicing), to identify specific rows that contain missing data.\n","\n","Subsetting dataframes is a technique that allows you to select specific rows based on a set of criteria you define. The criteria can be quantitative or qualitative. We'll look at one criterion for now: `isnull() == True` will help us locate rows that have null values in specific columns.\n","\n","First we'll consider the `petal_width` column:"]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":359},"id":"IRX2NMSSl2Xb","executionInfo":{"status":"ok","timestamp":1616171762665,"user_tz":300,"elapsed":233,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"0b889771-5242-4e17-9b37-3e7c780309d5"},"source":["iris_missing[iris_missing['petal_width'].isnull() == True]"],"execution_count":40,"outputs":[{"output_type":"execute_result","data":{"text/html":["<div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>sepal_length</th>\n","      <th>sepal_width</th>\n","      <th>petal_length</th>\n","      <th>petal_width</th>\n","      <th>species</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>2</th>\n","      <td>4.3</td>\n","      <td>3.0</td>\n","      <td>1.1</td>\n","      <td>NaN</td>\n","      <td>setosa</td>\n","    </tr>\n","    <tr>\n","      <th>7</th>\n","      <td>4.7</td>\n","      <td>3.2</td>\n","      <td>1.3</td>\n","      <td>NaN</td>\n","      <td>setosa</td>\n","    </tr>\n","    <tr>\n","      <th>14</th>\n","      <td>5.8</td>\n","      <td>4.0</td>\n","      <td>1.2</td>\n","      <td>NaN</td>\n","      <td>setosa</td>\n","    </tr>\n","    <tr>\n","      <th>25</th>\n","      <td>5.0</td>\n","      <td>3.2</td>\n","      <td>1.2</td>\n","      <td>NaN</td>\n","      <td>setosa</td>\n","    </tr>\n","    <tr>\n","      <th>26</th>\n","      <td>5.5</td>\n","      <td>3.5</td>\n","      <td>1.3</td>\n","      <td>NaN</td>\n","      <td>setosa</td>\n","    </tr>\n","    <tr>\n","      <th>27</th>\n","      <td>4.4</td>\n","      <td>3.0</td>\n","      <td>1.3</td>\n","      <td>NaN</td>\n","      <td>setosa</td>\n","    </tr>\n","    <tr>\n","      <th>29</th>\n","      <td>4.4</td>\n","      <td>3.2</td>\n","      <td>1.3</td>\n","      <td>NaN</td>\n","      <td>setosa</td>\n","    </tr>\n","    <tr>\n","      <th>38</th>\n","      <td>5.0</td>\n","      <td>3.5</td>\n","      <td>1.3</td>\n","      <td>NaN</td>\n","      <td>setosa</td>\n","    </tr>\n","    <tr>\n","      <th>39</th>\n","      <td>4.5</td>\n","      <td>2.3</td>\n","      <td>1.3</td>\n","      <td>NaN</td>\n","      <td>setosa</td>\n","    </tr>\n","    <tr>\n","      <th>43</th>\n","      <td>5.4</td>\n","      <td>3.9</td>\n","      <td>1.3</td>\n","      <td>NaN</td>\n","      <td>setosa</td>\n","    </tr>\n","  </tbody>\n","</table>\n","</div>"],"text/plain":["    sepal_length  sepal_width  petal_length  petal_width species\n","2            4.3          3.0           1.1          NaN  setosa\n","7            4.7          3.2           1.3          NaN  setosa\n","14           5.8          4.0           1.2          NaN  setosa\n","25           5.0          3.2           1.2          NaN  setosa\n","26           5.5          3.5           1.3          NaN  setosa\n","27           4.4          3.0           1.3          NaN  setosa\n","29           4.4          3.2           1.3          NaN  setosa\n","38           5.0          3.5           1.3          NaN  setosa\n","39           4.5          2.3           1.3          NaN  setosa\n","43           5.4          3.9           1.3          NaN  setosa"]},"metadata":{"tags":[]},"execution_count":40}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":49},"id":"TCrxspIhl2Xc","executionInfo":{"status":"ok","timestamp":1616171764568,"user_tz":300,"elapsed":224,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"03030e87-a2d7-47bf-910c-9f7b9e56c1ac"},"source":["iris_missing[iris_missing['sepal_length'].isnull() == True]"],"execution_count":41,"outputs":[{"output_type":"execute_result","data":{"text/html":["<div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>sepal_length</th>\n","      <th>sepal_width</th>\n","      <th>petal_length</th>\n","      <th>petal_width</th>\n","      <th>species</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","  </tbody>\n","</table>\n","</div>"],"text/plain":["Empty DataFrame\n","Columns: [sepal_length, sepal_width, petal_length, petal_width, species]\n","Index: []"]},"metadata":{"tags":[]},"execution_count":41}]},{"cell_type":"markdown","metadata":{"id":"N1uBvko_l2Xc"},"source":["Now we'll consider the remaining columns."]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":49},"id":"TS5r-Awyl2Xc","executionInfo":{"status":"ok","timestamp":1616171766651,"user_tz":300,"elapsed":281,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"e2336463-8704-4503-e8de-9f7faf0072ec"},"source":["iris_missing[iris_missing['sepal_width'].isnull() == True]"],"execution_count":42,"outputs":[{"output_type":"execute_result","data":{"text/html":["<div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>sepal_length</th>\n","      <th>sepal_width</th>\n","      <th>petal_length</th>\n","      <th>petal_width</th>\n","      <th>species</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","  </tbody>\n","</table>\n","</div>"],"text/plain":["Empty DataFrame\n","Columns: [sepal_length, sepal_width, petal_length, petal_width, species]\n","Index: []"]},"metadata":{"tags":[]},"execution_count":42}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":49},"id":"4k5oObf7l2Xc","executionInfo":{"status":"ok","timestamp":1616171768195,"user_tz":300,"elapsed":224,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"2d8e1400-3177-4fc8-a06c-45d47eddaf26"},"source":["iris_missing[iris_missing['petal_length'].isnull() == True]"],"execution_count":43,"outputs":[{"output_type":"execute_result","data":{"text/html":["<div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>sepal_length</th>\n","      <th>sepal_width</th>\n","      <th>petal_length</th>\n","      <th>petal_width</th>\n","      <th>species</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","  </tbody>\n","</table>\n","</div>"],"text/plain":["Empty DataFrame\n","Columns: [sepal_length, sepal_width, petal_length, petal_width, species]\n","Index: []"]},"metadata":{"tags":[]},"execution_count":43}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":49},"id":"zHSCzCBrl2Xc","executionInfo":{"status":"ok","timestamp":1616171771498,"user_tz":300,"elapsed":237,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"cc17e518-93c1-469a-b103-1574906ced82"},"source":["iris_missing[iris_missing['species'].isnull() == True]"],"execution_count":44,"outputs":[{"output_type":"execute_result","data":{"text/html":["<div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>sepal_length</th>\n","      <th>sepal_width</th>\n","      <th>petal_length</th>\n","      <th>petal_width</th>\n","      <th>species</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","  </tbody>\n","</table>\n","</div>"],"text/plain":["Empty DataFrame\n","Columns: [sepal_length, sepal_width, petal_length, petal_width, species]\n","Index: []"]},"metadata":{"tags":[]},"execution_count":44}]},{"cell_type":"markdown","metadata":{"id":"xpOlPYibl2Xd"},"source":["We can see that 10 rows are missing `petal_width` data, but otherwise all data are present.\n","\n","What can we do about the missing `petal_width` data?\n","\n","#### 3.2.4 Describe the distribution of missing data <a name=\"distr\"></a>\n","\n","We can use graphs to help identify patterns in the missingness of data.\n","\n","First we will make a new column that indicates whether `petal_width` data are present using a boolean (`bool`) data type:"]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"w_IJfTb2l2Xd","executionInfo":{"status":"ok","timestamp":1616171787524,"user_tz":300,"elapsed":231,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"4ccb5f2e-148b-4e31-85da-9bb6b5bcc885"},"source":["iris_missing['no_petal_width'] = iris_missing['petal_width'].isnull()\n","iris_missing.dtypes"],"execution_count":45,"outputs":[{"output_type":"execute_result","data":{"text/plain":["sepal_length      float64\n","sepal_width       float64\n","petal_length      float64\n","petal_width       float64\n","species            object\n","no_petal_width       bool\n","dtype: object"]},"metadata":{"tags":[]},"execution_count":45}]},{"cell_type":"markdown","metadata":{"id":"ux9Yv1kVl2Xd"},"source":["Next we will subset the dataframe into two new dataframes: one where the rows have `petal_width` values, and another where the rows do not. (This doesn't change the `iris_missing` dataframe; it makes copies of the selected rows to new variables.)"]},{"cell_type":"code","metadata":{"id":"p-HX9ZOBl2Xd","executionInfo":{"status":"ok","timestamp":1616171789360,"user_tz":300,"elapsed":322,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}}},"source":["no_petal_width = iris_missing[iris_missing['no_petal_width'] == True]\n","yes_petal_width = iris_missing[iris_missing['no_petal_width'] == False]"],"execution_count":46,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"XnVuEmGxl2Xd"},"source":["Now we can look at the distribution of the other measurements (sepal width, sepal length, and petal length) by whether or not there is a corresponding petal width measurement.\n","\n","We will use methods two plotting libraries together to make a single image with three panels:\n","- `matplotlib.pyplot`, which we imported as `plt`, has methods to organize the panels (subplots). We'll use `subplots()` to request the number of panels and the overall figure size.\n","- `seaborn`, which we imported as `sns` has a `distplot` method that makes smoothed histograms. We will use this function to populate the subplots. We'll use the argument `ax=` to tell `distplot()` which of the subplots each smoothed histogram should plot to.\n","- The `bins` parameter can specify the number of bins for each histogram. Most of the observations have petal widths, so we'll use `bins=10` to give those histograms 10 bins. Since there are fewer observations without petal withds, we'll use `bins=3`."]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":476},"id":"_YzFAR5tl2Xd","executionInfo":{"status":"ok","timestamp":1616171799621,"user_tz":300,"elapsed":1348,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"0121978d-8b33-4708-a7a0-f086ccf50c9a"},"source":["fig, ax = plt.subplots(1, 3, figsize=(20, 6))\n","\n","# panel 0: sepal width\n","sns.distplot(yes_petal_width['sepal_width'], bins=10, hist = True, kde = False, label = 'Petal Width Measured', ax=ax[0])\n","sns.distplot(no_petal_width['sepal_width'], bins=3, hist = True, kde = False, label = 'Missing Petal Width', ax=ax[0])\n","ax[0].set_title('Sepal Width')\n","ax[0].legend()\n","\n","# panel 1: sepal length\n","sns.distplot(yes_petal_width['sepal_length'], bins=10, hist = True, kde = False, label = 'Petal Width Measured', ax=ax[1])\n","sns.distplot(no_petal_width['sepal_length'], bins=3, hist = True, kde = False, label = 'Missing Petal Width', ax=ax[1])\n","ax[1].set_title('Sepal Length')\n","ax[1].legend()\n","\n","# panel 2: petal length\n","sns.distplot(yes_petal_width['petal_length'], bins=10, hist = True, kde = False, label = 'Petal Width Measured', ax=ax[2])\n","sns.distplot(no_petal_width['petal_length'], bins=3, hist = True, kde = False, label = 'Missing Petal Width', ax=ax[2])\n","ax[2].set_title('Petal Length')\n","ax[2].legend()\n"],"execution_count":47,"outputs":[{"output_type":"stream","text":["/usr/local/lib/python3.7/dist-packages/seaborn/distributions.py:2557: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n","  warnings.warn(msg, FutureWarning)\n"],"name":"stderr"},{"output_type":"execute_result","data":{"text/plain":["<matplotlib.legend.Legend at 0x7f2c02ae5890>"]},"metadata":{"tags":[]},"execution_count":47},{"output_type":"display_data","data":{"image/png":"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\n","text/plain":["<Figure size 1440x432 with 3 Axes>"]},"metadata":{"tags":[],"needs_background":"light"}}]},{"cell_type":"markdown","metadata":{"id":"yr3x7hxol2Xe"},"source":["Most of the observations lacking petal width data seem to come from flowers with short petals and sepals."]},{"cell_type":"markdown","metadata":{"id":"CSlhR-oyl2Xe"},"source":["We can also see how missing data is distributed between the three species in the dataset. We'll use the pandas `groupby()` and `count()` methods to handle this categorial data."]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":142},"id":"Hfgfi-Kql2Xe","executionInfo":{"status":"ok","timestamp":1616171816229,"user_tz":300,"elapsed":518,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"bf4f7952-71d9-4c39-c684-4d678c97d55e"},"source":["iris_missing.groupby(['species'])['petal_width'].count().reset_index()"],"execution_count":48,"outputs":[{"output_type":"execute_result","data":{"text/html":["<div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>species</th>\n","      <th>petal_width</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>0</th>\n","      <td>setosa</td>\n","      <td>40</td>\n","    </tr>\n","    <tr>\n","      <th>1</th>\n","      <td>versicolor</td>\n","      <td>50</td>\n","    </tr>\n","    <tr>\n","      <th>2</th>\n","      <td>virginica</td>\n","      <td>50</td>\n","    </tr>\n","  </tbody>\n","</table>\n","</div>"],"text/plain":["      species  petal_width\n","0      setosa           40\n","1  versicolor           50\n","2   virginica           50"]},"metadata":{"tags":[]},"execution_count":48}]},{"cell_type":"markdown","metadata":{"id":"oBUY_msgl2Xe"},"source":["We see that species `setosa` has only 40 observations of `petal_width`, which the other two species have 50 each. It seems all the data are missing from only the `setosa` species.\n","\n","We can also visualize this using the `seaborn` method `countplot()`. For this graph, we'll use the data frame with the missing `petal_width` observations removed. The `countplot()` method will perform the same aggregation as the `groupby().count()` operation we did above, but will produce a barplot instead of a dataframe."]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":296},"id":"3RrmR8O2l2Xe","executionInfo":{"status":"ok","timestamp":1616171819797,"user_tz":300,"elapsed":1554,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"942baad8-c87d-44bf-ea04-5bb328501ff0"},"source":["sns.countplot(x='species', data=yes_petal_width)"],"execution_count":49,"outputs":[{"output_type":"execute_result","data":{"text/plain":["<matplotlib.axes._subplots.AxesSubplot at 0x7f2bf9978750>"]},"metadata":{"tags":[]},"execution_count":49},{"output_type":"display_data","data":{"image/png":"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\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"tags":[],"needs_background":"light"}}]},{"cell_type":"markdown","metadata":{"id":"1wepF1cVl2Xe"},"source":["### 3.3 Handle missing data <a name=\"handle\"></a>\n","\n","There are three options for handling missing data once you've found it and described it:\n","\n","A. Ignore it\n","\n","B. Fill it in (interpolate)\n","\n","C. Remove it\n","\n","#### 3.3.1 Ignoring missing data <a name=\"ignore\"></a>\n","\n","We have pretty compelling evidence that the missing data in the `iris_missing` dataset are not randomly distributed and therefore have meaning. That makes ignoring the missing data (pretending they don't exist) unwise. But this is the easiest of all the options, because you don't have to do anything.\n","\n","#### 3.3.2 Interpolate missing data <a name=\"interp\"></a>\n","\n","If your missing data are randomly distributed you can sometimes reasonably fill in the mean value (or some other meaningful measure of center, like median or the expected value of a different distribution).\n","\n","We know this isn't an appropriate strategy for the `iris_missing` data, but here's what the code would look like if we wanted to fill in the missing sepal widths with the mean sepal width for the setosa species."]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":204},"id":"Jb7344Qil2Xe","executionInfo":{"status":"ok","timestamp":1616171826004,"user_tz":300,"elapsed":326,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"e9c35697-0d44-43c9-feca-f378dfd71bd9"},"source":["sepal_width_mean = np.mean(iris_missing[iris_missing['species'] == 'setosa']['petal_width'])\n","iris_filled = iris_missing.fillna(value=sepal_width_mean)\n","iris_filled.head()"],"execution_count":50,"outputs":[{"output_type":"execute_result","data":{"text/html":["<div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>sepal_length</th>\n","      <th>sepal_width</th>\n","      <th>petal_length</th>\n","      <th>petal_width</th>\n","      <th>species</th>\n","      <th>no_petal_width</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>0</th>\n","      <td>4.9</td>\n","      <td>3.1</td>\n","      <td>1.5</td>\n","      <td>0.10</td>\n","      <td>setosa</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>1</th>\n","      <td>4.8</td>\n","      <td>3.0</td>\n","      <td>1.4</td>\n","      <td>0.10</td>\n","      <td>setosa</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>2</th>\n","      <td>4.3</td>\n","      <td>3.0</td>\n","      <td>1.1</td>\n","      <td>0.25</td>\n","      <td>setosa</td>\n","      <td>True</td>\n","    </tr>\n","    <tr>\n","      <th>3</th>\n","      <td>5.2</td>\n","      <td>4.1</td>\n","      <td>1.5</td>\n","      <td>0.10</td>\n","      <td>setosa</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>4</th>\n","      <td>4.9</td>\n","      <td>3.6</td>\n","      <td>1.4</td>\n","      <td>0.10</td>\n","      <td>setosa</td>\n","      <td>False</td>\n","    </tr>\n","  </tbody>\n","</table>\n","</div>"],"text/plain":["   sepal_length  sepal_width  petal_length  petal_width species  no_petal_width\n","0           4.9          3.1           1.5         0.10  setosa           False\n","1           4.8          3.0           1.4         0.10  setosa           False\n","2           4.3          3.0           1.1         0.25  setosa            True\n","3           5.2          4.1           1.5         0.10  setosa           False\n","4           4.9          3.6           1.4         0.10  setosa           False"]},"metadata":{"tags":[]},"execution_count":50}]},{"cell_type":"markdown","metadata":{"id":"Di8qP13Xl2Xf"},"source":["You can see one entry where `no_petal_width` is `True` that has the mean value of 0.25 filled in.\n","\n","#### 3.3.3 Remove missing data <a name=\"remove\"></a>\n","\n","The final option, and the one most suitable for our `iris_missing` data, is to remove missing data. There are a few options for removing missing data:\n","\n","- By row: we can remove all rows that are missing the `petal_width` measures. This is not recommended for our data, because we know there is a pattern to our missing data. This can be done with the `dropna()` method for dataframes.\n","- By column: we can remove the entire `petal_width` column and only consider the other measures.\n","- By subset: we can remove a meaningful data subset to clear up the missing data. In the case of our `iris_missing` data, we may elect to remove all of the `setosa` species data. But since our original goal was to classify plants into the three possible species, we probably don't want to get rid of all the `setosa` data.\n","\n","We will remove the `petal_width` column before we proceed. We'll use the argument `axis=1` to let pandas know this is a column we want to drop, not a row."]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":204},"id":"x-jonsEgl2Xf","executionInfo":{"status":"ok","timestamp":1616171886779,"user_tz":300,"elapsed":463,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"bf5a509f-e586-42f5-83b5-7aa8cd1fc328"},"source":["iris_drop_missing = iris_missing.drop('petal_width', axis=1)\n","iris_drop_missing.head()"],"execution_count":51,"outputs":[{"output_type":"execute_result","data":{"text/html":["<div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>sepal_length</th>\n","      <th>sepal_width</th>\n","      <th>petal_length</th>\n","      <th>species</th>\n","      <th>no_petal_width</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>0</th>\n","      <td>4.9</td>\n","      <td>3.1</td>\n","      <td>1.5</td>\n","      <td>setosa</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>1</th>\n","      <td>4.8</td>\n","      <td>3.0</td>\n","      <td>1.4</td>\n","      <td>setosa</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>2</th>\n","      <td>4.3</td>\n","      <td>3.0</td>\n","      <td>1.1</td>\n","      <td>setosa</td>\n","      <td>True</td>\n","    </tr>\n","    <tr>\n","      <th>3</th>\n","      <td>5.2</td>\n","      <td>4.1</td>\n","      <td>1.5</td>\n","      <td>setosa</td>\n","      <td>False</td>\n","    </tr>\n","    <tr>\n","      <th>4</th>\n","      <td>4.9</td>\n","      <td>3.6</td>\n","      <td>1.4</td>\n","      <td>setosa</td>\n","      <td>False</td>\n","    </tr>\n","  </tbody>\n","</table>\n","</div>"],"text/plain":["   sepal_length  sepal_width  petal_length species  no_petal_width\n","0           4.9          3.1           1.5  setosa           False\n","1           4.8          3.0           1.4  setosa           False\n","2           4.3          3.0           1.1  setosa            True\n","3           5.2          4.1           1.5  setosa           False\n","4           4.9          3.6           1.4  setosa           False"]},"metadata":{"tags":[]},"execution_count":51}]},{"cell_type":"markdown","metadata":{"id":"OL69CufHl2Xf"},"source":["## 4. Classification Model <a name=\"class\"></a>\n","\n","Now that we've cleaned up our data, we can proceed to the main event: building a model that classifies plants into species given measurements on their flowers. Because we are classifying observations by a categorial value (species name), we will use logistic regression. Other classification techniques we won't discuss include decision trees, random forests, and support vector machines. Neural networks are also used for classification on more complex datasets, like images.\n","\n","Before we get started, let's import some modules specific to this part of the tutorial:"]},{"cell_type":"code","metadata":{"id":"RX8P1D9Rl2Xf","executionInfo":{"status":"ok","timestamp":1616171891590,"user_tz":300,"elapsed":336,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}}},"source":["from sklearn.preprocessing import label_binarize\n","from sklearn.linear_model import LogisticRegression\n","from sklearn import datasets\n","from sklearn.preprocessing import StandardScaler\n","from sklearn.model_selection import train_test_split\n","from sklearn.multiclass import OneVsRestClassifier\n","from sklearn.metrics import roc_curve, auc\n","from scipy import interp\n","from itertools import cycle"],"execution_count":52,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"cFKWpKMhl2Xf"},"source":["### 4.1 Class Balance <a name=\"bal\"></a>\n","\n","It is much easier to train a classification algorithm when there are similar numbers of observations from each class. In our case, we have three species each with 50 observations. This is perfect class balance. If we had unbalanced classes, we may need to tweak the data a bit to even things out a bit.\n","\n","Downsampling the majority class refers to excluding some observations from the category with the most observations. This is the most commonly used technique. \n","\n","Upsampling minority classes refers to duplicating data from categories that are underrepresented. This technique is less popular as it inflates sample size and makes your dataset appear larger than it truly is.\n","\n","In either case, look to `np.random.sample()` for methods on how to randomly select data from your dataset. Once you've randomly selected data, you can discard or duplicate them as desired.\n","\n","### 4.2 Standardizing Measure Scales <a name=\"std\"></a>\n","\n","Our measurements, sepal width, sepal length, and petal length, are observed on different scales:"]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":227},"id":"whpacHxLl2Xf","executionInfo":{"status":"ok","timestamp":1616171931341,"user_tz":300,"elapsed":317,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"7a533af9-d902-415a-dace-dc8114ed37a3"},"source":["iris_drop_missing.groupby('species')['sepal_width', 'sepal_length', 'petal_length'].mean()"],"execution_count":53,"outputs":[{"output_type":"stream","text":["/usr/local/lib/python3.7/dist-packages/ipykernel_launcher.py:1: FutureWarning: Indexing with multiple keys (implicitly converted to a tuple of keys) will be deprecated, use a list instead.\n","  \"\"\"Entry point for launching an IPython kernel.\n"],"name":"stderr"},{"output_type":"execute_result","data":{"text/html":["<div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>sepal_width</th>\n","      <th>sepal_length</th>\n","      <th>petal_length</th>\n","    </tr>\n","    <tr>\n","      <th>species</th>\n","      <th></th>\n","      <th></th>\n","      <th></th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>setosa</th>\n","      <td>3.428</td>\n","      <td>5.006</td>\n","      <td>1.462</td>\n","    </tr>\n","    <tr>\n","      <th>versicolor</th>\n","      <td>2.770</td>\n","      <td>5.936</td>\n","      <td>4.260</td>\n","    </tr>\n","    <tr>\n","      <th>virginica</th>\n","      <td>2.974</td>\n","      <td>6.588</td>\n","      <td>5.552</td>\n","    </tr>\n","  </tbody>\n","</table>\n","</div>"],"text/plain":["            sepal_width  sepal_length  petal_length\n","species                                            \n","setosa            3.428         5.006         1.462\n","versicolor        2.770         5.936         4.260\n","virginica         2.974         6.588         5.552"]},"metadata":{"tags":[]},"execution_count":53}]},{"cell_type":"markdown","metadata":{"id":"qQySJYRtl2Xf"},"source":["Just from considering the average values by species, we see that sepal length can be much larger than sepal width or petal length. If we use the raw data, it's possible that sepal length might have more influence on the performance of the model because it has greater magnitude, even it it's not necessarily the best predictor of species.\n","\n","We can standardize the values so that they are all on the same scale. We will do this using the `sklearn` function `StandardScaler.fit_transform()`. This function calculates the mean and standard deviation of each column in the data frame and then uses those values to calculate the standard score of each observation.\n","\n","Before we standardize the data, we'll split the `iris_drop_missing` dataframe into `X` and `y` dataframes. The `X` dataframe will contain the measures (the x-values in our regression), and the `y` dataframe will contain the species names (the y-values we are trying to predict with our regression)."]},{"cell_type":"code","metadata":{"id":"jZThmsAYl2Xg","executionInfo":{"status":"ok","timestamp":1616171941457,"user_tz":300,"elapsed":252,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}}},"source":["y = iris_drop_missing['species']\n","X = iris_drop_missing[['sepal_length', 'sepal_width', 'petal_length']]"],"execution_count":54,"outputs":[]},{"cell_type":"code","metadata":{"id":"5rAxEdH-l2Xg","executionInfo":{"status":"ok","timestamp":1616171942943,"user_tz":300,"elapsed":387,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}}},"source":["scaler = StandardScaler()\n","X_std = scaler.fit_transform(X)"],"execution_count":55,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"js-rJJnYl2Xg"},"source":["### 4.3 Multinomial Logistic Regression <a name=\"multi\"></a>\n","\n","#### 4.3.1 Fit the Model <a name=\"fit1\"></a>"]},{"cell_type":"code","metadata":{"id":"UVcM6WtLl2Xg","executionInfo":{"status":"ok","timestamp":1616171945339,"user_tz":300,"elapsed":265,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}}},"source":["X_train, X_test, y_train, y_test = train_test_split(X_std, y, test_size=0.33, random_state=42)\n"],"execution_count":56,"outputs":[]},{"cell_type":"code","metadata":{"id":"hrHKlyuXl2Xg","executionInfo":{"status":"ok","timestamp":1616171946267,"user_tz":300,"elapsed":291,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}}},"source":["log_reg = LogisticRegression(random_state=0, multi_class='multinomial', solver='newton-cg')"],"execution_count":57,"outputs":[]},{"cell_type":"code","metadata":{"id":"ed8LvNp8l2Xg","executionInfo":{"status":"ok","timestamp":1616171947379,"user_tz":300,"elapsed":279,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}}},"source":["model1 = log_reg.fit(X_train, y_train)"],"execution_count":58,"outputs":[]},{"cell_type":"code","metadata":{"id":"C4ToIw4Il2Xg","executionInfo":{"status":"ok","timestamp":1616171948133,"user_tz":300,"elapsed":267,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}}},"source":["test_results = model1.predict(X_test)"],"execution_count":59,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"U-QyUUm5l2Xg"},"source":["#### 4.3.2 Evaluate the Model: Accuracy Metrics <a name=\"acc\"></a>"]},{"cell_type":"code","metadata":{"id":"ugNUQwynl2Xg","executionInfo":{"status":"ok","timestamp":1616171951520,"user_tz":300,"elapsed":437,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}}},"source":["outcomes = {\n","    'versicolor': {'total': 0, 'test': 0}, \n","    'setosa': {'total': 0, 'test': 0}, \n","    'virginica': {'total': 0, 'test': 0}}\n","for test, true in zip(test_results, y_test):\n","    outcomes[true]['total'] += 1\n","    if test == true:\n","        outcomes[true]['test'] += 1"],"execution_count":60,"outputs":[]},{"cell_type":"code","metadata":{"id":"n8ygHZXql2Xh","executionInfo":{"status":"ok","timestamp":1616171952229,"user_tz":300,"elapsed":280,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}}},"source":["in_class_accuracy = {key: val['test']/val['total'] for key, val in outcomes.items()}"],"execution_count":61,"outputs":[]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"Vw4479ITl2Xh","executionInfo":{"status":"ok","timestamp":1616171953706,"user_tz":300,"elapsed":348,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"1835fdbc-0574-4f90-a913-553739961c04"},"source":["in_class_accuracy"],"execution_count":62,"outputs":[{"output_type":"execute_result","data":{"text/plain":["{'setosa': 1.0, 'versicolor': 0.9333333333333333, 'virginica': 0.875}"]},"metadata":{"tags":[]},"execution_count":62}]},{"cell_type":"markdown","metadata":{"id":"hsSycQ74l2Xh"},"source":["### 4.4 One-vs-Rest Logistic Regression <a name=\"ovr\"></a>\n","\n","#### 4.4.1 Fit the Model <a name=\"fit2\"></a>"]},{"cell_type":"code","metadata":{"id":"EZ2YPh8ml2Xh","executionInfo":{"status":"ok","timestamp":1616171957410,"user_tz":300,"elapsed":252,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}}},"source":["integer_classes = {\n","    'setosa': 0,\n","    'versicolor': 1,\n","    'virginica': 2\n","}\n","\n","iris_drop_missing['integer_species'] = [integer_classes[sp] for sp in iris_drop_missing['species']]"],"execution_count":63,"outputs":[]},{"cell_type":"code","metadata":{"id":"IUPjmFFal2Xh","executionInfo":{"status":"ok","timestamp":1616171958704,"user_tz":300,"elapsed":330,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}}},"source":["binary_y = label_binarize(iris_drop_missing['integer_species'], classes=[0, 1, 2])"],"execution_count":64,"outputs":[]},{"cell_type":"code","metadata":{"id":"R1pBVzF4l2Xh","executionInfo":{"status":"ok","timestamp":1616171960752,"user_tz":300,"elapsed":411,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}}},"source":["X_train_2, X_test_2, y_train_2, y_test_2 = train_test_split(X_std, binary_y, test_size=0.33, random_state=42)"],"execution_count":65,"outputs":[]},{"cell_type":"code","metadata":{"id":"gdEWzYssl2Xh","executionInfo":{"status":"ok","timestamp":1616171961953,"user_tz":300,"elapsed":362,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}}},"source":["ovr_log_reg = OneVsRestClassifier(LogisticRegression(random_state=0, multi_class='multinomial', solver='newton-cg'))"],"execution_count":66,"outputs":[]},{"cell_type":"code","metadata":{"id":"x0H3HjG9l2Xh","executionInfo":{"status":"ok","timestamp":1616171963628,"user_tz":300,"elapsed":443,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}}},"source":["model2 = ovr_log_reg.fit(X_train_2, y_train_2).decision_function(X_test_2)"],"execution_count":67,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"7Lgg6VHNl2Xh"},"source":["#### 4.4.2 Evaluate the Model: Receiver-Operator Criterion <a name=\"roc\"></a>\n","\n","Following from the `skearn` documentation: https://scikit-learn.org/stable/auto_examples/model_selection/plot_roc.html#multiclass-settings"]},{"cell_type":"code","metadata":{"id":"SS0BbRLyl2Xi","executionInfo":{"status":"ok","timestamp":1616171966839,"user_tz":300,"elapsed":748,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}}},"source":["# Compute ROC curve and ROC area for each class\n","n_classes = binary_y.shape[1]\n","fpr = dict()\n","tpr = dict()\n","roc_auc = dict()\n","for i in range(n_classes):\n","    fpr[i], tpr[i], _ = roc_curve(y_test_2[:, i], model2[:, i])\n","    roc_auc[i] = auc(fpr[i], tpr[i])\n","\n","# Compute micro-average ROC curve and ROC area\n","fpr[\"micro\"], tpr[\"micro\"], _ = roc_curve(y_test_2.ravel(), model2.ravel())\n","roc_auc[\"micro\"] = auc(fpr[\"micro\"], tpr[\"micro\"])"],"execution_count":68,"outputs":[]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":295},"id":"i6ODFBlKl2Xi","executionInfo":{"status":"ok","timestamp":1616171969236,"user_tz":300,"elapsed":405,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"e43e8cfa-0987-4bae-bd37-b16419c73fdc"},"source":["plt.figure()\n","lw = 2\n","plt.plot(fpr[2], tpr[2], color='darkorange',\n","         lw=lw, label='ROC curve (area = %0.2f)' % roc_auc[2])\n","plt.plot([0, 1], [0, 1], color='navy', lw=lw, linestyle='--')\n","plt.xlim([0.0, 1.0])\n","plt.ylim([0.0, 1.05])\n","plt.xlabel('False Positive Rate')\n","plt.ylabel('True Positive Rate')\n","plt.title('Receiver operating characteristic example')\n","plt.legend(loc=\"lower right\")\n","plt.show()"],"execution_count":69,"outputs":[{"output_type":"display_data","data":{"image/png":"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\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"tags":[],"needs_background":"light"}}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":675},"id":"7CbJARxbl2Xi","executionInfo":{"status":"ok","timestamp":1616171973825,"user_tz":300,"elapsed":580,"user":{"displayName":"Kelly Pierce","photoUrl":"","userId":"14154256606034993895"}},"outputId":"a0dbba20-5191-49c7-9237-09bc51d96784"},"source":["rev_integer_classes = {\n","    0: 'setosa',\n","    1: 'versicolor',\n","    2: 'virginica'\n","}\n","\n","# Compute macro-average ROC curve and ROC area\n","\n","# First aggregate all false positive rates\n","all_fpr = np.unique(np.concatenate([fpr[i] for i in range(n_classes)]))\n","\n","# Then interpolate all ROC curves at this points\n","mean_tpr = np.zeros_like(all_fpr)\n","for i in range(n_classes):\n","    mean_tpr += interp(all_fpr, fpr[i], tpr[i])\n","\n","# Finally average it and compute AUC\n","mean_tpr /= n_classes\n","\n","fpr[\"macro\"] = all_fpr\n","tpr[\"macro\"] = mean_tpr\n","roc_auc[\"macro\"] = auc(fpr[\"macro\"], tpr[\"macro\"])\n","\n","# Plot all ROC curves\n","plt.figure(figsize=(10,10))\n","plt.plot(fpr[\"micro\"], tpr[\"micro\"],\n","         label='micro-average ROC curve (area = {0:0.2f})'\n","               ''.format(roc_auc[\"micro\"]),\n","         color='deeppink', linestyle=':', linewidth=4)\n","\n","plt.plot(fpr[\"macro\"], tpr[\"macro\"],\n","         label='macro-average ROC curve (area = {0:0.2f})'\n","               ''.format(roc_auc[\"macro\"]),\n","         color='navy', linestyle=':', linewidth=4)\n","\n","colors = cycle(['aqua', 'darkorange', 'cornflowerblue'])\n","for i, color in zip(range(n_classes), colors):\n","    plt.plot(fpr[i], tpr[i], color=color, lw=lw,\n","             label='ROC curve of class {0} (area = {1:0.2f})'\n","             ''.format(rev_integer_classes[i], roc_auc[i]))\n","\n","plt.plot([0, 1], [0, 1], 'k--', lw=lw)\n","plt.xlim([0.0, 1.0])\n","plt.ylim([0.0, 1.05])\n","plt.xlabel('False Positive Rate')\n","plt.ylabel('True Positive Rate')\n","plt.title('Some extension of Receiver operating characteristic to multi-class')\n","plt.legend(loc=\"lower right\")\n","plt.show()"],"execution_count":70,"outputs":[{"output_type":"stream","text":["/usr/local/lib/python3.7/dist-packages/ipykernel_launcher.py:15: DeprecationWarning: scipy.interp is deprecated and will be removed in SciPy 2.0.0, use numpy.interp instead\n","  from ipykernel import kernelapp as app\n"],"name":"stderr"},{"output_type":"display_data","data":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAAmkAAAJcCAYAAACixjPMAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdd3gU5d7G8e+T0Hsv0qUKCKg0pbdQJEcEpYhgFxAUBcuxvWAXKyJ6RBE8wJGmCAhIE6lWQEBEBKQ3qQkQCYHkef+YIWzCpgBJZje5P9e1VzIzz87cs/W3zzRjrUVEREREAkuI1wFERERE5GIq0kREREQCkIo0ERERkQCkIk1EREQkAKlIExEREQlAKtJEREREApCKNJFUMMZ8Y4y5y4PlvmyMOWKMOZjRy/bHGNPMGPOn1zkCgTHmlDHm6gxepjXGVMnIZaaXy31PZYbXoDGmpTFmbzLTy7uvr9DLmPdOY0zbK0sogUJFWhZjjGlqjPneGBNpjDlmjFlljGngda7LlVEfSNbajtba/6b3cnwZY8oDQ4Ga1tpSfqa3NMbEuR/mJ40xfxpj7knPTNbaFdba6um5jEBkjFlqjLnfd5y1Np+1drtXmbyUFu+71L6nEheml/saNMYMN8ZMutT7ZYTEj6e1drf7+or1Mpd4T0VaFmKMKQDMAd4HigBlgBeAM17mkiSVB45aaw8l02a/tTYfUAB4DPjEGBN0RZQxJltWXLZXPH68jTFG3z0iqWGt1S2L3ID6QEQy00OA54BdwCFgAlDQnVYRsMA9wB7gONAfaABsACKA0Ynmdy/wh9t2AVAhmWU3Br5357MeaOmOvwk4ApRzh+u686sBTATigNPAKeDJ5OblTlsKvASsAk4CC4Fi7rRcwCTgqHvfX4CSPve7/xIep7uA3W72Z5NZ74Lu/Q+783vOnX9bd73i3HX7zM99WwJ7E407BNzuk/PfwF/uOk0Divi0berzOO0B7nbH5wTecvP/DXwE5E68TOAp4ItEy38PGOWzbp8CB4B9wMtAqDvtbvc5eNfN9rKf9csJjAT2u7eRQE7fHMAz7mO8E+id6L7JroOb/yDO66gwzg+YwzivrzlAWbf9K0AsEO0+F6Pd8Rao4v7/GfABMBfndfUTUNknTxjwJxAJfAgsw309+VnvUHe9/nLntYYLr3+L877b6j5vHwDGnVYZWOI+nkeA/wGFfOa7013nDTg/zLJx4fVxEtgE3JooywM47+Hz06/n8t93r7jP+WmgCgnfU1XcxyTSzT7VHb/cXecod1k9SPS6B8oBM9zn7iiJPofcNh2AGOCsO5/17virgNnAMWAb8EAy79XP3OfuG3ceq4BSOK/L48Bm4Dqf9vGvD5/7v+znfXTR48mFz5FsyeS56LnxeZ7buv83BH5wn5MDwGgghzvN4Lz/DgEngN+A2u60Tu48T+K8dx9Py+8i3VJ/8zyAbhn4ZDu9LUeB/wIdgcKJpt/rflBdDeRzP/gmutPOf2h8hFPMhOF8ac0ESuD0yh0CWrjtb3HndQ3Ol8FzwPdJ5Crj5uqEU1i0c4eLu9Nfwfnyye1+kAzyuW/8B1Iq57UU50upmju/pcDr7rR+wNdAHpwvyhuAAj73u/8SHqdP3PnXxflCvCaJdZ8AzALyu/fdAtznTmtJoiIs0X3jp7vr+i+cD/vr3HGDgR+BsjhFyxhgsjutgvsB3AvIDhQF6rnT3sX54iri5voaeM3PMisA/wD53eFQnC+Cxu7wV+4y8+K8Rn4G+rnT7gbOAQ/jvD5y+1m/F938JYDiOAXASz45zgHvuOvWAueLvHoq1+EcMMK9b253/bu5z31+YDow0ydL/PPvMy5xkXYU50sxG06BNMWdVgznS7CrO20wTrGQVJH2BM7rvDrOF2ldoKjPMucAhXB6Wg8DHdxpVXBe7zndx2s5MDLRe2UdTlFzvmC9HadQCcEpgKKA0j7T9uH8EDPu/CtcwftuN1DLfQyyk/A9NRl41r1vLqCpv8fZz2swFKcgfBfndZbgvoke1+HApETjluMUXrmAeu7j2TqJ+3+GU0De4LZfAuwA+ro5Xga+Syb3Z/gp0pJ4PCuSTJGW2ufGzdrYfcwr4hR1j7rT2uP8ACjkzuMan+f+ANDM/b8wbgGoW8bfPA+gWwY/4c4b8TOcnoRzOF9k53uLvgUe8mlbHefL5Pwb3AJlfKYfBXr4DH/p8wHwDW6x4Q6H4HyhV/CT6SncIsdn3ALgLvf/7O6HyW/AfNyeA3da4g+3lOa1FHjOZ9pDwHz3/3txCoE6fjIu5cIXSmoep7I+038GevqZZyjOr/uaPuP6AUvd/1uScpEWh/Mr+QxOb8+jPtP/ANr4DJf2yfk08JWfeRqcL2rfXqAbgR3+MgErgb7u/+2Av9z/S7qZcvu07YX7JYZTpO1O4bX6F9DJZ7g9sNMnxzkgr8/0acDzqVyHGCBXMsuuBxz39/z7jEtcpI31mdYJ2Oz+3xf4IdFjvCfx/Hym/wncksQ0S8ICZhrw7yTadgF+TfReuTeFx3zd+WXjvG8GJ9FuJ5f+vnsxmffUBOBjfN43/h7nxK9B93k9TDI9Tj73G45PkYZTrMbi/shwx72Gn15rn+f4E5/hh4E/fIavxWdLhZ/cn5F2RVqqn5tE0x7Ffd8DrXF+FDYGQhK1243zWVQgpcdVt/S9ab+ALMZa+4e19m5rbVmgNs6v6JHu5KtwNrmdtwvnC72kz7i/ff4/7Wc4n/t/BeA9Y0yEMSYCZ3OCwfnFnVgF4Pbzbd32TXGKCqy1Z3E+4GoDb1v3UyQJyc7L5Xuk5D8+mSfifPhNMcbsN8a8YYzJ7mcZqXmcklqGr2I4BWjiefl7jJKy31pbCKeXdBTOB+95FYCvfB6HP3C+lErifEH95Wd+xXF6k9b43G++O96fz3GKL4A73OHzy84OHPCZzxicXrHz9qSwbv4e56t8ho9ba6P8TE/NOhy21kafHzDG5DHGjDHG7DLGnMDpYSl0iUfXJfWcX4XPurqv3ySP7CPp5ybZ5RhjShpjphhj9rnrMAnnNeYrwWNujOlrjFnn8zjV9rlPSjl8peZ9l9zz/STO58PPxpjfjTH3pnK55YBd1tpzqWzv6yrgmLX2pM+4lN5/qf38SzPu0ayn3Nvv7uhUPTfGmGrGmDnGmIPua+JV3OfXWrsEZ/PnB8AhY8zH7n7L4PQqdwJ2GWOWGWNuTOv1ktRRkZaFWWs3c6H4AWe/nwo+Tcrj9Fb8zaXbg7Npq5DPLbe19vsk2k5M1DavtfZ1AGNMGWAYMB542xiT03c1LmVeybHWnrXWvmCtrYmzL1xnnF6QxNLqcTqC07OVeF77LnE+WGvP4PRmXGuM6eKO3gN0TPRY5LLW7nOnVU4i02mgls99Clrn4AR/pgMtjTFlgVu5UKTtwelJK+YznwLW2lq+sVNYLX+P836f4cLGmLx+pqdmHRIveyhOj2gja20BoLk73qQya3IO4GxydmZojPEd9iOp5yYlr+LkvNZdhzu5kP+8+PUwxlTA2Sw/CGdzaiFgo899kstxOe+7JB9Da+1Ba+0D1tqrcHpwPkzlqUb2AOVTeSBE4uXvB4oYY/L7jLus918S/sH5sXDeRUdoJ5PtwgTnaNZ87u38+ye1r5H/4OwrV9V9TTyDz2vCWjvKWnsDUBNnF5An3PG/WGtvwflRNROnx1Y8oCItCzHG1DDGDHW/UDHGlMPpBfnRbTIZeMwYU8kYkw/nQ3/qZf5K/Qh42hhTy11WQWPM7Um0nQSEG2PaG2NCjTG53NNLlHW/0D7D2QH9PpwvvJd87vs3zr5hKc4rpcDGmFbGmGvd3pMTOAVUnJ+mafI4Wefw+mnAK8aY/O6X5hB3HS6ZtTYGeBv4P3fUR+68KwAYY4obY25xp/0PaGuM6W6MyWaMKWqMqWetjcP54n7XGFPCvV8ZY0z7JJZ5GGez1XiczYl/uOMP4ByU8bYxpoAxJsQYU9kY0+ISVmky8Jybu5i7XokfmxeMMTmMMc1wiurpl7oOrvw4hV2EMaYIzo8CX4lfZ5diLm7x7BYTA0n+C3ss8JIxpqp7JGQdY0zRVCwnP86O55HuD5snUmifF6c4OAxgnNO31PaZPhZ43Bhzg5ujyvnXEmn4vnOXfbtP2+NurvPvveQe+59xPhNeN8bkdZfbJIm2fwMVjXtkqbV2D87uDa+596uD8xmTVqfpWAfc4T4eHXD2m0zKpb6+kntufOXH+Sw7ZYypAQw4P8EY08AY08g4WwuicPYxjnPfT72NMQXdrRgn8P85KBlARVrWchJoBPxkjInCKc424vQiAIzD2eS3HGeH2Gic/S4umbX2K5wds6e43ewbcQ5W8Nd2D86BBs/gfGHswfmCCQEewfk197y7mege4B73SxmcfUieM84mlsdTmFdKSgFf4Hwo/YFztNlEP+3S7HFy7xcFbMfZv+tzd/6XaxxOz0I4zpGWs4GFxpiTOM93I3DOw4SzOWMozqbodTg7qIPTI7cN+NF97hbj9DIl5XOco1E/TzS+L5AD5yix4ziPbWlS72VgNc7RiL8Ba91x5x1057sfp+js7/YOX846jMQ5gOAIzuM0P9H094DbjDHHjTGjLmEdsNYewdnR+w2c/ThruuuV1Klv3sEp3hfivBY/dbOl5AWcoy8jcQrDGSnk2oRT1P+AUyRci3PE4vnp03EO2vkc57NjJs6BGJC27ztwdoD/yRhzCuc1O9heOAfdcOC/7rK6J1qHWCAcZ8f53TibkXsksYzp7t+jxpi17v+9cPb/2o9zoMswa+3iVGZOyWA3WwTQG+fxS0qCxzOlGafw3Ph6HGc3hJM4P1ym+kwr4I47jrOZ9yjwpjutD7DTfe/0d/OLB84fui0iEjSMMS1xdgJPVU9NIHF7cvbinDLkO6/ziEjgUk+aiEg6czcDFjLO/pTn9wv6MYW7iUgWpyJNRCT93YhzNN4RnE1gXay1p72NJCKBTps7RURERAKQetJEREREAlDQXVi4WLFitmLFil7HEBEREUnRmjVrjlhrkzoheLKCrkirWLEiq1ev9jqGiIiISIqMMbtSbuWfNneKiIiIBCAVaSIiIiIBSEWaiIiISABSkSYiIiISgFSkiYiIiAQgFWkiIiIiAUhFmoiIiEgAUpEmIiIiEoBUpImIiIgEIBVpIiIiIgFIRZqIiIhIAFKRJiIiIhKAVKSJiIiIBCAVaSIiIiIBSEWaiIiISABSkSYiIiISgFSkiYiIiAQgFWkiIiIiAUhFmoiIiEgAUpEmIiIiEoBUpImIiIgEoHQr0owx44wxh4wxG5OYbowxo4wx24wxG4wx16dXFhEREZFgky0d5/0ZMBqYkMT0jkBV99YI+I/7VyRpEzbC0KUXhvvUhHda+2/bZipsOHxheHF3qFvi4nbrD0HbaReG6xSHb3v4n+eQJTBx04Xht1tC39r+2xYfnXD48CD/7bROQbFOy5/5jtteXxE/3KxyEb7c6r/tA1eNZNbBE/HDH7/dni6PXfzxFrlyD1Wbj48fLpArG9v+ecbvPEe3nsSLS7fHDw+8pQbDvurut22D7C+zKzYufvjn7Y9QsWIhrZPWSesUIOuUWsZae0UzSHbmxlQE5lhrL/p0NMaMAZZaaye7w38CLa21B5KdZ/36ltWr0yFt1hE25yTld5/1OoaIiEimN3Zg0TXW2vqXc18v90krA+zxGd7rjruIMeZBY8xqY4yqszSgAk1ERCT9REcdZ8XnjxFz+kTKjZORnps704y19mPgY3B60tKv7y9reMD9+8lDRTzNcVky4WY0rZP363Q0Lo73TkczOjqaZev7ce21JRO2edsAYB4fHj+qdbmCfLv7Ub+z7FHyHaYdOhk/PGVEO3o8edNF7SJW7KFw83HxwwWzhRBx9nm/83yj2X95auXO+OEnOlXjjbm9/LatHPoS2+MubJ7Ztu1hKle++P3+7VNLaPvGhc0zWietk9YpLdbpBDAJOESDSnEXtbsUQbm502pz5xV54MNjQJAWaSJp7MMPf+HJJxcRFeX0MPfsWZvJk7slbOQWaYf6nIoflT17CIUL5/Y7z8jIaM6ciY0fLlAgJ7lyXfybOC7OcuTIP/HDxkDx4nn9zjMqKiY+I0CePNnJly+H37ZHj/5DbOyFz/YiRXKTLdvFG05iYmKJiIjWOmmdtE5ptE7r1m3i9tv/xe7dO6lWrQbffruIcuXKXfbmTi+LtJuBQUAnnAMGRllrG6Y4TxVpV0xFmsgFs2ZtpkuXqfHDISGGzZsHUrVq0QuN3CKNoerHFxH/NmzYQPv27Tl48CANGjRg3rx5FCtWDGNM4O2TZoyZDPwAVDfG7DXG3GeM6W+M6e82mQdsB7YBnwAPpVcWEZGkhIdXp3btC5tXCxXKxebNRzxMJCLB5vvvv6dFixYcPHiQ1q1b8+2331KsWLErnm+67ZNmrfW/cffCdAsMTK/li4ict2HD30RFxXDjjeUumhYSYnjmmaYMGbKQoUNvpH//+klu9hAR8WfKlClERERw66238vnnn5MrV640mW9QHDggkuod1kV8/PTTXl55ZQVff72FunVL8uuv/TDGXNSue/dadOlSg9y5s3uQUkSC3bvvvkutWrW47777yJYt7UorXRZKRDKl3bsjufHGT/n66y0ArF//N3PnbvXbNjQ0RAWaiFySKVOmcOyYs493aGgo/fr1S9MCDVSkiUgmVb58Qbp1q5lg3CuvrCA9D5YSkczPWsvrr79Or1696Ny5M+fOnUu3ZalIE5FM69lnm8X/36hRGZ55pqmHaUQk2FlrefLJJ3n66acxxtC3b9807z3zpX3SJDhoHzRJ5OzZWCZP3sjHH6/hm296kz9/zova1KtXiueea0bLlhVp3bqS3/3RRERS49y5c/Tr149x48aRLVs2Jk2aRI8eSZxQO42oSBORoDNx4nr+7/+WsnNnBABjxqzh8ccvPgM5wEsvJXGlAxGRVDpz5gx33HEHM2bMIHfu3MyYMYMOHTqk+3K1uVNEgs62bcfiCzSAt9/+gdOndU1aEUkfn376KTNmzKBgwYIsWrQoQwo0UJEmIkHokUcakTfvhaMxDx+OYuXK3R4mEpHMrH///gwePJhly5bRpEmTDFuuijQRCUiHD0dx4MBJv9OKFs3DgAH1yZEjlH79bmDLlodp165yBicUkcxs3759HD58GICQkBBGjhxJ3bp1MzSDijQRCSj79p3gscfmU7Hiezz77JIk2z31VFO2b3+Ejz7qzNVXF87AhCKS2W3dupUmTZrQsWNHTpw44VkOHTggwWHCxoTDfWt7k0PS1erV+2nSZBwxMbEATJy4gWHDWlChQqGL2hYrliej44lIFrBu3Trat2/PoUOHKFWqVLqeBy0l6kmT4DB0acKbZErXXVeKChUKxg+fOxfHm29+72EiEclKVq5cScuWLTl06BBt27Zl8eLFFClSxLM8KtJEJGCEhobw739fOOHsVVflp2bN4h4mEpGsYt68eYSFhREZGUnXrl2ZM2cO+fLl8zSTijQRyVDff7+Hxx9fmOTlme68sw4tWlTgo49uZvv2R3jooQYZnFBEspp169Zxyy23cPr0ae677z6mTp1KzpwXnyA7o2mfNAkOfWqm3EYC2pIlO3j55eV8991OADp2rEKbNldf1C5HjlCWLr07Y8OJSJZWt25d7r77bgoXLsyIESMC5uokKtIkOLyjs8YHu3fe+SG+QAPnYuf+ijQRkYxgreXUqVPkz58fYwxjxowhJCSwNjAGVhoRybR8L3YO8N13O1m37qBHaUQkK4uLi2PIkCE0adKE48ePAwRcgQYq0kQkDcTFWZYu3cn998/mjz8O+21z443laN26EgCtW1diyZK+1K1bMiNjiohw7tw57r33XkaOHMnmzZv55ZdfvI6UJG3uFJErMn78rwwbtpQ9e5wTPpYsmZdXXmnjt+1bb7XjzJlYGjcum5ERRUQAiI6OpmfPnsyaNYs8efIwY8YMwsLCvI6VJPWkicgVCQkx8QUawKRJvxEX5//IzeuuK60CTUQ8cfLkSTp16sSsWbMoVKgQixcvpn379l7HSpaKNBFJ0alTMZw54/+s2127XkPu3Bc65XfvjtTFzkUkoERFRdG6dWu+++47SpUqxfLly7nxxhu9jpUiFWkSHNpMTXiTdHfuXBzz52/jzjtnULLkW8ya9affdvnz56RLlxrkzp2NO+64lm++6c1NN5XL4LQiIknLkycPjRs35uqrr2bVqlVce+21XkdKFe2TJsFhg/+d0SX9DBv2Ha++ujJ+eOLEDXTvXstv2zffbMeYMZ3Jn9/7kz+KiJxnrcUYgzGG9957j+HDh1O0aFGvY6WaetIkfUzYCMVHX7gNWZJ02zZTE7Zdf8h/u7CK6RJV/OvWLeEJhOfP38bhw1F+25YpU0AFmogElLVr19KyZUsOH3Z+5IeEhARVgQYq0iRYPNEAnmzodYpMJSIimrFj17J9+3G/06+7rhTXXFMsfrhUqXxs3Xoso+KJiFy25cuX06pVK5YvX85rr73mdZzLlqU2d46ac5Lfdp/1OoZcjicbXehhq1DA2yxBbuXK3Ywc+SNff72FmJhYnn++OS++2OqidsYYBgyoz7p1B+nTpy7Nm1cgJCQwLpUiIpKUOXPmcPvttxMdHU337t15/fXXvY502bJUkaYC7YJry2e/8pkUH51w+PCgK59nSioUgLdapv9yMrEtW47y5Zd/xA9PmrSBF15o6fdadQ8/3Cgjo4mIXJH//e9/3HXXXcTGxvLggw/y4YcfEhoa6nWsy5alirTzPnmoiNcRMr++tZ1banzbI3Xt6paA1X0vP1MWc36H2cS6dbuGgQPnER3tnFJjx44Ivv9+D02alM/oiCIiaeb999/nkUceAeDpp5/mlVdeCZgLpV8u7ZMmkokcOhTF++//RMOGnzBjxh9+2xQsmIt//as6AJUrF2bYsBZUqFAoI2OKiKS5PXv2APDmm2/y6quvBn2BBlm0J00kMxozZjUDB84jNtY52/+kSb9ddITmec8+24zHHmtMo0ZlMsUHmYjIiBEj6Ny5M82bN/c6SppRkSaXLyP2QZNUu/760vEFGsDcuVs4duw0RYrkvqhtnTq6sLmIBLdz587x3HPPMXjwYEqXLo0xJlMVaKDNnSJBZcOGvzlw4KTfafXrX0W1ahfOAXT2bBzLlu3MoGQiIhnn9OnTdOvWjREjRtCtWzes9X+94GCnIk0kwO3ff5I331xFnTr/oW7djxgzZo3fdsYY+vSpQ61axXn99Tbs2vUot956TQanFRFJXydOnKBjx47Mnj2bwoUL8+6772ba3Ta0uVMkwM2du4Unn1wcPzxp0gaGDWvh90Ppqaea8OyzzTLtB5aIZG2HDx+mQ4cOrF27lquuuoqFCxdSq5b/y9VlBupJEwlwt91Wkxw5Lpzn56+/jvPTT/v8ts2ePVQFmohkSrt376ZZs2asXbuWypUrs3LlykxdoIGKNBFPWWv5+ed9PPLINyxYsM1vm8KFcxMeXg2AkBBDhw5VdOZ/Eclypk+fzp9//kmdOnVYuXIllSpV8jpSutPmTrl8EzYmHE7tyWsFcC5YPnjwfLZsOQo4+561b1/Fb9uHH25I06bl6dmzNqVK5cvImCIiAWHIkCFkz56dvn37UqhQ1ji3o4o0uXxDlyYcVpF2SQoVyhVfoAF8/fUWIiKiKVQo10VtW7SoSIsWFTMwnYiI95YvX06lSpUoV64cxpj4KwpkFdrcKeKRRo3KUKXKhUuUxcTE8sUXmzxMJCISOGbPnk1YWBhhYWEcP37c6zieUJEmko6WLdvJoUNRfqcZY7jzzmvJnj2ELl1q8MUXt3PnnXUyOKGISOCZMGECXbt25cyZM7Rs2ZICBQp4HckT2twpl6+P/0sOCZw9G8vw4Ut57bWVdOhQhblz7/B71OWgQQ15+OFGfq8KICKSFb333ns8+uijADz77LO89NJLWfaodRVpcvneae11goB0+HAU4eGT40+T8c032xg16icGD258UduiRfNkdDwRkYBkrWX48OG8+OKLALz99tsMGTLE41Te0uZOkTRWuHBusmVL+Nb697+/5e+/T3mUSEQk8C1evJgXX3yRkJAQxo0bl+ULNFCRJpLmsmUL4X//60rBgjkBKF06H3Pm9KJkSZ06Q0QkKW3btuWZZ57hiy++4J577vE6TkDQ5s7MasLGhKfI6FMz6c2TbabChsMXhhd3h7olLm63/hC0nXZhuE5x+LZHmsTNbCpUKMTHH4czadIGxo27hWLFtFlTRCSx06dPc+TIkfhTbLzyyiteRwoo6kkTuUxxcZYdO5I+LLx791rMmtVTBZqIiB+RkZG0b9+eVq1acfDgQa/jBCQVaSKXYf/+k7RrN5GmTcdz5Mg/SbbLqkckiYgk5++//6Zly5asWLGC6OhoIiIivI4UkFSkiVyi2bP/pE6d/7BkyQ727z/JfffNxlrrdSwRkaCwa9cumjVrxrp166hatSqrVq2iRo0aXscKSNonLbPqWzv1l2lK7X5ldUvA4UGXnymT+PHHvRw9ejp+ePbsP/n44zX061ffw1QiIoFv06ZNhIWFsW/fPurVq8f8+fMpWbKk17EClnrSRC7RCy+0pFGjMvHDrVtXonPnah4mEhEJfIcOHaJ58+bs27ePpk2bsnTpUhVoKVBPmoiP6dN/59dfD7Jly1G2bDnK11/3okKFQgnaZM8eyuefd6NBg0944ombeOKJmwgN1e8dEZHklChRgoEDB7J69WqmT59Onjw6qColKtIyk+KjEw5r02QCUVExbNt2DGuhXr1Sftu8++6P/PDD3vjhP/88elGRBnD11YXZsWMwBQrkTLe8IiKZQXR0NLly5QJg+PDhxMbGki2byo/U0M9/yfS++24H5cq9S758r1Gv3hiefXZJkm2rVy+WYHjLlqNJtlWBJiKSvM8++4xatWqxd6/z49cYowLtEqhIk6D1ww97+PjjNQwduoDw8Mns2uX/EO5ChXKxd++J+OHkCq9q1YokGE6urYiIJO3dd9/lnnvuYfv27cyaNcvrOEFJ5WwwWrobHl8Kj1yf+iM4g1BERJ9ceyMAACAASURBVDQnT56hXLmCfqc/9dRiVqzYHT88YEB9v5smq1YtmmB4x47jxMTEkiNH6EVtW7WqxNNPN6VataJUq1aUGjWKXdRGRESSZq3l+eefj796wMiRIxk4cKDHqYKTirRg9PhS2HXi4vGZYB+0NWv2M3jwfLZuPcahQ1G0b1+Z+fPv9Nu2WrWiCYq0LVuO0qlT1Yva5cuXgzJl8nPgwCkqVSpEtWpFiYyMpnjxvBe1bdy4LI0bl027FRIRyULi4uIYNGgQ//nPfwgNDWXcuHH07dvX61hBS0VaMPIt0N74CZ5s5F2WVNq9O5I//jgcf9TkE080oXz5i3vIcuQIZdWqPfHDyW+aTNhDllzb77+/j5Il85Izp17yIiLpwVrLnXfeyeTJk8mZMyfTpk3jX//6l9exgpq+sYLZGz/D3/8ERJFmreX06XPkyZPd7/S77prJ0qU744fDwir7LdKqVEm4T9jOnRFER58jV66LX6qNGpXhjjuupVq1IlSrVpTrry+dZD5/yxIRkbRjjKFOnTrMmTOH2bNn07JlS68jBb2gK9KKHT7HAx8e8zpGYPg76WtGZoQ//zzCSy8tj+8da9CgDIsW9fHbtnr1ogmKtKR6vXLnzk758gXZvTuSHDlCqVKlCIcPR/ndL61Fi4q0aFExLVZFRETSwFNPPcWdd95J2bLabSQtBF2RdqWuLe+/pyeo9KmZrrOPiYllx47j8cVXjx61KVu2wEXtYmMt//vfb/HDabVp8osvbqdo0TxUqFBQJ4kVEQlgBw8e5N577+WDDz6gUqVKGGNUoKWhoCzSPnmoSMqNMrN3Wqfr7Dt3/pxFi7bHD1euXMRvkVa5cmFCQgxxcc7FxXfvjuT06bPkzn1xIVynTkluuqmce9RkEZo0KZ/k8hs0KJPkNBERCQw7d+6kXbt2bNu2jUGDBjF37lyvI2U6QVmkyeXbvv04gwfPJy7OMnfuHX7bVK1aJEGRllSvV86c2ahYsRDbtx8HIE+e7OzZc+KiXjOAtm2vpm3bq9NgDURExGubNm2iXbt27N+/n+uuu47x48d7HSlTUpGWxQwePJ85c7Zw1VX5k2xzKZsmR43qQJ482alWrShXXZUfY0yaZZVMZMbNsGOe1ylEJA38/PPPdOzYkWPHjtG8eXNmz55NwYI6OCs9qEjLQs6cOcfChX8BsH//SXbujKBs2QJky5Zwv6/q1YtRpkz++BO6tm5dKcl53nxztXTNLJlEZijQKnXyOoGI5xYvXkyXLl2IiooiPDycqVOnkjt3bq9jZVoq0rKQX389SExMbPxw48ZjOXjw8YvadehQhb17h2RkNMkqhlqvE4jIFdi0aRNRUVH06dOHTz/9lOzZM8HBeAFMRVp6mbARhi69MNynZtI7/LeZChsOXxhe3B3qlri43fpD0Haa8/8TDS75/Gg7d0aQM2coZ844hVrz5hUu6f4iIpK1PfLII1SpUoUOHToQEqKj79ObirRgtWCncwP4tkeq7tKzZ226dr2G9esP8tNP+6hcuXC6xRMRkcxh9OjRhIWFUa2as3tLp07a9J9RVKQFq/M9bxUuPjVGcnLkCKVBgzI6zYWIiCTLWsuzzz7La6+9RsWKFdm0aZP2P8tgKtKCWYUC8FZLr1OIiEgmExsby8CBAxkzZgyhoaG89NJLKtA8YKwNrh15i1eoZw/vWud1DBG5FG+7p2bRgQMiAS8mJoY+ffowbdo0cuXKxbRp0wgPD/c6VtAyxqyx1ta/nPuqJy0tFB+dcPjwIG9yiIiIXIGoqCi6devGggULKFCgAF9//TXNmzf3OlaWpSJNREREAFiwYAELFiygePHizJ8/n+uvv97rSFmajp/NAiIiorn++jG8+uoKlizZwcmTZ7yOJCIiAahr16588MEHrFixQgVaAFBPWia3du0BwsMns3//SX799SAAJUrk5eDBobqEk4iIsH37dqKjo6lZsyYADz30kMeJ5DwVaWkhgPdBy5EjlGPHTicY17JlRRVoIiLCxo0bCQsLwxjD999/T4UKOsl5INHmzkyudu0SvPNOWILht95q52EiEREJBD/++CPNmzfnwIEDVK9enSJFingdSRJRT1oW0L9/fRYt2k7p0vl4660wcufWtdZERLKyhQsXcuutt/LPP/9wyy23MGXKFHLlyuV1LElERVomcebMOWJjLXnyXFyAGWOYNu12smVTx6mISFY3ffp0evfuzdmzZ7n77rv55JNPyJZN5UAg0rd2JrBly1FuumkcjzzyTZJtVKCJiMi2bdvo1asXZ8+e5dFHH+XTTz9VgRbA9MwEMWstEyasZ+DAeURFnWXt2gO0a3c1PXrU9jqaiIgEoCpVqvDWW29x6tQpnn32WR1EFuBUpKWFCRsTDvfNmCIpJiaW119fRVTU2fhxDz44h4YNy1CpUuEMySAiIoHNWsu+ffsoW7YsAI8++qjHiSS1tA0sLQxdmvCWQXLmzMaUKd3IkSM0fly7dldTqJB2/hQREedC6Q8++CA33HADW7du9TqOXCIVaUGubt1SvPVWO3LnzsbHH3dm+vTbKVw4t9exRETEY2fOnKFnz56MHTuWEydOsGPHDq8jySXS5s4gYa1Nct+BQYMaEh5enYoVC2VwKhERCUSnTp2ia9euLFq0iIIFCzJnzhyaNm3qdSy5RCrS0kKfmuk6+2++2cobb3zP3Ll3JHmKDRVoIiICcOzYMTp16sRPP/1EiRIlWLBgAfXq1fM6llwGFWlp4Z3W6TLbM2fO8cwz3/LOOz8CMGTIAj76qHO6LEtERIJfTEwMrVq1YsOGDVSoUIFFixZRtWpVr2PJZdI+aQHstddWxhdoAGPGrOHLLzd5mEhERAJZjhw5GDBgADVr1mTVqlUq0IKcirQANnTojVSteuFaaqGhhl27Ij1MJCIigejcuXPx//fv3581a9ZQpkwZDxNJWlCRFsDy58/J5MndyJ49hPLlC7Js2d0MGXKj17FERCSArFq1iho1avD777/Hj9N1ODMHFWkB7oYbruKrr3qwbl0/mjQp73UcEREJIPPnz6ddu3b89ddfjB492us4ksZUpHksLs7yxhur+P33Q0m2ufnmajr3mYiIJDBlyhTCw8M5ffo099xzD++//77XkSSN6ejOtNBmasLhb3uk6m4HDpykb9+ZLF68nUmTNvDTT/eTO/fFp9gQERHx9dFHH/HQQw9hrWXo0KG8+eabug5nJqSetLSw4XDCWyrs33+SOnU+YvHi7QD89tshnnhiUXqmFBGRTGDEiBEMGDAAay2vvvqqCrRMTEWaR0qXzkf79pUTjJsyZSOHDkV5lEhERIJBqVKlCAkJ4T//+Q9PP/20CrRMTEWaR4wxfPjhzVx9dWEAWrSowLp1/SlRIq/HyUREJJDdddddbN68mf79+3sdRdJZuu6TZozpALwHhAJjrbWvJ5peHvgvUMht829r7bz0zJQuFne/rLsVKJCTKVO6sWDBXzz9dFNCQ1Uzi4hIQtHR0QwYMIDBgwfHX95JJ6nNGtKtSDPGhAIfAO2AvcAvxpjZ1lrfU+Y/B0yz1v7HGFMTmAdUTK9MaWLpbnh8Kew64RRndUs4tyRERkazdu0BWrWq5Hd6gwZlaNBAJxwUEZGLnTx5ki5durBkyRJ++OEHfv/9d0JDQ72OJRkkPbtuGgLbrLXbrbUxwBTglkRtLFDA/b8gsD8d86SN8wVaKvz0016uu24MnTtPZvPmI+mbS0REMpWjR4/Spk0blixZQsmSJZk2bZoKtCwmPYu0MsAen+G97jhfw4E7jTF7cXrRHvY3I2PMg8aY1caY1ekR9JL4Fmhv/Jxks3ff/YGmTcezY0cE//xzlp49vyA6+lyS7UVERM7bu3cvzZo145dffqFSpUqsWrWKOnXqeB1LMpjXO0H1Aj6z1pYFOgETjTEXZbLWfmytrW+trZ/hCZOzcGeSkwoUyMm5c3Hxw+vX/83IkT8m2V5ERARg69atNG3alD/++INatWqxcuVKKleunPIdJdNJzwMH9gHlfIbLuuN83Qd0ALDW/mCMyQUUA5I+/b7X6hRPVbN7772ORYu2M3Wqcy21hx9uyKOPNk7PZCIikgmsX7+e3bt307hxY+bOnUuRIkW8jiQeSc8i7RegqjGmEk5x1hO4I1Gb3UAb4DNjzDVALiB1Z4P1SiqvJmCMYcyYzvz113H+7/+aEx5ePZ2DiYhIZnDbbbcxa9YsWrVqRb58+byOIx5KtyLNWnvOGDMIWIBzeo1x1trfjTEvAquttbOBocAnxpjHcA4iuNtaa9MrU3o4cOAkpUvn9zutYMFc/Pzz/TrRoIiIJGvevHkUK1aMhg0bAhAeHu5xIgkE6bpPmrV2nrW2mrW2srX2FXfc/7kFGtbaTdbaJtbautbaetbahemZJy1Zaxk//leqVHmfiRPXJ9lOBZqIiCRn8uTJ3HLLLXTs2JG9e/d6HUcCiNcHDgSlyMhoeveewb33zuaff87y0EPz2Lr1qNexREQkyHz44Yf07t2bc+fOcf/991OmjM6bKReoSLsM+/adZObMzfHDp07F0KvXl5w9G+thKhERCRbWWl5++WUGDhyItZbXX3+dESNGaOuLJKAibeluqD8Bio92bkmZsDG+Tc052xk5skP8pNy5szFgQH2yZdPDKSIiyYuLi2PIkCE8//zzGGP4+OOPeeqpp7yOJQEoXa/dGRQu4QoC8Q5G8cABy8IqRdmaJxtTpnTjmmtSd2oOERHJ2tauXcuoUaPInj07n3/+ObfddpvXkSRAqUhLoUCLi7OEhCTqfp64CQOMK5uPHD/cSa5cehhFRCR16tevz/jx4yldujTt2rXzOo4EMFUXSThz5hxPPrmIEydiGD8+8SVHgQoFKPBWS1CBJiIiKTh58iRbt27l+uuvB6Bv374eJ5JgoArj8KCLRm3efIRevb5k3bqDALRrdzV39L0W+tbO6HQiIhLkjhw5QseOHdm6dStLly6lXr16XkeSIKE93ROx1nLXXTPjCzSA/v3n8NdfxzxMJSIiwWjPnj00a9aM1atXU7RoUQoUKOB1JAkiKtISMcYwdmw4OXOGxo8rWjQPkZFnPEwlIiLB5s8//6RJkyZs3ryZa6+9lpUrV3L11Vd7HUuCiIo0P669tiTvvNMegO7da/Hrr/24/vrSHqcSEZFgsXbtWpo1a8aePXu48cYbWbZsGaVL63tELo32SUvCgAH1qVy5MGFhlXVyQRERSbWTJ08SFhbG0aNHad++PV9++SV58+b1OpYEoSzbk7Z//0keeGA2UVExfqcbY2jfvooKNBERuST58+dn9OjR9OzZk9mzZ6tAk8uW+XrS1h+CttMuDNcpDt/2SNBkzpwt3H33TI4ePQ0TN/FJlRLwZENnoo7gFBGRy3D48GGKF3dObN6zZ0969OihH/pyRbJcT9qUKRsJD5/sFGjA2DNnmLb7OAxd6txEREQu0ahRo6hcuTI///xz/DgVaHKlslyRFh5ejerViyYYN/WMu8mzgg6NFhGR1LPW8sILLzB48GBOnjyZoEgTuVJZrkjLmzcHU6bcRo4coYQaeC1PHqblz+cUaG+19DqeiIgEibi4OB599FGGDx9OSEgIY8eOZdCgi0+QLnK5Mt8+aXVL+L2KgK969UrxySfhVKtWlMaNy2ZQMBERySzOnj3Lvffey6RJk8iRIweTJ0+ma9euXseSTCb4i7T1hxIO1y2Rqrv17Vs3HcKIiEhW0Lt3b6ZPn07evHmZOXMmbdu29TqSZELBX6T5HskJKfaiiYiIXKk77riDZcuWMXv2bBo1auR1HMmkgr9IExERyQBxcXGEhDi7cnfp0oW2bduSL18+j1NJZpblDhwQERG5VLt37+aGG25g5cqV8eNUoEl6C/6etDrFU9WsS5cpbNjwN3nz5iBfvhx89NHN1K1bKp3DiYhIsNu8eTPt2rVj7969PPfcc3z33Xc6B5pkiOAv0hJdTSApu3dHsmNHRPzw2bNx6ZVIREQyidWrV9OxY0eOHDlCkyZNmDlzpgo0yTDBubmz+GgYsuSS7hIVdTbBcN682dMykYiIZDJLly6lVatWHDlyhI4dO7Jw4UIKFSrkdSzJQoKzSLsMiS+kni9fDo+SiIhIoJs9ezYdOnTg1KlT9OzZk5kzZ5InTx6vY0kWE/ybO1NpzZoHOXkyhqioGKKizlKypHb4FBER/7Jly0ZsbCwDBgzg/fffJzQ01OtIkgVlmSKtZMl8lCzpdQoREQkGnTp1YvXq1dSpU0f7oIlngrNI0wlrJdjMuBl2zPM6hYgkwVrLSy+9RJMmTWjTpg0AdevqyjTirUy5T9rOnRF07Pg/tm8/7nUUEYcKNKjUyesEIn7FxcXx8MMPM2zYMLp160ZERETKdxLJAMHZk5aMqVM30q/fHCIjz9C79wyWL7+b7Nm1L4EEiKHW6wQi4uPs2bPcfffdfP755+TIkYPPPvtMR3BKwMhUPWmLF2+nZ88viYw8A8CPP+7lhReWeZxKREQC0T///MOtt97K559/Tr58+fjmm2/o0qWL17FE4mWqIq1Nm0qEh1dLMG7Dhr+Ji1PvhYiIXBAREUH79u2ZO3cuRYsWZcmSJbRu3drrWCIJZKoizRjDuHG3ULp0PnLkCOW99zowa1ZPQkJ0ZI6IiFzw+++/8/PPP1OmTBlWrFhBgwYNvI4kcpHg3Cet+Ogkj/AsViwPU6feRoECOXVtThER8atJkyZ89dVX1KpViwoVKngdR8SvoCzS5sbEcHbmZrp0qeF3erNmesOJiEhCmzZtYu/evYSFhQHOudBEAllQbu7sfOIkd989k127dJi0iIik7JdffqFZs2Z06dKFNWvWeB1HJFWCskgDiIw8w513fsW5c3FeRxERkQB2/qCAY8eO0bp1a6655hqvI4mkStAWaQA//LCHH37Y43UMEREJUDNnzqRjx46cOnWKO+64g6+++koXSpegEbRFWsWKhVix4h7tfyYiIn6NHz+ebt26ERMTw6BBg5g4cSLZs2f3OpZIqgVlkTZgQH3WrevHjTeW8zqKiIgEoIMHDzJo0CDi4uIYNmwYo0aNIiQkKL/yJAsLyqM7P/zwZq8jiIhIACtVqhTTp09n27ZtPPLII17HEbksQVmkiYiIJBYbG8uGDRu47rrrAJ1iQ4Kf+n5FRCToxcTE0Lt3bxo3bsy3337rdRyRNBGcRdqEjV4nEBGRAPHPP/9wyy23MHXqVHLmzEloaKjXkUTSRHAWaUOXep1AREQCQEREBGFhYcyfP59ixYrx3Xff0bJlS69jiaQJ7ZMmIiJB6eDBg3To0IH169dTrlw5Fi5cSI0a/i8XKBKMgrMnTUREsrS4uDg6derE+vXrqV69OitXrlSBJplOcBZpfWp6nUBERDwUEhLCm2++yU033cSKFSsoX76815FE0pyx1nqd4ZIUr1DPHt61zusYIpfmbeP8HRpc7zeRQBMZGUnBggXjh621GGM8TCSSPGPMGmtt/cu5b3D2pImISJazePFiKlWqxPz58+PHqUCTzExFmoiIBLwvv/ySm2++mePHjzNz5kyv44hkCBVpIiIS0D799FO6d+9OTEwMjzzyCB9++KHXkUQyhIo0EREJWG+++Sb3338/cXFxvPDCC4wcOVIXSpcsQ+dJExGRgPTiiy8ybNgwAN5//30GDRrkcSKRjBWcP0faTPU6gYiIpLNWrVqRP39+Jk2apAJNsqTg7EnbcNjrBCIikg58T6nRrFkzduzYQdGiRT1OJeKN4OxJExGRTCcqKorOnTsnOHpTBZpkZcHZkyYiIpnKsWPH6Ny5Mz/88AO//fYbHTp0IFeuXF7HEvFUcBZpi7t7nUBERNLIgQMHCAsLY+PGjZQvX55FixapQBMhWIu0uiW8TiAiImlg+/bttGvXju3bt1OjRg0WLVpE2bJlvY4lEhC0T5qIiHjit99+o2nTpmzfvp369euzYsUKFWgiPlSkiYiIJ06fPs2JEydo3bo1S5YsoVixYl5HEgkowbm5U8QLM26GHfO8TiGSaTRs2JDly5dTs2ZN7YMm4oeKNJHUutICrVKntMkhEsSmT59OXFwcPXr0AOD666/3OJFI4ArOIm39IR08IN4Zar1OIBKUPvnkE/r160doaCh16tThmmuu8TqSSEALzn3S2k7zOoGIiFyCESNG8OCDD2KtZfjw4dSoUcPrSCIBLzh70kREJChYa/n3v//NG2+8gTGGDz74gAEDBngdSyQoqEgTEZF0ERsbS//+/Rk7dizZsmVjwoQJ9OrVy+tYIkEjOIu0OsW9TiAiIinYsWMH06ZNI3fu3HzxxRd06qSDZ0QuRXAWad/28DqBiIikoEqVKsyZMwdjDE2bNvU6jkjQCc4iTUREAtLRo0f5/vvvCQ8PB6BZs2YeJxIJXsF5dKeIiASc/fv306JFC2699VbmzdOJn0WulIo0ERG5Ytu2baNJkyb8/vvvVK9enbp163odSSToqUgTEZErsmHDBpo2bcrOnTvjL/VUpkwZr2OJBD0VaSIictlWrVpF8+bN+fvvv2nTpg3ffvstRYsW9TqWSKYQnEXakCVeJxARyfLOnDlDr169iIyMpGvXrsydO5d8+fJ5HUsk0wjOIm3iJq8TiIhkeTlz5uSLL75g0KBBTJ06lZw5c3odSSRT0Sk4RETkkvzxxx/xF0dv2LAhDRs29DiRSOYUnD1pIiKS4ay1vPrqq9SuXZupU6d6HUck0wvOnrS3W3qdQEQkS7HW8vjjj/POO+9gjCEyMtLrSCKZXnAWaX1re51ARCTLOHfuHA8++CDjx48ne/bsTJw4kR49dHk+kfQWnEWaiIhkiOjoaHr16sXMmTPJkycPM2bMoH379l7HEskSVKSJiEiS7rnnHmbOnEmhQoWYO3cuN910k9eRRLIMHTggIiJJevLJJ6levTrLli1TgSaSwdSTJiIiCZw+fZrcuXMDcN111/H7778TGhrqcSqRrEc9aSIiEm/r1q3UrFmT//73v/HjVKCJeCM4i7Tio71OICKS6axbty7+Quljx44lLi7O60giWVpwFmkiIpKmVqxYQYsWLTh06BBhYWHMnz+fkBB9RYh4Se9AEZEsbu7cuYSFhXHixAluv/12Zs+eTd68eb2OJZLlqUgTEcnCvvjiC7p06UJ0dDQPPPAAkydP1oXSRQJEcB7deXiQ1wlERDKFa665hvz58/Pggw/y2muvYYzxOpKIuIKzSBMRkTRRq1YtNm7cyFVXXeV1FBFJRJs7RUSykLi4OB577DHGjBkTP04FmkhgSnVPmjEmj7X2n/QMIyIi6efcuXPcd999TJgwgVy5chEeHq4CTSSApdiTZoy5yRizCdjsDtc1xnyYmpkbYzoYY/40xmwzxvw7iTbdjTGbjDG/G2M+v6T0IiKSKtHR0XTr1o0JEyaQN29eZs+erQJNJMClpiftXaA9MBvAWrveGNM8pTsZY0KBD4B2wF7gF2PMbGvtJp82VYGngSbW2uPGmBKXsQ4iIpKMEydOcMstt7B06VIKFy7MvHnzaNy4sdexRCQFqdrcaa3dk+iIn9hU3K0hsM1aux3AGDMFuAXY5NPmAeADa+1xdzmHUpOHCRuhb+1UNRUfM26GHfO8TiEiGejw4cN07NiRNWvWULp0aRYuXEjt2vr8FAkGqTlwYI8x5ibAGmOyG2MeB/5Ixf3KAHt8hve643xVA6oZY1YZY340xnTwNyNjzIPGmNXGmNUADF2aisXLRVSgXblKnbxOIHJJIiIi2L17N5UrV2bVqlUq0ESCSGp60voD7+EUWPuAhcBDabj8qkBLoCyw3BhzrbU2wreRtfZj4GOA4hXq2TRadtY1VA+hSFZRtWpVFi9eTIkSJShVqpTXcUTkEqSmJ626tba3tbaktbaEtfZO4JpU3G8fUM5nuKw7ztdeYLa19qy1dgewBadoExGRy7R27doEp9ioU6eOCjSRIJSaIu39VI5L7BegqjGmkjEmB9AT9+ADHzNxetEwxhTD2fy5PcU596mZisWLiGQ9y5Yto2XLlvTv358FCxZ4HUdErkCSmzuNMTcCNwHFjTFDfCYVAEJTmrG19pwxZhCwwG0/zlr7uzHmRWC1tXa2Oy3MPcVHLPCEtfZoiqnfaZ1iExGRrGb27Nl0796dM2fO0KNHD1q1auV1JBG5Asntk5YDyOe2ye8z/gRwW2pmbq2dB8xLNO7/fP63wBD3JiIil2nixIncc889xMbG0q9fPz744ANCQ1P8PS0iASzJIs1auwxYZoz5zFq7KwMziYjIJRg1ahSDBw8G4JlnnuHll1/WhdJFMoHUHN35jzHmTaAWkOv8SGuttjmKiHgsIiKC119/HYC33nqLoUOHepxIRNJKaoq0/wFTgc44p+O4CzicnqFERCR1ChUqxMKFC1m7di19+/b1Oo6IpKHUHN1Z1Fr7KXDWWrvMWnsvoF40ERGPnD17ljlz5sQP165dWwWaSCaUmiLtrPv3gDHmZmPMdUCRdMyUsjZTPV28iIhXTp8+TdeuXQkPD2fs2LFexxGRdJSazZ0vG2MKAkNxzo9WAHg0XVOlZIO2topI1hMZGcm//vUvli9fTpEiRahTp47XkUQkHaVYpFlrz/epRwKtAIwxTdIzlIiIJHTo0CE6dOjAr7/+SpkyZVi4cCE1a+rE3iKZWXInsw0FuuNcs3O+tXajMaYz8AyQG7guYyKKiGRtu3btIiwsjC1btlClShUWLVpExYoVvY4lIuksuZ60T3GuvfkzMMoYsx+oD/zbWjszI8IlaXF3TxcvIpJRrLX07t2bLVu2UK9ePebPn0/JkiW9jiUiGSC5Iq0+UMdaG2eMyQUcBCqn6rJN6a1uCa8TiIhkCGMMn376KU8//TTjxo2jUKFCXkcSkQyS3NGdMdbaOABrbTSwPSAKNBGRLGDXrgsXeqle03MLPAAAIABJREFUvTozZsxQgSaSxSRXpNUwxmxwb7/5DP9mjNmQUQFFRLKaWbNmUb16dd577z2vo4iIh5Lb3HlNhqUQEREA/vvf/3LfffcRGxvL1q1bsdbqOpwiWVRyF1jXRdVFRDLQyJEjeeyxxwB47rnnePHFF1WgiWRhqbniQOBZf8jrBCIiacZay/PPPx9foL377ru89NJLKtBEsrjgLNLaTvM6gYhImnnllVd4+eWXCQ0N5bPPPuPRR729qIuIBIZUFWnGmNzGmOrpHUZEJCvq3bs3lSpV4ssvv+Suu+7yOo6IBIgULwtljAkH3gJyAJWMMfWAF621/0rvcCIimVVMTAw5cuQAoFKlSmzevDl+WEQEUteTNhxoCEQAWGvXAZXSMVPK6hT3dPEiIlciIiKCNm3aMGLEiPhxKtBEJLHUFGlnrbWRicbZ9AiTat/28HTxIiKX6++//6Zly5asXLmS0aNHExmZ+ONVRMSRmiLtd2PMHUCoMaaqMeZ94Pt0ziUikuns3LmTpk2bsn79eqpWrcrKlSspWLCg17FEJEClpkh7GKgFnAE+ByIBHXokInIJNm3aRNOmTdm2bRv16tVj5cqVVKhQwetYIhLAUjxwAKhhrX0WeDa9w4iIZEZr166lXbt2HDt2jGbNmvH111+rB01EUpSaIu1tY0wp4AtgqrV2YzpnEhHJVIoXL07evHm56aabmDZtGrlz5/Y6kogEgRSLNGttK7dI6w6MMcYUwCnWXk73dCIimUC5cuVYtWoVpUqVInv27F7HEZEgkaqT2VprD1prRwH9gXXA/6VrqpQMWeLp4kVEUjJ+/HhefPHF+OFy5cqpQBORS5JikWaMucYYM9wY8xtw/sjOsumeLDkTN3m6eBGR5Lz99tvce++9DBs2jJ9++snrOCISpFKzT9o4YCrQ3lq7P53ziIgELWstzz33HK+++ioA7733Ho0aNfI4lYgEq9Tsk3ZjRgQREQlmsbGxDBw4kDFjxhAaGsr48ePp06eP17FEJIglWaQZY6ZZa7u7mzl9rzBgAGutrZPu6ZLydkvPFi0iklhMTAx9+/Zl6tSp5MyZk+nTpxMeHu51LBEJcsn1pA12/3bOiCCXpG9trxOIiMSLiIjgl19+IX/+/Hz99de0aNHC60gikgkkWaRZaw+4/z5krX3Kd5oxZgTw1MX3EhH5f/buPS7H+/8D+OsqUc6nmFnIqda5CIlEEossp5zmzIYNc7bNGNuXYTbHzWHmuDmOOSzSRIuZihzGnHOsVJSSdLjfvz/S/dNyCN1d3fV6Ph73Q9d9X4dX1113b5/rc30+xU+1atWwf/9+3Lt3D40aNVI7DhEVEXkZgsPzKc91yO8gRET6JDo6GgsXLtQu161blwUaEeWr5/VJGw5gBIC6iqKceuKlcgAO6zoYEVFhdfXqVXh6euLy5cswNjbGsGHD1I5EREXQ8/qk/QzAH8AsAJOfeD5JRO7qNBURUSF15swZtGvXDlFRUWjUqBF8fX3VjkRERdTzLneKiEQCGAkg6YkHFEWprPtoRESFy9GjR+Hm5oaoqCi4u7vjwIEDMDU1VTsWERVRzyvSfn78bziAsMf/hj+xrB7TxaoenoiKn/3796Nt27a4d+8efHx84O/vj/Lly6sdi4iKsOfd3dnx8b/mBReHiKjwyczMxPjx4/HgwQP0798fK1euRIkSeZmwhYjo1eVl7k5XRVHKPP66r6Io8xVFqaX7aEREhYOhoSF2796NGTNmYNWqVSzQiKhA5GUIju8BpCiKYg9gHIDLANbpNBURUSEQGBgIkawJV8zMzDB16lQYGOTlY5OI6PXl5dMmQ7I+pToDWCwiS5A1DId6Yj9U9fBEVLSJCCZPngxPT0988cUXaschomIqL232SYqiTAHwHoCWiqIYADDSbSwiInVkZmZi+PDhWLFiBQwNDdGgQQO1IxFRMZWXljQ/AI8ADBKRaABvAZir01RERCpIS0tDr169sGLFChgbG+O3335Dnz591I5FRMXUC4u0x4XZBgAVFEXpCCBVRNbqPBkRUQF68OABOnXqhC1btqB8+fIICAiAt7e32rGIqBjLy92dPQAcA9AdQA8AfyuK0k3XwYiICtKoUaMQEBCAatWq4eDBg2jZsqXakYiomMtLn7RPATiLyB0AUBTFFEAggK26DEZEVJC+/PJLREZG4vvvv0fDhg3VjkNElKc+aQbZBdpj8XncTnfWnlH18ERUNMTExGiH2KhRowb++OMPFmhEVGjkpdjaqyjKPkVRBiiKMgDAHgC/6zbWC4w7qOrhiUj/nT59Gg4ODpgyZYraUYiInuqFlztFZIKiKF0AtHj81HIR2a7bWEREunPkyBF4e3sjISEBx44dQ1paGkqWLKl2LCKiHJ5ZpCmK0gDAPAD1AJwGMF5EbhVUMCIiXdi3bx+6dOmClJQUvPvuu/jll19YoBFRofS8y52rAOwG0BVAOIBFBZIoL96zUjsBEemhzZs3o1OnTkhJScGAAQOwZcsWGBsbqx2LiOipnne5s5yIrHj89XlFUY4XRKA8md9G7QREpGe2bduGnj17QkQwduxYzJ07l/NwElGh9rwizVhRFEcAyuNlkyeXRaTwFG1ERC/g5uaGhg0bol+/fpgyZQoURXnxRkREKnpekRYFYP4Ty9FPLAsANmcRUaEmIhARGBgYwNTUFOHh4ShTpozasYiI8uSZRZqItC7IIERE+SkzMxPvv/8+ypQpg++++w6KorBAIyK9wg4ZRFTkPHr0CH5+fvjxxx+xYsUKXLp0Se1IREQvLS/TQhER6Y3k5GT4+voiMDAQFSpUwJ49e9CgQQO1YxERvTT9bEnz2KR2AiIqhOLj4+Hh4YHAwEBUr14dhw4dgqurq9qxiIheyQuLNCVLX0VRPn+8XEtRlCa6j/Ycp2JVPTwRFT63b9+Gm5sbjh07hjp16iAkJAT29vZqxyIiemV5aUlbCsAFQK/Hy0kAlugsERHRKzA2NoaBgQGsrKwQEhKC+vXrqx2JiOi15KVPWlMRcVIU5QQAiMg9RVE4hwoRFSqVK1fG/v37YWRkhCpVqqgdh4joteWlJS1dURRDZI2NBkVRTAFodJrqRQJ7qHp4IiocDh8+jHHjxkFEAABvvPEGCzQiKjLy0pK2EMB2ANUURfkKQDcAn+k01YvYV1P18ESkPn9/f3Tt2hUPHz6Eo6Mj+vbtq3YkIqJ89cIiTUQ2KIoSDsADWVNCvSsi53SejIjoGX755Rf069cPGRkZGDx4MHr16vXijYiI9Exe7u6sBSAFwC4AOwE8ePwcEVGB+/7779GnTx9kZGRg4sSJWLFiBQwNDdWORUSU7/JyuXMPsvqjKQCMAZgDOA/AWoe5iIhyEBH873//w2efZfW2mD17NiZNmqRyKiIi3cnL5U7bJ5cVRXECMEJniYiInuLRo0f49ddfoSgKfvjhBwwbNkztSEREOvXS00KJyHFFUZrqIkyenbzDmweIihljY2Ps3bsXR48eRadOndSOQ0Skc3npkzb2icd4RVF+BnC7ALI9W9vNqh6eiApGamoqFi9eDI0ma9QfU1NTFmhEVGzkpSWt3BNfZyCrj9o23cQhIsqSlJSEzp07IygoCLdu3cKsWbPUjkREVKCeW6Q9HsS2nIiML6A8RESIi4tDhw4dEBYWhjfeeAO9e/dWOxIRUYF7ZpGmKEoJEclQFMW1IAPliZ2p2gmISEdu3rwJT09P/PvvvzA3N8f+/ftRr149tWMRERW457WkHQPgBCBCUZSdALYAeJD9ooj8quNsz/aHn2qHJiLduXDhAjw9PXH9+nXY2Nhg3759ePPNN9WORUSkirz0STMGEA+gDf5/vDQBoF6RRkRF0sSJE3H9+nU0a9YMe/bsQeXKldWORESkmucVadUURRkL4Az+vzjLJjpNRUTF0qpVq/D555/j66+/RpkyZdSOQ0SkqucNwWEIoOzjR7knvs5+EBG9ttDQUGRmZgIAKleujMWLF7NAIyLC81vSokRkRoElIaJiZ8OGDejfvz8GDRqEZcuWQVGUF29ERFRMPK8ljZ+WRKQzixcvRt++fZGZmYkqVaqoHYeIqNB5XpHmUWApXtbYA2onIKJXJCKYMWMGPvroIwDAnDlzMGvWLLaiERH9xzOLNBG5W5BBXsq6s2onIKJXoNFoMGbMGEybNg0GBgZYsWIFJkyYoHYsIqJC6aUnWCcielXz5s3DwoULUbJkSfz888/o2rWr2pGIiAqtF06wTkSUX95//324ublhz549LNCIiF5AP1vSvnFXOwER5VFSUhKMjY1hZGSEChUq4ODBg+x/RkSUB/rZktbPRu0ERJQHsbGxcHd3x+DBg6HRaACABRoRUR7pZ0saERV6N27cgKenJ86fP4/ExETExcWhWrVqasciItIb+tmSRkSF2vnz5+Hq6orz58/Dzs4OISEhLNCIiF4SizQiylfHjx9HixYtcOPGDbi6uuLQoUN444031I5FRKR3WKQRUb4JDw+Hu7s74uLi0KFDBwQEBKBixYpqxyIi0kvsk0ZE+cbCwgJWVlYwNzfHmjVrULJkSbUjERHpLf0s0kwXA7Efqp2CiB4TESiKgrJlyyIgIABlypSBoaGh2rGIiPQaL3cS0WtZuHAhevXqhczMTABA+fLlWaAREeUDFmlE9EpEBNOnT8fo0aOxadMmHDhwQO1IRERFin5e7iQiVWk0GowePRqLFy+GgYEBVq5cCU9PT7VjEREVKfpZpLE/GpFq0tPTMXDgQGzYsAElS5bExo0b4evrq3YsIqIiRz+LNCJSxcOHD9G9e3fs2bMHZcuWxY4dO+Dh4aF2LCKiIkmnfdIURWmvKMp5RVEuKYoy+TnrdVUURRRFaazLPET0+pKSklC5cmX88ccfLNCIiHRIZy1piqIYAlgCwBPATQChiqLsFJGz/1mvHIDRAP7WVRYiyh8mJibYuXMnoqKiYGlpqXYcIqIiTZctaU0AXBKRKyKSBmAjgM5PWW8mgK8BpOowCxG9omvXrmHUqFHIyMgAAFSoUIEFGhFRAdBlkVYTwI0nlm8+fk5LURQnAGYisud5O1IUZZiiKGGKooTlf0wiepazZ8/C1dUVixYtwpdffql2HCKiYkW1cdIURTEAMB/AuBetKyLLRaSxiGT1WVt7RsfpiCg0NBRubm64desWWrRogTFjxqgdiYioWNFlkXYLgNkTy289fi5bOQA2AA4qihIJoBmAnXm6eWDcwXwLSUS5BQUFoU2bNoiPj8c777yDffv2caJ0IqICpssiLRRAA0VRzBVFKQmgJ4Cd2S+KSKKIVBWROiJSB8BRAD4iwkuaRCrasWMHOnTogOTkZPTu3Rs7duxA6dKl1Y5FRFTs6KxIE5EMAB8C2AfgHIDNIvKPoigzFEXx0dVxiejViQiWL1+OR48eYeTIkVi3bh2MjIzUjkVEVCzpdDBbEfkdwO//ee7zZ6zrnucdv2f1WrmI6OkURcHmzZvx888/Y+jQoVAURe1IRETFln5OsD6/jdoJiIoMEcGaNWuQlpYGAChbtiyGDRvGAo2ISGX6WaQRUb7QaDQYOXIkBgwYgIEDB6odh4iInsC5O4mKqbS0NPTv3x8bN25EqVKl4Ofnp3YkIiJ6Aos0omIoJSUF3bp1g7+/P8qVK4edO3fC3d1d7VhERPQEFmlExUxCQgI6deqEkJAQVK1aFXv37kWjRo3UjkVERP/BIo2omJk5cyZCQkLw1ltvYf/+/ZyHk4iokNLPGwc8NqmdgEhvffnllxg4cCAOHz7MAo2IqBDTz5a0U7FqJyDSKxcvXoSZmRmMjY1hYmKCVatWqR2JiIheQD9b0ogoz44dO4ZmzZrBz88PGRkZaschIqI8YpFGVIQFBgaiTZs2uHv3LjQaDYs0IiI9op9FWmAPtRMQFXq//vorvL298eDBA/Tt2xe//vorjI2N1Y5FRER5pJ9Fmn01tRMQFWqrVq1C9+7dkZaWho8++ghr1qzhROlERHpGP4s0InqmnTt3YvDgwdBoNJg+fToWLFgAAwP+qhMR6Rv9vLuTiJ7Jy8sL7dq1Q8eOHfHRRx+pHYeIiF4RizSiIiAzMxNpaWkwMTFBqVKl4O/vz9YzIiI9x09xIj2XlpaG3r17w9fXF2lpaQDAAo2IqAjQz0/yk3fUTkBUKDx48ACdO3fG5s2bceTIEZw/f17tSERElE/0s0hru1ntBESqu3fvHtq1a4e9e/fC1NQUBw8ehK2trdqxiIgon7BPGpEeioqKgpeXF06fPg0zMzPs378fFhYWasciIqJ8xCKNSM/cvn0bbm5uuHz5MiwtLREQEAAzMzO1YxERUT7TzyLNzlTtBESqMTU1haWlJSpVqgR/f39UrVpV7UhERKQD+lmk/eGndgIi1RgZGWHLli1IT09H+fLl1Y5DREQ6op83DhAVM/v370enTp2QmpoKADAxMWGBRkRUxLFIIyrktm7dCm9vb+zevRsrV65UOw4RERUQFmlEhdiKFSvg5+eH9PR0jBkzBiNGjFA7EhERFRAWaUSF1Ndff41hw4ZBo9Fg5syZmD9/PmcSICIqRvTzxgGiIkxEMHnyZMyZMweKomDx4sVsQSMiKob0s0gbewCY3+blt/vVG7j6e/7nIcpHGo0Gly9fRokSJbBmzRr07t1b7UhERKQC/bx2su7sq23HAg0wf0ftBPQChoaG2LBhAw4ePMgCjYioGNPPlrTXNU7UTkCUQ3JyMmbMmIFp06ahTJkyKFWqFFxdXdWORUREKiqeRRpRIXL37l14e3vj6NGjiI6Oxtq1a9WOREREhYB+FmnfuKudgChf3L59G15eXjhz5gxq166NqVOnqh2JiIgKCf0s0vrZqJ2A6LVdvnwZnp6euHr1KqysrBAQEICaNWuqHYuIiAoJ/bxxgEjPnTp1Ci1atMDVq1fh7OyM4OBgFmhERJQDizQiFXz//feIjo5GmzZt8Mcff6BKlSpqRyIiokJGPy93Eum5hQsXok6dOhg9ejSMjY3VjkNERIUQW9KICoi/vz+SkpIAAEZGRpg0aRILNCIieiYWaUQFYNmyZfD29kbnzp2Rnp6udhwiItID+lmkmS5WOwFRnogIZs2ahQ8++AAiAk9PT5QowV4GRET0YvxrQaQjIoIJEybgm2++gaIoWLp0KT744AO1YxERkZ5gkUakAxkZGXj//fexatUqlChRAuvXr4efn5/asYiISI+wSCPSgZUrV2LVqlUwMTHBr7/+ivbt26sdiYiI9Ix+FmmxH6qdgOi5hgwZgtDQUAwaNIgTpRMR0SvRzyKNqBCKj4+HoaEhKlasiBIlSuDHH39UOxIREekx/by7k6iQuXnzJlq2bAlvb288ePBA7ThERFQEsEgjek0XL15EixYtcO7cOSQmJmoHrCUiInodLNKIXkNERARatGiBa9euoWnTpggODsYbb7yhdiwiIioCWKQRvaKQkBC4u7vjzp07aNu2LQIDA1G5cmW1YxERURGhn0Xa2jNqJ6Bi7vTp02jXrh0SExPRtWtX7N69G2XLllU7FhERFSH6eXfnuINAPxu1U1AxZm1tDV9fX5iYmGDZsmUwNDRUOxIRERUx+lmkEank0aNHKFWqFAwMDLBmzRoYGhpCURS1YxERURGkn5c7iQqYiODLL79Ey5YttXdvlihRggUaERHpjH4Wae9ZqZ2AihGNRoOxY8di6tSpCAsLw8GDB9WORERExYB+Xu6c30btBFRMZGRkYMiQIVizZg2MjIywYcMGdOrUSe1YRERUDOhnkUZUAFJTU9GzZ0/89ttvKF26NLZv34527dqpHYuIiIoJFmlET5GSkoKOHTsiKCgIlSpVwp49e+Di4qJ2LCIiKkb0s08akY6ZmJjA3NwcNWrUQHBwMAs0IiIqcGxJI3oKRVGwfPlyREdHo2bNmmrHISKiYogtaUSPXbhwAT4+PkhISAAAGBoaskAjIiLV6GeR5rFJ7QRUxBw/fhwtWrTArl278Pnnn6sdh4iISE+LtFOxaiegIiQ4OBitW7dGbGws2rVrh1mzZqkdiYiISE+LNKJ8snv3bnh5eeH+/fvo0aMHdu3ahTJlyqgdi4iISE9vHJj8EfDNR2qnID23YcMG9O/fH5mZmRg2bBiWLl3KidKJiKjQKH4taebvqJ2AComjR48iMzMTU6ZMwQ8//MACjYiIChVFRNTO8FJMaztI7LUItWNQEaDRaPD777+jY8eOakchIqIiSlGUcBFp/CrbFr+WNCq2NBoNZs+ejbi4OACAgYEBCzQiIiq0WKRRsZCeno4BAwZgypQp8PX1hb61IBMRUfGjnzcOEL2Ehw8fws/PT3vn5rRp06AoitqxiIiInotFGhVpiYmJ6Ny5Mw4dOoTKlSvj999/R9OmTdWORURE9EL6ebnz5B21E5AeuHPnDlq3bo1Dhw7hzTffRHBwMAs0IiLSG/pZpLXdrHYC0gOrV6/GiRMnUK9ePRw+fBjW1tZqRyIiIsozXu6kImvChAl49OgRhg4dijfeeEPtOERERC9FP1vSiJ7hxIkTiImJAQAoioKpU6eyQCMiIr2kn0WananaCagQOnjwIFq1agUvLy8kJiaqHYeIiOi16GeR9oef2gmokNm5cyfat2+PpKQkvP322zAxMVE7EhER0WvRzyKN6Alr165Fly5d8OjRIwwfPhzr169HyZIl1Y5FRET0WlikkV5bsGAB+vfvj8zMTHz22WdYsmQJJ0onIqIigXd3kt46cOAAxowZAwCYP38+Pv74Y5UTERER5R8WaaS3WrdujVGjRsHR0REDBgxQOw4REVG+UvRtomnT2g4Sey1C7RikkvT0dNy7dw/VqlVTOwoREdELKYoSLiKNX2Vb/eyTNvaA2glIBSkpKfD19UXr1q0RHx+vdhwiIiKd0s8ibd1ZtRNQAUtISICXlxf27NmDmJgY3LhxQ+1IREREOsU+aVToxcTEoH379oiIiEDNmjUREBAAKysrtWMRERHpFIs0KtSuXbsGT09PXLx4EQ0aNMD+/ftRu3ZttWMRERHpnH4Wad+4q52ACkBcXBxcXV1x69YtODg4YO/evahevbrasYiIiAqEfhZp/WzUTkAFoEqVKvDz80NoaCh27dqFChUqqB2JiIiowHAIDip0MjIyUKJE1v8fRASPHj2CsbGxyqmIiIheXvEbgoOKrB07dsDR0RExMTEAAEVRWKAREVGxxCKNCo2ffvoJXbt2xZkzZ7B27Vq14xAREamKRRoVCvPnz8egQYOg0Wjw+eefY/z48WpHIiIiUpV+3jhARYaIYOrUqfjqq68AAN999x1Gjx6tcioiIiL16WdLmulitRNQPhARjBgxAl999RUMDQ2xZs0aFmhERESPsSWNVKMoCqpUqYJSpUph8+bN8PHxUTsSERFRoaGfQ3CkDAFiP1Q7CuUDEcGFCxdgYWGhdhQiIqJ8V2iH4FAUpb2iKOcVRbmkKMrkp7w+VlGUs4qinFIU5Q9FUTjfTxGXkJCA3r174+bNmwCyWtNYoBEREeWms8udiqIYAlgCwBPATQChiqLsFJGzT6x2AkBjEUlRFGU4gDkA/F64c7ai6aXo6Gi0b98eJ0+exN27d7F37161IxERERVaumxJawLgkohcEZE0ABsBdH5yBREJEpGUx4tHAbylwzykoqtXr6JFixY4efIkGjZsiOXLl6sdiYiIqFDTZZFWE8CNJ5ZvPn7uWQYD8H/aC4qiDFMUJUxRlLB8zEcF5J9//kGLFi1w+fJlODk54c8//0StWrXUjkVERFSoFYohOBRF6QugMYC5T3tdRJaLSONX7XhH6vn777/h5uaG27dvo1WrVggKCkK1atXUjkVERFTo6bJIuwXA7Inltx4/l4OiKG0BfArAR0Qe6TAPqeDQoUO4e/cufHx84O/vj/Lly6sdiYiISC/ocpy0UAANFEUxR1Zx1hNA7ydXUBTFEcAyAO1F5I4Os5BKJkyYgFq1aqFbt24oUYLD8hEREeWVzlrSRCQDwIcA9gE4B2CziPyjKMoMRVGyRy2dC6AsgC2KokQoirIzTztfe0YXkSmfrF+/HpGRkQCyhtjo2bMnCzQiIqKXxMFsKV/NnTsXEydORP369REREYEyZcqoHYmIiEg1hXYwWyo+RARTpkzBxIkTAQCjRo1igUZERPQaeA2KXltmZiZGjhyJZcuWwdDQEKtXr0bfvn3VjkVERKTX9LNIe89K7QT0WFpaGt577z1s3rwZxsbG2Lx5Mzp16qR2LCIiIr2nn0Xa/DZqJ6DHdu3ahc2bN6N8+fLYtWsX3Nzc1I5ERERUJOhnkUaFRteuXTF79mx4enrCyclJ7ThERERFhn7e3XktQu0YxVpUVBQePHiA+vXrqx2FiIioUOPdnVRgrly5ghYtWqBt27a4dSvXBBJERESUT1ikUZ6dPn0aLVq0wJUrV2BqaopSpUqpHYmIiKjIYpFGefLXX3/Bzc0NUVFRaN26NQ4cOICqVauqHYuIiKjI0s8izWOT2gmKlYCAALRt2xYJCQno3Lkzfv/9d5QrV07tWEREREWafhZpp2LVTlBsXLlyBR07dkRKSgr69++PrVu3wtjYWO1YRERERR6H4KDnqlu3LqZNm4a4uDh88803MDDQz7qeiIhI37BIo6eKi4vT9jn75JNPAACKoqgZiYiIqFjRz2aRwB5qJyiyRAQTJ06Eg4MDrl27BiCrOGOBRkREVLD0syXNvpraCYqkzMxMvP/++/jxxx9RokQJnDhxArVr11Y7FhERUbGkn0Ua5btHjx6hT58+2LZtG0xMTLB161a88847asciKhLS09Nx8+ZNpKamqh2FiHTE2NgYb731FoyMjPJtnyzSCMnJyfD19UVgYCAqVKiA3bt3o0WLFmrHIioybt68iXLlyqFOnTrsOkBUBIkI4uPjcfPmTZibm+cLWIPzAAAgAElEQVTbfvWzTxrlm/T0dHh6eiIwMBDVq1fHoUOHWKAR5bPU1FRUqVKFBRpREaUoCqpUqZLvreVsSSvmjIyM0L17d0RHR2P//v2cNJ1IR1igERVtuvgd18+WtJN31E6g90RE+/XYsWNx8uRJFmhERESFiH4WaW03q51Ar506dQqOjo64ePGi9rny5curmIiICoOdO3di9uzZasdQ3erVq2FqagoHBwdYWlri22+/zfH68uXLYWlpCUtLSzRp0gQhISHa19LT0zF58mQ0aNAATk5OcHFxgb+/f0F/Cy80ZswYBAcHqx3jmcLDw2Fra4v69etj1KhRORoWst27dw++vr6ws7NDkyZNcObMGQDA+fPn4eDgoH2UL18e3333HQBg/PjxOHDgQIF+L69FRPTqUbWWvUjVRUKv5vDhw1KxYkUBIAMGDFA7DlGxcPbs2ZxPVF2U8/Esa07nXO/jP3Qb9CVoNBrJzMxU7fjp6ek62/dPP/0kI0eOFBGRuLg4qVKlily/fl1ERHbt2iVOTk4SGxsrIiLh4eFiZmYmUVFRIiIyadIk6devn6SmpoqISHR0tGzatClf82VkZLzW9nFxcdK0adOX2kaX5/tpnJ2d5a+//hKNRiPt27eX33//Pdc648ePl+nTp4uIyLlz56RNmza51snIyJDq1atLZGSkiIhERkaKp6enznLn+l0XEQBh8oo1j362pNEr2bt3r3ai9C5duuCHH35QOxIRFYDIyEhYWlpiwIABaNiwIfr06YPAwEC4urqiQYMGOHbsGICsFqQPP/wQABATEwNfX1/Y29vD3t4eR44cQWRkJCwsLNCvXz/Y2Njgxo0bmDBhAmxsbGBra4tNmzY99fjHjh2Di4sLHB0d0bx5c5w/fx4A0KxZM/zzzz/a9dzd3REWFoYHDx5g0KBBaNKkCRwdHfHbb79p8/n4+KBNmzbw8PBAcnIyPDw84OTkBFtbW+16ADBz5kxYWFigRYsW6NWrF+bNmwcAuHz5Mtq3b49GjRqhZcuW+Pfff5977qpUqYL69esjKioKAPD1119j7ty52hlZnJyc0L9/fyxZsgQpKSlYsWIFFi1ahFKlSgEAqlevjh49cg/AHhoaiubNm8Pe3h5NmjRBUlJSjvMPAB07dsTBgwcBAGXLlsW4ceNgb2+PWbNmoXv37tr1Dh48iI4dOwIAAgIC4OLiAicnJ3Tv3h3Jycm5jr1t2za0b99euzxjxgw4OzvDxsYGw4YN07Zaubu7Y8yYMWjcuDEWLFiA8PBwtGrVCo0aNYKXl5f2nKxYsQLOzs6wt7dH165dkZKS8txz+iJRUVG4f/8+mjVrBkVR0K9fP+zYsSPXemfPnkWbNm0AAJaWloiMjERMTEyOdf744w/Uq1dPO+Zn7dq1ER8fj+jo6NfKWGBetbpT61G1lr1Im42vWuQWWxs3bhQjIyMBIIMGDSrw/xURFWdqt6RdvXpVDA0N5dSpU5KZmSlOTk4ycOBA0Wg0smPHDuncubOI5GxB6tGjh3z77bciktUakZCQIFevXhVFUeSvv/4SEZGtW7dK27ZtJSMjQ6Kjo8XMzExu376d6/iJiYnaz5z9+/dLly5dRERk/vz58vnnn4uIyO3bt6Vhw4YiIjJlyhRZt26diIjcu3dPGjRoIMnJyfLTTz9JzZo1JT4+XkSyWncSExNFRCQ2Nlbq1asnGo1Gjh07Jvb29vLw4UO5f/++1K9fX+bOnSsiIm3atJELFy6IiMjRo0eldevWufI+eR6uXbum3ZeISKVKlSQhISHH+jt27BBfX185efKkODg4vPD9ePTokZibm8uxY8dynJ8njysi4u3tLUFBQSIiAkDbIpeeni5mZmaSnJwsIiIffPCBrFu3TmJjY6Vly5ba52fPni1ffPFFruP369dPdu7cqV3OPp8iIn379tW+1qpVKxk+fLiIiKSlpYmLi4vcuXNHRLL+pgwcOFBEslrmsn366aeycOHCXMc8cOCA2Nvb53q4uLjkWjc0NFQ8PDy0y8HBweLt7Z1rvSlTpsiYMWNEROTvv/8WQ0NDCQsLy7HOwIEDZdGinL9jQ4YMka1bt+baX37I75Y0/by78w8/tRPolWXLlmH48OEQEYwfPx5z5szhnWZExYy5uTlsbW0BANbW1vDw8ICiKLC1tUVkZGSu9Q8cOIC1a9cCAAwNDVGhQgXcu3cPtWvXRrNmzQAAISEh6NWrFwwNDVG9enW0atUKoaGh8PHxybGvxMRE9O/fHxcvXoSiKEhPTwcA9OjRA+3atcMXX3yBzZs3o1u3bgCyWoN27typbf1KTU3F9evXAQCenp6oXLkygKxGhk8++QTBwcEwMDDArVu3EBMTg8OHD6Nz584wNjaGsbExOnXqBCBrTMgjR47kaIV69OjRU8/Xpk2bEBwcjH///ReLFy+GsbHxy5/0Zzh//jxq1KgBZ2dnAHnrE2xoaIiuXbsCAEqUKIH27dtj165d6NatG/bs2YM5c+bg0KFDOHv2LFxdXQEAaWlpcHFxybWvqKgomJqaapeDgoIwZ84cpKSk4O7du7C2ttaeMz8/P23mM2fOwNPTE0DWDDU1atQAAJw5cwafffYZEhISkJycDC8vr1zHbN26NSIiIvJ8jvJi8uTJGD16NBwcHGBrawtHR0cYGhpqX09LS8POnTsxa9asHNtVq1YNt2/fztcsuqKfRRq9FI1GAxHBrFmzMGnSJBZoRGqL/fDF6wBAP5usRz7IvvwGAAYGBtplAwMDZGRk5Hk/ZcqUeeE6S5YswYoVKwAAv//+O6ZOnYrWrVtj+/btiIyMhLu7OwCgZs2aqFKlCk6dOoVNmzZpu2CICLZt2wYLC4sc+/37779zHH/Dhg2IjY1FeHg4jIyMUKdOneeOU6XRaFCxYsU8FQt+fn5YvHgxwsLC0K5dO/j4+OCNN96AlZUVwsPDtZfZgKxO7tbW1qhfvz6uX7+O+/fvv9LNWCVKlIBGo9EuP/m9GBsb5yhAevbsicWLF6Ny5cpo3LgxypUrBxGBp6cnfvnll+cex8TERLvv1NRUjBgxAmFhYTAzM8P06dNzHDf7fIsIrK2t8ddff+Xa34ABA7Bjxw7Y29tj9erV2ku0TwoKCsLHH3+c6/nSpUvjyJEjOZ6rWbMmbt68qV2+efMmatasmWvb8uXL46efftLmMzc3R926dbWv+/v7w8nJCdWrV8+xXWpqKkxMTHLtrzBin7RiYPjw4Th+/DgmT57MAo2I8sTDwwPff/89gKxWk8TExFzrtGzZEps2bUJmZiZiY2MRHByMJk2aYOTIkYiIiEBERATefPNNJCYmav/Irl69Osc+/Pz8MGfOHCQmJsLOzg4A4OXlhUWLFmn7Rp04ceKpGRMTE1GtWjUYGRkhKCgI165dAwC4urpi165dSE1NRXJyMnbv3g0g64+6ubk5tmzZAiDrD/vJkyefex4aN26M9957DwsWLAAATJw4EZMmTUJ8fDwAICIiAqtXr8aIESNQunRpDB48GKNHj0ZaWhoAIDY2Vnu8bBYWFoiKikJoaCgAICkpCRkZGahTpw4iIiKg0Whw48YNbV/Bp2nVqhWOHz+OFStWoGfPngCy+vgdPnwYly5dAgA8ePAAFy5cyLXt22+/rV0nuyCrWrUqkpOTsXXr1qcez8LCArGxsdoiLT09XdufMCkpCTVq1EB6ejo2bNjw1O2zW9L++/hvgQYANWrUQPny5XH06FGICNauXYvOnTvnWi8hIUF7nleuXAk3N7ccxfEvv/yCXr165druwoULsLHJn//86BqLtCIoIyMDH3/8Mc6dO6d9ztHRUcVERKRvFixYgKCgINja2qJRo0Y4e/ZsrnWyhz+wt7dHmzZtMGfOHLzxxhu51ps4cSKmTJkCR0fHXK123bp1w8aNG3N0rp86dSrS09NhZ2cHa2trTJ069akZ+/Tpg7CwMNja2mLt2rWwtLQEADg7O8PHxwd2dnbo0KEDbG1tUaFCBQBZrW8//vgj7O3tYW1tneNmg2eZNGkSfvrpJyQlJcHHxweDBg1C8+bNYWlpiaFDh2L9+vXaS39ffvklTE1NYWVlBRsbG3Ts2DFXq1rJkiWxadMmfPTRR7C3t4enpydSU1Ph6uoKc3NzWFlZYdSoUXBycnpmJkNDQ3Ts2BH+/v7amwZMTU2xevVq9OrVC3Z2dnBxcXnqjRHe3t7a1q6KFSti6NChsLGxgZeXl/YS7H+VLFkSW7duxaRJk2Bvbw8HBwdtgTVz5kw0bdoUrq6u2vfgdS1duhRDhgxB/fr1Ua9ePXTo0AEA8MMPP2hbXM+dOwcbGxtYWFjA399fW0gDWQXq/v370aVLlxz7TU9Px6VLl9C4ceN8yalrSvb/VPSFaW0Hib2Wv9e1i5LU1FT07t0b27dvR8OGDfHPP/+gRAle1SZS07lz5/D222+rHaNYSU5ORtmyZZGSkgI3NzcsX778uUVPcdOiRQvs3r0bFStWVDtKgdq+fTuOHz+OmTNn6mT/T/tdVxQlXEReqSrkX+8iJCkpCe+++y4OHDiAihUr4qeffmKBRkTF0rBhw3D27Fmkpqaif//+LND+45tvvsH169eLXZGWkZGBcePGqR0jz/TzL/jYA8D8Ni9erxiJj49Hhw4dEBoaiurVqyMgIEDbv4OIqLj5+eef1Y5QqDVt2lTtCKp48s5efaCffdLW5e4bUZzdvHkTLVu2RGhoKMzNzXH48GEWaERERHpOP4s0yuHPP//EuXPnYG1tjZCQENSrV0/tSERERPSa9PNyJ+XQq1cviAjat2+vHeSRiIiI9Jt+FmnfuKudQHUhISEoV64c7O3tAQC9e/dWORERERHlJ/283JlPI3Drqz179sDT0xNeXl45RmUmIqL8FRkZCRMTEzg4OMDKygr9+vXTTmsFZP2HuUmTJrC0tISlpSWWL1+eY/u1a9dqJ6B3dHTUTnVVmOzYsQMzZsxQO8Yz3b17F56enmjQoAE8PT1x7969p643adIk2NjYwMbGBps2bcr1+qhRo1C2bFnt8uLFi7Fq1Sqd5c4P+lmkFWM///wz3n33XaSmpqJTp07aARSJSH8oyhc5Hs+yfHl4jvWGDdtVgClfTmZmpmrHFpEc0ynlt3r16iEiIgKnT5/GzZs3sXnzZgBAdHQ0evfujR9++AH//vsvQkJCsGzZMuzZswdA1rRE3333HQICAnD69GkcPXpUO6hufnmZKb2eZc6cORgxYkSBHvNlzJ49Gx4eHrh48SI8PDwwe/bsXOvs2bMHx48fR0REBP7++2/MmzcP9+/f174eFhaWq7gbNGgQFi1apPP8r4NFmh5ZsmQJ+vbti4yMDEyaNAnLly/PMZcbEdHTREZGwtLSEgMGDEDDhg3Rp08fBAYGwtXVFQ0aNNBOP3Ts2DG4uLjA0dERzZs3x/nz5wFkFWDjx4+HjY0N7OzstH/Y6tSpg0mTJsHJyQlbtmzBL7/8AltbW9jY2GDSpElPzZKcnAwPDw84OTnB1tZWO+L/5MmTsWTJEu1606dP17Y6zZ07F87OzrCzs8O0adO035OFhQX69esHGxsb3LhxA8OHD0fjxo1hbW2tXQ/Imj/U0tISjRo1wqhRo7Qj9D948ACDBg1CkyZN4Ojo+MLZBwwNDdGkSRPcunULQNZn8oABA7RjsFWtWhVz5szRFhGzZs3CvHnz8OabbwLImj916NChufYbExMDX19f2Nvbw97eHkeOHEFkZGSOqYvmzZuH6dOnAwDc3d0xZswYNG7cGF999RVq166tLVIfPHgAMzMzpKen4/Lly2jfvj0aNWqEli1bPnX2gQsXLqBUqVKoWrUqAGDXrl1o2rQpHB0d0bZtW8TExGjfj/feew+urq547733EBsbi65du8LZ2RnOzs44fPgwgGf/DL2O3377Df379wcA9O/fHzt27Mi1ztmzZ+Hm5oYSJUqgTJkysLOzw969ewFk/fxOmDABc+bMybFN6dKlUadOnedOv6U6EdGrR9Va9lLcaDQamTFjhgAQAPL111+rHYmIXsLZs2dzLAPTczyeZdmysBzrDR2685WOf/XqVTE0NJRTp05JZmamODk5ycCBA0Wj0ciOHTukc+fOIiKSmJgo6enpIiKyf/9+6dKli4iILF26VLp27ap9LT4+XkREateurf08unXrlpiZmcmdO3ckPT1dWrduLdu3b8+VJT09XRITE0VEJDY2VurVqycajUaOHz8ubm5u2vXefvttuX79uuzbt0+GDh0qGo1GMjMzxdvbWw4dOiRXr14VRVHkr7/+0m6TnSsjI0NatWolJ0+elIcPH8pbb70lV65cERGRnj17ire3t4iITJkyRdatWyciIvfu3ZMGDRpIcnJyrnNnbW0tIiIPHz4Ud3d3OXnypIiI+Pr6yo4dO3Ksn5CQIJUqVRIRkUqVKklCQsIL358ePXrIt99+q82ekJCQ47giInPnzpVp06aJiEirVq1k+PDh2td8fHzkwIEDIiKyceNGGTx4sIiItGnTRi5cuCAiIkePHpXWrVvnOvaqVatk7Nix2uW7d++KRqMREZEVK1ZoX5s2bZo4OTlJSkqKiIj06tVL/vzzTxERuXbtmlhaWorIs3+GnnT//n2xt7d/6uOff/7JtX6FChW0X2s0mhzL2fbt2yfNmzeXBw8eSGxsrJibm8u8efNEROS7776T+fPni4hImTJlcmz35ZdfatfLD//9XRcRARAmr1jz6OeNA8XM8ePHMW3aNBgYGGDZsmUYMmSI2pGISM+Ym5vD1tYWAGBtbQ0PDw8oigJbW1tERkYCyJqwvH///rh48SIURdH2vQoMDMQHH3ygncHkybvI/fz8AAChoaFwd3eHqakpgKx5NYODg/Huu+/myCEi+OSTTxAcHAwDAwPcunULMTExcHR0xJ07d3D79m3ExsaiUqVKMDMzw4IFCxAQEKCdfzg5ORkXL15ErVq1ULt2bTRr1ky7782bN2P58uXIyMhAVFQUzp49C41Gg7p168Lc3BxA1t3w2f3GAgICsHPnTm2LXWpqKq5fv55rWp/Lly/DwcEBV69ehbe3d76PQ3ngwAGsXbsWQFZrXYUKFZ7Z7ypb9nnP/nrTpk1o3bo1Nm7ciBEjRiA5ORlHjhzJMXjro0ePcu0nKipK+54BWeNu+vn5ISoqCmlpadrzBgA+Pj4wMTEBkPUz8eR8rvfv30dycvIzf4aeVK5cOUREvNr0joqiQFGUXM+3a9cOoaGhaN68OUxNTeHi4gJDQ0Pcvn0bW7Zs0c5V+l/VqlV7agtjYcEiTQ80atQIS5YsgampKbp166Z2HCJ6TSLTXrwSgGHDGmHYsEb5csxSpUppvzYwMNAuGxgYaPsYTZ06Fa1bt8b27dsRGRkJd3f3F+63TJkyz33977//xvvvvw8AmDFjBu7evYvY2FiEh4fDyMgIderUQWpqKoCs0eC3bt2K6OhobREiIpgyZYp2H9kiIyNzHPvq1auYN28eQkNDUalSJQwYMEC732cREWzbtg0WFhbPXS+7T1pcXBxcXV2xc+dO+Pj4wMrKCuHh4ejcubN23fDwcFhbWwPIKobDw8PRps3Lz5BTokSJHP3s/vu9PPm9+/j44JNPPsHdu3e1x3vw4AEqVqz4wmLIxMQEiYmJ2uWPPvoIY8eOhY+PDw4ePKi9xPrfY2o0Ghw9ehTGxsY59vfhhx++8GcoKSkJLVu2fGqen3/+GVZWVjmeq169OqKiolCjRg1ERUWhWrVqT932008/xaeffgoga8SDhg0b4sSJE7h06RLq168PAEhJSUH9+vVx6dIlAFnnNbvwLIz0s0+a6WK1E+hcamoqzp07p10ePnw4CzQi0qnExETUrFkTALB69Wrt856enli2bJm2mLt7926ubZs0aYJDhw4hLi4OmZmZ+OWXX9CqVSs0bdoUERERiIiIgI+PDxITE1GtWjUYGRkhKCgI165d0+7Dz88PGzduxNatW7UtQF5eXli1ahWSk5MBALdu3cKdO3dyHf/+/fsoU6YMKlSogJiYGPj7+wMALCwscOXKFW1r4ZN3/Xl5eWHRokXIuiIFnDhx4rnnp2rVqpg9ezZmzZoFABg5ciRWr16tLYTi4+MxadIkTJw4EQAwZcoUTJgwAdHR0QCAtLQ0rFy5Mtd+PTw88P333wPI6j+VmJiI6tWr486dO4iPj8ejR4+we/fuZ+YqW7YsnJ2dMXr0aHTs2BGGhoYoX748zM3NsWXLFgBZBenJkydzbfv2229rCxYg58/AmjVrnnnMdu3a5eh0n30OnvUz9KTslrSnPf5boAFZRWh2ljVr1uQoirNlZmYiPj4eAHDq1CmcOnUK7dq1g7e3N6KjoxEZGYnIyEiULl06x/d74cKFHH3/Chv9LNKKuPv37+Odd95By5YtcxRqRES6NHHiREyZMgWOjo457uAbMmQIatWqBTs7O9jb2z91XswaNWpg9uzZaN26Nezt7dGoUaOn/jHt06cPwsLCYGtri7Vr18LS0lL7mrW1NZKSklCzZk3tnevt2rVD79694eLiAltbW3Tr1g1JSUm59mtvbw9HR0dYWlqid+/ecHV1BZDVUrR06VJtB/py5cpp77CcOnUq0tPTYWdnB2tra0ydOvWF5+jdd99FSkoK/vzzT9SoUQPr16/H0KFDYWlpiebNm2PQoEHo1KkTAOCdd97Bhx9+iLZt28La2hpOTk457jjMtmDBAgQFBcHW1haNGjXC2bNnYWRkhM8//xxNmjSBp6dnjvP0NH5+fli/fn2Oy6AbNmzAjz/+CHt7e1hbWz/1xgg3NzecOHFCW6hOnz4d3bt3R6NGjbQ3EzzNwoULERYWBjs7O1hZWeGHH34A8OyfodcxefJk7N+/Hw0aNEBgYCAmT54MIOuOzezuP+np6WjZsiWsrKwwbNgwrF+/Xnt5/nkOHz4MT0/PfMmpC0r2G6MvTGs7SGzKECD2Q7Wj6ERsbCw6dOiA8PBw1KhRAwEBAYW6yieiFzt37lyufk5UcJKTk1G2bFmICEaOHIkGDRrg448/VjtWoTF69Gh06tQJbdu2VTtKgTpx4gTmz5+PdevW5ds+n/a7rihKuIg0fpX9sSWtELlx4wbc3NwQHh6OevXq4fDhwyzQiIhe04oVK+Dg4ABra2skJibm6t9W3H3yySdISUlRO0aBi4uLw8yZM9WO8Vz62ZJ27dXuCinMzp8/D09PT9y4cQO2trbYt28fB6olKiLYkkZUPOR3Sxrv7iwEkpOT4e7ujujoaLi4uGDPnj2oVKmS2rGIiIhIRbzcWQiULVsWX331Fdq3b4/9+/ezQCMiIiIWaWp68g6lQYMGYc+ePS8cc4iIiIiKBxZpKlm/fj3q1q2bY9waAwO+HURERJSFVYEKFi5ciPfeew9xcXHaAReJiHTJ0NAQDg4OsLGxQadOnZCQkKB97Z9//kGbNm1gYWGBBg0aYObMmXjypjJ/f380btwYVlZWcHR0xLhx49T4Fl5Jr169YGdnh2+//TZP65ctW1bHiV4sISEBS5cu1flxxowZg+DgYJ0f51V9+umnMDMze+F7MmvWLNSvXx8WFhbYt2+f9vm9e/fCwsIC9evX1056DwA9e/bExYsXdZY7X73qpJ9qParWshdZc/qlJjwtLDQajUybNk07UfrcuXPVjkREBeBpky4XtCcnlu7Xr598+eWXIiKSkpIidevWlX379omIyIMHD6R9+/ayePFiERE5ffq01K1bV86dOyciWROAL126NF+zZU/Ind+ioqKkXr16L7XNfyfgVsN/J1fXhbi4OGnatOlLbaOr9+lZ/vrrL7l9+/Zz35N//vlH7OzsJDU1Va5cuSJ169aVjIwMycjIkLp168rly5fl0aNHYmdnp528/eDBgzJkyBCdZM7vCdb1syVt3EG1E7w0jUaD0aNH44svvoCBgQFWrlyJ8ePHqx2LiAqYoqPHy3BxccGtW7cAZM2V6Orqinbt2gEASpcujcWLF2tbHubMmYNPP/1UO+K9oaEhhg8fnmufycnJGDhwIGxtbWFnZ4dt27YByNkytXXrVgwYMAAAMGDAAHzwwQdo2rQpJk6ciDp16uRo3WvQoAFiYmIQGxuLrl27wtnZGc7Ozjh8+HCuY6empmqP7ejoiKCgIABZsxXcunULDg4O+PPPP3NsExMTA19fX9jb28Pe3h5HjhzJ9f14eHjAyckJtra22tH6Hzx4AG9vb9jb28PGxkY7zdTkyZNhZWUFOzu7p362Hzp0CA4ODnBwcICjo6O2T/LcuXPh7OwMOzs7TJs2Tbuv7EndJ0yYABHBhAkTYGNjA1tbW+0xo6Ki4Obmpm0hzf4ehw8fjsaNG8Pa2lq7z//atm0b2rdvr12eMWMGnJ2dYWNjg2HDhmlbUt3d3TFmzBg0btwYCxYsQHh4OFq1aoVGjRrBy8sLUVFRALLGonN2doa9vT26du2aL+OuNWvW7IVDUf3222/o2bMnSpUqBXNzc9SvXx/Hjh3DsWPHUL9+fdStWxclS5ZEz549te9hy5YtERgYmG8zIujUq1Z3aj2q1rIXqbroFWtc9QwaNEgASMmSJWXbtm1qxyGiAvTk/6519eH4ItmtERkZGdKtWzfx9/cXEZGPP/5Yvvvuu1zrV6xYURITE8XR0VEiIiJeuP+JEyfK6NGjtct3797NcVwRkS1btkj//v1FRKR///7i7e0tGRkZIiIyatQoWbVqlYiIHD16VDw8PEREpFevXvLnn3+KiMi1a9fE0tIy17HnzZsnAwcOFBGRc+fOiZmZmTx8+PC5LVI9evSQb7/9VntOEhIScuRNT0+XxMREERGJjY2VevXqiUajka1bt+ZohUlISJC4uDhp2LChaDQaERG5d+9eruN17NhRQkJCREQkKSlJ0tPTZd++fTJ06FDRaDSSmZkp3t7ecujQoVy5t9FWkUIAACAASURBVG7dKm3btpWMjAyJjo4WMzMzuX37tsybN0/bIpqRkSH3798XEZH4+Hjtc61atZKTJ0/mytOvXz/ZuXOndjl7GxGRvn37al9r1aqVDB8+XERE0tLSxMXFRe7cuSMiIhs3btSe97i4OO32n376qSxcuDDXMQ8cOCD29va5Hi4uLrnWfdLzWtJGjhwp69at0y4PGjRItmzZIlu2bJHBgwdrn1+7dq2MHDlSu9y2bVsJCwt77nFfRX63pHGctALi4eGBLVu24Ndffy12U28Q0f9Ta/jwhw8fwsHBAbdu3cLbb7+d7/MVBgYGYuPGjdrlvAwl1L17dxgaGgLImntyxowZGDhwIDZu3KidgzIwMBBnz57VbnP//n3tNE/ZQkJC8NFHHwEALC0tUbt2bVy4cAHly5d/5rEPHDiAtWvXAshqHcyezzObiOCTTz5BcHAwDAwMcOvWLcTExMDW1hbjxo3DpEmT0LFjR7Rs2RIZGRkwNjbG4MGD0bFjR3Ts2DHX8VxdXTF27Fj06dMHXbp0wVtvvYWAgAAEBATA0dERQFbr3cWLF1GrVq0c24aEhKBXr14wNDRE9erV0apVK4SGhsLZ2RmDBg1Ceno63n33XTg4OAAANm/ejOXLlyMjIwNRUVE4e/Ys7OzscuwzKioKpqam2uWgoCDMmTMHKSkpuHv3LqytrbVzkGa/F+fPn8eZM2e0PzuZmZnalq4zZ87gs88+Q0JCApKTk+Hl5ZXrHLRu3Vo7EbvaqlWrhtu3b6NRo0ZqR3ku/bzc+Z6V2gnyROT/P4579+6NK1eusEAjIlWYmJggIiIC165dg4hgyZIlAAArKyuEh4fnWPfKlSsoW7YsypcvD2tr61yvvwxF+f+LsampqTlee3LIIRcXF1y6dAmxsbHYsWMHunTpAiCrq8jRo0cRERGBiIgI3Lp1q0A692/YsAGxsbEIDw9HREQEqlevjtTUVDRs2BDHjx+Hra0tPvvsM8yYMQMlSpTAsWPH0K1bN+zevTvHZcRskydPxsqVK/Hw4UO4urri33//hYhgypQp2u/t0qVLGDx4cJ4zurm5ITg4GDVr1sSAAQOwdu1aXL16FfPmzcMff/yBU6dOwdvbO9d5B7J+HrKfT01NxYgRI7B161acPn0aQ4cOzbFN9vskIrC2ttbmPX36NAICAgBkXb5evHgxTp8+jWnTpj31mEFBQdpLvk8+mjdvnufv+b9q1qyJGzduaJdv3ryJmjVrPvP5bKmpqTAxMXnl4xYU/SzS5rdRO8EL3blzB+7u7ggLC9M+V7VqVRUTERFl9TlbuHAhvvnmG2RkZKBPnz4ICQlBYGAggKwWt1GjRmHixIkAgAkTJuB///sfLly4ACCraPrhhx9y7dfT01Nb+AHAvXv3AADVq1fHuXPnoNFosH379mfmUhQFvr6+GDt2LN5++21UqVIFQFa/skWLFmnXe1pLTMuWLbFhwwYAwIULF3D9+nVYWFg89zx4eHjg+++/B5DVIpSYmJjj9cTERFSrVg1GRkYICgrCtWvXAAC3b99G6dKl0bdvX0yYMAHHjx9HcnIyEhMT8c477+Db/2vvzuOqqvaHj3+WaArmlANZlqYhgpxzQEFCRCsFJ8THKa5aDpW37HGozOFX3FLxllqOacPNSksTy3tR00e9ofg4a2gkjmiWXpGHyAEZQhDW88c57MsRkKMx2vf9evGSc/bea6+91wG+rrX2+i5YYLe0UoGffvoJk8nE1KlT8fPz4+TJk/Ts2ZPPPvuMjIwMAJKSkvj111+pV6+e3TqaQUFBrFmzhry8PFJTU9m5cyedOnXi3LlzuLq6MmbMGJ5//nkOHz7MtWvXqFu3Lg0aNCAlJaXEFQQ8PDw4c+YM8N/guUmTJmRkZLB27dpij3F3dyc1NZV9+/YBkJuby7FjxwDrup/NmzcnNzfXaIubFfSk3fx183zA2xEWFkZUVBTXr1/n559/5vTp03Tq1Ak/Pz9Onz7Nzz//TE5ODlFRUYSFhRnHJSYmVovc2NUzSKvizp07R1BQEDt37mTChAl2PWpCCFHZfHx8MJvNrF69GmdnZ9avX8+sWbNwd3fHZDLh5+fHuHHjADCbzSxcuJChQ4fi4eGBl5cXZ8+eLVJmREQEV65cwcvLC4vFYkzenz17NqGhoXTu3LnUSeDh4eGsXLnSGF4D65JFcXFxmM1mPD09iw0QX3rpJfLz8zGZTISHh7N8+XJq1659y3MtWrSI2NhYTCYTHTt2tBtSBRg+fDhxcXGYTCa++OIL48GJhIQEOnXqhLe3NzNmzCAiIoL09HRCQ0Mxm8106dKF+fPnFznfwoUL8fLywmw2U6tWLXr37k1ISAjDhg0jICAAk8nE4MGDSU9Pp3HjxgQGBuLl5cXkyZMZMGAAZrMZi8XCk08+ydy5c7n//vvZsWMHFosFHx8f1qxZw8SJE43X7dq1Y9iwYQQGBhZ7/X379mXHjh0ANGzYkDFjxuDl5UXPnj3x8/Mr9ph77rmHtWvXMnXqVCwWC97e3kaAFRkZib+/P4GBgca9+qOmTJlCixYtyMrKokWLFkyfPh2ADRs28OabbwLQvn17nnrqKTw9PenVqxdLly7FycmJmjVrsmTJEnr27ImHhwdPPfUU7du3B6wPjTg7O3P//feXST3LkyRYL2MnT54kODiYCxcuYLFY2Lp1K66urpVdLSFEJZIE66Iq6tKlCxs3bqRhw4aVXZUKtWDBAurXr39bQ8uOKusE69KTVobi4uIICgriwoULBAYGsmPHDgnQhBBCVEnz5s3j/PnzlV2NCtewYUNGjhxZ2dVwiDzdWUZiY2MJCwsjIyOD3r17s3btWlxcXCq7WkIIIUSx/P39K7sKlWL06NGVXQWHSU9aGbly5QpZWVkMHTqUdevWSYAmhBBCiD+kegZp3ddUdg2KGDhwIDt37mTlypXcc889lV0dIYQQQlRz1TNIO5Ja2TUA4P3337dLURIYGEiNGtXzlgohhBCiapE5aXdAa81bb71FZGQkjRo14syZM9x3332VXS0hhBBC3EWk2+c25efnM378eCIjI3FycmLBggUSoAkhqjwnJycjEXe/fv3skpkfO3aMJ598End3d9zc3IiMjLRb33Hz5s34+vri6emJj48PkyZNqoxLuCNDhw7FbDazYMECh/aviGwGZelOV+sfNWpUiYvWOmrhwoVGaq2q6Oeff8bf359HH32U8PBwcnJyiuyzatUqu+wHNWrUMBZMXrNmDWazmfbt2zN16lTjmCVLlvDZZ59VzEXcadLPyvpq8rBF6/iU20x5WjZycnL0sGHDNKBr166t161bVyn1EEJUL8UlXa5ohZNUjxgxwkjMnZWVpVu3bq23bt2qtdY6MzNT9+rVSy9ZskRrrXVCQoJu3bq1PnHihNbamrT7gw8+KNO65ebmlml5BZKTk3WbNm1u65hbJfOuLOVxf0aOHKm/+eYbh/e/ceOG3evc3FxtMpluq27l1c4lGTJkiF69erXWWusXXnih1M/tkSNHdOvWrbXW1oTxDz30kJFMfsSIETomJkZrbf0Z8fb2LraMsk6wXj170izNKvyUWVlZDBgwgK+++op7772XzZs3079//wqvhxCimpunyufrNgQEBJCUlATAV199RWBgICEhIYA1bdSSJUuYPXs2AHPnzuWNN94wVpF3cnJi7NixRcrMyMhg9OjRmEwmzGYz//znPwH7nqm1a9cyatQowNqT8+KLL+Lv78+UKVNo1aqVXe+em5sbKSkppKamMmjQIPz8/PDz87ObB1wgOzvbOLePj4+R7SAkJISkpCS8vb3ZtWuX3TEpKSkMGDAAi8WCxWIpkpooIyOD7t2706FDB0wmE+vXrwcgMzOTvn37YrFY8PLyYs0a64Ns06ZNw9PTE7PZzGuvvWZXVn5+/m1f3/Tp03nmmWcIDAzkmWee4dixY0amA7PZzOnTp4vc3zlz5mAymbBYLEybNg2wptF67LHHMJvNDBgwwEjXVdi2bdvw8fHBZDLx7LPPcv36dQBatWrF1KlT6dChA998843dMdu3b6dDhw7UrGmdNfXJJ5/g5+eHxWJh0KBBZGVlFdvOP/30E7169aJjx44EBQVx8uRJAL799lv8/f3x8fGhR48epKSkFKnn7dBas337dgYPHgzAyJEjWbdu3S2PWb16NX/5y18Aa/5aNzc3Iwl9jx49jM+0i4sLrVq14uDBg3+ojo6QOWkOiouLY8uWLTRu3JgtW7bg63tHiwcLIUSlysvLY9u2bcZq68eOHaNjx452+7Rp04aMjAyuXbvG0aNHHRrejIyMpEGDBiQkJAAUGwzc7MKFC+zduxcnJyfy8vKIjo5m9OjRHDhwgJYtW+Lq6sqwYcN45ZVX6NKlC+fPn6dnz56cOHHCrpylS5eilCIhIYGTJ08SEhJCYmIiGzZsIDQ0tNh8nxMmTKBbt25ER0eTl5dn5M8sUKdOHaKjo6lfvz6//fYbjz32GGFhYWzZsoUHHniATZs2AdYcn5cuXSI6OpqTJ0+ilLILxgBq1KhB//79b/v6jh8/zu7du3F2dmb8+PFMnDiR4cOHk5OTQ15ent05Nm/ezPr16zlw4AAuLi5cvnwZgBEjRvD+++/TrVs33nzzTWbMmMHChQuN47Kzsxk1ahTbtm2jbdu2jBgxgg8//JCXX34ZgMaNG3P48OEi92/Pnj12n5uBAwcyZswYwJoi7NNPP2X8+PFF2rl79+589NFHuLm5ceDAAV566SW2b99Oly5d2L9/P0opli1bxty5c5k3b57dOU+dOmWXMqywHTt22GVOuHTpEg0bNjSCyBYtWhj/MSnJmjVrjGD80Ucf5dSpU/zyyy+0aNGCdevW2Q2X+vr6smvXLjp16nTLMv8oCdIc1LVrV6Kiomjfvr2kdxFC3LlJlZOK7/fff8fb25ukpCQ8PDwIDg4u0/JjYmKIiooyXjdq1KjUY4YMGYKTkxNgzds5c+ZMRo8eTVRUlPHHOCYmxi6v5rVr18jIyLDrQdq9e7cRELRr146WLVuSmJhI/fr1Szz39u3bjflUTk5ONGjQwG671prXX3+dnTt3UqNGDZKSkkhJScFkMjFp0iSmTp1KaGgoQUFB3Lhxgzp16vDcc88RGhpKaGhokfPd7vWBNXm4s7MzYO39/Pvf/86FCxcYOHAgbm5uduXHxMQwevRoY43O++67j7S0NK5evUq3bt0Aa2/SkCFD7I47deoUjzzyCG3btjX2Wbp0qRGklRQUJScn2/0tPHr0KBEREVy9epWMjAx69uxpbCto54yMDPbu3WtXh4JeuwsXLhAeHk5ycjI5OTk88sgjRc7p7u5ebMBdFgqC24Kk640aNeLDDz8kPDycGjVq0LlzZ3766Sdj/2bNmhm9gOWpeg53VpBz586xe/du4/XgwYMlQBNCVEvOzs7Ex8dz7tw5tNYsXboUAE9PTw4dOmS379mzZ7n33nupX78+7du3L7L9dij136HY7Oxsu21169Y1vg8ICODMmTOkpqaybt06Bg4cCFiHCvfv3098fDzx8fEkJSVVyOT+VatWkZqayqFDh4iPj8fV1ZXs7Gzatm3L4cOHMZlMREREMHPmTGrWrMnBgwcZPHgwGzdupFevXkXKu5PrK3x/hg0bxoYNG3B2dqZPnz5s37693O/BzXUozNnZ2a49R40axZIlS0hISOCtt96y21ZQRn5+Pg0bNjSuNT4+3ug1HD9+POPGjSMhIYGPP/64yGcFrAFl4Un+hb9u7r1s3LgxV69e5caNG4A1CHzwwQdLvM6oqCiGDh1q916/fv04cOAA+/btw93d3QhkwfpZLgigy5MEaSU4fvw4gYGB9O7du9widyGEqGguLi4sXryYefPmcePGDYYPH87u3buJiYkBrD1uEyZMYMqUKQBMnjyZt99+m8TERMD6h/ajjz4qUm5wcLAR+MF/hztdXV05ceIE+fn5REdHl1gvpRQDBgzg1VdfxcPDg8aNGwPWeWXvv/++sV9xv4+DgoJYtWoVAImJiZw/fx53d/db3ofu3bvz4YcfAtYh4LS0NLvtaWlpNGvWjFq1ahEbG8u5c+cAuHjxIi4uLjz99NNMnjyZw4cPk5GRQVpaGn369GHBggX8+OOPZXp9YA2cW7duzYQJE+jfvz9Hjhyx2x4cHMznn39uzAW7fPkyDRo0oFGjRsZ8vC+//NLoVSvg7u7OL7/8wpkzZ0rcpzgeHh7GMQDp6ek0b96c3Nxcoy1uVr9+fR555BFjfpvW2rhXaWlpRhC1YsWKYo8v6Ekr7uvmJPFKKZ544gnjCdYVK1aUOI88Pz+fr7/+2piPVuDXX38FrJ/lDz74gOeff97YlpiYaPS6lafqGaT9+Gu5Fn/w4EG6du1KUlISPj4+xXa7CiFEdeXj44PZbGb16tU4Ozuzfv16Zs2ahbu7OyaTCT8/P8aNGweA2Wxm4cKFDB06FA8PD7y8vDh79myRMiMiIrhy5QpeXl5YLBZj8v7s2bMJDQ2lc+fONG/e/Jb1Cg8PZ+XKlXZDbIsXLyYuLg6z2Yynp2exAeJLL71Efn4+JpOJ8PBwli9fTu3atW95rkWLFhEbG4vJZKJjx452Q44Aw4cPJy4uDpPJxBdffGE8OJGQkGBM4J8xYwYRERGkp6cTGhqK2WymS5cuzJ8/v0yvD+Drr7/Gy8sLb29vjh49yogRI+y29+rVi7CwMHx9ffH29ua9994DrMHJ5MmTMZvNxMfH8+abb9odV6dOHT7//HOGDBmCyWSiRo0avPjii7e8dwC9e/dm586dxuvIyEj8/f0JDAw07lVxVq1axaefforFYqF9+/bGHLDp06czZMgQOnbsSJMmTUo9vyPmzJnD/PnzefTRR7l06ZIxD3PDhg1292Hnzp089NBDtG7d2u74iRMn4unpSWBgINOmTbPrSduzZ0+ZTxkojtK6cuZH3KmmLb11atbzkDquXMrftm0b/fv3JzMzk9DQUL7++usK6dIUQty9Tpw4IVMlxF1nwIABzJ07t8j8uLvdDz/8wPz58/nyyy+LbCvuZ10pdUhrfUdPG1bPnrRyEh0dTZ8+fcjMzGT48OH861//kgBNCCGEKMbs2bNJTk6u7GpUuN9++43IyMgKOZc83WmTkpJiPNo8fvx4Fi5cKHk4hRBCiBK4u7uXOvfvblQRw5wFqmeQZm5a5kW6urry5ZdfGk+mFH4iSQghhBCiolXPIG1b8eu23C6tNadPnzYmAw4aNIhBgwaVSdlCCCGEEH/En3Y8Ly8vj7Fjx+Lj41MkHYgQQgghRGWrnj1pf1BOTg4jRoxgzZo11K5dm0uXLlV2lYQQQggh7PzpetKysrLo378/a9asoV69emzZsoV+/fpVdrWEEKJcOTk54e3tjZeXF/369bNbof3YsWM8+eSTuLu74+bmRmRkJIWXZ9q8eTO+vr54enri4+PjUC7PqmLo0KGYzWYWLFjg0P4Vkc2gsI8++shIT+WouLg4JkyYUOp+nTt3vtNq2Vm3bh0zZ84sk7LKw+XLlwkODsbNzY3g4OAS88ZOnToVLy8vvLy8WLNmjfF+QbJ4Ly8vRo4caWQp2LhxY5F15Sqc1rpafTV52KLv1OXLl3Xnzp01oJs0aaLj4uLuuCwhhHDU8ePHK7sKum7dusb3I0aM0LNmzdJaa52VlaVbt26tt27dqrXWOjMzU/fq1UsvWbJEa611QkKCbt26tT5x4oTWWusbN27oDz74oEzrlpubW6blFUhOTtZt2rS5rWMK36fKVF735E4EBATo1NRUh/ev6LpPnjxZv/POO1prrd955x09ZcqUIvts3LhR9+jRQ+fm5uqMjAzt6+ur09LSdF5enm7RooU+deqU1lrrv/3tb3rZsmVaa63z8/O1t7e3zszMdLguxf2sA3H6DmOeP01PmtaaPn36sHfvXh566CF27dpFx44dK7taQog/mTEfXC6Xr9sREBBAUlISAF999RWBgYGEhIQA1rRRS5YsYfbs2QDMnTuXN954w1hF3snJibFjxxYpMyMjg9GjR2MymTCbzfzzn/8E7Hum1q5dy6hRowBrrscXX3wRf39/pkyZQqtWrex699zc3EhJSSE1NZVBgwbh5+eHn58fe/bsKXLu7Oxs49w+Pj5GtoOQkBCSkpLw9vY2UiMVSElJYcCAAVgsFiwWS5G5yRkZGXTv3p0OHTpgMpmMlfEzMzPp27cvFovFrkdm2rRpeHp6Yjabee211+zKys/PL/H6pk+fbmQHePzxx3n55Zfx9fVl0aJFfP/995jNZry9vZk8ebKRhmjHjh1GEvfp06fz7LPP8vjjj9O6dWsWL15snKPwvZ8zZw4mkwmLxcK0adMA+OSTT/Dz88NisTBo0CAjpVRhiYmJ1K5d28gC8O233+Lv74+Pjw89evQgJSXFqMczzzxDYGAgzzzzTIntdvDgQQICAvDx8aFz586cOnWqyDlv1/r16xk5ciRgTRC/bt26IvscP36crl27UrNmTerWrYvZbGbLli1cunSJe+65x3iAMDg42PjsKqV4/PHH2bhx4x+u452qnnPSXt0O85+8rUOUUkRERDBt2jQ2bdrEww8/XE6VE0KIqisvL49t27YZKXKOHTtW5D+sbdq0ISMjg2vXrnH06FGHhjcjIyNp0KABCQkJACUOORV24cIF9u7di5OTE3l5eURHRzN69GgOHDhAy5YtcXV1ZdiwYbzyyit06dKF8+fP07NnTyMpd4GlS5eilCIhIYGTJ08SEhJCYmIiGzZsIDQ0tNh8mBMmTKBbt25ER0eTl5dHRkaG3fY6deoQHR1N/fr1+e2333jssccICwtjy5YtPPDAA2zatAmw5py8dOkS0dHRnDx5EqVUkWTfNWrUoH///sVe381ycnKIi4sDwMvLi08++YSAgAAjsCrOyZMniY2NJT09HXd3d8aOHUutWrWM7Zs3b2b9+vUcOHAAFxcXLl+2BvUDBw5kzJgxgDWt16effsr48ePtyt6zZw8dOnQwXnfp0oX9+/ejlGLZsmXMnTuXefPmAdZAaPfu3Tg7O5fYbu3atWPXrl3UrFmTmJgYXn/9dSMoKpCenk5QUFCx1/rVV1/h6elp915KSoqRcuz+++83AsfCLBYLM2bMYNKkSWRlZREbG4unpydNmjThxo0bxMXF4evry9q1a/nPf/5jHOfr68uuXbt46qmnSrz/5al6BmlfHnc4SMvOzqZOnToA9O3bl549e1KzZvW8bCFE9ffJS/dVynl///13vL29SUpKwsPDo8wX5IyJiSEqKsp43ahRo1KPGTJkCE5OToA1r+XMmTMZPXo0UVFRRn7LmJgYu7ya165dIyMjw66XaPfu3UZw0a5dO1q2bEliYiL169cv8dzbt2835oI5OTnRoEEDu+1aa15//XV27txJjRo1SEpKIiUlBZPJxKRJk5g6dSqhoaEEBQVx48YN6tSpw3PPPUdoaKjRy1VYSddX3H4AV69eJT09nYCAAACGDRtWYo9O3759qV27NrVr16ZZs2akpKTQokULY3tMTAyjR4/GxcUFgPvus34Gjx49SkREBFevXiUjI4OePXsWKTs5OZmmTf+7NumFCxcIDw8nOTmZnJwcu9zWYWFhRpaektotLS2NkSNHcvr0aZRS5ObmFjlnvXr1Skw0XxqlVLHrnIaEhPD999/TuXNnmjZtSkBAAE5OTiiliIqK4pVXXuH69euEhIQYn0mAZs2acfHixTuqS1m4q4c79+/fT5s2bdixY4fxngRoQog/I2dnZ+Lj4zl37hxaa5YuXQqAp6cnhw4dstv37Nmz3HvvvdSvX5/27dsX2X47Cv/BzM7OtttWt25d4/uAgADOnDlDamoq69atY+DAgYB1qHD//v3Ex8cTHx9PUlJShUzuX7VqFampqRw6dIj4+HhcXV3Jzs6mbdu2HD58GJPJREREBDNnzqRmzZocPHiQwYMHs3HjRnr16lWkvJKu72aF74mjCieTd3JyMia+l2bUqFEsWbLEWMT95vYB6+em8Pvjx49n3LhxJCQk8PHHH9ttK1z3ktrtb3/7G0888QRHjx7l22+/Lfac6enpeHt7F/tVOPAr4OrqaqSnSk5OplmzZsVe7xtvvEF8fDzfffcdWmtjiDMgIIBdu3Zx8OBBunbtapdIPTs7u1LTQ961Qdp3331Hjx49uHjxIsuWLavs6gghRJXg4uLC4sWLmTdvHjdu3GD48OHs3r2bmJgYwNrjNmHCBKZMmQLA5MmTefvtt0lMTASsf3w/+uijIuUGBwcbgR/8d7jT1dWVEydOkJ+fT3R0dIn1UkoxYMAAXn31VTw8PGjcuDFg7QF5//33jf2K62EJCgpi1apVgHUO1fnz50tNV9S9e3c+/PBDwDoEnJaWZrc9LS2NZs2aUatWLWJjYzl37hwAFy9exMXFhaeffprJkydz+PBho4eoT58+LFiwgB9//NHh6ytJw4YNqVevHgcOHACw66W8XcHBwXz++efGnLOC4c709HSaN29Obm6ucf9u5uHhwZkzZ4zXaWlpPPjggwCsWLGixHOW1G6Fj1++fHmxxxb0pBX3dfNQJ1h78ArqsmLFCvr3719kn7y8PGO5rSNHjnDkyBFjHuavv/4KwPXr15kzZw4vvviicVxiYqIxF7AyVM8gbd7jt9y8du1a+vbtS2ZmJiNGjCjxgyCEEH9GPj4+mM1mVq9ejbOzM+vXr2fWrFm4u7tjMpnw8/Nj3LhxAJjNZhYuXMjQoUPx8PDAy8uLs2fPFikzIiKCK1eu4OXlhcViMSbvz549m9DQUDp37mzMGypJeHg4K1eutBsKXLx4MXFxcZjNZjw9PYsNEF966SXy8/MxmUyEh4ezfPlyu96l4ixatIjY2FhMJhMdO3Ys0kMzfPhw4uLiMJlMfPHFF8aDEwkJCXTq1Alv8nh7ZgAAC3RJREFUb29mzJhBREQE6enphIaGYjab6dKlC/Pnz3f4+m7l008/ZcyYMXh7e5OZmVlkSNZRvXr1IiwsDF9fX7y9vY0HFSIjI/H39ycwMNC4vpt17dqVH374wViSZfr06QwZMoSOHTsaDxMUp6R2mzJlCv/zP/+Dj4+Pwz1+pZk2bRrfffcdbm5uxMTEGPP34uLieP755wHIzc0lKCgIT09P/vrXv7Jy5UpjZO3dd9/Fw8MDs9lMv379ePLJ/06nio2NpW/fvmVSzzuhCm58ddG0pbdOPVfyWPWyZct44YUXyM/PZ+LEicyfP18SpQshKtWJEyfw8PCo7GqIaqbw3LvZs2eTnJzMokWLKrweEydOpF+/fvTo0aPCz12ZUlJSGDZsGNu2bXP4mOJ+1pVSh7TWvndSh7sqelm4cCFjxowhPz+fmTNnsmDBAgnQhBBCVEubNm0yFiDetWsXERERlVKP119/vdjlOe5258+fN55crSx31Sz69u3bU7t2bd577z2jq14IIYSojsLDwx0eGi1Prq6uhIWFVXY1Kpyfn19lV+HuCtKCg4M5c+aM3aPHQghRFWiti10aQAhxdyiP6WPVeiwwJyeHp59+mq1btxrvSYAmhKhq6tSpw6VLl8rll7gQovJprbl06ZKxLmtZqbY9aZmZmQwcOJB///vfbN++nZ9++qlS1zIRQoiStGjRggsXLpCamlrZVRFClJM6deqUeUdRtQzSLjeeQ6j7evbt20fTpk3ZtGmTBGhCiCqrVq1adiuzCyGEI8p1uFMp1UspdUopdUYpVSTxmFKqtlJqjW37AaVUq9LKzM+7Qbe0Rezbt4+HH36Y3bt34+PjUx7VF0IIIYSoNOUWpCmlnIClQG/AExiqlLp5qeDngCta60eBBcCc0sq9mnKao3nJtGvXjj179tilbxBCCCGEuFuUZ09aJ+CM1vqs1joHiAJuztXQHyjIK7EW6K5KefwpPy8X35oPs2vXLnlIQAghhBB3rfKck/Yg8J9Cry8A/iXto7W+oZRKAxoDvxXeSSn1V+CvtpfX426cP9q0adNyqbQod024qX1FtSFtV71J+1Vf0nbV260Tyd5CtXhwQGv9D+AfAEqpuDtNryAqn7Rf9SVtV71J+1Vf0nbVm1Iq7k6PLc/hziTgoUKvW9jeK3YfpVRNoAFwqRzrJIQQQghRLZRnkPY94KaUekQpdQ/wF2DDTftsAEbavh8MbNey2qMQQgghRPkNd9rmmI0DtgJOwGda62NKqZlAnNZ6A/Ap8KVS6gxwGWsgV5p/lFedRYWQ9qu+pO2qN2m/6kvarnq74/ZT0nElhBBCCFH1VOvcnUIIIYQQdysJ0oQQQgghqqAqG6SVR0opUTEcaLtXlVLHlVJHlFLblFItK6OeoniltV+h/QYppbRSSpYGqEIcaT+l1FO2n8FjSqmvKrqOongO/O58WCkVq5T6wfb7s09l1FMUpZT6TCn1q1LqaAnblVJqsa1tjyilOjhSbpUM0sorpZQofw623Q+Ar9bajDXTxNyKraUoiYPth1KqHjAROFCxNRS34kj7KaXcgP8BArXW7YGXK7yioggHf/YigK+11j5YH7T7oGJrKW5hOdDrFtt7A262r78CHzpSaJUM0iinlFKiQpTadlrrWK11lu3lfqxr6ImqwZGfPYBIrP8xyq7IyolSOdJ+Y4ClWusrAFrrXyu4jqJ4jrSdBurbvm8AXKzA+olb0FrvxLpKRUn6A19oq/1AQ6VU89LKrapBWnEppR4saR+t9Q2gIKWUqFyOtF1hzwGby7VG4naU2n62bvqHtNabKrJiwiGO/Py1BdoqpfYopfYrpW71v39RcRxpu+nA00qpC8D/AcZXTNVEGbjdv41ANUkLJe5OSqmnAV+gW2XXRThGKVUDmA+MquSqiDtXE+uQy+NYe7F3KqVMWuurlVor4YihwHKt9TylVADWdUa9tNb5lV0xUT6qak+apJSqvhxpO5RSPYA3gDCt9fUKqpsoXWntVw/wAnYopX4BHgM2yMMDVYYjP38XgA1a61yt9c9AItagTVQuR9ruOeBrAK31PqAO1uTroupz6G/jzapqkCYppaqvUttOKeUDfIw1QJP5MFXLLdtPa52mtW6itW6ltW6FdU5hmNb6jhMIizLlyO/OdVh70VBKNcE6/Hm2IispiuVI250HugMopTywBmmpFVpLcac2ACNsT3k+BqRprZNLO6hKDneWY0opUc4cbLt3gXuBb2zPepzXWodVWqWFwcH2E1WUg+23FQhRSh0H8oDJWmsZhahkDrbdJOATpdQrWB8iGCWdE1WDUmo11v/8NLHNGXwLqAWgtf4I6xzCPsAZIAsY7VC50r5CCCGEEFVPVR3uFEIIIYT4U5MgTQghhBCiCpIgTQghhBCiCpIgTQghhBCiCpIgTQghhBCiCpIgTQhR5pRSeUqp+EJfrW6xb0YZnG+5Uupn27kO21Zjv90ylhUktFZKvX7Ttr1/tI62cgruy1Gl1LdKqYal7O+tlOpTFucWQlQ/sgSHEKLMKaUytNb3lvW+tyhjObBRa71WKRUCvKe1Nv+B8v5wnUorVym1AkjUWv/9FvuPAny11uPKui5CiKpPetKEEOVOKXWvUmqbrZcrQSnVv5h9miuldhbqaQqyvR+ilNpnO/YbpVRpwdNO4FHbsa/ayjqqlHrZ9l5dpdQmpdSPtvfDbe/vUEr5KqVmA862eqyybcuw/RullOpbqM7LlVKDlVJOSql3lVLfK6WOKKVecOC27MOWYFkp1cl2jT8opfYqpdxtq87PBMJtdQm31f0zpdRB275F7qMQ4u5RJTMOCCGqPWelVLzt+5+BIcAArfU1Wyqi/UqpDTetlj4M2Kq1/rtSyglwse0bAfTQWmcqpaYCr2INXkrSD0hQSnXEuqq3P6CAA0qp/wu0Bi5qrfsCKKUaFD5Yaz1NKTVOa+1dTNlrgKeATbYgqjswFmtOxTSttZ9SqjawRyn1b1tuzCJs19cda+YUgJNAkG3V+R7A21rrQUqpNynUk6aUehtrCrxnbUOlB5VSMVrrzFvcDyFENSVBmhCiPPxeOMhRStUC3lZKdQXysfYguQL/r9Ax3wOf2fZdp7WOV0p1AzyxBj0A92DtgSrOu0qpCKy5DJ/DGgRFFwQwSql/AUHAFmCeUmoO1iHSXbdxXZuBRbZArBewU2v9u22I1ayUGmzbrwHWpOU3B2kFweuDwAngu0L7r1BKuWFN91OrhPOHAGFKqddsr+sAD9vKEkLcZSRIE0JUhOFAU6Cj1jpXKfUL1gDDoLXeaQvi+gLLlVLzgSvAd1rroQ6cY7LWem3BC6VU9+J20lonKqU6YM2jN0sptU1rfaueucLHZiuldgA9gXAgquB0wHit9dZSivhda+2tlHLBmqPxfwOLgUggVms9wPaQxY4SjlfAIK31KUfqK4So3mROmhCiIjQAfrUFaE8ALW/eQSnVEkjRWn8CLAM6APuBQKVUwRyzukqptg6ecxfwv5RSLkqpusAAYJdS6gEgS2u9EnjXdp6b5dp69IqzBuswakGvHFgDrrEFxyil2trOWSytdRYwAZiklKqJ9f4k2TaPKrRrOlCv0OutwHhl61ZUSvmUdA4hRPUnQZoQoiKsAnyVUgnACKxzsG72OPCjUuoHrL1Ui7TWqViDltVKqSNYhzrbOXJCrfVhYDlwEDgALNNa/wCYsM7ligfeAmYVc/g/gCMFDw7c5N9ANyBGa51je28ZcBw4rJQ6CnxMKSMVtrocAYYCc4F3bNde+LhYwLPgwQGsPW61bHU7ZnsthLhLyRIcQgghhBBVkPSkCSGEEEJUQRKkCSGEEEJUQRKkCSGEEEJUQRKkCSGEEEJUQRKkCSGEEEJUQRKkCSGEEEJUQRKkCSGEEEJUQf8faI7U8fWI2OYAAAAASUVORK5CYII=\n","text/plain":["<Figure size 720x720 with 1 Axes>"]},"metadata":{"tags":[],"needs_background":"light"}}]},{"cell_type":"code","metadata":{"id":"NKR8H7Nxl2Xi"},"source":[""],"execution_count":null,"outputs":[]}]}