{ "cells": [ { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "# Introducing Scikit-Learn" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "\n", "\n", "*This notebook contains an excerpt from the [Python Data Science Handbook](http://shop.oreilly.com/product/0636920034919.do) by Jake VanderPlas; the content is available [on GitHub](https://github.com/jakevdp/PythonDataScienceHandbook).*\n", "\n", "*The text is released under the [CC-BY-NC-ND license](https://creativecommons.org/licenses/by-nc-nd/3.0/us/legalcode), and code is released under the [MIT license](https://opensource.org/licenses/MIT). If you find this content useful, please consider supporting the work by [buying the book](http://shop.oreilly.com/product/0636920034919.do)!*" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "\n", "< [What Is Machine Learning?](05.01-What-Is-Machine-Learning.ipynb) | [Contents](Index.ipynb) | [Hyperparameters and Model Validation](05.03-Hyperparameters-and-Model-Validation.ipynb) >" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "Python machine learning \n", "\n", "- [Scikit-Learn](http://scikit-learn.org) provides efficient versions of a large number of common algorithms.\n", " - Scikit-Learn is characterized by a clean, uniform, and streamlined API, as well as by very useful and complete online documentation.\n", "\n", "> # Once you understand the basic use and syntax of Scikit-Learn for one type of model, switching to a new model or algorithm is very straightforward.\n" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "Python machine learning \n", "\n", "A solid understanding of these API elements will form the foundation for understanding the deeper practical discussion of machine learning algorithms and approaches.\n", "\n", "This section provides an overview of the Scikit-Learn API \n", "\n", "- The *data representation* in Scikit-Learn\n", "- The *Estimator* API\n", " - a more interesting example of using these tools for exploring a set of images of hand-written digits." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "## Data Representation in Scikit-Learn" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "Machine learning is about creating models from data: \n", "- How data can be represented in order to be understood by the computer.\n", " - Tables of data." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "### Data as table\n", "\n", "A basic table is a two-dimensional grid of data\n", "- the rows represent individual elements of the dataset\n", "- the columns represent quantities related to each of these elements.\n", "\n", "For example, consider the [Iris dataset](https://en.wikipedia.org/wiki/Iris_flower_data_set), famously analyzed by Ronald Fisher in 1936.\n", "\n", "We can download this dataset in the form of a Pandas ``DataFrame`` using the [seaborn](http://seaborn.pydata.org/) library:" ] }, { "cell_type": "code", "execution_count": 39, "metadata": { "ExecuteTime": { "end_time": "2018-05-15T07:26:46.002170Z", "start_time": "2018-05-15T07:26:45.988082Z" }, "slideshow": { "slide_type": "subslide" } }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
sepal_lengthsepal_widthpetal_lengthpetal_widthspecies
05.13.51.40.2setosa
14.93.01.40.2setosa
24.73.21.30.2setosa
34.63.11.50.2setosa
45.03.61.40.2setosa
\n", "
" ], "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" ] }, "execution_count": 39, "metadata": {}, "output_type": "execute_result" } ], "source": [ "import seaborn as sns\n", "sns.set_context(\"talk\", font_scale=1.5)\n", "\n", "iris = sns.load_dataset('iris')\n", "iris.head()" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- Each row of the data refers to a single observed flower\n", "- The number of rows is the total number of flowers in the dataset.\n", " - the rows of the matrix as *samples*\n", " - the number of rows as ``n_samples``.\n", "- each column of the data refers to a particular quantitative piece of information that describes each sample.\n", " - the columns of the matrix as *features*\n", " - the number of columns as ``n_features``." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "#### Features matrix\n", "\n", "This table layout of the information can be thought of as a ``two-dimensional numerical array or matrix``, which we will call the **features matrix**.\n", "\n", "- The features matrix is often stored in a variable named ``X``.\n", "- The features matrix is assumed to be two-dimensional, with shape ``[n_samples, n_features]``, \n", "- The features matrix is most often contained in a NumPy array or a Pandas ``DataFrame``\n", "- some Scikit-Learn models also accept SciPy sparse matrices.\n", "\n", "The samples (i.e., rows) always refer to the individual objects described by the dataset.\n", "- For example, the sample might be a flower, a person, a document, an image, a sound file, a video, an astronomical object, or anything else you can describe with a set of quantitative measurements.\n", "\n", "The features (i.e., columns) always refer to the distinct observations that describe each sample in a quantitative manner.\n", "- Features are generally real-valued, but may be Boolean or discrete-valued in some cases." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "#### Target array\n", "\n", "In addition to the feature matrix ``X``, we also generally work with a *label* or *target* array, which by convention we will usually call ``y``.\n", "- The target array is usually one dimensional, with length ``n_samples``\n", "- The target array is generally contained in a NumPy array or Pandas ``Series``.\n", "- The target array may have continuous numerical values, or discrete classes/labels.\n", "\n", "While some Scikit-Learn estimators do handle multiple target values in the form of a two-dimensional, ``[n_samples, n_targets]`` target array, we will primarily be working with the common case of a one-dimensional target array." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "#### Target array\n", "\n", "The target array is usually the quantity we want to *predict from the data*: in statistical terms, it is the dependent variable.\n", "> For example, in the preceding data we may wish to construct a model that can predict the species of flower based on the other measurements; in this case, the ``species`` column would be considered the target array.\n", "\n", "With this target array in mind, we can use Seaborn (see [Visualization With Seaborn](04.14-Visualization-With-Seaborn.ipynb)) to conveniently visualize the data:" ] }, { "cell_type": "code", "execution_count": 95, "metadata": { "ExecuteTime": { "end_time": "2018-05-15T12:51:31.747920Z", "start_time": "2018-05-15T12:51:26.463393Z" }, "slideshow": { "slide_type": "subslide" } }, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "%matplotlib inline\n", "import seaborn as sns; \n", "sns.set()\n", "sns.set_context(\"talk\", font_scale=1)\n", "sns.pairplot(iris, hue='species', size=1.5);" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "For use in Scikit-Learn, we will extract the features matrix and target array from the ``DataFrame``\n", "- we can use some of the Pandas ``DataFrame`` operations discussed in the [Chapter 3](03.00-Introduction-to-Pandas.ipynb):" ] }, { "cell_type": "code", "execution_count": 40, "metadata": { "ExecuteTime": { "end_time": "2018-05-15T07:31:02.519569Z", "start_time": "2018-05-15T07:31:02.513989Z" }, "slideshow": { "slide_type": "fragment" } }, "outputs": [ { "data": { "text/plain": [ "(150, 4)" ] }, "execution_count": 40, "metadata": {}, "output_type": "execute_result" } ], "source": [ "X_iris = iris.drop('species', axis=1)\n", "X_iris.shape" ] }, { "cell_type": "code", "execution_count": 41, "metadata": { "ExecuteTime": { "end_time": "2018-05-15T07:31:11.307830Z", "start_time": "2018-05-15T07:31:11.298124Z" }, "slideshow": { "slide_type": "subslide" } }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
sepal_lengthsepal_widthpetal_lengthpetal_width
05.13.51.40.2
14.93.01.40.2
24.73.21.30.2
\n", "
" ], "text/plain": [ " sepal_length sepal_width petal_length petal_width\n", "0 5.1 3.5 1.4 0.2\n", "1 4.9 3.0 1.4 0.2\n", "2 4.7 3.2 1.3 0.2" ] }, "execution_count": 41, "metadata": {}, "output_type": "execute_result" } ], "source": [ "X_iris[:3]" ] }, { "cell_type": "code", "execution_count": 42, "metadata": { "ExecuteTime": { "end_time": "2018-05-15T07:31:27.159081Z", "start_time": "2018-05-15T07:31:27.153904Z" }, "slideshow": { "slide_type": "subslide" } }, "outputs": [ { "data": { "text/plain": [ "(150,)" ] }, "execution_count": 42, "metadata": {}, "output_type": "execute_result" } ], "source": [ "y_iris = iris['species']\n", "y_iris.shape" ] }, { "cell_type": "code", "execution_count": 9, "metadata": { "ExecuteTime": { "end_time": "2018-05-15T07:06:05.380136Z", "start_time": "2018-05-15T07:06:05.374539Z" }, "slideshow": { "slide_type": "fragment" } }, "outputs": [ { "data": { "text/plain": [ "0 setosa\n", "1 setosa\n", "2 setosa\n", "Name: species, dtype: object" ] }, "execution_count": 9, "metadata": {}, "output_type": "execute_result" } ], "source": [ "y_iris[:3]" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "To summarize, the expected layout of features and target values is visualized in the following diagram:" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "![](figures/05.02-samples-features.png)\n", "[figure source in Appendix](06.00-Figure-Code.ipynb#Features-and-Labels-Grid)" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "## Scikit-Learn's Estimator API" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "With this data properly formatted, we can move on to consider the *estimator* API of Scikit-Learn:" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "Guiding principles outlined in the [Scikit-Learn API paper](http://arxiv.org/abs/1309.0238):\n", "\n", "- *Consistency*: All objects share a common interface drawn from a limited set of methods, with consistent documentation.\n", " - Every machine learning algorithm in Scikit-Learn is implemented via the Estimator API, which provides a consistent interface for a wide range of machine learning applications.\n", "\n", "- *Inspection*: All specified parameter values are exposed as public attributes.\n", "\n", "- *Limited object hierarchy*: \n", " - Only algorithms are represented by Python classes; \n", " - datasets are represented in standard formats (NumPy arrays, Pandas ``DataFrame``s, SciPy sparse matrices) and \n", " - parameter names use standard Python strings.\n" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "Guiding principles outlined in the [Scikit-Learn API paper](http://arxiv.org/abs/1309.0238):\n", "\n", "- *Composition*: Many machine learning tasks can be expressed as sequences of more fundamental algorithms,\n", " and Scikit-Learn makes use of this wherever possible.\n", "\n", "- *Sensible defaults*: When models require user-specified parameters, the library defines an appropriate default value.\n", "\n", "> In practice, these principles make Scikit-Learn very easy to use, once the basic principles are understood.\n", "\n" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "### Basics of the API\n", "\n", "Most commonly, the steps in using the Scikit-Learn estimator API are as follows\n", "(we will step through a handful of detailed examples in the sections that follow).\n", "\n", "1. Choose a class of model by importing the appropriate estimator class from Scikit-Learn.\n", "2. Choose model hyperparameters by instantiating this class with desired values.\n", "3. Arrange data into a features matrix and target vector following the discussion above.\n", "4. Fit the model to your data by calling the ``fit()`` method of the model instance.\n", "5. Apply the Model to new data:\n", " - For supervised learning, often we predict labels for unknown data using the ``predict()`` method.\n", " - For unsupervised learning, we often transform or infer properties of the data using the ``transform()`` or ``predict()`` method.\n", "\n", "We will now step through several simple examples of applying supervised and unsupervised learning methods." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "### Supervised learning example: Simple linear regression\n", "\n", "As an example of this process, let's consider a simple linear regression—that is, the common case of fitting a line to $(x, y)$ data.\n", "We will use the following simple data for our regression example:" ] }, { "cell_type": "code", "execution_count": 98, "metadata": { "ExecuteTime": { "end_time": "2018-05-15T12:53:05.100098Z", "start_time": "2018-05-15T12:53:04.956259Z" }, "slideshow": { "slide_type": "subslide" } }, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZUAAAEcCAYAAAAP5CkrAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzt3Xt0U1W+B/Bv0leSvtLUIjDWUl6lFBAoQkGktToXGVuQguOA81gqQpc4zFLBkTUyOuNyDS6WAypXcHB0ROYO9xa4ykNUYNCrgvJwoCMtglpaxoJNmzZgk7S0OfePTiKF5OQkPcnJyfl+1prV6UlysneL+XXv/du/rRMEQQAREZEM9Eo3gIiIYgeDChERyYZBhYiIZMOgQkREsmFQISIi2cQr3QAlWa0XRR/X6XTIzExGS0s7tJgkx/6z/+w/+++r/1lZqX5fx5GKCL2+5wer1+hPif1n/9l/9j/Y/qvix3XkyBHcddddKCwsxG233YbNmzcDAOx2OxYvXozCwkKUlJSgqqpK4ZYSEWlb1E9/2e12PPjgg3jiiSdQVlaG2tpa3Hvvvbj++uuxefNmmEwmHDhwAF988QUeeOABjB49GiNGjFC62UREmhT1I5XGxkYUFxdj5syZ0Ov1KCgowKRJk/DZZ59h7969WLJkCZKSkjBmzBiUlZVxtEJEpKCoH6nk5+dj1apV3u/tdjuOHDmCvLw8xMfHIzs72/tYbm4u3nvvPcn3DjRfqNfren3VGvaf/b/8q9aouf/WNieaWh3ol2FCltkY0j1C7X/UB5XLXbx4EZWVld7RysaNG3s9bjAY4HK5JN8vMzMZOl3gH5jZnBx0W2MJ+8/+a5ma+n/8tBVV+07h9Nk2OFxdMBniMSzbjB/fOhxjhmWFdM9g+6+aoHL27FlUVlYiOzsba9aswVdffXVVAHG5XDCZTJLv2dLSHnCkYjYno62tHW639lIK2X/2n/1XT/9r6mz4044atF7s8F5zuLpw/HQzGs5fxMLykRiZa5F8P7H+Wywpfl+niqBy4sQJLFiwADNnzsSvf/1r6PV65OTkoKurC42NjRg4cCAAoK6uDkOHDpV8X0EQ0N0d+Hlut4Du7uj/RxUu7D/7z/5Hf/+3f3ymV0C5XOvFDmw/cAZ512cEfd9g+x/1C/XNzc1YsGAB7r33Xixfvhz6fw8tUlJScOutt+K5556D0+lEdXU1du7cifLycoVbTEQUWU1tTtSduyD6nDPnLsDa5gx7W6J+pLJlyxbYbDasW7cO69at817/+c9/jqeffhpPPvkkiouLYTKZsGzZMtxwww0KtpaIKPKaW51wdYpPuzg7utFsd4a8cC9V1AeVyspKVFZW+n38+eefj2BriIiizzUZRhgS40QDizEpDtekhzegACqY/iIiInH9zEbkDkgTfc6gAWlhH6UADCpERDGhbEoOMlKTfD6WkZqEssk5EWkHgwoRUQzIz7FgQVk+8gdlwJgUB6Bnyit/UEbP9Rzp6cR9EfVrKkREJE1+jgX5ORZY25xotjtxTboxIlNel2NQISKKMVnmyAcTD05/ERGRbBhUiIhINgwqREQxyNrmRE2dLSK76C/HNRUiohhSW2/DzgP1OHP+Apwd3TAmxWFQ/zSUTcnxLuJbW53IygjPuguDChFRjKitt+GVnbW9Cks6O7pRW9+KhqaLyEwzwNrm9Bls5MLpLyKiGLHzQL3fSsXtzi40fPsdnB09pVw8weaVnbWorbfJ1gYGFSKiGCClUrEvrRc7sPNgvWztYFAhIooBUioV+yNnWXwGFSKiGOCpVBwKT1l8OTCoEBFFSDjTfKVUKvZHzrL4zP4iIgqzQGm+cimbkoPzNoffxXp/5CyLz5EKEVEYedJ8a+tbw5555a9S8fXXpiDVmODzNXKXxedIhYgojMTSfD2ZV3KOVvxVKq6tt2HnwXqcOXfZaGlAGsomyztaYlAhIgoTKWm+nswruXe3X1mpOFJl8RlUiIjCREqaryfzKlKl6sNdFp9rKkREYSIlzVfOzKtowKBCRBQmUtJ8pWReKVVxOBSc/iIiCiOxNN9AmVeRSkWWE0cqRERh5C/NN39QRs91P8EhkqnIcuJIhYgozELJvIp0KrJcGFSIiCJEauaVkqnIfcXpLyKiKBNMKnK0YVAhIs2LtuwqNacic/qLiDRLLLtq1OBMAD0B53yzI2xnuvviSUWurW/1+xw5i0DKiUGFiDRJ7Dz38zYHpt94HWrP2nGqoVWRdN6+pCIridNfRKRJgbKrtnzwNY6fblYsnTfUVGSlcaRCRJojJbuqq1vweT2S6bxyF4G0tjlhbXWGdSqPQYWINKcv57kDkU/n7WsRyEjuzOf0FxFpTl/OcweiN53Xl0jvzGdQISLN6ct57kD0pvP6ImVnvpwYVIhIk8qm5CAjNcnnY3FxOtHXRms675WC2ZkvFwYVItIkseyqucWD/QacaE7nvZISO/O5UE9EmiWWXTWofxreOXy29z6VMJzpHk6etSOxwCL3VB6DChFpnq/sqpG5FkwtvB5ffG3FtzaH33TeSKTphkqJnfmqCirV1dV48MEH8dFHH3m/v/vuu2EwGLzPWbRoESorK5VqIhHFmCyzEZZUw1XX1XKAVqR35qsiqAiCgK1bt2LlypWIi/s+DfDkyZOYNm0aXn75ZQVbR0RaE6jESzTtePesHe08WI8z5y6EfSpPFUFl/fr12L17NyorK7Fhwwbv9ZqaGowYMULBlhGRFqntAC25d+aLUUVQmTNnDiorK3Ho0KFe12tra5GYmIjS0lK43W7MmDEDDz/8MBITEyXdV6fTQS+S/6bX63p91Rr2n/2//KvW+Ot/U6sTZ84HTtO1XXRF3RpL/0wT+meaJD031N+/KoJKv379fF7PyMjApEmTcPfdd6OlpQW/+tWv8MILL2Dp0qWS7puZmQydLvAPzGxODqq9sYb9Z/+17Mr+N7Q4vTvT/XF2dKOjG7BYUsLZtIgI9veviqDiz/r1673/32QyYdGiRfjjH/8oOai0tLQHHKmYzcloa2uH2+27uFwsY//Zf/b/6v4b9D1puGKBxZgUh6Q4wGb7LhJNDQux379YsFRtULHb7Vi/fj0WL16MlJSeDnZ0dCApyfeGJV8EQUC3hJpybreAbj8VS7WA/Wf/2f/v+5+ZZsCg/oHTdC2phpj4uQX7+1ftjvrU1FTs2bMHa9euxaVLl1BfX4/169ejoqJC6aYRUYwTK/Giph334aDaoKLX67F+/XqcPHkSRUVFmD9/Pm6//Xb84he/ULppRBTj1HqAViToBEFQ//gsRFbrRdHH4+J0sFhSYLN9FxPD2GCx/+w/+x+4/5FI01WCWP+zslL9vk61aypERNGgrwdoxRrVTn8REVH0YVAhIiLZMKgQkapZ25yoqbPJetAUhY5rKkSkSmqpEqw1DCpEpDpqqhKsNZz+IiLVkVIlmJTBoEJEqtLU5kTducBVgrnGogwGFSJSleZWp+iZ60DPVFiznUFFCQwqRKQq12QYYUiME32OMSkO16RzQ6ISGFSISHHBpAX3MxuROyBN9DmDBqRxl7tCmP1FRIoJNS24bEoOztscPhfrtV4lWGkcqRCRIjxpwbX1rd4DrzxpwT3XbX5fyyrB0YsjFSJShJS0YLHgkJ9jQX6OJWarBKsVgwoRRVwwacGBAgWrBEcXTn8RUcQxLTh2MagQUcQxLTh2MagQUcQxLTh2MagQkSLKpuQgIzXJ52NMC1YvBhUiUgTTgmMTs7+IKCKsbU5YW53Iyvg+W4tpwbGHQYWIwkrKrnmmBccOBhUiChsepqU9XFMhIlF9OQOeh2lpD0cqROTTgX+ew56j/8I31u/Q0ekO+gx4uXbN+1qLoejFoEJEXtY2J4580YSPqs/hXIuj12PBTlsFs2veV7AItYIxKYtBhYi8H+Bff2NHR5db9LlSij0C3++aFwss/nbNcy1GvbimQqRxl5egDxRQPKScAd+XXfNci1EvBhUijRP7APdHarHHUHbNB7MWQ9GHQYVIw6R8gPsitdhjKLvmWcFY3bimQqRhUj7AfQmm2GOwu+b7shZDyuNIhUjDpJSgv1J8nA5jBge/SJ5lNiI/xxIwGLGCsboxqBBpmJQP8Ct1dQt47/C/RM+Q7ytWMFYvBhUijRP7APcn3BlYrGCsXlxTIVKJcO0s93yA/+f/fg6Hq0vy66SeId+XdrGCsfowqBBFuUjsLM9MN8LtFoJ6jdhueDmxgrG6MKgQRbFI7SwPJQuMGVjkC9dUiKJYpHaWh5IFxgws8oVBhShKRXJnebBZYMzAIn9UFVSqq6sxdepU7/d2ux2LFy9GYWEhSkpKUFVVpWDriOQV6Z3lUrLAmIFFgahiTUUQBGzduhUrV65EXNz3Q/QVK1bAZDLhwIED+OKLL/DAAw9g9OjRGDFihIKtJZJHpHeWe7LAdh6sx5lzPUkBJkM8BvVPxU2j+8OcksQMLApIFUFl/fr12L17NyorK7FhwwYAQHt7O/bu3Yt3330XSUlJGDNmDMrKylBVVYUVK1ZIuq9Op4NeZKym1+t6fdUa9l/Z/g/INGHwwDTUnGn1+xydTofWiy70zzTJ8p6jBmdi1ODMnjTeC04Myc6EIQ5BZ4bFAqV//0oLtf8hBZVVq1ahoqICQ4YMCeXlQZszZw4qKytx6NAh77X6+nrEx8cjOzvbey03Nxfvvfee5PtmZiZDpwv8AzObk4NrcIxh/5Xr/z235+OPf/sMLXaXz8cdri5s2FmLh+eNxw3DsmR7X4slBXmy3U3d+O8/uP6HFFT+/Oc/49VXX8XIkSNRUVGBO+64A2azOZRbSdKvX7+rrjkcDhgMhl7XDAYDXC7f//H50tLSHnCkYjYno62tXbN/qbH/yvb/ukwjFtyRjxe3VsPR4XsarMXuwn+9U4vszO+npaxtTjS1OtAvwxTydFU09F9J7L///lssKX5fF1JQ0ev1cLvdOHHiBGpqarBy5UrccsstmDVrFkpKSnqte4SL0Wi8KoC4XC6YTNKnAQRBQLeE1Hy3W0B3t/b+UXmw/5Ht/5U75zPSDAj0mVZ37gLOtzjQbHfKvlGSv3/2P5j+hxRUPvjgA+zYsQPbt2/HyZMncenSJezZswd79uyB2WxGeXk57rzzTowcOTKU20uSk5ODrq4uNDY2YuDAgQCAuro6DB06NGzvSRRO/nbOjxpskZQFdvRUE/Yc/heP4CVFhZRSnJWVhfvuuw9vvvkmtm/fjvvuuw/9+vWDIAhobW3FG2+8gTlz5mDmzJn4y1/+gubmZrnbjZSUFNx666147rnn4HQ6UV1djZ07d6K8vFz29yIKt8uP9HX+e5rLExDePXQWifHi/6nqdcBHx8/xCF5SXJ/3qQwfPhyPPfYY3n//fbz22mu48847YTQaIQgCTp06hWeffRYlJSVYtGgR3n33XXR2dsrRbgDA008/ja6uLhQXF2PJkiVYtmwZbrjhBtnuTxQpYjvnL7R3Ij5OPKHELQCNLQ7R5/AIXooEnSAIsk8Wulwu7N+/H/v27cOHH34Iu93uzbJKS0vDj370I1RUVGD06NFyv3VQrNaLoo/HxelgsaTAZvtOk3Oq7L88/Q9UXbipzYmnXj0kOsWVGK+DMSke9vZLIbcDAJbNGyt5Coy/f/bfX/+zslL9vi4s+1QMBgNmzJiB0tJS7N+/H6tXr0ZDQwOAnl3wmzdvxubNmzFq1Cg89NBDKC4uDkcziBQltbqwlJ3znV0CbisciN2H6hHqn4EsAEmRIHtQcblc2LNnD3bv3o0DBw6go6NnSC8IAhISEjB58mTU1tbCarXin//8JyorK1FRUYFnnnlG7qYQKSaY6sJSd84PvMYUckABWACSIkOWoOJ2u/HRRx9hx44d2Lt3rzfV1zOzNnLkSMyePRvl5eUwm81wu9348MMPsWbNGtTW1mLbtm3IycnBwoUL5WgOkeKkVBf2BBVPMcfaev875wcNSMPQbHPA4OMPC0BSpPQpqFRXV2P79u3YvXs3bLae86o9gcRisaC8vBwVFRXIy+u9N1ev16O4uBjjxo3DnXfeicbGRlRVVTGoUEwIprqwZ+RQNiUH520On4HIExCkBJ/rr01GsjHRW7vLmBSHQQPSUDZZvgO9iMSEFFTWrl2LHTt2eNdJPIEkPj4e06ZNQ0VFBUpKShAfL377tLQ0FBUVYdu2bfj2229DaQpR1AmmurAnqPgq5ugrIAQKPneXDuMRvKSokIOKTqfzBpPhw4ejoqICM2fOhMUS3F9DDkdPGqRnAyOR2oVaXVjKmexSgw+P4CWlhDz9lZaWhvLycsyePRsFBQUhN2DWrFmYM2cOBg8eHPI9iKKJ1DUSfx/6gQKClOBDpJSQgsrzzz+P0tJSJCQk9LkBt9xyS5/vQRRtpKyR9BVHIxSNQtpRP336dFkCClEssLY5UVNn67Vb3TNNlT8oA8akngKrPDWRtEAVh3QRRaNAmxs5TUVaxKBCFIJgNjdymoq0pM8FJYm0SMrmRiItYlAhClIwmxuJtIZBhShIwWxuJNIaBhWiIHk2N4phRWDSKgYV0ixfqcBSeDY3imFFYNIqZn+R5kg950RMJDY3EqkRRyqkKWJnwfdct0m6Dzc3EvnGkQppSjDnnATCzY1EV2NQIc0I5ZwTKbi5keh7nP4izWAqMFH4MaiQZjAVmCj8GFRIM5gKTBR+DCqkKWVTcpCRmuTzMaYCE/UdgwppClOBicKL2V+kOUwFJgofBhVSDWubE9ZWJ7Iy5AkCTAUmkh+DCkU9OcqqEFFkMKhQVAvmhEUiUh4X6imqXFk5WG0nLIZa+ZgoVnCkQlHB1xTXwMxknG36TvR1oZRVCQdO0RH1YFAhxfmb4vqqUbxOl+d5zXZlgwqn6Ii+x+kvUpzYFFcg0VBWRW1TdEThxKBCipJSOViM0mVVgql8TKQFDCqkKCmVg/2JhrIqrHxM1BuDCilKSuXgOD2Q0z8lKsuqsPIxUW9cqCdFeSoH19a3+n1OtxtoanViYGYyCkdkoXB4v5CnvOTelS+l/UpP0RFFEoMKKa5sSg7O2xyii/WebDDbxQ7kXJsa9Id0OFN+xdofDVN0RJGk+umvV155BaNGjcK4ceO8/zty5IjSzaIg+Koc7E8o2VSelN/a+lY4O3rWPzwpvz3XbSG3HWDlY6LLqX6kUltbi4cffhj333+/0k2hPvBUDj7Z0Io1/3McnV1uv88NdsOjlJTfvn7ws/IxUQ/Vj1Rqa2uRn5+vdDNIJu5uQTSgAMFlU0U65TfLbER+joUBhTRL1SMVp9OJM2fOYOPGjVi2bBnS0tJw//33Y+7cuZJer9PpoBcJq3q9rtdXrVGi/9dmmmBMivNOU/liTIrDtRYT4uICt8t2QVrKr+2iC/0zTb2u8/fP/l/+VWtC7b+qg0pzczPGjx+PefPm4YUXXkB1dTUqKyuRlZWF4uLigK/PzEyGThf4B2Y2J8vRXNWKZP8tlhQMvz4Dx083+33O8OszkDc4S9L9hgk6GJPi4ezo8vsckyEeQ3MyYbH47id//+y/lgXbf1UHlezsbGzatMn7/YQJEzBr1izs27dPUlBpaWkPOFIxm5PR1tYOt1uQo8mqolT/Z0zMRsP5i36zqWZMzIbNJl5o0iNRB+QOSEXNGZGU3/6pSNQJV92Tv3/2n/333X+LJcXv61QdVE6cOIGPP/4YCxcu9F7r6OiAwWCQ9HpBENAtYTO32y2gu1t7/6g8It3/4dk9WVM7D9bj62/s6LjkRlKCHoN/kI6yyTkYnp0RVHvumJyDcy3+U37vmJwjej/+/tl/9l96/1W9UG8ymbB27Vq88847cLvdOHjwIHbt2oXZs2cr3TSSgwDAMzup+/f3IWDKL1HkqHqkkpubizVr1mD16tV4/PHHce211+IPf/gDCgoKlG4a9YGvUvIdne4+lZJnyi9RZKg6qABAaWkpSktLlW6GqshdqkRu4dxXkmWOzj4TxQrVBxWSrq+lSiIRjILZV8LgQBR9GFQ0oi+nE9bU2bD94zMROSr3y7NtkkvJM6gQRR8GFY0INKX03/u+xOKK0Vd9UB8/bcWfdtSE/ahczyjq60Z7wOeylDxR9FJ19hdJI2VKqaHpOzyx4SCe2XikV4HFqn2nwn5U7uUFHzsuiZdoAVhKniiaMahogNTTFS91A181XsDq/zmOdw/Vo6nViVMNbaKvkaNuVjBn1LOUPFF04/SXBnhOJ5R6bG9Xt4CtH3wNAKLlTXoel7a+4W+RX+oZ9UmJegwe2LP5kftKiKIXg4oGSDmd8Epd3QI+rWkKWDcr0PpGoIwzqaOon//HCEwe1V9y+4lIGZz+0oiyKTnISE0K6jXnmtuRMyBV9Dli6xtSDseSesb70OvSg2o7ESmDQUUjLi9VkpQg7dfe0eXGlFED/QajQOsbUjYxekZRYrgwT6QenP7SkMtLlaypOoZzLeIL7DodkJ6aiB/fMgT/V30OZ85dNoU1IE10fSOYTYw8450odjCoaFCW2YiyKYOwYUet6PMEAVj9t39410F++h95MKckSqqbJWWtxLPI7xlF7TxYH1TgIqLow6CiUekm6esrV252lDIVJSXj7PJFfhZ8JIoNXFPRKCkL5FcKZrNjqGslPOOdSN0YVDRKyoe+L8FsdhTLOONaCVFsYlDRsFDSjD3rIFLwcCwi7eGaiob5WiAPJNhijlwrIdIWBhWNu/JDf9sHX+OrRv+pwKHuGeHhWETawOkvAvD9AnlF8WCugxBRyBhUqBeugxBRX3D6i67imRKzXXShoxtIigMsqQalm0VEKsCgQn5lmY2wWFJgs32H7m5B6eYQkQpw+ouIiGTDoBJDrG1O1NTZ+nwSIxFRqDj9FQMCHYRFRBQpDCoq5zkI6/Ky8VcWgLwm3ejzKF8iIrkxqKhcoIOw/vN/P4cgCBzBEFFEMKjIxNrmjOhowNrmxOl/teGrRrvo8xyu78+Xv3IEw8BCRHJjUOmjSK9nXPl+ofCUsGdQISK5Maj0gZT1DDk/uH29X6g8Jey5xkJEcmJKcR8EWs+QeqCVHO8XrGBK2BMRScWgEqKmNifqzvmv5gsEd6CVHO8XjGBL2BMRScGgEqLmVqfo+etA30YDV25klPJ+HsakOJgM4jOboZawJyISwzWVEHnOeBf7oA9lNOBv4X/qmP4B3y8pQY+f356HoT8wo9nuxPo3T+Ci89JVz0s1JrCEPRGFBYNKiDxnvNfWt/p9TrCjgUAL//0yjGj49ju/rx/8g3RMLhgAAGi2O+GG7yKQgp/rRER9xemvPhA74z2UA60CLfx77ivl/XYeqEe7s8vnc79zdsmeREBEBDCo9EkoB1r5K/ooZSHe2ubE3JIhAd8v0kkEREQenP7qoyvPeL8m3feO+kCbJKUu/JtTErHsJ+NE3y+YJAIu1hORnBhUZJJl9l+eRVLRxyAX/sXeL1xJBEREgXD6KwKkbJL0LPyLkbrwL+e9iIiCofqgUlNTg7lz52Ls2LGYNWsWjh07pnSTeglmfUPOhX+5kwiIiKRQdVDp6OhAZWUlKioqcPjwYfzsZz/DQw89hM7OzrC+bzAnLAazvhHKwr8/ct6LiEgqVa+pfPLJJ9Dr9Zg/fz4AYO7cuXj99dexf/9+TJ8+Xfb3C6Ui8TUZRiQl6NFxye33vpevb0hd+JdCznsREUmh6qBSV1eHIUOG9LqWm5uL06dPSwoqOp0OepGxml6v836tqRNfbF9YPhIjc3sHlpo6G3YcOINLXf4DCgDkDkhD/0xTr2v9M01XXQtVqPe6vP9axP6z/5d/1ZpQ+6/qoOJwOGA09v7L22AwwOVySXp9ZmYydLrAPzCzORnvHD4mutj+zuGzmFp4vffa8dNWvLKrFi128bZkphtwz+35sFhSJLVZCWZzstJNUBT7z/5rWbD9V3VQMRqNVwUQl8sFk0naX+UtLe0BRypmczJO1VlxqsF/ORYAONXQii++tnqnl/7rHfGAotcBI3IyMPOmQbgu0wibzX/5FaV4+t/W1g63W3ulXdh/9p/9991/sT+CVR1UBg8ejE2bNvW6VldXh7KyMkmvFwQB3RIK/35rcwQ8ZdHZ0Y1vbQ5YUg1oanPiy2/Ej/lNSNDjF7ePQJbZiO7u6P4H63YLUd/GcGL/2X/2X3r/VZ39NXnyZHR2duKNN97ApUuXsGXLFjQ3N2Pq1Kmyvk+W2QRDYpzoczyL7bX1Nry07Z/oFFmYB4COTjcPySKimKPqoJKYmIgNGzZg165dmDhxIjZt2oR169ZJnv6Sql+GtM2EzXYnXtlZi4amwFNZ3NFORLFI1dNfADBixAhs3rw57O9TNiUH520On4v1ns2EwRz3yx3tRBSLVD1SiaRAmwkz042Sj/vljnYiilWqH6lEkthmwpo6m6TjfrOvTcFPSodyRzsRxSQGlRD4qhAspTJwUoIeD80ezWkvIopZnP6SiZTKwIN/kM6AQkQxjUFFRqwMTERax6AiI1YGJiKt45qKzFgZmIi0jEElTMSO+yUiilWc/iIiItkwqBARkWwYVIiISDY6QRC0W9OZiIhkxZEKERHJhkGFiIhkw6BCRESyYVAhIiLZMKgQEZFsGFSIiEg2DCpERCQbBhUiIpINgwoREcmGQcWPmpoazJ07F2PHjsWsWbNw7NgxpZsUUUeOHMFdd92FwsJC3Hbbbdi8ebPSTYq45uZmTJ48Gfv371e6KRF3/vx5LFq0COPHj8e0adOwceNGpZsUUZ999hkqKiowfvx4TJ8+HTt27FC6SRFRXV2NqVOner+32+1YvHgxCgsLUVJSgqqqqsA3EegqLpdLuPnmm4W//vWvQmdnp1BVVSXcdNNNQkdHh9JNi4i2tjbhxhtvFN566y2hu7tb+Pzzz4Ubb7xR+Pjjj5VuWkQtXLhQGDFihPD3v/9d6aZElNvtFmbPni2sXLlS6OzsFE6dOiXceOONwtGjR5VuWkR0dXUJRUVFwu7duwVBEITDhw8LI0eOFM6ePatwy8LH7XYLVVVVQmFhoTBx4kTv9V/+8pfC0qVLBZfLJRw/flyYOHGiUFtbK3ovjlR8+OSTT6DX6zF//nwkJCRg7ty5yMjKL6QqAAAGqklEQVTI0MxfrI2NjSguLsbMmTOh1+tRUFCASZMm4bPPPlO6aRHzt7/9DUajEQMGDFC6KRF3/PhxNDU1YenSpUhISMCwYcOwefNm5ObmKt20iLhw4QJsNhu6u7shCAJ0Oh0SEhIQFxendNPCZv369di4cSMqKyu919rb27F3714sWbIESUlJGDNmDMrKygKOVhhUfKirq8OQIUN6XcvNzcXp06cValFk5efnY9WqVd7v7XY7jhw5ghEjRijYqsg5c+YMXnvtNTz11FNKN0URJ06cwLBhw7Bq1SrcdNNNmD59Oo4fP46MjAylmxYRGRkZmD9/Ph555BEUFBTgnnvuwYoVK2L6D4w5c+bgrbfewujRo73X6uvrER8fj+zsbO81KZ+DPPnRB4fDAaOx96mNBoMBLpdLoRYp5+LFi6isrERBQQFKS0uVbk7YdXV1YdmyZfjNb34Ds9msdHMUYbfb8emnn6KoqAj79+/H559/jgULFiA7OxsTJkxQunlh53a7YTAY8Pzzz6O0tBQHDhzAo48+ioKCgpj9w6pfv35XXXM4HDAYDL2uSfkc5EjFB6PReNUPzuVywWQyKdQiZZw9exY/+clPkJ6ejrVr10Kvj/1/Li+99BLy8/NRXFysdFMUk5iYiPT0dCxatAiJiYnexep9+/Yp3bSIeO+991BdXY3bb78diYmJKCkpQUlJCd58802lmxZRoX4Oxv6nRAgGDx6Murq6Xtfq6uowdOhQhVoUeSdOnMCPf/xjTJ06FS+99NJVf7HEqrfffhu7du3ChAkTMGHCBDQ2NuKRRx7Bn/70J6WbFjG5ublwOp3o6uryXvOsL2jBuXPn0NnZ2etafHw84uO1NbGTk5ODrq4uNDY2eq9J+hwMb06BOnV0dAhTp04VNm7c6M3+KioqEtrb25VuWkRYrVahqKhIePnll5VuiuJuueUWzWV/OZ1O4eabbxZWrlwpXLp0STh69KgwduxY4R//+IfSTYuIkydPCgUFBcKWLVsEt9stfPrpp8K4ceOE6upqpZsWdp988kmv7K+HHnpIeOSRRwSHw+HN/jp27JjoPThS8SExMREbNmzArl27MHHiRGzatAnr1q3TzPTXli1bYLPZsG7dOowbN877v9WrVyvdNIoAg8GAN954A6dOncKUKVOwdOlSPPHEExg7dqzSTYuIvLw8vPDCC9i4cSMKCwvx+9//Hs8++2yvRWytePrpp9HV1YXi4mIsWbIEy5Ytww033CD6Gh4nTEREsuFIhYiIZMOgQkREsmFQISIi2TCoEBGRbBhUiIhINgwqREQkGwYVIiKSDYMKERHJhkGFiIhkw6BCRESyYVAhIiLZMKgQEZFsGFSIFGK1WjFp0iTk5eUhLy8PBw8e9Pm8L7/8EmPGjEFeXh6KiorQ1NQU4ZYSScegQqSQrKwsPPXUU97vf/vb31510t6lS5ewbNkydHR0AACeeeYZn0e/EkULBhUiBc2YMQNlZWUAgIaGBrz44ou9Hn/xxRdRU1MDAJg3bx5uvfXWiLeRKBg8T4VIYXa7HWVlZWhqakJ8fDyqqqowcuRIHD16FD/96U/hdrsxZMgQbNu2TTPHOpN6caRCpLD09HQ888wzAICuri48+eSTcDgcWL58OdxuNxISEvDcc88xoJAqMKgQRYFp06bh7rvvBgBUV1dj3rx5qK+vBwA8+uijyM/PV7J5RJJx+osoSjgcDsyaNQsNDQ3ea1OmTMGrr74KnU6nYMuIpONIhShKmEwmPPHEE72uLV++nAGFVIVBhSiKbN++vdf369atU6glRKFhUCGKErt378bOnTsBAKmpqQCAt99+G2+//baSzSIKCoMKURSwWq3ejZBZWVnYunUrLBYLAOB3v/sdd9GTajCoEEWBFStWoK2tzfv/c3JysHz5cgBAW1sbVqxYoWTziCRjUCFSWFVVFfbv3w8A+OEPf4jp06cDAGbOnIni4mIAwPvvv4+qqirF2kgkFVOKiRT0zTffoLy8HO3t7UhLS8OuXbt61fZqbGzEHXfcAYfDgeTkZGzfvh3XXXedgi0mEseRCpFCBEHA448/jvb2dgDAY489dlWxyIEDB+LRRx8FALS3t2P58uXg34EUzRhUiBTy+uuv49ChQwCAoqIi3HXXXT6fd88992D8+PEAgEOHDuH111+PWBuJgsXpLyIikg1HKkREJBsGFSIikg2DChERyYZBhYiIZMOgQkREsmFQISIi2TCoEBGRbBhUiIhINgwqREQkGwYVIiKSDYMKERHJ5v8BR41I3Ujk6Q0AAAAASUVORK5CYII=\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "import matplotlib.pyplot as plt\n", "import numpy as np\n", "\n", "rng = np.random.RandomState(42)\n", "x = 10 * rng.rand(50)\n", "y = 2 * x - 1 + rng.randn(50)\n", "plt.scatter(x, y)\n", "plt.xlabel('x', fontsize = 30)\n", "plt.ylabel('y', fontsize = 30);" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "With this data in place, we can use the recipe outlined earlier. Let's walk through the process: " ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "#### 1. Choose a class of model\n", "\n", "In Scikit-Learn, every class of model is represented by a Python class.\n", "So, for example, if we would like to compute a simple linear regression model, we can import the linear regression class:" ] }, { "cell_type": "code", "execution_count": 44, "metadata": { "ExecuteTime": { "end_time": "2018-05-15T07:35:54.202165Z", "start_time": "2018-05-15T07:35:54.199317Z" }, "slideshow": { "slide_type": "fragment" } }, "outputs": [], "source": [ "from sklearn.linear_model import LinearRegression" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "Note that other more general linear regression models exist as well; you can read more about them in the [``sklearn.linear_model`` module documentation](http://Scikit-Learn.org/stable/modules/linear_model.html)." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "#### 2. Choose model hyperparameters\n", "\n", "An important point is that *a class of model is not the same as an instance of a model*.\n", "\n", "Once we have decided on our model class, there are still some options open to us.\n", "Depending on the model class we are working with, we might need to answer one or more questions like the following:\n", "\n", "- Would we like to fit for the offset (i.e., *y*-intercept)?\n", "- Would we like the model to be normalized?\n", "- Would we like to preprocess our features to add model flexibility?\n", "- What degree of regularization would we like to use in our model?\n", "- How many model components would we like to use?\n" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "#### 2. Choose model hyperparameters\n", "\n", "\n", "These are examples of the important choices that must be made *once the model class is selected*.\n", "- These choices are often represented as *hyperparameters*, or parameters that must be set before the model is fit to data.\n", "- In Scikit-Learn, hyperparameters are chosen by passing values at model instantiation.\n", "\n", "We will explore how you can quantitatively motivate the choice of hyperparameters in [Hyperparameters and Model Validation](05.03-Hyperparameters-and-Model-Validation.ipynb).\n", "\n", "For our linear regression example, we can instantiate the ``LinearRegression`` class and specify that we would like to fit the intercept using the ``fit_intercept`` hyperparameter:" ] }, { "cell_type": "code", "execution_count": 47, "metadata": { "ExecuteTime": { "end_time": "2018-05-15T07:40:02.740297Z", "start_time": "2018-05-15T07:40:02.735314Z" }, "slideshow": { "slide_type": "subslide" } }, "outputs": [ { "data": { "text/plain": [ "LinearRegression(copy_X=True, fit_intercept=True, n_jobs=1, normalize=False)" ] }, "execution_count": 47, "metadata": {}, "output_type": "execute_result" } ], "source": [ "model = LinearRegression(fit_intercept=True)\n", "model\n", "#help(LinearRegression)" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "Keep in mind that when the model is instantiated, the only action is the storing of these hyperparameter values.\n", "- In particular, we have not yet applied the model to any data: \n", "- the Scikit-Learn API makes very clear the distinction between *choice of model* and *application of model to data*." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "#### 3. Arrange data into a features matrix and target vector\n", "\n", "Previously we detailed the Scikit-Learn data representation, which requires a two-dimensional features matrix and a one-dimensional target array.\n", "\n", "- The target variable ``y`` is already in the correct form (a length-``n_samples`` array)\n", "- The feature matrix ``x`` should be transformed to a matrix of size ``[n_samples, n_features]``.\n", "\n", "In this case, this amounts to a simple reshaping of the one-dimensional array:" ] }, { "cell_type": "code", "execution_count": 49, "metadata": { "ExecuteTime": { "end_time": "2018-05-15T07:41:32.840436Z", "start_time": "2018-05-15T07:41:32.835772Z" }, "slideshow": { "slide_type": "fragment" } }, "outputs": [ { "data": { "text/plain": [ "(50, 1)" ] }, "execution_count": 49, "metadata": {}, "output_type": "execute_result" } ], "source": [ "X = x[:, np.newaxis]\n", "X.shape" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "#### 4. Fit the model to your data\n", "\n", "Now it is time to apply our model to data.\n", "This can be done with the ``fit()`` method of the model:" ] }, { "cell_type": "code", "execution_count": 50, "metadata": { "ExecuteTime": { "end_time": "2018-05-15T07:41:45.689495Z", "start_time": "2018-05-15T07:41:45.684155Z" }, "slideshow": { "slide_type": "fragment" } }, "outputs": [ { "data": { "text/plain": [ "LinearRegression(copy_X=True, fit_intercept=True, n_jobs=1, normalize=False)" ] }, "execution_count": 50, "metadata": {}, "output_type": "execute_result" } ], "source": [ "model.fit(X, y)" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "This ``fit()`` command causes a number of model-dependent internal computations to take place, and the results of these computations are stored in model-specific attributes that the user can explore.\n", "\n", "In Scikit-Learn, by convention all model parameters that were learned during the ``fit()`` process have trailing underscores; \n", "- for example in this linear model, we have the following:" ] }, { "cell_type": "code", "execution_count": 16, "metadata": { "ExecuteTime": { "end_time": "2018-05-15T07:07:27.190067Z", "start_time": "2018-05-15T07:07:27.185547Z" }, "slideshow": { "slide_type": "fragment" } }, "outputs": [ { "data": { "text/plain": [ "array([1.9776566])" ] }, "execution_count": 16, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# The parameters represent the slope of the simple linear fit to the data.\n", "model.coef_" ] }, { "cell_type": "code", "execution_count": 17, "metadata": { "ExecuteTime": { "end_time": "2018-05-15T07:07:28.062749Z", "start_time": "2018-05-15T07:07:28.058403Z" }, "slideshow": { "slide_type": "fragment" } }, "outputs": [ { "data": { "text/plain": [ "-0.9033107255311164" ] }, "execution_count": 17, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# The parameter represent the intercept of the simple linear fit to the data.\n", "model.intercept_" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "Comparing to the data definition, we see that they are very close to the input slope of 2 and intercept of -1.\n", "\n", "One question that frequently comes up regards the uncertainty in such internal model parameters.\n", "\n", "In general, Scikit-Learn does not provide tools to draw conclusions from internal model parameters themselves:\n", "- interpreting model parameters is much more a *statistical modeling* question than a *machine learning* question.\n", "- Machine learning rather focuses on what the model *predicts*.\n", "\n", "If you would like to dive into the meaning of fit parameters within the model, other tools are available, including the [Statsmodels Python package](http://statsmodels.sourceforge.net/)." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "#### 5. Predict labels for unknown data\n", "\n", "Once the model is trained, the main task of supervised machine learning is to evaluate it based on what it says about new data that was not part of the training set.\n", "In Scikit-Learn, this can be done using the ``predict()`` method.\n", "For the sake of this example, our \"new data\" will be a grid of *x* values, and we will ask what *y* values the model predicts:" ] }, { "cell_type": "code", "execution_count": 51, "metadata": { "ExecuteTime": { "end_time": "2018-05-15T07:43:57.243102Z", "start_time": "2018-05-15T07:43:57.237670Z" }, "slideshow": { "slide_type": "fragment" } }, "outputs": [ { "data": { "text/plain": [ "array([-1. , -0.75510204, -0.51020408, -0.26530612, -0.02040816,\n", " 0.2244898 , 0.46938776, 0.71428571, 0.95918367, 1.20408163,\n", " 1.44897959, 1.69387755, 1.93877551, 2.18367347, 2.42857143,\n", " 2.67346939, 2.91836735, 3.16326531, 3.40816327, 3.65306122,\n", " 3.89795918, 4.14285714, 4.3877551 , 4.63265306, 4.87755102,\n", " 5.12244898, 5.36734694, 5.6122449 , 5.85714286, 6.10204082,\n", " 6.34693878, 6.59183673, 6.83673469, 7.08163265, 7.32653061,\n", " 7.57142857, 7.81632653, 8.06122449, 8.30612245, 8.55102041,\n", " 8.79591837, 9.04081633, 9.28571429, 9.53061224, 9.7755102 ,\n", " 10.02040816, 10.26530612, 10.51020408, 10.75510204, 11. ])" ] }, "execution_count": 51, "metadata": {}, "output_type": "execute_result" } ], "source": [ "xfit = np.linspace(-1, 11)\n", "xfit" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "As before, we need to coerce these *x* values into a ``[n_samples, n_features]`` features matrix, after which we can feed it to the model:" ] }, { "cell_type": "code", "execution_count": 52, "metadata": { "ExecuteTime": { "end_time": "2018-05-15T07:44:17.061898Z", "start_time": "2018-05-15T07:44:17.058567Z" }, "slideshow": { "slide_type": "fragment" } }, "outputs": [], "source": [ "Xfit = xfit[:, np.newaxis]\n", "yfit = model.predict(Xfit)" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "Finally, let's visualize the results by plotting first the raw data, and then this model fit:" ] }, { "cell_type": "code", "execution_count": 20, "metadata": { "ExecuteTime": { "end_time": "2018-05-15T07:07:37.931538Z", "start_time": "2018-05-15T07:07:37.811851Z" }, "slideshow": { "slide_type": "fragment" } }, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "plt.scatter(x, y)\n", "plt.plot(xfit, yfit);" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "Typically the efficacy of the model is evaluated by comparing its results to some known baseline, as we will see in the next example" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "### Supervised learning example: Iris classification\n", "\n", "> ### Question: given a model trained on a portion of the Iris data, how well can we predict the remaining labels?\n", "\n", "For this task, we will use an extremely simple generative model known as **Gaussian naive Bayes** \n", "- which proceeds by assuming each class is drawn from an axis-aligned Gaussian distribution\n", "- see [In Depth: Naive Bayes Classification](05.05-Naive-Bayes.ipynb) for more details).\n", "- it is so fast \n", "- it has no hyperparameters to choose\n", "\n", "Gaussian naive Bayes is often a good model to use as a baseline classification, before exploring whether improvements can be found through more sophisticated models." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "# To evaluate the model on data it has not seen before\n", "\n", "- we will split the data into a *training set* and a *testing set*.\n", " - Using the ``train_test_split`` utility function:" ] }, { "cell_type": "code", "execution_count": 53, "metadata": { "ExecuteTime": { "end_time": "2018-05-15T07:46:59.245968Z", "start_time": "2018-05-15T07:46:59.240652Z" }, "slideshow": { "slide_type": "fragment" } }, "outputs": [], "source": [ "from sklearn.cross_validation import train_test_split\n", "Xtrain, Xtest, ytrain, ytest = train_test_split(X_iris, y_iris,\n", " random_state=1)" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "With the data arranged, we can follow our recipe to predict the labels:" ] }, { "cell_type": "code", "execution_count": 59, "metadata": { "ExecuteTime": { "end_time": "2018-05-15T07:50:17.006283Z", "start_time": "2018-05-15T07:50:17.000757Z" }, "slideshow": { "slide_type": "fragment" } }, "outputs": [], "source": [ "from sklearn.naive_bayes import GaussianNB # 1. choose model class\n", "model = GaussianNB() # 2. instantiate model\n", "model.fit(Xtrain, ytrain) # 3. fit model to data\n", "y_model = model.predict(Xtest) # 4. predict on new data" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "Finally, we can use the ``accuracy_score`` utility to see the fraction of predicted labels that match their true value:" ] }, { "cell_type": "code", "execution_count": 60, "metadata": { "ExecuteTime": { "end_time": "2018-05-15T07:50:19.472731Z", "start_time": "2018-05-15T07:50:19.468189Z" }, "slideshow": { "slide_type": "fragment" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "('setosa', 'setosa') ('versicolor', 'versicolor') ('versicolor', 'versicolor') ('setosa', 'setosa') ('virginica', 'virginica') ('versicolor', 'versicolor') ('virginica', 'virginica') ('setosa', 'setosa') ('setosa', 'setosa') ('virginica', 'virginica') ('versicolor', 'versicolor') ('setosa', 'setosa') ('virginica', 'virginica') ('versicolor', 'versicolor') ('versicolor', 'versicolor') ('setosa', 'setosa') ('versicolor', 'versicolor') ('versicolor', 'versicolor') ('setosa', 'setosa') ('setosa', 'setosa') ('versicolor', 'versicolor') ('versicolor', 'versicolor') ('versicolor', 'virginica') ('setosa', 'setosa') ('virginica', 'virginica') ('versicolor', 'versicolor') ('setosa', 'setosa') ('setosa', 'setosa') ('versicolor', 'versicolor') ('virginica', 'virginica') ('versicolor', 'versicolor') ('virginica', 'virginica') ('versicolor', 'versicolor') ('virginica', 'virginica') ('virginica', 'virginica') ('setosa', 'setosa') ('versicolor', 'versicolor') ('setosa', 'setosa')\n" ] } ], "source": [ "print(*zip(ytest, y_model))" ] }, { "cell_type": "code", "execution_count": 61, "metadata": { "ExecuteTime": { "end_time": "2018-05-15T07:50:48.890800Z", "start_time": "2018-05-15T07:50:48.885683Z" }, "slideshow": { "slide_type": "subslide" } }, "outputs": [ { "data": { "text/plain": [ "0.9736842105263158" ] }, "execution_count": 61, "metadata": {}, "output_type": "execute_result" } ], "source": [ "from sklearn.metrics import accuracy_score\n", "accuracy_score(ytest, y_model)" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "With an accuracy topping 97%, we see that even this very naive classification algorithm is effective for this particular dataset!" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "### Unsupervised learning example: Iris dimensionality reduction\n", "\n", "Reducing the dimensionality of the Iris data to more easily visualize it:\n", "- Iris data is four dimensional: \n", " - there are four features recorded for each sample.\n", "\n", "The task of dimensionality reduction is to ask:\n", "\n", "> ## whether there is a suitable lower-dimensional representation that retains the essential features of the data." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "### Unsupervised learning example: Iris dimensionality reduction\n", "\n", "Dimensionality reduction is often used as an aid to visualizing data: \n", "- it is much easier to plot data in two dimensions than in four dimensions or higher!\n", "\n", "Here we will use ``principal component analysis`` (PCA; see [In Depth: Principal Component Analysis](05.09-Principal-Component-Analysis.ipynb))\n", "- It is a fast linear dimensionality reduction technique.\n", "\n", "We will ask the model to return \n", "- two components\n", " - a two-dimensional representation of the data.\n", "# Following the sequence of steps outlined earlier:" ] }, { "cell_type": "code", "execution_count": 62, "metadata": { "ExecuteTime": { "end_time": "2018-05-15T07:52:04.614637Z", "start_time": "2018-05-15T07:52:04.609500Z" }, "slideshow": { "slide_type": "subslide" } }, "outputs": [], "source": [ "from sklearn.decomposition import PCA # 1. Choose the model class\n", "model = PCA(n_components=2) # 2. Instantiate the model with hyperparameters\n", "model.fit(X_iris) # 3. Fit to data. Notice y is not specified!\n", "X_2D = model.transform(X_iris) # 4. Transform the data to two dimensions" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "To plot the results:\n", "- A quick way to do this is to insert the results into the original Iris ``DataFrame``, \n", "- use Seaborn's ``lmplot`` to show the results:" ] }, { "cell_type": "code", "execution_count": 100, "metadata": { "ExecuteTime": { "end_time": "2018-05-15T12:53:24.796325Z", "start_time": "2018-05-15T12:53:24.563624Z" }, "slideshow": { "slide_type": "subslide" } }, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "sns.set_context(\"talk\", font_scale=1.5)\n", "iris['PCA1'] = X_2D[:, 0]\n", "iris['PCA2'] = X_2D[:, 1]\n", "sns.lmplot(\"PCA1\", \"PCA2\", hue='species', data=iris, fit_reg=False);" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "In the two-dimensional representation, the species are fairly well separated, even though the PCA algorithm had no knowledge of the species labels!\n", "\n", "A relatively straightforward classification will probably be effective on the dataset." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "### Unsupervised learning: Iris clustering\n", "\n", "Let's next look at applying clustering to the Iris data.\n", "\n", "> ### A clustering algorithm attempts to find distinct groups of data without reference to any labels.\n", "\n", "We will use a powerful clustering method called a ``Gaussian mixture model (GMM)``\n", "- more detail in [In Depth: Gaussian Mixture Models](05.12-Gaussian-Mixtures.ipynb).\n", "- A GMM attempts to model the data as a collection of Gaussian blobs.\n", "\n", "We can fit the Gaussian mixture model as follows:" ] }, { "cell_type": "code", "execution_count": 64, "metadata": { "ExecuteTime": { "end_time": "2018-05-15T07:53:16.203796Z", "start_time": "2018-05-15T07:53:16.170761Z" }, "slideshow": { "slide_type": "subslide" } }, "outputs": [], "source": [ "from sklearn.mixture import GMM # 1. Choose the model class\n", "model = GMM(n_components=3,\n", " covariance_type='full') # 2. Instantiate the model with hyperparameters\n", "model.fit(X_iris) # 3. Fit to data. Notice y is not specified!\n", "y_gmm = model.predict(X_iris) # 4. Determine cluster labels" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "As before, we will \n", "- add the cluster label to the Iris ``DataFrame`` and \n", "- use Seaborn to plot the results:" ] }, { "cell_type": "code", "execution_count": 78, "metadata": { "ExecuteTime": { "end_time": "2018-05-15T10:05:36.294395Z", "start_time": "2018-05-15T10:05:36.289669Z" }, "slideshow": { "slide_type": "skip" } }, "outputs": [ { "data": { "text/plain": [ "{'axes.labelsize': 11.0,\n", " 'axes.titlesize': 12.0,\n", " 'font.size': 50.0,\n", " 'grid.linewidth': 1.0,\n", " 'legend.fontsize': 10.0,\n", " 'lines.linewidth': 1.75,\n", " 'lines.markeredgewidth': 0.0,\n", " 'lines.markersize': 7.0,\n", " 'patch.linewidth': 0.3,\n", " 'xtick.labelsize': 10.0,\n", " 'xtick.major.pad': 7.0,\n", " 'xtick.major.width': 1.0,\n", " 'xtick.minor.width': 0.5,\n", " 'ytick.labelsize': 10.0,\n", " 'ytick.major.pad': 7.0,\n", " 'ytick.major.width': 1.0,\n", " 'ytick.minor.width': 0.5}" ] }, "execution_count": 78, "metadata": {}, "output_type": "execute_result" } ], "source": [ "sns.plotting_context()" ] }, { "cell_type": "code", "execution_count": 101, "metadata": { "ExecuteTime": { "end_time": "2018-05-15T12:53:39.330142Z", "start_time": "2018-05-15T12:53:38.835696Z" }, "slideshow": { "slide_type": "subslide" } }, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "sns.set_context(\"talk\", font_scale=1.5)\n", "iris['cluster'] = y_gmm\n", "sns.lmplot(\"PCA1\", \"PCA2\", data=iris, hue='species',\n", " col='cluster', fit_reg=False);" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "By splitting the data by cluster number, GMM algorithm recovered the underlying label without an expert: \n", "- the measurements of these flowers are distinct enough\n", "- we could *automatically* identify the presence of these different groups of species \n", " - with a simple clustering algorithm!\n", "- might further give experts in the field clues as to the relationship between the samples they are observing." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "## Application: Exploring Hand-written Digits" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "In the wild, this problem involves \n", "- locating characters in an image. \n", "- identifying characters in an image. \n", "\n", "Here we'll take a shortcut and use Scikit-Learn's set of pre-formatted digits, which is built into the library." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "### Loading and visualizing the digits data\n", "\n", "We'll use Scikit-Learn's data access interface and take a look at this data:" ] }, { "cell_type": "code", "execution_count": 28, "metadata": { "ExecuteTime": { "end_time": "2018-05-15T07:09:24.416431Z", "start_time": "2018-05-15T07:09:24.256577Z" }, "slideshow": { "slide_type": "fragment" } }, "outputs": [ { "data": { "text/plain": [ "(1797, 8, 8)" ] }, "execution_count": 28, "metadata": {}, "output_type": "execute_result" } ], "source": [ "from sklearn.datasets import load_digits\n", "digits = load_digits()\n", "digits.images.shape" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "The images data is a three-dimensional array: \n", "- 1,797 samples \n", "- each consisting of an 8 × 8 grid of pixels.\n", "\n", "Let's visualize the first hundred of these:" ] }, { "cell_type": "code", "execution_count": 102, "metadata": { "ExecuteTime": { "end_time": "2018-05-15T12:59:40.709794Z", "start_time": "2018-05-15T12:59:38.045852Z" }, "slideshow": { "slide_type": "subslide" } }, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "import matplotlib.pyplot as plt\n", "\n", "fig, axes = plt.subplots(10, 10, figsize=(8, 8),\n", " subplot_kw={'xticks':[], 'yticks':[]},\n", " gridspec_kw=dict(hspace=0.1, wspace=0.1))\n", "\n", "for i, ax in enumerate(axes.flat):\n", " ax.imshow(digits.images[i], cmap='binary', interpolation='nearest')\n", " ax.text(0.05, 0.05, str(digits.target[i]),\n", " transform=ax.transAxes, color='green')" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "In order to work with this data within Scikit-Learn, \n", "- we need a two-dimensional, ``[n_samples, n_features]`` representation.\n", "- treating each pixel in the image as a feature: \n", " - so that we have a length-64 array of pixel values representing each digit.\n", "- target array gives the previously determined label for each digit.\n", "\n", "Features and targets are represented as the ``data`` and ``target`` attributes in the `digits` dataset respectively:" ] }, { "cell_type": "code", "execution_count": 107, "metadata": { "ExecuteTime": { "end_time": "2018-05-15T13:02:52.799145Z", "start_time": "2018-05-15T13:02:52.794763Z" }, "slideshow": { "slide_type": "subslide" } }, "outputs": [ { "data": { "text/plain": [ "(1797, 64)" ] }, "execution_count": 107, "metadata": {}, "output_type": "execute_result" } ], "source": [ "X = digits.data\n", "X.shape" ] }, { "cell_type": "code", "execution_count": 108, "metadata": { "ExecuteTime": { "end_time": "2018-05-15T13:02:53.880019Z", "start_time": "2018-05-15T13:02:53.875168Z" }, "slideshow": { "slide_type": "fragment" } }, "outputs": [ { "data": { "text/plain": [ "array([[ 0., 0., 5., ..., 0., 0., 0.],\n", " [ 0., 0., 0., ..., 10., 0., 0.],\n", " [ 0., 0., 0., ..., 16., 9., 0.],\n", " ...,\n", " [ 0., 0., 1., ..., 6., 0., 0.],\n", " [ 0., 0., 2., ..., 12., 0., 0.],\n", " [ 0., 0., 10., ..., 12., 1., 0.]])" ] }, "execution_count": 108, "metadata": {}, "output_type": "execute_result" } ], "source": [ "X" ] }, { "cell_type": "code", "execution_count": 110, "metadata": { "ExecuteTime": { "end_time": "2018-05-15T13:03:30.604467Z", "start_time": "2018-05-15T13:03:30.600002Z" }, "slideshow": { "slide_type": "subslide" } }, "outputs": [ { "data": { "text/plain": [ "(1797,)" ] }, "execution_count": 110, "metadata": {}, "output_type": "execute_result" } ], "source": [ "y = digits.target\n", "y.shape" ] }, { "cell_type": "code", "execution_count": 111, "metadata": { "ExecuteTime": { "end_time": "2018-05-15T13:03:31.333742Z", "start_time": "2018-05-15T13:03:31.329466Z" }, "slideshow": { "slide_type": "fragment" } }, "outputs": [ { "data": { "text/plain": [ "array([0, 1, 2, ..., 8, 9, 8])" ] }, "execution_count": 111, "metadata": {}, "output_type": "execute_result" } ], "source": [ "y" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "We see here that there are 1,797 samples and 64 features." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "### Unsupervised learning: Dimensionality reduction\n", "\n", "We'd like to visualize our points within the 64-dimensional parameter space\n", "- it's difficult to effectively visualize points in such a high-dimensional space.\n", "- Instead we'll reduce the dimensions to 2, using an unsupervised method.\n", "\n", "Here, we'll make use of a manifold learning algorithm called *Isomap* (see [In-Depth: Manifold Learning](05.10-Manifold-Learning.ipynb)), and transform the data to two dimensions:" ] }, { "cell_type": "code", "execution_count": 32, "metadata": { "ExecuteTime": { "end_time": "2018-05-15T07:09:47.457538Z", "start_time": "2018-05-15T07:09:45.592768Z" }, "slideshow": { "slide_type": "subslide" } }, "outputs": [ { "data": { "text/plain": [ "(1797, 2)" ] }, "execution_count": 32, "metadata": {}, "output_type": "execute_result" } ], "source": [ "from sklearn.manifold import Isomap\n", "iso = Isomap(n_components=2)\n", "iso.fit(digits.data)\n", "data_projected = iso.transform(digits.data)\n", "data_projected.shape" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "We see that the projected data is now two-dimensional.\n", "Let's plot this data to see if we can learn anything from its structure:" ] }, { "cell_type": "code", "execution_count": 112, "metadata": { "ExecuteTime": { "end_time": "2018-05-15T13:10:41.495284Z", "start_time": "2018-05-15T13:10:41.273837Z" }, "slideshow": { "slide_type": "subslide" } }, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "plt.scatter(data_projected[:, 0], data_projected[:, 1], c=digits.target,\n", " edgecolor='none', alpha=0.5,\n", " cmap=plt.cm.get_cmap('nipy_spectral', 10))\n", "plt.colorbar(label='digit label', ticks=range(10))\n", "plt.clim(-0.5, 9.5);" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "This plot gives us some good intuition into how well various numbers are separated in the larger 64-dimensional space. \n", "\n", "- zeros (in black) and ones (in purple) have very little overlap in parameter space.\n", " - Intuitively, this makes sense: a zero is empty in the middle of the image, while a one will generally have ink in the middle.\n", "- There seems to be a more or less continuous spectrum between ones and fours: \n", " - we can understand this by realizing that some people draw ones with \"hats\" on them, which cause them to look similar to fours.\n", "\n", "Overall, however, the different groups appear to be fairly well separated in the parameter space: \n", "- this tells us that even a very straightforward supervised classification algorithm should perform suitably on this data.\n", "\n", "Let's give it a try." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "### Classification on digits\n", "\n", "Let's apply a classification algorithm to the digits.\n", "\n", "- split the data into a training and testing set\n", "- fit a Gaussian naive Bayes model" ] }, { "cell_type": "code", "execution_count": 117, "metadata": { "ExecuteTime": { "end_time": "2018-05-15T13:15:12.114546Z", "start_time": "2018-05-15T13:15:12.110268Z" }, "slideshow": { "slide_type": "fragment" } }, "outputs": [], "source": [ "Xtrain, Xtest, ytrain, ytest = train_test_split(X, y, random_state=0)" ] }, { "cell_type": "code", "execution_count": 118, "metadata": { "ExecuteTime": { "end_time": "2018-05-15T13:15:12.845573Z", "start_time": "2018-05-15T13:15:12.836504Z" }, "slideshow": { "slide_type": "subslide" } }, "outputs": [], "source": [ "from sklearn.naive_bayes import GaussianNB\n", "model = GaussianNB()\n", "model.fit(Xtrain, ytrain)\n", "y_model = model.predict(Xtest)" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "Now that we have predicted our model, we can gauge its accuracy by comparing the true values of the test set to the predictions:" ] }, { "cell_type": "code", "execution_count": 119, "metadata": { "ExecuteTime": { "end_time": "2018-05-15T13:15:16.281720Z", "start_time": "2018-05-15T13:15:16.276351Z" }, "slideshow": { "slide_type": "fragment" } }, "outputs": [ { "data": { "text/plain": [ "0.8333333333333334" ] }, "execution_count": 119, "metadata": {}, "output_type": "execute_result" } ], "source": [ "from sklearn.metrics import accuracy_score\n", "accuracy_score(ytest, y_model)" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "With even this extremely simple model, we find about 80% accuracy for classification of the digits!\n", "\n", "However, this single number doesn't tell us *where* we've gone wrong\n", "- one nice way to do this is to use the *confusion matrix*, \n", " - which we can compute with Scikit-Learn and plot with Seaborn:" ] }, { "cell_type": "code", "execution_count": 123, "metadata": { "ExecuteTime": { "end_time": "2018-05-15T13:15:41.399490Z", "start_time": "2018-05-15T13:15:41.036420Z" }, "slideshow": { "slide_type": "subslide" } }, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "from sklearn.metrics import confusion_matrix\n", "sns.set_context(\"notebook\", font_scale=1.7)\n", "\n", "mat = confusion_matrix(ytest, y_model)\n", "\n", "sns.heatmap(mat, square=True, annot=True, cbar=False)\n", "plt.xlabel('predicted value')\n", "plt.ylabel('true value');" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "This shows us where the mis-labeled points tend to be: \n", "- a large number of twos here are mis-classified as either ones or eights.\n", "\n", "Another way to gain intuition into the characteristics of the model:\n", "- to plot the inputs again, with their predicted labels.\n", "- using green for correct labels, and red for incorrect labels:" ] }, { "cell_type": "code", "execution_count": 124, "metadata": { "ExecuteTime": { "end_time": "2018-05-15T13:17:10.256934Z", "start_time": "2018-05-15T13:17:07.529831Z" }, "slideshow": { "slide_type": "subslide" } }, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAcsAAAHDCAYAAABLUTEqAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzsnXd8VGX2/99pk97pIAlSpFcBkRZAQMSVoq6rohR3dXVXMbJi21XAL8KqS1lZLCtLEXdVQIpShQBrQKRIUYq0JJRAaAkkIXUyvz/GBMKcZG7IZO5kf+f9evkKPvfOzGee+9znmXvP557jZbPZbCiKoiiKUibeZgtQFEVRFE9HF0tFURRFcYIuloqiKIriBF0sFUVRFMUJulgqiqIoihN0sVQURVEUJ/i64k2GDh3q0LZ8+XKHtvDwcIe2TZs2ObS1b9/eFbJKyMjIcGiLjY11aIuIiHBok/RJr3U1ycnJDm1SP+/du9ehbeTIkQ5t8+bNc4WsCvP88887tEn9J+1nFnv27HFoGzVqlEObpFnazx1I55s0Xnr37u3QJo0Nd4xxCan/pOMhtbkL6dycMGGCQ5ukUZpjZsyY4dDm6jlQ6ldpbouLi3Nok/RJ38PVSJ9hdN6WqOw41ytLRVEURXGCLpaKoiiK4gRdLBVFURTFCbpYKoqiKIoTKmzwkYwEUptkJJACsUYD+pVh2bJlhrRIgXsJo4H7ypgkpH6R9I0dO9ahbebMmQ5tkgHBLBOHZCzwJIOPZGiQ+l4yQ5iFZJ4bMmSIQ5t0rkpjwyxDmHRuSSY26XtI37cqkMaqUUPexIkTHdpcbXKU3k9qk467NPalseDq81XSd/nyZYc2aSxIa400f1Z2vtMrS0VRFEVxgi6WiqIoiuIEXSwVRVEUxQm6WCqKoiiKEyps8ImJiXFoa9eunUObZKrZvHmzQ5uUbcbVSEaMlJQUhzajxh0peOxqQ4QUjJa+hxRolww+UgDdHdlmpH6RxoZZSMYMSZ9ZmZyMIo0NqU3SLGW4MgvJwCEhGZrchdRfUl8b7VdX9780pqXjLp3/Zo0Fo2a5jRs33vRrK4teWSqKoiiKE3SxVBRFURQn6GKpKIqiKE7QxVJRFEVRnFBhg4+UWcJoxp3p06c7tLkjOCsFt6VsH1LGDSngLQXGXV1SR8qkIeFJ5gzJWCBl4XB1X1UGSbNk6vIkzRLS96hMRiqjJpHKUJlMXdJr3WX0MFruStIoGSQ7dOjgClklSPPY6NGjHdqk7yGZ24yOI1cjzdFmHne9slQURVEUJ+hiqSiKoihO0MVSURRFUZygi6WiKIqiOKHCBh+jSMYTKYOPZPpxB1JQ2GjZH6Pmm8ogmUwkpAw+UnYTdwTBjWZfMVoGSDIquKPvJSRzi5QFymg2K1ebZaS+Mno8pCxa7jD4SKYYo5hlOgHZGGM0G5Z0Hu7evduhrTLlxqS5QxqD0piR5g6pr91heJP6Sjr/3VXeT68sFUVRFMUJulgqiqIoihMqfBs2LSuNSZsn8fWRrzmTeYYw/zB6xfTijd5v0K6O4y0oT6DAWsDUxKks2LeAlIwUIgIiuKXgFgYHDCbKO8pseSLVsZ8BtmZs5avzX5GUk4S/tz+MADYCp81WVjabz29m8anFHMs+hr+3P7n9c4nYHYH/BX+zpZXCa6JX2RuH2f/UPlibOofquEfQTZKdn03r91tTN6cuvw78tdlyyuRSziUmbZ7E0kNLSctKIyYihkfbPMq4buMItgSbLU9Emuv6N+5P78Le1PavbbY8kT8n/JnJ304WtzXwacC4kHFuViRTocUyNTOVrh935dSVU/SJ7cODLR8kOSOZZYeWsfroataNWEfPmJ5VpfWmeXDRgyz/eTltarXh2S7PkpSRxLJDyzhceJj4kHiPWzCraz8vSlvEp2c+paZfTQZEDyDLmkVCTAKMAeYDJ8xW6MjClIXMSZ5Dbf/aDK4zmKzCLNbmrOXsPWepvaY2AecCzJZYwhu93xDbC6wFTNk8BZuXjeALnjmJF1NYVMgjXz5CckYydS11zZZTJhevXqTbnG4cuXSENrXaMLz5cFIupzBh0wS+OvwV3zz2DREBxnwF7kSa6/7z439Y6buSt5u+7ZEL5r60ffh4+fDnXn8u1b5t2zbCvMNMUuWIl81msxndeczyMczdM5e373qbF7u/WNK+/vh6Bi4cSNOophz64yHAeHaIqs5Ak5CUQL8F/YiLjWP9Y+vx8fYBILB3ILl9c/Hb70fQuiDAePmiqjYXlNXPC7cuZOT6kcSGxrJh2AZADrRLfV/VxpjdZ3Zz+0e30zioMRMbTyTIx96nQ58ZCr8FzgIf2feVTARSkF7qe6ntZsujFWtuEdGCD+78gBC/EAA6DO7goHns2LEOr58wYYJDm5RhRNqvMuaWGxn/zXje2foOzzR/ht/d9ruSdumYS5/rLrPM+ezzPLT4ITYm28sstc5rTb+r/UrtI5WXk5BMSa4sk/fE8if4155/0dbWliEMwfuXiNVBDvKF1xeMbjOaaX2nlexv1JAn7SeNmZsxU5U11/1j+z/44+o/EpkSScMfGpbsb9QYI+03f/58h7abNSXFzIgh2C+YA3844HRfqV+k86sqyg8ajlnabDaWHFxCzaCajLuz9GXxXbfeRe+Y3vx88WeOXTrmcpGVYfvp7QA80vqRksEDYPnJAjaw1rGaJU2kvH7uUa8HXWt35fiV46RccXRimsms7bMooohnbnmmZKEEIBX4FjiJx0XIizX/pf1fShZKwKM130jiiUTe3fouLcJbMKbZGLPllMmcH+bQ4h8t2Ji8kQGNB5gtp1ysRVYWHVhEsF8wgxhUslACtKAFvW/pzSf7PyE9N91ElY6UNdeN6TAGbHA16qpZ0sokIzeDE5dP0LZ2W7OlOMXwbdiCogImxk3E4mPB28txBgn0CwQgKz/LdepcQHRgNADJGcml2otCisALvHLKiQOZgLN+DvCx3xbMLsx2t7RyWXlkJfX869EosJHjRsd6rR7ByiMraRjckNvCb3Pc6KGar8dms/H8muexYeOVtq/g4+Xj/EUm8d729wgPCOeTYZ8Q6BfIumPrzJZUJueyz5GZn0mX+l0IOOV4G751jdZsPrmZ7We2M7DRQBMUypQ116VmpoIX+OZV2ZOCN82+tH0A/1uLpcXHwvN3yJftF65eIPFEIv4+/jSJauIyca5geIvhvJbwGjO/n0m7Ou24p+k9nLh8gpy7c8AG/j94lomjvH6+lHuJned2YvG2EBsa615h5XA++zxp2WncEX4HaXlpLDyzkD2Ze8i35cNjwHrgjNkqS1OsuW/dvpzOPs17B9/j+/Pfk2vN9VjNN/Lpj5+y68wu7m9xP22i2pgtp1ze6f8OvWN7Y/GxsCl5k9lyysXf1z4n5BXmiduv5F8B8Li7O2XNdb9d8VuwQc2jNc2W6MDes/ZngTNyMxj06SC2n95OgbWA7g27MzFuIl3qdzFZ4TVccpNp3LpxXMm7wmNtH/M4l1h0UDRbxmyhVa1WPLT4IUKnhNJqdiusNa0EfRWE3zE/syUaZvLOyWQWZDKs8TCC/IKcv8BNpGamAnC58DLjDo8jOTeZ3pG96RTaCRphN/g0LPct3E6x5vS8dB7Z/AhHrhzhngb30KN2D4/VfCPvbH0HgJe6v2SyEuf0b9wfi4/FbBmGiAqMonFkY/af3885zpXalk8+G1LsfoEreVfMkFcmZc11e87uIfb7WMLPOHoFzKb4ynLad9Ow+Fj4bYff0rdRX9YdW0ePf/Vg5eGVJiu8RqWvy1/f+DoL9i7AP9efA38/QNy0OEDO1iMhZZGQAvVGA+g3klOQw7ivx7Hj9A5ur3U77Wu058zVM6w8tpKcvjl4ZXnhm2bvBskQ0adPH4c2KSDv6owWN5oupu2expfHvsTriher4lexJm8NIJfAkpCC4FKw/GYC48W33g9mH6RPnT78tfNf8fO2/wjp8OsO8CgwBJgF2IwbGowaVG6GYs27L+1maPOhfP7A5yWTee0etTnX/xy+D/hSb0k9vPAynJFq7ty5Dm03O3bLIyEpgX1p+4iLjaNz/c5Q33EfyVAnGTPMIj8v/6YNfq40SEmM7z6ep75+iuWBy3m+6fO0Dm/N2dyzzD42mzPnzoAfzPrHLBY+uRCQ5zGpTTpfXZUZKacgh9c3vc6O0ztoH9WeNpFtSMtJI+FMAimtUwg6HlQy14HxUmPS3JaUlOQSzT7ePsSEx/BszLN0iuwEwKA6g+jl34sX973Io4se5cMWH5aEnyTcZVC76cXSZrPxp3V/Ytq2afjl+9Fmbxv8Cj3vKi1+bTxfJX/Fc+2eI759fEn75smbyfp1FleHXiX0X6F4FXhW7LIYm83G5J2TmXNgDl5XvQheFox3nme5Tq6Prb7U9qWShRKAI8BRoAlQF7t5xgO4XvN7g94rddUTeCqQgFMB5DbIJb9Gvsc9bwkwb888AJ7q9JS5Qv5HebLTkyRnJPPXxL8y/sfxJe2twloRuTOSS90u4V3oWedh/Np4PvvpM5687Umebv50SfuBjAM8uv5Rj5zrPrj3A8Bxke4Y2ZFekb3YlL6JvVl76Rre1QR1pbmpxTK3MJfHlj7G4gOLqRdaj3rr6xF81bNuvwIU2YqYt2cetQJr8Vzb50pt8znng2WfhfxO+RTcWoDlZ8+7RZRnzSP+23hWp6ymdmBtsudn43PJ80wc4QH22ztR/lHUDhSe4zqDfbGMwmMWy2LNtYJr0SCsgcN2yyULuQ1yKQwt9LjFsrCokK8Pf02gbyD3NrvXbDn/s7zV7y1a5LVgR/oO8ovyaRrSlI4RHXlg7QMA+OR4zrlYPNfVDanLk7c9WWpby4iWHj/XSTQOasym9E2k5aWZLQW4icXycu5l7v3PvSSeSKRFjRasGbGGx1c8XhXaKk1aVhp51jwahTUqZaUuxueiva0orMjd0pxyJf8KT2x4gp3ndtIkvAnz7prH4LcGmy1LpHFkY3y9fSksKpR3KO76ArdJckqx5gJrGaJ+uWjwKvScX+HFbDmxhfTcdIY1H0aIJcT5C5Sb5pagW7gl6JZSbXk17MYfv3TPuZNWPNc1i24muqI9ca4rsBaw5+werDb58b28Ins/+3t7xo/VCt1HyC3MZfC/B5N4IpFuDbqROCaRhuGe64KICozC4mMh6UoSRTbHQWKNtB8k72zPup2SW5hbslB2rNmRRYMWUT9ECEp5CP6+/nSp34UrBVfYn77fcYd6v/z1jB+IwDXN6bnp7Ezd6bC9eEK0pHver/CtJ7cCEBcbZ66Q/2EeXvIwNd+paXdHX0euNZecW3LwyfbxqMWyeK47fPFwtZnr8q35dJvTjYELB4o/tA9k2ZMUNAnyjCcsKnRl+dI3L7Hl5Ba61O/CN499U+J8lYL0UqkiKeAtmRAiIyMd2t54wzHVl2RauR5/X3+GNh/KF/u/YO6RuYzrcu0hf2u0lfx2+ZAHvsft3SCZeXr37u3Q5upSRTfy0jcvsfPcTjrV7sTy+5cT7Bdc5udKZZiMZkGRkIL5RsxLv+/0e7ae3MqHyR+ytsfakuduR709inlX59HEpwl/eOMPgFx6Ryp3JfW9NF5ulmLNL6x9gbUjrmn+4+w/8uKOF+lcozMfLbSn8DGakUrKduJsnFaUnWfsi/v1tnrJFCf1c2VKP7maO++8k+n9SpfoGzZsmMN+kplE6lOjbUZoUaMFn/30GYdDDvNcV3sIx1pkZeSykRQFFDGx60Qef+baHTVpfEjnYUxMzE3pccb1c92arDW82vPVkm37z+3H2sGKT6EPHYI64Nfevsh7eTneNTGaXcsVZbGCLcEMbjaYFT+vYF3eOl7tdk3z54c+54fMH/BJ9WHS9Ekl7ZJByh3lB6ECi2VKRgqzd84GoGXNliW2dYCzzc+W/DsqOQpLruf8Gp8xcAbfn/ye//vu/9iQsoHOdTtzJusMWY9kgRcErgnEO9dzfm1d38+3Rd3Ge7veK9nmyf38WLvHWHV0FZ/99Blt3m/DkNuGcPLKSRZfXUywVzAPBj5otkQHytK85MASIiwRvNr2VedvYgJHLx0FoFl0M5OV/O8Sf0c88/fOJ35tPIknEmkU0YiE5AR2pu7k7oZ380izR8yW6MCMgTPYcXoHryW8xpqja+jWoBunM0+z5OASiryLaH6wuceZMGfePZPtp7fzzvZ3+C71O9rXas/BiwfZkLIBr2wvAtcEmi2xBMOLZUJSQsmlcrETr4QW1/4ZmhbqUZN43dC6JPwmgb/t+Bsrj63k/d3vE+wXjG+KL/7b/fE961lZLa7v538f/HfpjR7czwCfDv+UXg178eGuD5m9czZh/mF08OvAIP9B1PCpYbY8EUnzwAYDebr509wSfIvzNzCB89nnCfQNJCrQswoA/C8R6h/KljFbeG3Da6xPWs+qI6toGt2UWYNmMbDGQHy9PWveAPtct+N3O5j87WSWHVrG9G3TCbGE0P/W/pz6zynCrnhOUvJiYiNi2fXkLl5d9yrrktaxLXUbtYJqMar1KJY8t8SjbhsbPuKjO4xmdIfR4jZ3VM2uDFGBUUzuNZnJva6VganqW6k3S3E/S7e23XW74Wbx9vLm6c5P83Tna7b1qk7gXlkkzdJzoJ5E6jgPsRTfJHGxcdjesFV5EYXKUiekDnOGzHFod9dzfTdDdFA00wZOY9rAaaXa4z6MM0eQAeqF1mNGP8d5Ymn2UhPUlI3nLNuKoiiK4qFUqESXoiiKovz/iF5ZKoqiKIoTdLFUFEVRFCfoYqkoiqIoTqiw/1lyCkouTanSgvSQrlkPSUvZ9aWkCRLSw+jucKpK7kHpoWvpAXWjlT7cgdRX0nfzJFeq1M/SGJLGRlVUHTGC9ND59OnTHdpc8YB5VSK57aU+repKJMVI41KaO6RxLrnD3TE+pPEraTFaxUiqrnMzFYvKQ+pnaaxK/SftV9k5Wq8sFUVRFMUJulgqiqIoihN0sVQURVEUJ+hiqSiKoihOqLDBRwpkGw1Qjxw50qHNHSmvJMPL6NFy6r4bkSpfSIFsd6TA2rx5s0ObZJqSTBxmmUyWL1/u0CZ9D6mqjFlIY1IyQ3hSmkdpjEt4kmYJ6dySKutUVfWOG5HOa2kOlMxyktlLMiFJRpbKnK+S5okTJzq0SeZKaV50hwnTqOHSqClRqiBV2fw7emWpKIqiKE7QxVJRFEVRnKCLpaIoiqI4QRdLRVEURXFChQ0+UgBeypYgBbel4LEUjHZ1ZhlJixS0lgLKUuYLszLfjB071qFNMsZ4UkYWo8Ydo9mT3IE0XqTMJmZlY5GQNIeHhzu0mVUTVTJNScdcMn9JuMuoJBlPjGYPMprpRxpH0rxjFKOmSbPMPBJGM18ZzeoljfPKrjV6ZakoiqIoTtDFUlEURVGcoIuloiiKojhBF0tFURRFcUKFDT5GA8+eXl5JCrQbzcIjBaPdgRS4lwLU0veVAt5mlRVr166dQ5snZZYxapAwy+glIWn2pD6VjIGSmUfKmCXt567vFhkZ6dBm1MQlaXSH7pSUFEP7SXOgdG66Yy6XTE5Gs/pI85ikubLZ4vTKUlEURVGcoIuloiiKojhBF0tFURRFcYIuloqiKIrihAobfCSkwKnRrDnuMElIQXXpcyUjgdHMEmYhZeGQgvRSCScp04+rs/8YNSVJAf7KlOgxiqQvPj7e0GulcSWVTHNHVhTJ0CBlHfLy8nJok7IsVSaDjIR0LKWxJn0P6bx0l7lKyoJUmXJ80nhz9XyydOlShzZpTjB6/ktl9swa01JfSfqkMoWVNVfplaWiKIqiOEEXS0VRFEVxQoVvw6ZlpTFp8yS+PvI1ZzLPEOYfRrd63RjfdTxtarapCo2VpsBawNTEqSzYt4CUjBQiAiLwauFFo6RGBOQGmC2vTJYcWML0bdPZm7aXQN9Asodl47/VH980l9w9rxKqo+YVR1Ywe/ds9l/YT4BvAIwANgKnzVZWNtWxn4vJK8rjz6l/plt+N15o9oLZcsqkyFbE0tNL+erMV6TmpBLqG0rPGj0p8ivCu8AzrzP+nPBn5jea77ihEQSmB9JsUzP3izKANEf3b9yfyX0nExsRa7Y8oIJXlqmZqdz+z9uZvXM2jSMb81zX54iLjWP18dUM+HwAW09vrSqdleLBRQ/y+qbXCfQN5Nkuz9KjYQ/O1T7HD51+IDcg12x5Im99+xYPLHqAk1dO8ruOv2Nws8EUNigk+6FsCusVmi1PpDpq/tv2vzFy1UhOZ53m8daPMyB2AMQAY4CGZquTqY79XIzVZuWDCx9wwXrBbClOGbt6LH8/+ncs3haG1x/OrcG3sjR1KWcHnaXIp8hseSL70vbhZfOiXXq7Uv/VPlib6ORos+WViTRH/+fH/9Dln11Izkg2Wx4AXjabzWZ05zHLxzB3z1zevuttXuz+Ykn78D8NZ1nIMiKKInj8yuOAbHRISkpyaJMC9a4MgickJdBvQT96NOjBsmHL8PH2ASByQCQMBvYAv3h4pCC4lB1CMqO4kt1ndnP7P2+nU91OrH98PWH+YQDUvb0uZwefxXLJQt2v6gKyAUkKZEvBcimYL2XNqIzmZnHNONL7CIGXr/2q3bt3r8Prpb6X9tu9e7dD280G7os1t6/VnqXDlpZo7vtYX3Z33E1IVgiddnUCZJPJ3LlzHdqMZou62TFUVj93/FVHQ5qlfpYMK9LYqKzp53z2eR5a/BAbkzcC8FSnp/jg3g9K7SMZ0SQTm1ED4c2y/fR2un7clbti7uLzIZ/j7WW/rpjy3RTe3v42EbsiCN93zfwj6ZbmrA4dOji0uXJMx8yIwcfqw5f9vizVLs2zkj5prEpaXGkAK56j42LjWP/Y+pI5+qmPn+Kj0x/RJ7IPY2PGlqlv5MiRDm3SPCa1VUmJLpvNxpKDS6gZVJNxd44rta1hYUPqF9Yn3SedDO/KpRRyNdtPbwfggdseKDkIAOwGbEB9U2SVy6ztsyiyFfHPX/2zZDIE8L/gT/i+cPzP+WPzMvwbxy2UpTkoI4jah2sTdCnIYzXP6DejlOawzDAanmhI2JUwirw86wqirH72ZM0Ac36YQ4t/tGBj8kYGNB5gthynvL/zfQBevuPlkoUS4IXOL+Cd601WsyyzpJVJRm4GJy6foGl4U7OlVIjiOfqR1o+UmqP7RffDCy+OXD1ilrRSGA5wFBQVMDFuIhYfS6nBc+2N7G9V4FXgOnUuIDrQfuvhxJUTpTeEAl7AVbdLcsrKIytpGtWUdnUcrwIidnvOYyvXU57mOgfrmKDIOSuPrKRxRGMx1t4oqZEJipxTXj97qmaA97a/R3hAOJ8M+4RAv0DWHVtntqRy2XJiC6GWUDrULn0l6O/rj/85f3Ia5lAYXIhvtufEiPel7QOgWZhnxiXLoniOvvF266WCS9iwEeYbJrzK/Ri+srT4WHj+jud5pvMzDttyvHJI9U3Fx+ZDhNWzJvPhLYZTI7AGH+75kKWHl5KVn8Whi4dgOPYry+/MVlia89nnSctOo3Wt1iSlJ/Hwkoep8XYNgt8KJm1AGnnReWZLdKA8zcfuPMbVCM/7RVKsuUV0C1Iup/DE6ido/GFj6v+jPvva7SMzJNNsiQ6U18+eqrmYd/q/w8E/HGRQ00FmS3GKtcjK0UtHaRTZSL4wyPzlwiDcsy4M9p61hy0yCzL5w3d/IG5VHD1W9uAP3/2BXWd3mayubIa3GE7NoJrM/H4mX+z/gqz8LA6cP8CMlBl44cV9Ne8zWyLgokdHvg38lnyvfJrnN8cPP1e8pcuIDopmzYNraB7VnDGrx3DL+7fQbWE3qAN8DhwyW2FpUjNTATh/9TydPurEvrR9jGg7gnua3kNu3VzS7kkjt7ZnmZLK05xVM4ujvY6SFe1Zt62KNV/IuUDcf+I4cOEAv27+a/rH9ic9Ip09HfeQEe5ZIYXy+tlTNRfTv3F/LD4Ws2UY4nLeZWzYiAxwrDgC4J1vnzaL/DzrdnfxleXCYwvx8/ZjWMwwOtfozLZz2xi0aBBrk9aarFAmOiiaLWO20KpWKx5a/BChU0JpNbsVSTlJvBT7EndE3GG2ROAmHh25MUj6r6R/cdD/IGTA/g/3sz9nPyBnvpCC4JKZx+h+RsgpyOHdH97lh7Qf6H5Ld+5ocAcnr5xk0Y+L8BniQ6M6jQhKDwLkQLakpSoNPln59kUl8UQiQ5sP5fMHPi+ZZOIvxDPj3Axs99l4vd7reHt5i/qkoLVk4nBV6Z3yND+Z/CT/vPpPLsdd5pmQZ/D28hYz5BjVbNTQZFTzttRtDprnJs5lzIYxXO59mS+HfYm3lzeNGhm7xSkZwlw1hsrr5zb/acNP7X7icPPDdP6+M144ZuoBuVSWUX2uzuojYdRwIZliXGXwyc7PBuy3XCUTTKf2nUgggZ59etKWtoDxvpFKkLmqbJePtw8x4THMuW8O/W7tV9KekJTAgE8G8Oz6Z0kam0SwJRiQj7uUDcfVWb1uJKcgh9c3vc6O0zsc5ugZh2fQJakLkVftP1yMZieTjofR/cripq8sbTYbs4/N5pMTn0A28CmQc7PvVnXEr43ns58+4y+9/kLimETeHfAunz/wOU02N8HqZ+V4t+NYfaxmyyzh+ts+7w16r9Sv8XaB7Wgd0JpzhedIzk82QZ1MeZpb+rWkuW9zLhRd4JT1lBnyRMrT3KdBH3rV60VyZjI/XvzRDHki5WmOvhRN5MVIcoJyyAz13Nux1YEAX/uz1/nWfHF7IfbHczztLtoH935A8vPJpRZKgL6N+vJIm0c4f/U864+vN0ld2ZQ1R/c63IsC3wK23bqNQm/zH4m6qcUyvyifiQcmsujUImpYasA84LxrhbmCIlsR8/bMo25IXV7v/XqpbUEZQUQnR2P1t3Kl7hWTFDoSHmC/Iq8VXIsGYQ0ctje02B/+O1/oOR3uTHN9H7vl+EKR5zxb50xzq6hWAJzIPOGwzSycaQ7JCgEgN9CzbtNXN8IDwvH28uZyrmNuXYA87L6BADw3ocmNdKprf5woKcPx8T0zKW+OjsiJIPZCLPl++ZwNP2uSwmtUeLHMKsziT3v/xOYLm4kJimFWh1keuVCCPdtQnjUDQFmqAAAgAElEQVSPZtHNSj828gsBV375BRkk/4I0g8aRjfH19qXAKpsHrNivgi1enhP/carZVv00FxTZ2wN8PGdCdKa5+NEcb6tnZpepLlh8LDSKaERSRhLSY+jppANQgxrullYmBdYCdpzewbZT28TtVwvsJrtA30B3ynKKszk6NCcUgKsW802CFTqrcgtzeeXHV/jxyo+0CmvF39v/ndoBtatKW6WJCozC4mPh8MXDFNkcg/F5IfZfiH65nnM7xd/Xny71u5Cem87O1J0O25PzkgG4xXKLm5WVjTPNJ60nAajnU8/d0srEmeZ9F+1miRZRLdwtrUycac4Ms99+DckOcbe0/zl6xvQkIzeDvWmlE2PkW/M5yUkibBGEEmqSOkfyrfl0m9ONgQsHij+m/nvivwB0rt/Z3dLKxdkcnRVgj9MHFJj/o7VCBp+XvnmJn678ROuI1nzU/aNyf6VI5YEmTpzo0CYZgebPF3Ib3gT+vv4MbT6UL/Z/wdTEqbza89WSbX/5+194/L+PE+wVzII/LyDCEiGaM8zg951+z9aTW3lh7QusHbGWQD97P4d2CeXnRT/TJ7YPfxr5J0A2bBg1j7jK4FOe5qjuURxbdYyeDXrywv32PKCScUcKvkvHQ8rg4WrNu67uYnvadvrE9qFH6x6AnK1Hyiwj4crMMmVpHvPOGEauGknPBj1Z8dkKQB4bkrlKOgel17oD6ZhL+lw5diVGtRvFvD3zeHXDq6x4eAW+3vapcsq3U8jxyuG5Ts8xqvWokv2NlpirKrNMsCWYwc0Gs+LnFbz53zeZ1GdSybaF+xay5uga6hbWZcFfF7CABYCcZU0q11aV82J5c/QzbzzD8FXDCfEKYc74OUQGRIrjUio7JyGV7aoIhhfLlIwUZu+cDcCtobcy/+h1C1rcdTv+AHhOCJAZA2ew4/QOXkt4jTVH19CtQTdOZ55m8f7FWG1W/q/T/xFh8axnQx9r9xirjq7is58+o837bRhy2xBOXjnJkoNLiA6M5v3B75st0YHyNEcFRPG3vn8zW6ID2s9KWfSO7c3IdiOZv3c+t390O3c3uZsfz/3IqiOraFejHSObO6ZYM5uZd89k++ntvPnfN/n2xLd0qtuJ/ef3s+boGoKKghiQ7ZmZk8qcow8sxlpkZXrP6WU+xuNODC+WCUkJFBbZHUkrTq4ovTHuun8fxaMWy7qhddnxux1M/nYyyw4tY/q26YRYQuhaqytPNH2CtlFtzZYo8unwT+nVsBcf7vqQ2TtnE+Yfxm9a/4ZJcZNoHNXYbHkikubhzYbz6h2v0ijCM7PLaD8rZfHxfR/TsmZL5uyew/Rt06kXWo/4O+IZ1WiUx8X+AGIjYtn15C4mbJrAyiMrSTyRSO3g2jzV6SmsG6wE24LNlihS1hzdo24P/tD2D3So6ZhP1wwML5ajO4xmdIfR4u0PKTmwJxEdFM20gdOYNnBaSVtV38apLN5e3jzd+Wme7vy02VIMI2m+2edj3YX2s/uJi43D9oZn5QmW8PX2ZXz38YzvPr5UuxRG8BTqhdbjo1995ND+/PqqfVayskhztKf1s9rmFEVRFMUJFSrRpSiKoij/P6JXloqiKIriBF0sFUVRFMUJulgqiqIoihNcUrlUylQvPdwqIT24647qBpLTqjIPYkuVA6QHkI1WVZCQ9EkJHKS+d0efSkiuY+kh5xur2YDrqjFUFGlMSmNcqiAhsXnzZoc26QHpyjywLh1fKQmIxNixYx3aqrKyTnlIfS+NA3eNZ8llLB0nqRqOhDvGuTS3SZqlMS0lgJDGgjuSVkj6Ro50fL5VqgzjqupE16NXloqiKIriBF0sFUVRFMUJulgqiqIoihN0sVQURVEUJ7gkKYGU9V2qtCAFnvfu3evQJlV4cHVAWQp4S5UvpKCwFPSXvocUZDZagUIKqkv6jAbfpe9bVRUQrkcyNKWkpDi0paenO7RJgXtXYzR9o2SCkfRJphrJlCAdS6NI4y8y0jHRtGTWkMxVZhlopD6QxqR0jCpjlKsIRs85o+erNAdKpp/KIJmkpM+Q5glpzpLGuavThUr9Io1VaXwYNVdWVrNeWSqKoiiKE3SxVBRFURQn6GKpKIqiKE7QxVJRFEVRnFDhDD5G6+aNHj3aoa1du3YObVIQ1x21JqXPkILCktFBapOCzEbNPJVBMjpI38Md9Q6lfpHMPJKBSzIRSO8nmW8q089S/0mGHKMGjpiYGIc2d5hlpM+Vzq3KGItcjdSnRrNeSYYVo+d0RZA+W2ozOo8ZzfRTGYwed+mck/pVej93IM1ZjRo5Fje/fPmyQ5tkuJTeryImQr2yVBRFURQn6GKpKIqiKE7QxVJRFEVRnKCLpaIoiqI4wSUZfKQg+syZMw29Vsoy4mozitGMJ7t37zb0fkaD+a7OMiIZIuLj4w29tjLZhIwi9YuUmUNqk8rxSJlI3FF+TBovRrO2SEj9Uhmjh9FxIBmVJC1mlVuSjm9SUpKh95NeK323ypYaM5pZRjKySUhGLMlU444MRUbnRbNK/knHTvpcyeAjIc3vFSnbpVeWiqIoiuIEXSwVRVEUxQm6WCqKoiiKE3SxVBRFURQnuMTgIxkE5s+f79AmBbclpKB6ZZAC2dJnSMFeKZgvmQtcXe5K0ix9rtHvZrSUV2UwanySsmZIBgkp45M7sjtJ32Pz5s0ObZI5TTIgSMetMgYO6fhKx1LSLOHqEmISRstBSeeg0TFkdL/KIp1zRj9bOg/dYbqSNEufYbSEmDvK50kYne+MZniqyDjXK0tFURRFcYIuloqiKIrihAonUi+wFjA1cSoL9i0gJSOFiIAIQmqG0DG9IyGFIVWh8abxmujldJ+nbnuK3zf/vRvUVIwCawEzds7gs4OfcTLzJOH+4Xi38KZRUiMCcgPMlmeIk7ec5HKfywSuDcRywGK2nFKUOzb62P/EJMUQmxzrFj0V4d0f3uUfP/6jdOMvmkOuhNBpVyf3i6ogP4X/xOV4zxwbxVzKucSkzZNYemgpaVlpxETE8GibR3mi5RME+wWbLU8kLSuNSZsn8fWRrzmTeYYw/zB6xfQivmM8bWq2MVtemZTV1+O6jSPY4hl9XeHF8sFFD7L85+W0qdWGZ7s8S1JGEkuzl3Im8AyDUwd71IL5Rm/7w7S5ubml2guKCnhv13sU2YroGN3RDGlOGbVqFKuOr6JldEuebPckKVdS+Prq16RHpdNxV0ePXzCvBl4luVGy2TLKpHhs3EiBtYAp/52CDRvhGY4xSU/gUPohfLx8+GPbP5a0Fcf/LHmeufBcz2W/y+yJqPrYc2W4ePUi3eZ048ilI7Sp1YbhzYeTcjmFCZsmsOzAMpYOX0q4v2eNj9TMVLp+3JVTV07RJ7YPD7Z8kOSMZJYdWsbqI6tZMmwJd9a/02yZDpTX118d/opvHvuGiABzYqTXU6HFMiEpgeU/L6d1SGsm1J2AT64PvQJ6sXrjanL75rLiygqC1gUBsHHjRofXS4FsLy/HX/hS0PVmAt4T4iaI7eO/GU9BUQFv9nmTJ3s9WdIuBYWlgLKrzTw3kpCUwKrjq7i9xu18cOcH+Hj5APD137+mYHAB3wd+D2vt+0oZT6SAt1HNlTEWFE/YRbYiBi0aRNGZIgCaN29Ow1oNS/YzatKZOHHiTWtxRnljw+ZtY0zsGB7r81hJu9R/Ul9V9dgAOJJ5hGbRzZgx9JpZ5Pp/FyONXencckc2luJzv8hWxJS0KVjzrADk5OSQczmnZD/J4CMZlSQDoStNJ+O/Gc+RS0cY0XYEc4fMxdfbPlV+efBL7v/ifqbumMr7975fsr/RMlZSX0vj6GbOwz8n/JlTV07x9l1v82L3F0va1x9fz8BPBvLkV0/yjxbX7khIWZUk01VVm3mc9fUr618p6WtJizRmjJqrKkKFYpbbT28HoFdkr5IJHMDykwVsYK1jrZQYd5B4IpF3t75Lx7odeaXHK2bLESnu50H1B5XqZ3YDNqC+KbIM8/7u99lxZgf9Y/ubLaVCFI+NpiFNeaThI2bLEcnIzeDE5RO0rd3WbCk3xTeZ33As7xhtAz1Xv7XIyqIDiwj2C2bWoFklkzfA8BbD6deoHx/v/phLOZdMVFkam83GkoNLqBlUk3F3jiu17a5b76JVSCtO553mTN4ZkxTKVKe+rtBiGR0YDcC5/HOl2otCisALvHKcxwjNxGaz8fya57FhY/Y9s/Hx9nH+IhMo7ufUq6mlN4QCXsBVt0syzLH0Y0z+bjJj2o6hW71uZssxzPVj4/mmz5f+keJB7EvbB1AtF8uzBWf5MuNL+oT0oZl/M7PllMm57HNk5mfSqlYrwgMcb7W2q92OwqJCvjv5nQnqZAqKCpgYN5EJcRPw9nKc1i3e9tvzuUW5DtvMpDr1dYUWy+EthhPuG87X578mMT2RHGsOJ3NPknN3DtjA/wf/qtLpEj798VN2ndnF/S3up2uDrmbLKZPhLYYTaYnk38f/zbrT67haeJVjV47BcOxXluaPGxGbzcYf1/+R6MBo3uguxwQ9levHRsuwlmbLKZO9Z+0V4DNyMxj06SCi344mbEoYgz4dVHJHwhOx2Wz86+K/CPUO5YHIB8yWUy7+vvZ5LK8wT9x+Oc+euDspw1jSd3dg8bHw/B3P80znZxy2Xbh6gYNZB/Hz8qOupa4J6sqmOvV1xa4sg6KZ0nQKtwTcwrsp7/Lwjw/z7KFnsda0EvRVEH7H/KpKp0t4Z+s7ALzU/SWTlZRPdFA0c3vOpXFoY17a+RLdV3bngY0PQB3gc+CQ2QplPtzzIdtStzG973RCLaFmy6kQ1WVsFF9ZTvtuGhYfC7/t8Fv6NurLumPr6PGvHqw8vNJkhTLfZH7DkbwjjIweSaB3oNlyyiUqMIrGkY3Zf34/+8/tL7UtOz+btcfshoHLucaqXZjNuHXjuFp0lbjIOAJ8PMsYWJ36ukIGn5yCHOanzOfo1aM09W/Krf63cqnwEjsyd5DXL48wrzD8L9h/KUgmDqOZeaqiPE1CUgL70vYRFxtH5/qdxX0kc4bRwL0rySnI4e/7/s7+jP20DmtNy7CWnMs7x6azm+AeIBM4bd/XaCYiqU9dZaQCOJ5+nDe/e5OR7UbyQHv7lUNgoH1SHDBgAHfXubtkXymjjZRFxtUlxMrixrGx6cgmh3327t3r0GbG2PDx9iEmPIY5982h3639Stq7j+jOd02+Y/iC4fQ/0B/fIl9Gjx7t8HrpmFd1Oajj6cdZnrWcke1G8rehfwPgbOJZFm9YzNy5cxnVflS5+qTMX717964quQCM7z6ep75+ivs+u4/3B7/PnbfcSXJGMi+sfYFL2fb4WeqZ1HLNakbLP7m6JOH1c8K03dNYsG8BXle82DZ7G9vzrt19GDt2rMNrpaxZkvFR4mbPh7L6esjsIZzjHHjBhoQNFCQUGM5sJo1pKUuY9H5lGZoqtFjGr43n+6vfc1/4fQyLGFbS/spnr5B2Txrn+p+j/qL6eBd6Xq6DeXvmAfBUp6fMFWKA+LXxJJxP4PGYxxkde23C2/R/m2AM8AgwE8g3SeAN2Gw2nljxBKGWUKYNnGa2nApTncbGB/d+ILbXzKpJ/fT6nIo6xfnQ89S97Bm326rr2Hiy05MkZyTz1y1/ZeDCgSXt3Rp047mWzzH1x6ked5V2PTabjck7JzPnwByiA6LJW5CHd57nzctQdl83oAH96MdqVuOH+XctDS+WRbYi5u2ZR4RPBEPCS//i97/oT8ihEDJbZ5LTMIfg457xEGkxhUWFfH34awJ9A7m32b1myymX4n6OtkTzeMzjpTeeAXYC3YDbgB/dr09i9o7ZbErexOIHFxMVGGW2nApRncaGMyKuRnAq6hRXLZ7jAKvOY+Otfm8xqv0o1h5dS25hLh3rdqRvo768uvRVAGoE1DBZoUyeNY/4b+NZnbKa2oG1+WTAJzw45UGzZZWL1Nf/XfBfdrADgBDMf37f8GKZlpVGnjWPWP9Y0W3ll2Ff+QuDC12nzkVsObGF9Nx0hjUfRojF/E4vj+J+bh7SXHZkFhuRPeh56EUHFgHwwCLZuPHXn//KX3/+Ky/d9lKp27GeQHUaGwXWAvac3YPVZuWOBnc4bLd62x/d8inyHCevs7ExevloRi8fzdwhpW/HegrNopvRLLq0c3d/hj221ji0sRmSyuVy7mVGrBvBznM7aRLehHl3zaN+iIc/a/YLN/b1t3xLKvYnAmpRyyxZJRheLKMCo7D4WEgrTKPIVuSwYBaG2xdJnxzPOVGL2XpyKwBxsXHmCjFAcT+fyjkl9jPRv/zNdLu0MhnVfpTYt4knEtmQtIHu0d1pEtKEJiFN3C/OCdVpbORb8+k2pxvBlmAuvHgBP5/St6YuhlwE7FeYnoKzsTHktiG0r9Oe9nWMxffcxcNLHmb98fWkPJ9CkF9QSfvVgqt8m/YttQJq0TSsqYkKHcktzGXwvwez89xOOtbsyJx+c4jw95yxUBZl9XUBBRzmMKG20Oq1WPr7+jO0+VC+2P8FZ5uc5dWer5ZsO3jxIDOzZuKPP/H3xBPsHSxmqrh82dHRJBk7XG2c2HlmJwBd6ncpaTNqQHJ1qRxnXN/P23y3lern8e+OZ2bWTAD+8vhfCPYOFoPbUj9LJgmjgXtnlHVFMDVxKhuSNtCjRo9SV5SSvmHDhjm0uQNpbEhjVzKUGDVwuIpgSzCDmw1mxc8rePO/bzKpz6SSbScjT3Iu7ByRWZFE5NgnSGlsSMydO7dK9ILzsTG0+dBS+0h9LxnvqjqrTIsaLfjsp8/4+IePea7rc4D9Afonv3qSjPwMul7oysyZM0v2lzIKGcVVRraXvnmJLSe30K5GOz7p/0mphUfiev3ltRklPT3doc3IcSqrr7fX3k7OuRyea/Icferbkx9LWXik8SEZ8qR+rsg4qpDBZ8bAGew4vYPXEl5jzdE1dGvQjdOZp/ki6wusWBkROIJgb8+KVwIcvXQUwOF2iqdSXfu5OlLdxsbMu2ey/fR23vzvm3x74ls61e3E/vP7+SH2B/wL/OmY4pm5jqsb8XfEM3/vfOLXxpN4IpFGEY1ISE5gZ+pOGmY3pFmmZ42XlIwUZu+cDUDT8KZ8tP+jUttz77AnI7D8ZME7y7OMPmX29bmd9KzRk/vq3We2RKCCi2Xd0Lrs+N0OJn87mWWHljF923RCLCE0823GXf53EesbW0UyK8f57PME+gZWG4NBde3n6kh1GxuxEbHsenIXEzZNYOWRlSSeSKR2cG1iL8Ry25nbCCj0XIdmdSLUP5QtY7bw2obXWJ+0nlVHVtE0uimzBs3i+9nf4+1h1Q0TkhIoLLKHwhYfW+y4wy/JtHyTfT1usSyrr59r8hz31bvPY7JpVbjqSHRQNNMGTitlA69sgtqqJnVcqvOdPIzq2M838nKPl7mj0NGI4klUx7FRL7QeH/2q9JWD0VuunsLLPV7m5R4vmy2jXOqE1GHOkDkO7Ttm7zBBTfmM7jCa0R3sj5kZffbak5D62lVhIlfhWT8xFEVRFMUD8bLZbDazRSiKoiiKJ6NXloqiKIriBF0sFUVRFMUJulgqiqIoihMq7IY1iuTOkx4AlaoMuAPJWSq1SZnq33jDsVaj9DC1qzGaNGHZsmUObVX9EHdZSPokZ540DqTkFO5wBEvHcuLEiQ5t0sP80nczy4koVVSQqjFICTpcXYlEGrtSP0sOSKPjxawxDrJuaaxK/dqnTx+HNldX3Fm+fLlDm5QQRupr6Tx0x3xnFGksSMejsmuNXlkqiqIoihN0sVQURVEUJ+hiqSiKoihOcEnMUrp/Ld0jl2JpZlGZ2Jd079sd9/Cl+/AVqfRtBpIWKUZmNJ4tJU2uTHxNiqVJ41RqW7p0qUObNA6kz3AHmzdvNrSf1H/SWKtMgQOjfSodXyleLI0hVxdgKAtpbqtMBiUpZulqJJ+FVNBAGgvScZK+r1mxeemck+bFyqJXloqiKIriBF0sFUVRFMUJulgqiqIoihN0sVQURVEUJ1TY4CMF1qUAvPTAa3h4uEObFCh2hxHIqOlCeqjeLMOG0UQKnoR0fI0+rC2ZftzxsLxkDujdu7dD28aNGx3apCQWZiGZOiQDjfR9JdPJ7t27HdqMmjqk88hoQg1p3jCz5JSUjMIoY8eOdWhzRwKCvXv3GtIinZvuMtAYQRoz7jrn9MpSURRFUZygi6WiKIqiOEEXS0VRFEVxgi6WiqIoiuKECht8pGBvu3btHNqMZl+RgtHuQMpQYnQ/ybQiGUVcbUaRgvRmGh2MIOkzWgHBHRVppKwv0nGT9pPGgXQuuAOpT41WqZEy/bjaVGM0q5Q0R0ia3ZWlSjr/KzNnVSbTj1GkYyeNS6NZx6Sx7+oMTxKS2Wj+/PmGXiv1QWXRK0tFURRFcYIuloqiKIriBF0sFUVRFMUJulgqiqIoihMqbPAxmrmhMiW6pMBzZUpgSYHi+Pj4m36/mTNnOrRJphDJrGAUoyaCyMhIQ/tJBhqpX1xtnDBqgjHTxHEjknlBylwlGV6k/Vxt/jJqkJJKMDVq1MjQZ0jZf1yNNJdImo0avaQ5orJjSDrGMTExDm1SFhkpQ467yojdiHQeGu0bd5h5pPNGOp7Tp083tF9VGKn0ylJRFEVRnKCLpaIoiqI4QRdLRVEURXFChWOW13PylpMcb3KcW3bdQtSJKFdpchleE73K3jjhl7+bfvnPA9masZWvzn9FUk4S/t7+MALYCJw2W5lMgbWAqYlTWbBvASkZKUQERODVwotGSY0IyA0wW57IpZxLTNo8iaWHlpKWlUZMRAyPtnmUcd3GEWwJNlueIZaeW8rl+MsErg3EcsBithyRSzmXmPzfyXx1+CtOXjlJ/dD6DG0ylPjO8QT7eX4/Zxdkc+fCO+nbsK/ZUsqlyFbE0tNL+erMV6TmpBLqG0rPGj1pRjMC8Mxz8HpyCnN4YOMDdKvVzWwpDlR4sSw2hSRnJvObTb+BIggKCnIIFkvBbSkILpWncVVw9o3edpPCjcFjq83KyoyV2LDRNrwtkb3tJhkpk4mkTzIcuNqMkuiVyLvJ71IvuB6PNn+Uy3mXWZy7GMZA8OJgfFPth04y6Uj9J5mNjJbjMcqDix5k+c/LaVOrDc92eZakjCSWZi8lPSqdjrs6liyYRrMiGc2ydLNcvHqRbnO6ceTSEdrUasPw5sNJuZzChE0T+OrwV3zz2DdEBNiPq9HsNe4oXXT9uVUQVsCZIWfAFxo2bEgdS51yXyuZjaTzUjJc3SyZeZn0nNuTA+cPMLDxQIY2H8oPZ37gbzv+xoaTG0gcnVjywyQ9Pd3h9ZIxUBrP0jlYmfEM9rnDarPy+v7XOXHlBBcvXjR8jM3IrjV29VhmHZ1Fh1oduK/5ffx04SeWpizF75IfdVbWwdt67Wai0b6RzDeuNtC0b9+ewqJC7v/iflKvplIjuobhfpbm46o4D2/qyrLIVsSEPRPIK8pztR6XMiFuAgDzkueVav8i/Qts3jZij8cSmWHMTepOdp/ZzbQ902gb3ZaFAxYSagkFYPWbq8n+TTa5cbmE/DvEZJWlSUhKYPnPy4mLjWP9Y+vx8fYBoOmjTTna7CjJsck0P9TcZJWlGf/NeI5cOsKItiOYO2Quvt720+HLg19y/xf388r6V3j/3vdNVlk2Nmxc7HERm6/NbCnlMjVxKgfOH2BKvym83OPlkvbx34znna3v8MHODxh35zgTFZZNRn4Gkw5OYneGYy1PT2P76e3M2jGLu2Lu4vMhn+PtZV8Yp3w3hbe3v01mq0zC97k+DZwrOJ99nocWP8TGZMcasZ7CTcUsPz32Kfsu7aNH7R6u1lPlHM49zJorawjJDKFhSkOz5YjM2j6LIlsRU+6cUrJQAvim+eK/3R+fVB9s3p41QW4/vR2AR1o/UrJQAtQ5UwdskBmWaZY0EWuRlUUHFhHsF8ysQbNKFkqA4S2G069RPz7e/TGXci6ZqLJ8Mltlklcrj8CTgWZLKZekjCTqhtTl+TtKXw2OaDsCgO9OfWeGLKfM+WEOI3eMZHfGbm6PvN1sOU55f6f9h93Ld7xcslACvND5BbxzvclqlmWWtHKZ88McWvyjBRuTNzKg8QCz5ZRJhRfLlKwUZh+azYOxD9IhqkNVaKoybDYb/0n/DzZsND3cFC/KiWmayMojK4kNi6VlVEuHbQHfBRC4KRCvIs/SHh0YDUByRnKp9nxLPniBX4GfCarK5lz2OTLzM2lVqxXhAULi6drtKCwq5LuTnjmRF4QVkNExg5BDIfin+Zstp1z+ff+/SR2XSoBv6ZjZwfMHAagbUtcMWU55b/t7hPiGMLX1VB5t+KjZcpyy5cQWQi2hdKhdel729/XH/5w/haGFFAYXmqSubN7b/h7hAeGsemQVr/R4xWw5ZVKhxdJmszFx90QiLBE81/K5qtJUZXyX/R3J+cncHnQ7YVfCzJYjcj77PGnZadwWcRsnM0/y3Obn6PhZR1p+2pLs4dlYa1nNligyvMVwagbVZOb3M/li/xdk5Wdx4PwBDrU8BDZocLKB2RJL4e9rX2DyCuVQwuU8+8PxSRlJbtNkFBs2Lna/iHeuN5E7PS+M4IyLVy/yyd5PeGbVM4T7h/NcV8+cS97p/w7zOs+ja3RXs6U4xVpk5eilozSKbFTqqrIY30z7nZOC8AJ3S3PKO/3f4eAfDjKo6SCzpZRLhWKWf//+7+y+tJvVj66me5PufJv3LRyEl19+mVHtR5XaVzIISEFXo1l9KkOxlukf2LM/zH5kNjUfrOmwn5TdpCqygpRHamYqANlkM2TVEOqG1uXx9o9zOvM0SwqWkPNwDv3P9qd2Xm1ANsEYzYbjyiB4dFA0qx9YzVNrn+KhxQ+VtPuE+tDnXB8ahjSEkLI/VzKeVCVRgVE0jmzM/vP72X9uP61qtSrZtnrDalYcWAHAnoN72HR1k7cT7uAAACAASURBVGjWkPpU+h6uHi8vfPYCz699ntWPrubud+5mauJUXtnwinge3ojRkmSuLi9X3C8f/fQRU3ZNAcDP5scIRvDpe5+Wq88oru7n/o37Q+Nf/icZ2Av16tUTzy8pq5c7x/TlvMvYsBEZECn2Q1RQFJlkUqN+DWpa7HOf0Qxj0hzjSvNS/8b9K/V6qZ+rIlOS4SvL4+nHeTXhVUa2G8ndTe52uZCqJiEpgX1p+4iLjaNz/c5myymTrHx7XCHxRCK9Y3uz+6ndzLh7BoseXETftL5Yva1sqbkFG54Vs8wpyOGtbW/xQ9oPdK3blT92/CPDmg7D5mVjW/Q2zlvOmy3RgfHdx1NYVMh9n93HumPryMrP4qdzP/H6/tfJKrQfB0/r5+p+HtYJqsPTrZ9mcOxgrFhZyEKOctRsWdWe7Pxs4NodkxvxstnDNkXeRW7T9L+GocXSZrPxxIonCLWEMm3gtKrWVCXM2zMPgKc6PWWuECdcfwvlvUHvYfG59txcg5wG1Ltaj0y/TC5aLpohr0zi18bz5eEvebHLi6z59Rre7Pkm/7rnX9yTeg/53vlsqLOBAi/PugX0ZKcneaXHKyRnJDNw4UBCp4TS5v025Fhz+F2j3wHYn2/1EP4XzsP7br2P8Z3GM6v3LEYyEitWlrKUfPLNllatKY4H51vlfrR52X/0+Vh9xO2Kcwzdhp29Yzabkjex+MHFRAV6XvIBZxQWFfL14a8J9A3k3mb3mi2nXIrNJrWCa9EgzDHOF5UfRWpQKpl+mdTIr+FueSJFtiLm7ZlHneA6jO86vtS26Pxobsu8jQPhBzgZdJJbs281SaXMW/3eYlT7Uaw9upbcwlw61u2Id4o3y1Lt4YEoi+eM9+p+Ht5IQxrSilb86PUjp2ynuBXPGhvVifCAcLy9vLmc65iIHqDQ127s8SnUxfJmMbRYLjqwCIAHFj0gbh+9fDSjl49m7pC5TmMmZrDlxBbSc9MZ1nwYIRbPej7xRhpHNsbX25cCq3wVVuRlv43iY/OcQZ+WlUaeNY/GEY1LPTZSTES+PYaS7ZvtbmmGaBbdjGbRzUr+f9OJTfyc+TMAjYKNVelwB9XxPCywFrA5ZTOnz5ymd/3eDtsjsI+Nq1x1t7T/KSw+FhpFNCIpIwmbzYaXV2m3fG5gLgBBV4PMkPc/gaHFclT7UcTFxjm0J55IZEPSBobcNoT2ddrTvs61oK9k4jAjowXA1pNbAUp9ByloLZWOcrdmf19/utTvwtaTW9mZupPb6117vmvChAk8tOYhSINJz0yiQUgD0Wxw44kCEB7u+HiEqwwIUYFRWHwsHL98nLDwsFK3kjMyMjgXdA4AW6aNjIwMMfguGSSqmoeXPMz64+tJeT6FIL9rk8jPx39my/ktRPpE4n3em+QLyWLGEqmclFQqy1XGk4qeh9LxlbJUVaXJzoaNe/99L/4+/ux8aCd+3tceIVq2bBnHux2HOrBv0z6OZhxl4sSJDu8hjV1p3FemJF5FkDJfSYYoo+XQXEXPmJ7M2zOPvWl7S83F+dZ8CuoUUMe3DlNenVLS7kkmzMpgNDuZ1FYRDC+WElMTp7IhaQNDmw/1mF+yEjvP7ASgS/0uJisxxu87/Z6tJ7fywtoXWDtiLYF+9ofOV6esZnvadrrV6UaDEM95FMPf15+hzYfyxf4vmJo4lVd7vlqy7UrAFZJqJOFr9aXO5fJTsbmbFjVa8NlPn/HxDx+XPL5gLbIy79I8soqyGBE1QvzhYRbV8Ty0+Fh4oOUDfPrjp/xj3z94vv21iS2jXgaZdTIJyAggMMOzEytUB0a1G8W8PfN4dcOrrHh4RUmijSnfTuFK4RV+c8tvTFZYvalUIvXqwtFLdrfd9bfaPJnH2j3GqqOr+Oynz2jzfhuG3DaEk1dOsuTgEiL9I3nzjjfNlujAjIEz2HF6B68lvMaao2vo1qAbpzNPs/m2zdi8bHRK7oTF6llJvuPviGf+3vnEr40n8UQijSIakZCcwM7snXQK6kRcSJzZEv8neHfAu2xO2szMvTPZkbaDVtGtOJpxlJSuKfjm+RKzM8ZjE4RUJ3rH9mZku5HM3zuf2z+6nbub3M2P535k1ZFVNA9tzrD6w8yWWK35/2KxPJ99nkDfwGplivh0+Kf0atiLD3d9yOydswnzD+NXsb8ivn08MWGOldrNpm5oXXb8bgeTv53MskPLmL5tOiGWEGpl1qLp2aZEXfW8vg/1D2XLmC28tuE11ietZ9WRVTSNbsqIqBHEhcTh4+U5ceHqTJ2QOiwbvIwZe2aw/uR6tqdtJzogmqikKGofqo0l17N+RFVnPr7vY1rWbMmc3XOYvm069ULrEX9HPH19+hLg4/lVRzyZSi2WL/d4uVRiZE8ldVyq2RIqjLeXN093fpqnOz9d0ubuB/crSnRQNNMGTiv1WIOrqxO4mjohdZgzZE6pNqmKgSdTHc7D6IBo3rzjzVJ3RTx9bFxPXGwctjc865lbCV9vX8Z3H8/47qVd6VVdwcdVeHI/e9lsNs9UpiiKoigewk1VHVEURVGU/5/QxVJRFEVRnKCLpaIoiqI4wSVuWMl4Ij2QGxPj6OKUAs+urnggIWmWDAd79uypci0S0sPBkj7pIXMJ6cFuV1YdKQup/6SHoffu3evQNmTIEIc2sx6QlvpKehh6/vz5Dm1vvPGGQ5tUzcbVSPqk/pO0SMfI1RitFpGSkuLQZlafgqxb6mtpP+nBeFdXyDA6t0nnnJRUw5MMb1L/xcfHG3rt9OnTHdoqkshCrywVRVEUxQm6WCqKoiiKE3SxVBRFURQn6GKpKIqiKE6osMFHMjoYNQNIgXopKO+OgLKk2VXVIVyB1AdS30tBa2k/qZqDO5CMBZKBSzJ/LV++3KFN+m6VOW7S+xmtWCBVHTH6Ga5GGs+SuUoaV5LJQTKduNp4JxlRpM+Q5g2zjHcg97VktJOqGEnng6vHh9FsPZKZRzKoebrBxyiVrSClV5aKoiiK4gRdLBVFURTFCbpYKoqiKIoTdLFUFEVRFCdU2OAjBdal4Hbv3r0NvZ87yk5JnyFpTkpKqnItRpFMF0ZNSZJpSjLQuAOj2WGkwL30WlebsIxm4ZHMGpIZRTIluVqzNJ4lU4fRc0syPkjv5+qsPpKJSPpcLy/HwtDuME2VhTR3jB071qFNGr/umO+k4yS1edI8YTTrkGT2kpDWn8pmStIrS0VRFEVxgi6WiqIoiuIEXSwVRVEUxQm6WCqKoiiKEyps8JHMAFL5J8nUIJkf5s6dW1EJFUYKHkuBbClThfQ9jGalcTWSPilIbzSzjPQ9XF0Cy6gpRPpcV5cukpCMRZJmSYvR72ZWCSajSPrcYUSRMJqZp7LZWIxiNBuOpEdqk46TO4xskiFKmk8qM44qg3T+SyXEzESvLBVFURTFCbpYKoqiKIoTdLFUFEVRFCfoYqkoiqIoTqiwwUcKPEsmidGjRzu0SVlQXJ0VREIK0ksBb6P7SQF5yZhQGdOP9H7S50omAum1UgDdHaYkSbPRYL5kwpKOUWUMNNJ4Nvp+Rk0wrjb4SAauyhhCpO/hjqxD0jFfunSpofeTxrjRkl8VwajhSJrvJGbOnOnQJp3Drp4XpfOwMvu5GimTltEsZlLGraowB+qVpaIoiqI4QRdLRVEURXGCLpaKoiiK4oQKxywLrAVMTZzKgn0LSMlIISIggsY05v7I+6nhW6MqNLqEAxxgG9tIIw1ffMkflo//Vn980yrcBVWK10THCgslxNv/+H/nT8C2APcIMkC5mn/Z1NvWmzji3KLnZsnOz6b1+60Z2HggH9z7gdlyyqTIVsSaC2tYc3ENZ/POEuIbQmMa05e+BOA54wKcjI1feKj2Qzxc92E3qKk4ewv2sjlvM6nWVPy8/Gjg0wDqA6fNViaTlpXGpM2T+PrI15zJPEOYfxi9YnoR6BNITWtNs+U5xZPPwQqvFA8uepDlPy+nTa02PNvlWZIyklh6cCl7s/bym8zfEFYUBsjmDKNmGVcHmbM7ZrNo9yLqBdfjsZjHuJx3mcXWxRQ+VEjw4mB8U+3dIBlPjBqapGD0zWTDeaP3G2J7gbWAd797F2uRlS/e+YJet/QCjGcEqcpsOMWaDx06VKrdipXFpxeDN2z+ZDObkxxLGxUjjRd3GJCKKSwq5JEvHyE5I9ltn3mzLMtbxoLTC2gb3ZZ7mtzDwfSDbD69mcNXD9PjcA98bfbxLJ1v0niW9uvQoYNLtBaPjYkTJ5be4A3caf/7+duf83nS5xV6X8nM4+qyXW99+xbzrs6jQWgDnmjyBBm5GXx5+EsYA8wHTlzb12hGsD59+ji0uSpbUmpmKl0/7sqpK6fo2aAn9zW+jxNXTrDi5xUUBRVRa10tAtKu/ZiS5iyjxi6jY6siGDkHjR5jae6t7LpSocUyISmB5T8vJy42jvWPrcfH2weAPi/1YVPQJrYFbGPA1QGVEuRqdp/ZzbQ902gb3ZaFAxYSagn9f+y9e3xM1/7//5xcJnchEcQ1qKDu11I0cVqXoqWqdTla9Hra09NS5yi9oadKy4c6qrTfOqhylCraUre6VIKqUtT9kkQIEYlE7slk5vfHyBDzjtmRyezR33o+Hn04571nMq9Ze+31nrXXa78XAD/++0eyh2STF51H4LJAnVXeYFL0JDE+bvM4CooKeKvzW7ZE6S4Ua16bUbKc4aKkRdYethVwn61C7UjJTmHwN4PZFr9NbykO2XthL18e/5LoWtEseHABHgbrSkrvab05EX6Cs9XOEpkcqbPKGxT3jcndb0mWPXDrvnHg4gHe2fYObau3ZfVjq6nkY50EPNvyWR7834PQG/hcX4238vbWtzl/7TyTu07m1Xav2uLbz23nsVWPkXp/KrVW19JRYencDddgmdYs917YC8Cw5sNsiRKgWX4zsMAlr0vOVecEPtn7CWaLman3T7UlSgCvZC989vrgmeSJxcOio0LHxJyLYcauGbSq1oox7cfoLUcTR7OOsvbyWkgCduqtpnQW7F9A07lN2Ra/jZ4N3euHnsS8ffMAeK3Va7ZECdAouRHeJm8SQrVtjqsrdbHOKt24bxSPGx8/+LEtUQK0rdHWqjkRt3J8WCwWVh1bRZh/GK+0faXEsei60fgk+2CqbKIwqFAnhaVzt1yDZZpZhvqFAthNk7M8ssAAfmY/pwlzFutOrSOiUgT3htxrd8x3t3ut70hYLBZGbxiNBQszus8o8SPFXbFYLCy4sAALFlgHuPFvkTl75xDsG8ySx5bg5+3HpjOb9JZ0W2LPxRLoHUjLqi1LxD0tnoRmh3Ip+BI53jn4F/rrpFADvbGuZbtx31h3ah2NQhrRIqyF/UE3nPwUmguZHD0Zo6exxI+oYjxM1pjFy/0a/G65BsuULAc2HchbW99i9i+zaVWjFX0a9eFcxjk2BWwCC7TJd846h7NIyU4hOTuZXnV7kZiZyPT904m5GENeUR6FAwvxjfHF87J7J5+lh5fy28XfeLzp47Sv0V5vOZrYcXUHZ3LP0Dm4M7sv7NZbzm2Z3mM6URFRGD2NbI/frrec21JkLuJ02mkaV2ksDoj++dYEmeWb5b7JsiVQEziK25pkiseN++vcT0JGAu/teo/t57aTZ8rjvpr3QThwUW+VJTF6GhndyboGeeu6XmpuKnnV88AEXpnuZWiEu+caLNvM0j+U2GdiGb56OIO/GWyLe3p40v1yd6rmVCUd64mSzBlS9QVpkdlZBp+kzCQAssmm//r+hAeF83Trp7mQeYFvCr4he0g2DWIbEJhqXbOUjDuS0UFauK8oA830XdMBeKPLG+Li9o4d9qYZSZ8r6N+/PwDvzn8XgNlPzMY72tvudVKbSgvyrtiGqUfDHhX+GVIVmDv5bhn5GViwUL1Sdbvra8CAAVjSLZzNOEubTm3oENBB8/UmVUVydgWZbdus07Fn9z3L2eyzzPvrPJq81ETT50pjidats+6E4nEjJSeF7su7Ex4UzlOtnuJC5gVWH1uNx3MetDjYgsoZNwwt0nUotbWEs9v61v429fhULEYL/AaJZxJt8fr169u9VxoDpWtTMiXdqVGpLNegVhORVB2qvBWeypQscwtzeXf7u/x64Ve61OlCp9qdSLyWyDd/fMOe0D34mfwIK3Afe3JWQRZgXfMb0GQAXw/6GqOnEYAGXzQg7v44Etsm0mRzEww4tri7mq1xWzmUfIjoiGg61Oqg2x6DZeFWzb+naCsXpnBMdkE2AD5ePuJxb4P1h0mhxf3WpQD2X93P2eyztA5uTZNK9onSXbjduLH+1Hr6LuvLySYn6fBLB7ccN27mv3H/ZVPyJkgHtuit5u6mTEvUYzaOYfkfy3nngXeIeSaGGT1n8PWgr+mT1IcCjwJ+qvEThQb3uVBvvlU15+E5tg4PUCm5EkHJQRQEFpBbOVcPeQ5Z9PsiAF5s96K+QsrA3aj5bsHXy7rGXlBUIB4vTpI+HnIy1ZsNyRsAeKTmIzoruT23Gzf6NOpDldQq5PrnkhmUqYc8TVgsFj498ylLzi2hsndlWAq45zB316A5WZotZhb9vojwwHDejXq3xLHQglAaZzYm3zOfRP/EUv6C6wn2DQagWkA1aleqbXfcL91qSMoPyHepLi2YzCZ+OPkDfl5+9Ivsp7ccTdyNmu8mgn2D8TB4kJFnX0gdINdsHQ39DO5ntCuyFLEndQ8+Hj50Du2st5zb4mjcCMyyLtvk+eW5VJdWCswFTD46mZXnV1LVWJVZrWZBit6q7n40J8vkrGTyi/KJDI0UHZmVC6z3krO9sp2nrpw0rNIQLw8vCovk2W7xIyMeRW7kAb9O7LlYruZdpfc9vQk0us9zoLfjbtR8N2H0NFK/cn3i0uOwWOxdjSkm64hY07umq6U55HDGYTJNmXQM6Yifp/sl85txOG4Y3HfcyMjL4J8H/8mOKzuo51+PT9p8QkRAhN6y/hRoXrMM8QvB6GnkZOpJzBZziVsVkyZNYtpv09j1xy6GPTKMx+95XKwsI21jExUVdYfSHePj5UPHWh3ZlbiLfUn7aF/zhpt0zZo1DN4wmJTkFJbMWELtwNqiZrvKI9wwstyMs6sO7UrcBUB0RLQtJul77bXX7GKurHxzM5JmyRwgbdXmCjOPs5HaWTJ6SGaUO/2+3ep1Y9HviziYfJDWNW78jT6P9OH1L16nbqW6jBw0EpDbXqogI12Dzj4fuaHWWe+g9oOIvi8akA0XCQn2z4lqNco4i9uNGwA1O9TkfPJ55k+eT+1A68xTuv6lraNmzZplF3PW9ZpnyqPvsr4cvnaYllVa8p9O/yHYaJ0lBwcH271eGk8kRowYYRe7k+pkzkCrwaci8ormn0Y+Xj4MaDKAi1kXmRYzrcSxk1dP8tXxrwj0DuQvtf/idJHl4W/t/gbA6xtfJ7fwxk37HxN+ZG/yXjrX6Gzr8O7Evov7AOhYq6POSrRzN2q+2xjZaiQAb/70JiazyRaf9essruZd5ZkWz+ik7PbcbX2jtHFj1dFVbjtuvLH5DWITY2leuTnz759vS5QK51AmN+zHvT7m1wu/8tbWt9hwegOda3e2PoZx9BuKzEXM6jaLKr5VKkrrHfFUq6dYf3o9y/9YTot5LejfuD+J1xJZdWwVVXyq8O9O/9ZbosjptNMARIa6T+kyR9yNmu82oiKiGNFqBIsPLqb95+3pfU9vDl8+zPpT62lXvR3Pt3peb4kid1vfuNvGjYT0BD7d9ykADYIasPh0yVltXifr+qrxDyMeWe53+/huoEzJMjwonF+f/5UpO6ew5vgaZu2ZRaAxkK7hXfl7y7/TJsy9ihIUs3TgUh6o+wCf/fYZn+77lEo+lXgk4hHGtB5DvUr2BZDdgZTsFPy8/AjxC9FbimbuRs13I188+gX3ht3LggMLmLVnFjWDavJym5cZd984/L3dsxjB3dg3pHFjSPMhvHDPC243bmyN22q70/Bd4nf2L7juqfKK91LJ8g4pczmHUP9QZvaaycxeM20xd3/+z8PgwUsdXuKlDi/ZYu6uOWlskt4SyszdqPlmoiOisUx0v3Jgt+Ll4cW4LuMY12WcLebsHTeczd3YN6RxA9xz7BjVZhSj2lg9IVIRjIrcdciZuPM1aLBItjqFQqFQKBQ21HxcoVAoFAoHqGSpUCgUCoUDVLJUKBQKhcIBZTb4SEYCaZcL6aFVre/V6wF1qfq/pFmvB3Klh9ulB4ulB3elB5+l1zn7u0m7GEgxyYAg9QNXPKAuGSSkz5X0Sd/NFUjXkaRZ6gfSOXdFYQupP0sP90uFHiSkYiHS+Sjvd5PGBK3Xl9S3tD5oXx4kU5LWogTSeXK2ZqlNpf4rFXqQkPpCecc2NbNUKBQKhcIBKlkqFAqFQuEAlSwVCoVCoXCASpYKhUKhUDigzAYfyUig1VwgxaRFZr0qZEgL2e6EZA6QkBbfJZOEtJtAeVi7dq1dbMyYMXaxevXsS4VJC/fS+XCFwUfrZ0j9VDL9aD1vWtFqhpCuLem9krnKFdegpE8yykycONEuJrWp1PblNaJoNcZkZNjvMSrFtPYZZ6PVeCbtsuQKA5JWE5fUF6RxQhqLpNeVpbKRmlkqFAqFQuEAlSwVCoVCoXCASpYKhUKhUDhAJUuFQqFQKBxQZoOPtNgrLVBL1XCkReaDBw+WVYJTkBZ7ExIS7GIHDhxwgRptSKYGqeKJdI5GjRql6b3lIS4uzi6mtZKGZFDRWgWqPAYEyXAhtbMUk8wokmFAMhtI7aIVqQ0kM4nWCj5Vqthv2F5eM8StlMcwJJlOJKOHs/szyONTecYsvbZSk86n1K7l6ZflQfpcKSa1n9S3pFh5tylTM0uFQqFQKBygkqVCoVAoFA5QyVKhUCgUCgeoZKlQKBQKhQPKbPCRqldIxh3J9KN1kd/ZJg4JrVqk7yYZCVyxpZFWI5V0jqKiouxiztasdcsqKSadD8lwJVX6KI8pQfpcrcYYCel8SCYxZ2uWzq90PqRrS6qo5Gwkg4lkSpJo1aqVXUwyp0htX94+HhwcXK7334pe2w9KfUZqV3dHMv1J1b+2bdvm9M9WM0uFQqFQKBygkqVCoVAoFA5QyVKhUCgUCgeoZKlQKBQKhQOcUsFHqm4iLehLi9taDUPOrs4hbSsmodUAIlXIcXY1jPr169vFpLaSTDCSAcTZRiqpQoZ0frWeS0mzs80yUj8tTxtIhhJXbHel1cgifTdXGOqkz5DOm9atpKRrQTqXkumnLEh9etasWXYxaSs6yUDjiu2utKK1+pdEeUxw5UFrBSSt5rGyoGaWCoVCoVA4QCVLhUKhUCgcoJKlQqFQKBQOKPOaJWlpMGUKfP89JCZCrVowZAhMmAABARUg0TmsOrqKWXtmcTD5IH5efphamoiIi6BSZiW9pZXAMNng8DWDqw9maPhQF6gpG2m5aby34z1WH19NclYy9SrXI8Q3hLZ5bfHGW295IhYs/Mqv/MZvpJGGH360ym1FH98++Bn89JYnYraYWZu0lu8vfk9SbhJBXkFUDalKm6ttMFqMessTWZ+3ns35m0sG21j/Cc4JJvpEtMs1OaSwEKZNgy+/hIQEqFwZevSgUsOGXAsJ0Vtdqdw61rUNb4upugmv5LIP964iLTeNKT9P4fuT35N4LZFaQbUY0nwIE7pOIMDoHnmlbK2XmQndusHRo9CrFwwYAPv3W5PnunUQE2NLmOXdDsWZfLDzA97a+hZ1g+vyfNvnuZp3lS+zvuT3Kr/T5XQXQrNDAXnhXvoe0iK49Lo7MZ5MjLJuPXSrKcRkMbH07FLwgK8/+pqv474u09+VTCbONHak5qTSeUFnTqWdokW1FgxsMpCEjATWpK7B1NDE5qc2U9nX+re1Gnwks4eztzja6rOVmIIY6njW4QGvB0gqSiKmIIY4UxyvBr6K0VC25CNVGHG28eGT05+wOmk1jYMaM7DWQM5knWFvwV7Oe5yn9f7WeJo9nfp5zsC3ni+epzx5+4G37Y7VDKrJC+1eKPW90hZnUmUdp485TzwBa9dCixbwj39AXBz873/8tVIldnz0EbnVq9/27a4wvNyKNNYt/2M5eYPzYDFw7sZrpWtJ0iwZOCXjlGSw0kJmfibdFnbjaMpRejXsxYAmA9h/cT9Tdk5h3al1xIyKsSVM6XOla04yFkpVfcrSZ8qWLKdNsybKqVNh/Pgb8XHjYPp0mD8fxo4t05+saA5cPMA7296hQ80ObHl6C5V8rDPJs1+fZWfkTg7XPuxWv2onRU8C7Dve/DPzrWdrK2C/baTujNs8jlNppxjecjgL+y/Ey8Patb499i2Pr3icCVsmMK/fPJ1VlmTvhb3EFMTQ1Kspz/k/h4fBuirxY96PbMrfxI78HfTw7aGzypLsvbCX1UmruS/kPj5o/oFN88hFI0mon8D5Ouepl1Dx5evKysHkg0SGRtr6t9uzdas1UUZHw5Yt4Hn9B8jcufi88gpNli/nwGuv6SrxVkob615u/zIdP+8IvYHP9dUoMS1mGkdTjjL1wamM73ojr4zbPI7pu6Yzf998xt6vf14p25plXByEh8OttuHhw63/7t7tJFnO45O9n2C2mPl/j/w/W+cBqJJThUaXGhGSHYIZs44KHXM44zArzq+AJGCn3mrsKTIXsfLoSgK8A/jk4U9siRJgYNOBPFj/Qb448AVpuWk6qrRn3j5r8u7l08uWdAB6+PTA3+DPnoI9ekkrlWLNI+qNKKG5bkJdvAq9uBR+SS9ppZKel865jHO0rN5Sbyna2bvX+u+wYTcSJcAzz2AxGKhy6pQ+um5DaWNdh1odrONGIm7pUolLjyM8MJzRnUrmleEtrXll93n3yCtla7plyyApCXx9S8aPHbP+Gx7uJFnOY92pdTQKaUSrGvbPPDW91JSW51vi4Y496DoWi4VPTn+CBQusAyx6K7LncvZlMgsyaVatGcG+9rfHlxw2CAAAIABJREFUWlVvhclsYneie3T6YmLPxeKDD3U865SIexm8qO9ZnzRLGmlm90rwsedi8ff0p3FQ4xJxD4sHwRnB5PnlkeeTp5M6mUPJhwDurmQZal2a4dbli6QkDBYL+ZXcy+sAtx/r2Ab8CO44L1j2+DKSxibh61UyrxxLseaV8ED3yCvlyxKpqbBkCbz8MgQHw6uvOkmWc0jJTiE5O5nm1ZoTdzWOoauGUvWjqgR8EMCuhrtI93Pu+ldFsOXyFk5mneSBqg/ABb3VyPh4+QCQb8oXj2fkWx8Qjkt3n/vHReYiTqedJtQjtMQMrZhQD+tgmVKU4mpppVKsuYZvDVGzb651sMnxz3G1tNty8NJBwDrDfHjpw4R+FEqlqZV4eOnD7L2wV2d1pTBwIISFwezZsGIFZGVZl6CeegqLwcDZRx/VW2EJbjfW9VzSE9wj32giNSeVJQeX8PL6lwn2CebV+9wjr9y5PWrGDPjXvwDI9fDgXy1b8sfzz9sOS4vH0rY6FVndJCkzCYCUnBTafd6O8KBwhrcczoXMC3xb+C27gnfxz2r/JNI3EpANJVKlCmdXfZEoXnh+bb51XWTGgBl0+HsHTZ8rLYKXp1qHI0L8QmhYpSFHUo5w5PIRmlVrZjtW9566JA1MggB45/13mHFohljJSdIiLdyXtyJLMRn5GViwEF4lnOiu0XafcanJJWgKP277kd1Ju0XNkvFh9uzZdrG4OOf8SLBprhxupyc9PR1TlgmAa/nX8Ej3EPuudL1J1U6c2Z+LZ5Yzd8+kb2RfnmvzHCdST/Dd8e/YdGoT/bL6Ud9krcijdZs3ycDhVENNaCjExlqXmAYPvhH38+PIe++R07UrjjbvcmW1npvHuraftaVGQA2eaPwEF7Mu8sOZHzA8a6Dapmr4Jvve9u9IfUbq59LrnMGMXTP412ZrXvH38mdxj8V4Z3oTnxkPyP1X6jNSFbPycuczy1q1rCafJ5/E22Jh+qFDdEhzr1tWWQVZAMSciyEqIooDLx7g494fs/KJlbxa7VUKLYUsSF2A2eKG9yaArXFbOZR8iOiIaOu6gxszrss4TGYTjy5/lE1nNpFVkMUfl/8g5S8pmI3W9rUY3OcecnZBNgBGD9ntajBbH+GxeLqfZh9PH/G4O2oG8PTwpF5wPTYO38jaIWv5sMeHrBmyhseyHsOChc0BmymkUG+ZJcnNhXffhV9/hS5drMbFJ58Ek4lGs2cTVLz05CbcPNZ1qd2FHcN2MDVqKov6LmLZI8uweFlI7ZpqXc5xY2oF1WJ8l/E82exJCooKGLF5BDsuOD/x3Ql3PrMcOtT6HzD65ElmHTzIm8ePM/S++8jzdA/r+s23quY8PAej542BsZVfK5r7NuePvD+IL4ingU8DPSTelkW/LwLgxXYv6itEAy+0e4H49Hg+jP2QXl/1ssV9vH2osq8KaZ3T8DC5z9pw8fpIoUUepC0e1kHFHTUXmAvE4+6oGWB+v/livI6pDo0LGnPc5zjnvM/RsLChi5XdhjFjYPlya8K8+Y7Yb7/h1aULLSZM4Jf//Y8iP/d4Fvfmse7DqA9LjHU96/fE97wvebXzKKhagM8V+ceWOzC0xVCGtrDmlW/2fsOwjcMYGzOWnwf+jL+3v67anHJV/REczPawMKoUFtLs2jVn/EmnUGw2qRZQjdqVatsdr2usC0CKyX3WpYoxmU38cPIH/Lz86BfZT285mvjgwQ849vdj/Kf3f/jooY/Y8tQWqq+78SyaZ657/IgCa9/wMHiQVZglHi/yLgLAo9B9Ek+x5mv58jXmjpodUa2oGgDXPNxn3MBshkWLrIbFd98teaxdO5IeeQRjRgahsbG6yJO4eayrFVTL7rgxzZo8TUEml+oqD+2rtadPRB9S81LZn7JfbzllmFkWFsKOHVBUZC1IcAuXrjtkgwvd53ZKwyoN8fLworBI1lSEdXAp64PnriD2XCxX867yWJPHCDQG6i1HM5GhkUSGRtr+vwED+VWtxh/vq+5TxcfoaaR+5fpcyLqAxWLBYChZOakgwDp78828/RqPKynWnHAt4a7RXFhUyO+XfqfIUkSn2p3sjpsM1sHby+JG1WWSkyE/HyIjSz42cp3s62ujvpcvu1hY6Tga64qnRQaT4wphrqSwqJAdCTsoMhfR6x77vFI7wDrJuZp/1dXS7NDeQy0W6NcP/Pzg8mXwvjHwjRkzhvvefx/OnSPq2Wdpdc89YjUcaVFeqnLjrMVjHy8fOtbqyK7EXexL2kf7mu1tx0aOHMnCRQvhGowZNoaIyhHiQvatAxLI1UMWL17sFM3F7ErcBUB0RLQtJlUykZDar6LNBkNXDWXL2S0kjE4ocbuk74C+/Df4vwSaA3ki6gkMGDRX+pDMPFq3cNJCt3rdWPT7IgzhBlrXuGESeOW1Vxh9fjRVDVX510tWs0H37t3t3i/1g4qu4FOsOSE/oYTm6TOnM2j3IGp41WDuh3MB2dQlGSQqUnNBUQGdF3QmwBjAlX9dwdvzxrgRERHBxuyNYILW1VtTx7OOZiOaM/uBHSEhYDTCyZPWWaZHyZl6eGYmAEGNG9+2nZxdbep23DzWnc45XWKsA7gn6h72p+5n0YeLqOlfE5D7R5UqVexiWvv5nWDBQr9l/fD18uXU86dK9I/WrVuT3T8bGsCEv03gneR3xL4gGXzKW61HQvv9GqMRBg2C9HRrebubqBkbS419+8iIiCC9oRutOwB/a/c3AF7f+Dq5hbm2+Kqjq/g54We6R3QnonKETupKZ9/FfQB0rNVRZyXaaFq1KVdyrvDF/i9ssSJzET/5/0SeRx7t89pjwL1+1Y5sNRKAN396E5P5xu2pHzJ+INucTfcg+wSpN6VpXnZuGddM13g03L0eaQgwBtA3si/X8q/x75//XeLYvoJ9HDcdJ8Izwu5ZV13x8bGW8rx40Vq17GaOHCHoq68wBwaS+5e/6KOvFG431u1P3U+Hqh1sidJdMHoaGXTvIDLyM/i/X/+vxLHCRoWYGpjwuOyBZ7L+Szhlu/cxY4bVTj15MuzcCW3bwtGjdFi/nvzgYPaNHQvCTExPnmr1FOtPr2f5H8tpMa8F/Rv3J/FaIquOrSLUL5R5fd2rBFsxp9NOA5S4penOjOk0hsUHFzNm4xhizsVQv3J9tsZv5YTPCRoWNKRFfgu9JdoRFRHFiFYjWHxwMe0/b0/ve3pz+PJh1mesp4GxAQ8GPai3RDtK1ZywniZBTXis1mN6S7Rjdu/Z7L2wl3///G92nttJu/B2HEk5wobcDQQZghjmN0xvifZ8/LHVCfvWW7BhA3TuDBcuwKpVGEwmUmbNwizMwvTkdmNdZWNl3mz5pt4SRWb0nMHOhJ18+MuH7L6wm5bVWnIi7QQ5/XIw5Bjw/9HfLX5oly1Z1qhhLQM1aZK1buLPP0O1asT36MGJwYPJq1q1YlSWk6UDl/JA3Qf47LfP+HTfp1TyqcSQ5kN4L/o9Goa410y4mJTsFPy8/Ajxc9/dDW4myCeI2Gdieeunt9gSt4X1p9bTKLQR0TnRtMhv4bZVkr549AvuDbuXBQcWMGvPLGoG1aRnUE/6V+6Pj4d7ugYlzYNqDWJExAh8Pd1nvbKYiMoR/PbCb0zaPslaGPtcDNUDqtPZ2JlePr0I9nD0xKIOhIdbk+WUKbBmDcyaBYGB0KMHl0aNIr9NG70VipQ21g2pPoQ6AW40e7+JGoE1+GnwT0z7ZRo/nv2RXUm7CPMLw/uwN757fPHIco+xo+yr6mFhMHeu9b/rHNS4lqYXHgYPXurwEi91eElvKZpJGpukt4QyUyOwBgv6LygRq6iHl52Fl4cX47qMY1yXcbbYokWL9BOkAUnzne744CpqBtXk80dKVvGu0HVHZxAaCjNnWv+7ifwKLKRSXkob6yQ/hjtR1b8qM7rPYEb3GbZYxPsR+gkSMFgsFvd+SlWhUCgUCp1xj/mtQqFQKBRujEqWCoVCoVA4QCVLhUKhUCgc4JSyGdLisfTAq1SUwFm7SJQV6eFs6aFVrTt4uIKzZ8/axT788EO72ObNm+1iTz75pF1s2q3PkLkIqf0kU01F7khzOyTjiRSTTDVO3fmiFKR20brDiKRP+m6u+B5S+0l9Q+sOEhXxILqEVGxAGie0jouuGAO17hwyZswYu5hUOMYVSP1DKg4iMXHiRLtYecdtNbNUKBQKhcIBKlkqFAqFQuEAlSwVCoVCoXCASpYKhUKhUDjAKUUJpAVqaXE2ISHBLhYXF2cXc7a5QFrclsxGkhlA2k3k6lX77WKcvauH9BkNGthvUN2+fXu7WLt27exin332mabPcDbSTilad59x1s4GZUUyxtSvX98utnDhQruYK8wakiFHaivJ0CCZOqTz4QoTm6RFuvalXSWkdpbeWxGVjbQaT+rVq2cXk4xszjYhSQYkqW2kMUsao/Xq51JbjRo1StN7pZ1SpOu6LOO2mlkqFAqFQuEAlSwVCoVCoXCASpYKhUKhUDhAJUuFQqFQKBxQ5go+kllGMhdIi6nSorC0yO9sY4fWqhlSTNLibDOPxPjx4zW9TqrWU0XYlFaq9ONsJGPBiBEj7GJajQ9STK+KT5JhQK9tj6T+J7V9RkaGptdpNQc5m/Js0SWZflx1PrS2jV4VnqTzqbUvlMdU4+xrszxtJfWP8o7bamapUCgUCoUDVLJUKBQKhcIBKlkqFAqFQuEAlSwVCoVCoXBAmQ0+0qKw1sVUaUFfqoyidWshrWh9r6TZFQvyElLFnTfeeMMutmXLFruYdI5eeOEF5wi7DZI5QDIWSFWRpHMuGQuk17nCjCJ9D70MPpKRQroGtW6TJ7WpVHlJr62apHaW9EmVZsqL1DZatwyTzpMrtsCSDH4SUl/Q2qddMS6Wx5BTEdemmlkqFAqFQuEAlSwVCoVCoXCASpYKhUKhUDhAJUuFQqFQKBzglAo+WpEWhaOiouxiklGkPCYO6b3Swr1U+UIyTuiFtEWXZPBp27atXUzatmvlypWaPkMrkrFI4sCBA3ax1157TdN79argIyEZPfQyxkj9VGslLOn6kM6RK76H1Iekc+6qbaPKU01M6h9STGrr8ow70nmaOHGiXUyrkVLatsvZ24pJlKfty1MdqjTUzFKhUCgUCgeoZKlQKBQKhQPKfBvWbDGzq2AXuwp2ccV8BX+DPwGNAqh/tj5eRWX+cy7hcvZlPtr7ERvjNpKcnUyQMYjWVVrzQuMXaBzcWG95Imm5aUz5eQrfn/yexGuJ1AqqxZDmQ5jQdQIBxgC95WkilliYBKwB9HkksUxkF2TTfF5z7q92P1M6T9FbTqmk5abx3o73WH18NclZydSrXI+/tvgrYzuPdd++UVgI06bBl19CQgJUrgw9esCUKaDTs8xaOFh4kB35O0gqSsLb4E1tz9pQC7igt7JSKKWdK4eFke6CDSDuGEH3wBo12PqXv7iN7jLPLFfnrWZV3iq8DF508+lGuGc4SbWTONjmIEUeRRWhsVwkZSbRfXl3FhxaQP3g+rzY+kW61u7Kjks7GPHzCPan7tdboh2Z+Zl0W9iNmXtm0qBKA/7R8R9EVI5gys4pdF3YleyCbL0lOuQKV9jOdr1laMZkNjHs22HEp8frLeW2pOak0umLTsz+ZTbBPsG81P4lmoU1Y9L2SUQvjiY9T9u6sct54gl4913w84N//AO6doX//Q86dgRh7cwd+GDnByzKWUS6OZ1Oxk7c63UvZ0xn4Bmgrt7qSqGUdn7+//0/Kmv0FOiCoLvF4cNupbtMU8G9F/YSUxDDfSH38UHzD/AwWHPtgI8HkNEmgz8C/yD4kP1WRsVIi/eSYciZi8dvb32bpKwkPnroI/7V5V+2+JazW+j1VS8+OvYRx185DsjGIj0MPtNipnE05ShTH5zK+K43tuoat3kc03dNZ/6++Yy9fywgV+aRYi+++KJdTNq2S6ocpJVio4jZYqbbwm6YEk0APPvss3QN7Gr3upuRTASSUaEiKoekZKcw+JvBbIvfBkBQUJDd50gmDMlQIhkpnGmMGbd5HKfSTjG85XAW9l+Il4f1Ev722Lc8vuJxJmyZwLx+88r0N6VKLk6tirR1K6xdC9HRsGULeHoCkDN9Ov7jxlHw5pvkfPppqVqkWEUbvQ5cPMA7296hWeVmzL9/PoHegQAcuXqE4duH49nXk8BlgbbXS1WppPMuVaWRvssdGSlLaWfmziXglVd4+fJlchycV2kMlLRIY/kdV9y5rruwa1ey16yx6Z63dy8TLlygwZdf8m7dsv06qYhxokwzy3n7rBfhiHojbIkSIPhQMB55HmRFZjlXXTmxWCysOraKMP8wW3Ip5qEGDxFVL4oTqSc4k3ZGJ4UycelxhAeGM7pTyb0+h7ccDsDu87v1kKWZj/d8zO7E3fRp1EdvKQ5ZsH8BTec2ZVv8Nno27Km3nNtSZC5i5dGVBHgH8MnDn9gSJcDApgN5sP6DfHHgC9Jy03RUKbB3r/XfYcNuDOBAwfDhWAwGPH/7TSdhpfPJ3k8wW8y80/odW6IEaFalGT57ffBM8sTiYdFRoUAp7cwzz7htOwM23YWDBpXQvSYkBDPQPCdHJ2ElKdPMMvZcLP6e/jQOKrnOZzAb8LnsQ27dXEwBJryy3WPtstBcyOToyRg9jSWSezF+3n4AZBW4V5Jf9vgyMX4s5RgA4YHhrpRTJk6lnuLtrW/zUvuXqBNch/Wn1ust6bbM2TuHYN9gljy2BD9vPzad2aS3pFK5nH2ZzIJMOtbqSLCv/R2cVtVb8VPcT+xO3E3fyL46KCyF0FDrv7fcbvW4dAmDxYKl+Lgbse7UOhqFNBI9Db67fXVQpIFS2pmkJLdtZ8Cm2+PcuRLhsMJCPICrXu6RTzTPLIvMRZxOO00N3xpi4vHKtH6hwuBC56krJ0ZPI6M7jeblDi/bHbuSc4WYczH4ePpwT8g9OqjTTmpOKksOLuHl9S8T7BPMq/e9qrckEYvFwjPfPUNYQBjTHpqmtxxNTO8xnWN/P8bDjR7WW4pDfLx8AMg35YvHM/Ktxd7j0uNcpkkTAwdCWBjMng0rVkBWFhw9iv+LL2IxGMh/2f761JOU7BSSs5NpXq05F7IvMH7feLr/2J3OP3TmpV0vUVTN/bwZQKntzFNPuWU72xg4EHPVqvh89hneq1dDVhYex48z5dw5zMBXYWF6KwTKMLPMyM/AgoUgryDxuEeBNYGavc3OUVbBjN00lmv513iuzXPu6yAEZuyawb82W9daA7wD2Dh8I41CG+msSuY/v/yHmHMx/PjXHwnykfuJu9GjYQ+9JWgmxC+EhlUaciTlCEcuH6FZtWa2Y9kF2Ww8sxGAjDz7HVJ0JTQUYmNh+HAYPNgW9vTzI2fJEgr7utEsGKspECAlJ4VhO4ZR1bcqfWr34XLeZbZd3EbR4CICvg3A64J7zHhslNLOuGk72wgNJWvDBvxffJGAZ56xhSMNBsZGRLAtuHQfjCvRfLaLHZiVAirZLZ6mp6eTl51nfV1+NoXpheICq9YKL6NHj3b8onLw7rZ3+fLgl9QNrsuHPW6YXKSKEZK5YPv27XYxrUalshqGagXVYnyX8ZxNP8uKQyvo+llXAtYF4J3gDcjb8Xz++ed2satXr2qK3Slnr57lza1vMqLVCHrf07vEsa5duzKy9cjbvl/ajqui+0FpnDhxgkVXFpWIScYHqT9rrXZyp1s1jesyjhd/eJFHlz/KvL7zuL/O/cSnx/PSmpe4mmM9n2lX04iPj9d8DUrfrTzbI9mRm2t1Ov76K3TpAp06QWIi5lWr4JVXiDlxgivXK0dJFW6ka6YiKyUVL8vEnIthQJMBfD3oa4yeRgDWn1pP32V94VGIOhaFAQMgjxP16tWzi0lbvTnNRFhKO7N6NUUvvcTKwYO5UKuW7eVSBR8JSbNTDT65uRS99Rae+/eT1bo1OS1a4J2cTMCmTfwzPp5Y4NfbvD1YSKa6Gnx8vaz36QvNpdxmvZ52DYWGcouqKCwWC2M3juXfP/+bMP8w1g9bT4hfiN6ybsvQFkOZ+tBUvh70NYHfBIIn5PTKweLlPuYCi8XCs989S5AxiJm9Zuot50/NC+1eYELXCcSnx9Prq14ETQ2ixbwW5JhyGNduHAB+Xn46q7yFMWNg+XJ45x2IiYEZM+Drr1n/9tv45OTw0KxZeOXl6a3Sxs3LTHMenmNLlAB9GvWh2rVqZPtmk+7vHo802Cilndm9G7/cXIYtW4axoEBvlfaMGUOVjRtJfv55Ti9cSNLrr5Pw4Yd0BqoA6wB3uPenOVkG+wbjYfAgsyBTPG4xWgdvQ4F7Jss8Ux5PfvMkM/fMpGZQTbaP3F7iNtbdgNdFL7xPemPxt2AKN+ktx8anv37K9vjtzO0z1+1/fPwZ+ODBDzj292P8p/d/+Oihj9jy1Ba+efgb2/EwP/dY4wHAbIZFiyA83DrruYnUiAhOREfjm5VFHeHRHL0oNk9VC6hG7Uq17Y/nWI9nG93oeefbtDPt2rGvfXsCcnJofOKELvJK5bruwqpVuXTLI2/7gflAGPCoHtpuQfNtWKOnkfqV65OYlYjFYsFgKJkUzcHWtUqPNPeroJeRl0G///Uj5lwMTas2ZcPwDdQNds+niguLCtmRsIMicxG97ulld9zjmrV9LX7uM7NcedRakH3QykHi8VFrRzFq7SgW9l/o8HasQhuRoZFEhkba/n98fDyHrhyyHqscWdrbXE9yMuTnQ2RkyccZrpN+/bZgYJr7PO7SsEpDvDy8KCyS76KZDdaxztNs/310w0E7X75ukgkWbqnqynXd+c2bi7qPXP/XHUbrMq1Qd6vXjUW/L+LY1WPcG3KvLW7xsFBUswhDhgGPbPdKlnmmPPou60tsYiyda3fmh2E/uPXsx4KFfsv64eftx+V/Xsbb07vE8aIwqxPPI9192nlk65FER0TbxWPOxfBT3E/0b9yf1jVa07qG++zgcrcydNVQtpzdQsLoBPy9/W3xXFMuW89vpYZ/DZpUaaKjwlsICQGjEU6etM4iPEr220qXLgGQ4yYmDrC6jjvW6siuxF3sS9pH+5rtSxwvvv0anOc+mh21c2hqKgCZgYHSu/Xjum6fhARRd/HPvouuV2ZHmZLlyFYjWfT7Ij459gnfDf3O9lB0n6l9WH5pOU83eJqBiwcCciUTidmzZ9vFnGkueGPzG8QmxtKxVkc2P7XZ5nzVajaS9EmxVq1a2cWkz3C0mG/0NDLo3kEsPbyUKTunMCl6ku3Y+C/HM37/eCIrRbJ06VIMBoNYheeJJ57QpGXFihW31aKV0maL02Km8VPcTwxoMqDEaySDlLRI74ptgCR8fHzsjFOS8UQr0nu1VkW5laZVm7L8j+XMiZ3Di62tVZmKzEW0ersVhU0L8d3qS5spbQDZmCEhGVGkc3RH+PjAgAHWRxmmTYM337Qd6hQURKMdOygKCKDy8OG0rlxZ3LJPqxZnVkr6W7u/sStxF69vfJ2NwzfansledXQVqUGpdI/ozqaJN57JlfqqdD4rrFrSbdqZI0e4/+BBzIGBtJs4kTbX+7akRdI8a9Ysu5jTDDTXdXuvWEHrDRtK6F79/vv0ff99CoDe06fTPTBQNKO5ijIly6iIKEa0GsHig4tp/3l7et/Tm8OXD7P+0noa+Teib5h7WZMT0hP4dJ+1jNa9Yfcyfdd027G8mwwFTzV7ilpBtezerxczes4gNjGWyTsms/PcTtrWaMvRK0dZf2o9VYxVmNJmit1tcMX/PxjTaQwLDyzkzZ/fZE/SHupWqsvOxJ0UNi3E65QXxkNGx3/E1Xz8sdWh+dZbsGEDdO4MFy4Q+c03GIqKSHj/fYrcpFh2MU+1eor1p9ez/I/ltJjXgv6N+5N4LZFVx1YR6hfKvL5lKynoEkppZ1atwmAykTJrFmbBPa87pejut2IFHmYzPz//PPluMCMu84NCXzz6BfeG3cuCAwuYtWcWNYNq8mjYozxZ/Ul8PHwqQuMdszVuKyaz1Qiz6PdFpb7uwXoPulWyrBFYg73P7WXS9kmsPbGWnxN+plpANR6r+xjPNXqO6n7V9Zao0IkgnyA2PLGB93e9z/bE7WyO30yDyg3w3eqL8ZARg8UNf0SFh1sHwylTYM0amDULAgPJvO8+kp99lpyWLfVWKLJ04FIeqPsAn/32GZ/u+5RKPpUY0nwI70W/R8OQhnrLs6eUdqZHDy6NGkV+mzZ6K5QpRXdSs2Yc7tePlIbu0dZlTpZeHl6M6zKOcV3G2WLSM0/uwKg2oxjVxv75PdB+G1YvwgLCmNt3LnP7zrXFfnPX2o6lML7r+BKF4N2d6IhoLBMtbtufi6keUJ05PeaUiEW8GaGPGK2EhsLMmdb/rhN3J8XCXYiHwYOXOrzESx1e0luKdoR2Bsh3051dbAi6t+p4y1XCfVwiCoVCoVC4KQaLxeI+zyAoFAqFQuGGqJmlQqFQKBQOUMlSoVAoFAoHqGSpUCgUCoUDnLLHjPSg/cGDB+1ir732ml1MejDWFQ+jSw+FS7tcSA+USw9xa63grxXJkTlixAi7WHl2LJAehnZF22st1iAVp5AeUC9PEQutu8VID/gvXLjQLjZy5Mg71lIepP5Xv359u5jUhz7++GO7mFN3HUHWJ51zqZ2lghXSeauInSYkpO8ijWOSHmlnI1cgPcwvxaTvJr3OFeOENB5L512KSZS3z6iZpUKhUCgUDlDJUqFQKBQKB6hkqVAoFAqFA1SyVCgUCoXCAU4x+EiLwpKZR0JaGJf+nl4mDslc4ApWr15tFytPG0hmCq3t4goSEhI0xbQaRbSidccHaTcMyYCgl8FHq3Fk8eLFdjGpTZ2268h1JJOI1J+lNtVqTnHK7h0akDRqLZ8pvVcyWJUH6XxqNctI50QywTi7XKj097T2GWnMkgyS0jUitUtpqJmlQqFQKBQOUMlSoVBtVRt/AAAgAElEQVQoFAoHqGSpUCgUCoUDVLJUKBQKhcIBTjH4aDWFSKYB6b3Orh4iIZlCpAVgaXHbFSYOrUYWvSqCSEiL6lKVm/KYR8pj5pEoj7lCqpCjtY87m/J8htYKKOVBuqZnz55tF2vVqpVdbPLkyXaxNjpuZCydY8lcJH1nybTiCrQad6S+IBkfpfGpPBWUJH1aK+5IWqSxqLxjh5pZKhQKhULhAJUsFQqFQqFwgEqWCoVCoVA4QCVLhUKhUCgc4BSDz6hRo+xiUmUeaWsrVyx4Swu70iK9VM1BWtx2BWPGjLGLSdsrSWYjvUwEUrUjqZ21tqkrqie5alunikarKc4V28tJaK2UIhmVJNNP//79yyvpjpG+i2T22rZtmyvk2CH1aal/SCYYqfKa1Gek69oVxkfpc6XxTho7ynutq5mlQqFQKBQOUMlSoVAoFAoHqGSpUCgUCoUDVLJUKBQKhcIBTjH4xMXFaXqdXtV6tKJ12xmpoohEebYMkgwM0uK2VMlEMiA4u/KNhHR+pTaV2kVqU1dtuaQFyQwh4Yp2ltBqcpC2PZPe6wqzhnR+paotrjAglQVJt9TPJbNMWbaEciZSpS+tW59J58QVVZ8kk+jBgwfv+O9J1brKUsFLzSwVCoVCoXCASpYKhUKhUDigzLdhC4sKmRYzjS8PfUlCegKVfStTp7AOfX37EuIRUhEay01abhpTfp7C9ye/J/FaIrWCajGk+RAmdJ1AgDFAb3kib299myk7p4jHmgY3ZVn0MhcrKhvZBdk0n9ecXg17Mb/ffL3llEpyVjLv7XiPH079wMXMi1TyqcQD9R5gYtREWtWwf77P7cjOhubN6Rcayg/9+umtplSkdg4jjCiiqEENveU55no706sXzHff/pyWm8Z7O95j9fHVJGclU69yPf7a4q8MqjkIf29/veXZYZhsKP3gY9Z/qh+rTo3j+veRMifLJ1Y+wdoTa2lRrQX/6PgP4tLjWHN8DSdNJxkTOMbtEmZmfibdFnbjaMpRejXsxYAmA9h/cT9Tdk5h3al1xIyKccuEeSj5EJ4GT95+4G1b7NKlSwCE+YbpJUsTJrOJYd8OIz49Xm8ptyUpM4n7vriP89fO0z2iO0/c+wTx6fGsOb6GH0//yKbhm+hWr5veMkvHZIJhwyA+HkJD9VZTKqW187dHv+U0pxnOcOph/+C723BzO7sxqTmpdF7QmVNpp2hRrQUDmwwkISOBSdsnsSp0FUt6LqGSsZLeMkswMWqiGC8sKmTqjqlYDBYCrrjH+FymZLk1bitrT6wlOiKaLU9twdPDE4DQ3qGkdU5j5v6ZVI2pCsjVZhYvXmwXkxbLnVlVZVrMNI6mHGXqg1MZ33W8LT5u8zim75rO/H3zGXv/2FK1PPbYY3axiRPlE+xMDiYfJDI0kknR9ppuRjJxaDUqVQQp2SkM/mYw2+JLr16i1bBR0Vtbvb31bc5fO89HD33Ev7r8yxbfcnYLvb7qxfPfP8/xV44D8lZjUttLxgfJ9FNuY1tKCgweDNerxMTFx9uZOCQzidQ3JJODMw0+xe08uetkXm33qi2+4f0NZD+WzZfpXxK0OAjQvtWVy7ilnUtDMsZIZqqKNPiM2zyOU2mnGN5yOAv7L8TLwzq8f3vsWx5f8TjzTs5jXr95ttdrNdBV5DaApY1v4zaPw+JpocbRGgRdCbrt35ByjdSny9uPyrRmuffCXgCGNR9mS5QAAacCwAL5YfnlElMRxKXHER4YzuhOJTvp8JbDAdh9frcesm5Lel465zLO0bJ6S72llIkF+xfQdG5TtsVvo2fDnnrLuS0Wi4VVx1YR5h9m+7FUzEMNHiKqXhQnUk9wJu2MTgpvw4IF0LSpdQDveXe0c1W/qrzS9pUSx7zOeeF53hNziJmi4CKdFN6Gu6idi8xFrDy6kgDvAD55+BNbogQY2HQgD9Z/kC8OfEFabpqOKrURcy6GGbtm4HfVj2onquktx0aZZpahftZbPbfeXivyLwIDeOZ5Cu/Sl2WPy2t7x1KOARAeGO5KOZo4lHwI4K5LlnP2ziHYN5gljy3Bz9uPTWc26S2pVArNhUyOnozR04iHwf43o5+3HwBZBVmuluaYOXMgOBiWLAE/P9jk/u1syjeJ7WwwXV+z8naxMC3cRe18OfsymQWZdKzVkWBf+zseraq34qe4n9iduJu+kX11UKgNi8XC6A2jsWCh1sFaGLjNmqaLKdPMcmDTgYT5hzH7l9msOLKCrIIsjqYcJfWBVLBApSPudT9cIjUnlSUHl/Dy+pcJ9gnm1ftedfwmF3PwkvVZovS8dB5e+jChH4VSaWolHl76sG12745M7zGdY38/xsONHtZbikOMnkZGdxrNyx1etjt2JecKMedi8PH04Z6Qe3RQ54Dp0+HYMXj47mnn51o9Z3fM7GvGVMsEJvBId0Nj/l3Uzj5ePgDkm+S7exn51s0L4tK1PROvF0sPL+W3i7/xeNPHCbjqHmuVxZSph4b6hxL7TCzNqjVj8DeDCZoaRLNPm1EQUkDY1jD8z7mf2+pmZuyaQdXpVXl6zdPkm/JZN2wdjUIb6S3LjuKZ5czdMzF6GnmuzXP8pf5f2HRmE13/25V1J9fprFCmR8MeGD2NessoN2M3jeVa/jWeavmUW5q/6NEDjHd/O+dF5YEPeB/zvjHDdCfuonYO8QuhYZWGHEk5wpHLR0ocyy7IZuOZjQBk5Omzi5JWpu+aDsAbXd7QWYk9ZboNm1uYy7vb3+XXC7/SpU4XOtXuROK1RFYcWkFKhxQ4C1ywvlYyCEgLxVKsohbBawXVYnyX8ZxNP8u3R7+l55KefBr9KVG1osr0uZJBRaqkc6d4enhSJ6gOcx6aQ1TdKFv8rfNvMT97PkOWD+HtoLfxMfjotoVYedBq4NJqlikPtxpePtj9AV8e/JIAUwBZa7IY+e1IQHsFn+7du9vFpGpMzjZN+BiNdgYGqZ9K7VeeqihaufVc/jfuvxTeWwjpULi+kIxcaz+WNGs1okhjTkVswSaZpKTzGRUVZReTqnBJ3IkpbFyXcbz4w4s8uvxR5vWdx/117ic+PZ7XN77OlawrABw9dpS1V619WeqXUj/XOm6Xl61xWzmUfIjoiGg61OogjsfSdpAVYeaRKFOyHLNxDMv/WM67D7zL5O43ypOteHUFPAMMA2YDBc4V6SyGthjK0BZDAfhm7zcM2ziMsTFj+Xngz271DNL8fvPFC7KRVyPaerdlX+E+TppO0sK7hQ7q/pxYLBbe2fkOcw/MxbfIlwcvPYiP2UdvWX86LBYL887OY+X5lZANLAVy9Vb15+CFdi8Qnx7Ph7Ef0uurXrZ459qdeTr8aT6/8Dk+Hu7bpxf9vgiAF9u9qK+QUtB8G9ZsMbPo90WEB4bzbtS7JQ9eBPYBAUBjp+qrMNpXa0+fiD6k5qWyP2W/3nI0U9uzNgBpZvd3td0t5JnyGLV+FHMPzCU8IJxeF3tRpbCK3rL+dBSYC5h8dDIrz6+kqrEqLAJS9Fb15+KDBz/g2N+P8Z/e/+Gjhz5iy1NbiH0m1na8srf71OK+GZPZxA8nf8DPy49+ke5ZXEPzzDI5K5n8onwiQyNLPDZi4/L1fyt+c3vNFBYVsiNhB0XmInrd08vueO0Aa+K5mn/V1dJKpbCokN8v/U76tXQ6hHewP24pBMDbLe2Ddx8Z+RkM+W4Ie5L20DikMSv7r+T/3v0/vWX96cgyZfHm4Tc5fO0w9fzr8WGLDxmSMkRvWX9KIkMjiQyNLBE7lXMKgHq+7ln8IfZcLFfzrvJYk8cINAbqLUdE88wyxC8Eo6eRk6knMVvM9i8oLiCS6SRlTsCChX7L+jFk1RAKiwrtjh+7an18pF6Q+3SggqICOi/ozONrHhc1nymyPvdXx6uOq6X96cgz5TF47WD2JO2hQ3gHfnziR+pUUu3qbPJMeUw4PIHD1w7TrFIz/tP6P1T3ra63rD8dQ1cNJWx6GDmFOSXiOYU57Lu2j1DvULdNlrsSdwEQHRGtr5DboHlm6ePlw4AmA1hxZAXTYqbxZrc3bce6D+7Oz5E/YygyWB2REUbR6CBRkVVajJ5GBt07iKWHlzJl55QS1SJWHlnJ1vNbiawUSR3POqSnp2s2cUhb70iL5XdCgDGAvpF9+e7Ed4z9bizP1H/GduzN5W+S+3AunkmeLPh6ASCbCCq68k150Wr+khb4pe92pyaONza/wS8Xf6FJUBMmN5xMwvEEEkgQK03Vq2c/yEifK+mTthpyNvfffz913367RExrO0sVUJzJG5vf4I9rf9CqaiuW9Vxm8wdIFZBmz559x58jfTdXbCVV2udI5jut4+KsWbPsYo4MiE2rNmX5H8v5Yv8XtkfiisxFvPD9C2QWZfK49+MlDFRaTURSdSJnG+32XdwHQMdaHW0x6XxKY6+rKjyVyeDzca+P+fXCr7y19S02nN5A59qduZB5gR2Nd2AxWGgX3w5jkXtZrWf0nEFsYiyTd0xm57mdtK3RlqNXjrL+1HqqGKswtf1UDAb3sq3P7j2bmLgYlpxbwuGMw0QGRRKfHU/uw7kYsg34bfDTW+JdT0J6Ap/u+xSACP8Ivk78+sbB6JteuB+45kplfy5ubudGwY34/MjntmN5nfJs/9v4hxGPLDd81vIuYkynMSw+uJgxG8cQcy6G+pXrszV+K/uS9tHSqyWdjZ31llgqp9NOA9jdPnYnypQsw4PC+fX5X5mycwprjq9h1p5ZBBoDqZZZjUaXGhGS415F1AFqBNZg73N7mbR9EmtPrOXnhJ+pFlCNgfUG8kLjF6ju5363gyIqR/BZ289YlLCIX1J/4fC1w1TxroLxkBGfPT54ZKtBpbxsjduKyWwCYEPyhpIHo2/636dRybIc3NzO35z5puTBm8Zur3gvlSzLSZBPELHPxPLWT2+xJW4L60+tp1FoIz55+BPyYvPwNLhfhbViUrJT8PPyI8TP/XJIMWXedSTUP5SZvWYys9dMW8wVt5nKQ1hAGHP7zmVu37m2mKtuz9wpVX2q8s/If5aIDfjQvdv5ZqIjorFMtOgto1RGtRnFqDajxOfetN4qcwuio8Fi4byb9ufidtb6zKfbcr2dAdBxowJH1AiswYL+C+ziH++yfxbRnUgam6S3BIeon3IKhUKhUDjAYLFY3Pfnv0KhUCgUboCaWSoUCoVC4QCVLBUKhUKhcIBKlgqFQqFQOKDMblgJ6eFnyW0qPWRaEbsCaEGqVC89fCvtyCB9D2cVJSgrUiEF6cFdyfXpiraX2lTaOUAqriC1syseQJYe/pa0JCQk2MVatWql6b3ObnupnaXdOqSiCdJ7XYE0RkjXpVQkQmLbtm12sfIW6JD0jBkzplx/81ZcsSuNhNbvVp7dU7QibRwhnTutO+RURF9QM0uFQqFQKBygkqVCoVAoFA5QyVKhUCgUCgeoZKlQKBQKhQOcYvCRSllJC7GS4cAV5gKt5gcpJhlKKnLXkdshLYJPnDjRLiaVEXOFmUcybEhmHsm8oHVB3hVlCiXzgmRik2LSe6XrozznQ6tpStopRTLL6GXwkfqppEUyokjvlfpGeU0d0jUnIY0Jkkapz2j9jPLgCqNSeZDGXum6WbhwoV1M+m7SdagMPgqFQqFQVDAqWSoUCoVC4QCVLBUKhUKhcIBKlgqFQqFQOMApBh9pP0tpcVYvc4G08C9pkSpVSNVc9Nq/c8eOHXYxrRWGXIFU0UZqU8kMJcW0VpuRTBPlQWubSsYCyRDmbH1VqlSxi40YMcIuJhlHpHPk7kjfQ4pVxP6Y0rU+efJku5h0jrXqcXZVKmm8K4+Zp7zGGC1Ie8hK51hqZ2lMkPJPeVEzS4VCoVAoHKCSpUKhUCgUDlDJUqFQKBQKB6hkqVAoFAqFA5xi8NFqgqlfv75dTFrEdfaCt2TEkJCqSEhGIL0MNFevXrWLBQcH28Ukg4orKvgcOHDALiYttGutaKPVOFYeJDOE1KaSWUNrhSHpdeUxTWg1SEnXpYSkxdlbMElo/QzpupTIyMi4czGlIJ13rW3tijaU0Otzy4PWPi2ZeSTjY0VUJ1IzS4VCoVAoHKCSpUKhUCgUDlDJUqFQKBQKB6hkqVAoFAqFA5xi8PmzIC3SS8YTqVqKZDwpj1FJMj5J+iRTg6RPazWc8miW3lseY5FkrnC2eUGqaCO1qbQ1kFRNRGpnV5irJCRjm6S5TZs2mt6r1TAkIfVnycCl1aQjbYklGbMqAqkdtFagcbZ5UULSp9WEuXbtWruYXoYhSYu0FZ003lXElolqZqlQKBQKhQNUslQoFAqFwgF3fhs2OxuaN4devWD+fCdKcj5puWlM+XkK35/8nsRridQKqsWQ5kOY0HUCAcYAveWVynenvuPTA59y5MoRfL18yX4sG59dPnglu+fdc7PFzK6CXewq2MUV8xX8Df4ENAqg/tn6eBW5p+alF5eyMnllyeCk6/8mAZ+7WJAG7qb+bJhsKP3g9UfhfHb74LvH1zWCysDbW99mys4p4rHIwEg+a/eZixWVjeyCbJrPa06vhr2Y38+9x2gKC2HaNPjyS0hIgMqVadu0Kcf++ldyq1fXWx1wp8nSZIJhw6ACKrs7m8z8TLot7MbRlKP0atiLAU0GsP/ifqbsnMK6U+uIGRXjdgMMwP/t/T/e3/0+tYNq83Tzp0nPS2dZ5jJMg00EfBOAV5L7JZ/VeauJKYihjmcduvl0I6koieO1j3Mt+Bqt97fG0+ypt0Q74nPj8cCDJ6o/YYt9/fXX1v+RqZOo23C39eeJURMByMvLKxEvNBcyc/dM8ACvC+7XlwEOJR/C0+DJ2w+8XSIeHx9PqDFUJ1XaMJlNDPt2GPHp8XpL0cYTT8DatdCiBfzjHxAXR+01a6h28CA7PvrILRJm2XtpSgoMHgzbttlC0gKwFJMW4Ct6wXtazDSOphxl6oNTGd91vC0+bvM4pu+azvx98xl7/9hStUiGCGlhXIrd6cL4gYsH+GDPB7QMbclXPb8iyBgEwK+f/sqpqFOYe5qJ2B4ByCYCqYqMpE+qjKK12tGt7L2wl5iCGO4LuY8Pmn+Ah8F6h3/kopEk1E/gfJ3z1EuoB2g3vEiVkpxtlonPjaemT02Ghg+1xTr173TjBcOt/0j9QGpn6Xw4U3Nxf36+/vMMqz0MgD51+2BMNfLjpR95dv6z9K7Uu1TNWpG+x50wKXqSGB+3eRx4weutX+cfo/5hi0tGFMk8d6f9tCwcTD5Ibb/aRBNdIh49MtrutfXq1bOLaa3m5GxSslMY/M1gtsVvu+3rJC2SqUb6Hk5l61ZYu5bcTp1I/uor8LT+qJ6+ejVzMzJIevFFim09Wqv6VARlW7NcsACaNrUmyp49K0iSc4lLjyM8MJzRnUpehMNbWkfB3ed36yHrtnyy9xPMFjNT759qS5QA/un+VD9ZHf80fywGi44K7Zm3bx4AI+qNsCVKgLoJdfEq9OJS+CW9pJVKel46KYUpRPhF6C1FM3HpcYQaQxlUe1CJeOeAzgCcyT+jh6wyEXMuhhm7ZtA8pDkvt3hZbzki6XnpnMs4R4OABnpLKRML9i+g6dymbIvfRs+Gd8cYzd69AGQ/+qgtUQL8FzAD9+mjyo6yzSznzIHgYFiyBPz8YNOmCpLlPJY9vkyMH0s5BkB4YLgr5Whi3al1RFSK4N6Qe+2O1ThWQwdFjok9F4u/pz+NgxqXiHtYPAjOCCa1aip5Pnn45rvP2tSh5EMA1POznxW4K8seXybesUgqTAIg2NM1j0/cKRaLhdEbRmPBwnud3sPTw/1uzcONvnG3Jcs5e+cQ7BvMkseW4Oftx6Yz7j9GE2q9pe11/nyJcE2ss7kU1ysSKVuynD4doqLAaIS7sFgvQGpOKutPrWf0xtEE+wTz6n2v6i2pBCnZKSRnJ9Orbi8SMxOZvn86MRdjyCvKw/N+T8KPhuOf7q+3zBIUmYs4nXaa+gH1S8wqi/HNtSbIHP8ct0qWBy9ZC85nF2Xz3pn3OJVzCpPFRD2PevT26U09L/dPohmFGfyS9gtL0pbgZ/DjoaCH9JZ0W5YeXspvF3/j8aaP0ybM/vlOd6G4b2SZsnjj0BsczzyOyWKieaXmzGk0h461OuqsUGZ6j+lERURh9DSyPX673nK0MXAgRRMmUGnRIgqaNiW3e3e8kpJYgnVmOUtvfdcp223YHj2sifIuZcauGVSdXpWn1zxNvimfdcPW0Si0kd6ySpCUaZ0hpOal8sgPj3D86nEGNBhAdK1ossKyOP3AabJCs3RWWZKM/AwsWAjyChKPe5msv8mKvIpcKcshxbOH7y5/h5fBi4dCH6JFYAtOmE4wJ3sORwqP6Kzw9nyd+DUDdg1g6v/X3pmHRXWl+f9TLAXFIggBQVwwKNHEfWuNW6lBYmsbG5eok0y0J4lJOnZi/I3R7plEk6ZjGifGxxjTThi1p80YbdtohGnaJSrgCooYcYmKgBAV2QQEhKJ+f5QQ8R6oKii4V+d8nscHn/fWrfrWrXPve+853/Oe8x9RY65hYeBCgly12fNQR8zhGADeHfGuykqapq5tbLu2DRcnF34e/HMG+A4gpSiFkf81kriLcSorFBMRFoHe+SG7Rvv78+O2bVT36EHgggV07d2bkAkT6A9EAeqs8aTEITY0UfUKWxFVaREZIhyxLFaIdwhLRizhSvEVtmds55lNz7D8qeUM9bPcJYqqh/Tr108REy2B5ShDRNldSyJMuZnC1J5T+Xr61/WNf8X2FSz9fimVkZWsH7IeJ52T8NiLBrwdpU9E+d1yADz0HgqT1MGDBy23h90g42IGpItNBJmZmYqY6Dd3ZDURZydnuvp0JXZKLOMfH18f//AvH7Ls8jK23d3GM2HP4O7sLjSUiI5zS6rc2EpdO/UweTAtcBrX714nuSCZP+T8AY/dHrhmuQJiY8bq1asVMVEbb43vsT9zP+k30jGGGhkSMkTYJkVtQ40l8eraxsBrA+lS1AWADnSgg0sHdnjtYMbmGcwrmYcrlmMtqgTVFtV6WoKt+kTXRdHv1OyqORUV8P776NPTKejVi6LwcAy3bhGQnMxaLDO4Ttx7qcjsZesSbqJKVfZo1qZnu5WY3Wc2s/tYXI9rdq7hndPv8NH5j/jqZ19hcDaorM7C/d2YayauaXCXOMx/GEPaD+FE0Qkull6kZ7ueakhU4O5i6VqtNleLX1A3LHW3bfTYSmNzz/p692V0+9EcKDrA6bLT/MxHKxaDhoxuPxraW/6f9qc0yqeXU/FsBS6xLuhqmpjfqBIb0zYCMH/QfHWF2EBd23jwpqFzTWeeuPsE593Ok+2aTVh1mBryHi0WLqRTYiLnn3+eC7N/cqV/kJxMIhAHdAPK1dJ3j/+zFXz6+PTBGGCkuLqYjNsZasupx8fdYtAI9AykU7tOiu09vCzdxnmVeW2qqyl83H1w0jlRVt1I93DdMGVVm0lqMWEelovgjaobKiuxDZc8F1wvumL2MGPqqK3ubrDM+9t9cTcGFwOTwyerLadFBJoCAbjtdFtlJY8AtbWwcSOV7dtzYebMBptOAl8AAcAUNbQ9wCP9ZFltquZg1kFMtSYiu0cqtge5W8Z3Sqodv8J6cwlrH4aLkwvVJvFTWo25BgA3J7e2lNUkemc93Xy7kVuWi9lsRqd74Knm3tOPZmxtWNpG2vU0TGYTwzoNU2yvqrVkdi0d57r2fOr2KQa0U3YpOd223PvWute2tTSrJGcnU1RZxC97/hIvvZfacprk/rYhokZnOQddzI/05bNtuHEDqqoo6969wbSROupcA13aVpWQR/rJ0oyZyV9NZtb2WcLkc6nsEgAhhpC2ltYobi5uDA0ZSlFlESl5KYrt50vPAxDmpa3un1FdR1FaXcrF2xcbbnAGOgNFgIZ8SXdNdxkeO5zIv0QK20ZGmaW3obtH97aW1ih17Xll1sr6m6b7MQVYLu5OJdo7rQ/nHAbAGGpUV4gN3N82TCgTZq5LLgAdTOpXlXno8fMDvR6vvDzLU+YDhN/7+2PbqhLikFuj999/XxETmR9srYJia2UUawPUemc905+czuYzm4lOjG5QTWTb2W0cLTxKZ9fO6G/puVpwVaj54MGDipioWofIONFcXhv0GodzDvNOwjskvJCAwdUynloQWEB6STpjQ8cy69lZgNj8IKrWI6qe1JIKL4r36jeXjWkb2XRtE7tG7cLFydK0In8fSUJVApP9JjN+1fhG9YlMJq1ZwcdT78mk8EnsurCLDw99yAdjP6jftvIfKzkZeJKAygDSE9JJJ51NmzYp3kO0TJQjj+mD3N+ez/ieadCeR7wygniveB6reYw5E+egQydsz6Jz1VaDREtI+dFy43f/lAuRWWPMmDGKWGsst9QU97eNmz1vMid4Tv22A4UHyMrOwjnPmfiv4+vjovYrMi9qHVuXORN932bh5gZTp+K+dSsTTp6katGi+k1Lp0zhN/HxVOh0eEVF8ZK7u9Dgt3z58mZ/vNlse3GXR74fYeWElSTnJLP84HISsxMZGDSQjFsZxN+Kx9vJm9cCXlN2G6rMi/1eJP5SPFu+30KfdX147onnyLmdw/Zz2/E3+LNu0jq1JSoYEzqGl/q9xKbTmxi8fjDPdn+WMzfPkFCVQBfnLozSj1JbooLVz67meO5xPjz0IYnZiQwKHsTZ/LMkBSZhqDEwMn+k2hIVNNqeveIx1BqYWD4RHdpqzwCXCi29OOH+4VZeqQ3q2sbWG1vJKM8gzBBGTmUOJ0tPoivXYfi7NgyBjwSfforp2DEMv/89rvv2UTNkCE4//siyuDicamtZP2oU5e7qz89+5JNlkFcQx18+zrIDy9h5YSeHsg4R6BnIGK8xTPGZgp+Ln9oShWyO2szoLqP5U+qf+Dzlc9q5tWNW71l8YPyAMD9tdcHW8eWUL3ky4EliT8Wy6ugqOnp3ZIx+DJHuke3Vy1MAABIPSURBVOh12pv7FeobSuqrqSw7sMxShDw7iQ6eHQi/HU6/4n54mLRV/AEab8+9q3oztGIo3mbxXFe1yS/Px+BiwM+gzfPtQeraxtxNc0m9ncq5snP4uvoS6R/J4fWHcSrXXlf3Q0twMGX79+P2H/+Ba1wcbuvWYfb05HRwMLv79uVyQIDaCoGWJEujEeoeYdugG6clBHgGsHbSWtZOWlsfa6viu83FSefE60Ne5/Uhr6stxWZcnFxYPGIxi0csro+1RcHrltDRuyPrf9FwHa7W7Ep1BKL23BbzO1tC3iLtuLdtpaN3R37d+deK+NHyoyqosR9jqBHz+9qqId0YZj8/KqOjqYz+aUm01Rpr0/L2SCKRSCQSK+jM9oxwSiQSiUTyfxD5ZCmRSCQSiRVkspRIJBKJxAoyWUokEolEYgWHTB0pKipSxGY+UOcPoLCwUBEbPHiwIvbMM8o1+WbMmNFMdbYjmiQtcnOKChC89NJLilhLHLeiIgyiyeOizxBN3Bd9D6PR2AxlLUf03UQOVFu/h6OxdTUGUfEMRxVNsBfRbykqqCFCVAjAkau72IPoHBR9N9Hr2grRsbF1sryo0Iajiy6Ijk23bt0c+hmnTp1SxBxdhMHWFUZE1xPRdbGlq8DIJ0uJRCKRSKwgk6VEIpFIJFaQyVIikUgkEivIZCmRSCQSiRUcYvB5/PHHFbGPP/5YEROZdPz8lLUiRYYhRxt8REYR0eCxqIzYqlWrFDHRwH1LEJkzRIYS0eeKBrfVNEQ8iEjfzp07FTGHrWxgJ6LfXNReRK9zdDuwFZFBylZDmKNW+XEEonYq0qImot9YZPoTrfCyYcMGRczRBh9bfyfR+SVq0yLTWlusqHL69GlFTGSksnXlpZaa1uSTpUQikUgkVpDJUiKRSCQSK8hkKZFIJBKJFWSylEgkEonECnYbfK5cuaKIiQw+opioMs+KFSsUsdTUVHtlNYmt1XBEA8CigWzR4LGjEQ36i2K2mn5E37ctEB1TW6vwqFVhyNa2ITr2amHrGpwiA4foOLeFmUeEqG2IzjeRISwzM1MRa411PseOHauI2XqNEV1PRKamllSCsvVcF5m92sK4Yysio2dLqmu1FPlkKZFIJBKJFWSylEgkEonECjJZSiQSiURiBZksJRKJRCKxgt0Gn/bt2ytiooHYd999VxFbv369TfuKYi1BZBoQDRSLBrxFxg5RZQlRtY62QGTOEJkD2sKwITKZbNq0SRETGZWysrIUMS2ZTESINIsMEqLfyNFLjYkML999950iJmq7opjIIOHoZbtsrbIi0iLa11aTU0sRtV/R+S/6LlqqpCU6XqJYa5ikbEF0rGxdwq012oJ8spRIJBKJxAoyWUokEolEYgWZLCUSiUQisYJMlhKJRCKRWMEhBh9RVR9b2bt3b7P3tRXRALWt5hGR6Uc0eKxWtRmR+UFkHhEZT0QGhJZUDhEdK1uXhBIZVFqixVZEx2rhwoXNfj/RcRZVfGkJthpeRIjMKbZWmmkJonNG9JuLjDKiNiQyJallCAPx+S/63UXfz9HtXHSsbTXLiNq+6HWONqiJPmP58uWKmKj9io6fo/WBfLKUSCQSicQqMllKJBKJRGIFu7thCysK+eDgB+w4v4MbZTfo6tuVf+rzTywavghPvWdraGwxRZVF/PHYH9l9eTf5d/Lp7N2Z7pXdMboZcdO5qS1PSGFFIdGHovn24rfk3M4hxDuEWb1nsXTkUs0e5/spv1tO73W9iQyL5IvJX6gtR4Fuua7xjcvu/T1w759GeBg1N8b3Pt9bNH8DaKcefQOqTdWsSFrBn9P/TFZxFr7uvkSERRA9LppQ31C15Qn5t/3/RnRitHLDAPC544PxgrHNNdmD7s4dOkZGUjl6tNpSFNiVLAvuFDA8djg/FP5An8A+RPWMIqski2UHlvHtxW/Z8+IefN3VGzcQUXCngIivI7hcfJkn/Z/kF2G/IKc0h7jLcZytOcvrnq9j0BnUltmA0qpSRm0YRUZ+BpFhkUztOZWTP54kOjGauB/iSJqXpOmEWVNbw5y/zeFq8VW1pTTK+2MsY0dHjx5tEDdhYm/5Xkufi7LmgKrUab5+/XqDeE1tDbHnYjWpWUSJawlpvhrNkPcxY9sMdl7YSZ/APiwYuoDM4kz+58z/sOfyHo6/clyTCTP9RjrOOmdeDn+5QfzAgQO4V7urpMpGamoIeOstXK9do1JtLQLsSpaL9yzmh8IfeKHvC2x4bgMuTpbdY3bHsDh1MS9veZmlfZYC4io8oiW6RMtxiZb3ai6L9yzmcvFlhea/nfsb07ZOI7tHNusmr2t0f5EJQYQjDT4rklaQkZ/BR+M/YsnIJfXxxXsWE3M4hi9SvmDR04sA2we3RQPoIuNES5fyyi/P5/m/Ps93V5XVY+ylNaudLDMuA6C4f8Olld5Leo+9qXuZEzSHmatm1sdFBpquXbsqYqJj6ijqND/I4j2L4QeY6DaRCb+Z0KQW0fJtbUFdOzVjZgMbMOlMADzxxBME+QTVv05kOhMtf+Xj46OIOXLpvP2Z+9l5YSfGUCN7X9yLs5MzAGuPr+XN/32TqZ9OZSo/fZ7oWIvax4ABAxymUcTpG6cJ9w/ni1kP9ObMsm1/Ww1voutESwxWTgUFBCxYgOHIkfqY6PiJ9LXV8oM2j1maak1sy9iGp6snn038rD7pAIwLHsfQx4byTfY3lNwtaRWhzaEpzVG9ohjfbTxfnvqSwopCFVUqySzOJNgrmLeHNXTxvtD3BQCOXDsi2k11Yk/G0mttL767+h0TwiZY30FjHMk9wmepn/G44XGmdZimthybSMpOYuXhlXRy6sQzbsqbUa1xlKPkkEMPcw+1pTTJ8dzjAMzpPac+UQL8asCvwAy55KolrVGKK4vJLsmmb4e+akuxj9hYQiIiMBw5QsWoUWqraRSbk+XN8puU3i3lqcCn8HFX3tX1aNcDk9nEmaIzDhXYEqxp7tehHzW1NRzJ0Vby+WraV+QtysPdpWG3ybn8cwAEewWrIcsqa46vwcfdh/g58SwduVRtOXZhNpv57aHfYsbMa51ew1nnbH0nlTGbzbz997cxY2a6YTpOOm379QooYD/7GcxgutBFbTlN4m/wB1AMJeSV5oEOPPBQQVXTpN9IB3j4kuWaNdR6e3NjwwZK3nhDbTWNYvPZ5eZiMcJU1VQJt5dVlwGQe0c7d1zWNJdUWZ6CM4uVK6xriYI7Bfz36f/mjfg38HHz4Tc/+43akoTERMRw7tfnmNhjotpS7Gbrha2k3UxjSvcphHuGqy3HJjaf2Uzqj6lM6zWNri7KListYcbMLnbhiSfPoP0n4KheUQR4BLD62Gq2nt1K2d0yMvIzeHHHi2CGYQxTW6KC09ctc0+LK4uZuHki/n/0p91H7Zi4eWL9k7ImiYkhd88eKlSaq24rNidLP4MfYe3DOJt/lrM3zzbYVlFTwdF8i1GirKbMsQpbQFOay++Wk3A5AYCSSu10HT/IysMreSzmMf75m3+mqqaKuDlx9PDXZhdWRFgEeme92jKaxZrUNQC8NegtlZXYTszhGADeHaFc4UdrHOMY2bpsJjMZN7TpQL8ffw9/kn+VzFOBT/H8X5/H+yNvnvr8KdKup/E8z9OLXmpLVFD3ZPnJkU/QO+t5ecDLjOs2jn9c/gcj/2skcRfjVFbYCBERoNf+dcM+g8+IxczfPZ8pW6awbtI6nu78NFeLr/LzP/+ccr9ycLEYCI79cIyZM2cq9hdV/xFV8NmzZ489spql+Z2EdyiutBgHzJgb3X/Hjh2KWFtW6wnxDmHJiCVcKb7Ctu+3YYw1YrxpJKQipFEttlYEUaviiWhpKxFtsZyRr68v+zP3c/bWWYyhRsb1HMfbXygrPokMJWqZZcBiQEm/kY4x1MiQkCFEH1BOF3Ck4aUlXCm6QqI+kZeefImNUzcCFhPbvn37WLJkCXP7z21yf1FFGlGFIUe254rqCt478B4nck8wovMIhnUaRs7tHHac28HXpV/D13D/sKWoLbz1lvLGS9SOHIWzkzNdfboS/bNoRgSPsAR7wMyuM3kh4QWi/hxFREYELrWWy/68efMU7yH6HiKjjaOvHfXXrHvnvLe3t7Dymq1LuLUGdiXLVwe9ytXiq3yc/DGRf4msj3u6eBKSEUJO3xycTNoaN2lM8/BOw1kxfgVv/u+beLhqb/yhjtl9ZjO7z2wASveUkhCcQFJAElE5UbiaXVVW92iwMW0jAPMHzVdXiB08LJrNZjP/sutf8NZ780nkJ2rLsZmFCQvZ8v0W3hv9HsvH/lR2LTUvlcGfD4Y5wGrgrmoSFdTNZ37wJvPp4KcJKQrhmt818r3zCS7RpudB69hdlOAP4//A3P5zSbiUQGVNJQODB/KvM/6VW6G3AHCt0t4FXKR5XLdxfH7icwCCvIKsvIM2CKwKJLQ8lCteV8h3y6djZUe1JT301NTWsPvibgwuBiaHT1Zbjk08TJo/P/E5B64e4K8z/oqfwU9tOTZRa65lY9pGgr2CeW/Mew22Deo4CFKA4cATgHb8jE3ie8eXa37XuKO/o7aUhxa7kyVAuH844f4/mSB06Cj3LQfAvVSbE18f1AxwIu8EAL0De6shSUi1qZqDWQcx1ZqI7B6p2O5ZYylGUOUsNi1J7CM5O5miyiJ+2fOXeOm91JZjEw+T5m0Z2wCYvm26cPu8nfOYt3MeG57bYLU7tq24UXaDKlMV4f7hDaaN1HPz3t/W61G1m2pTNWnX0zCZTQShvPk3OVnmtTrXat/lrVXsSpazt89m75W9ZL2d1aDrsta5lpIOJbhWuGK4ra1qOI1pvlN9h90XdxPiHUKfwD4qKmyIGTOTv5qMwdXAzf93E1fnhk/qRXpLsQfvam815D1yHM45DIAx1KiuEDt4mDTP7T9XqDMpO4l9mft47onn6B/Un/5Bjl3lpCX4GfzQO+u5WHCRWnOtckqO/72/pW0urVHumu4yPHY4nnpPUmam4OrU8LpR4FUAWJ4wJc3DrmTZ67FebPl+C1+e/LJ++oKp1kTwq8Gk5aaxqPciZs6wGHtmzJih2H/JkiWKWEpKSnN0t1jzq9++SkFFAZ9N/AydrvGam6IB5das0qJ31jP9yelsPrOZ6MToBlVbykPLuaa7RgdzBwZ1HIQOnXAQvKRE6e4VmQ1Exom2oF+/foqYyETQFkaqQ5cOAdDZqXP9WM/q1asVr/vuO2VFIkcvY2UrKT9azpmhIUPrY6JKU21V2aQpGntaXJG0gn2Z+5jac2qD14iq9YgMHK15Drq5uDG151S2nt3KiqQV/HbUb+u3nb15FvdRlt6zmAUxeDlbnuxFZqoxY8YoYq3VZjz1nkwKn8SuC7uIORbDG71+mq8YlxPHzXY3IQcObjlYH7e1OplaRjbR9amtluMSYVeyXDhsIZtOb2JhwkKSspPo5tuN/Vf3k5KXwtigsUR1iWotnc2mKc1RvaKYP1h7BomVE1aSnJPM8oPLScxOZGDQQDJuZRCvi8fD7ME0pqGjiaLaEpu5WnoVgG7tuqkrxA4uFV4CUAwrSBzHp5GfciL3BL/b/zv+funvDO80nNzSXLaf2061uZpXHnulPlFqhdXPruZ47nH+8+J/cqrwFL18enG59DKHbx62PAUrjf0SO7ArWXq7eZP8q2R+t+937M3cS/wP8fTw78Hi3ouJ6hLVoJycVmhM82cTP2P+4Pma1BzkFcTxl4+z7MAydl7YyaGsQwR6BjLQPJAxjKEd7dSW+MhQWFmIu7M7vm4PT/dUfnk+BhfDQ2OYeRgJ9g7mxCsniE6M5pvz37Dq6Cq89F5EPB5Bv9v9CHMLU1uiglDfUFJfTWXB9gUk3kgkrSANPzc/poVOY/ub20E7U+AfSuzOFEFeQcQ+F9sgJiqGriVEmrVOgGcAayetZe2ktfUxLXSr2Yox1Ij5/cbnr2qFYzOPqS3BbvIW5aktocUsGbmkwSIBWsTfw59PIj9RTHlpzS7gltLRuyP/3v/fFfHtZdtVUNMMjEYw37tuCLrk1URbkyIlEolEItEgOrPZrP3bf4lEIpFIVEQ+WUokEolEYgWZLCUSiUQisYJMlhKJRCKRWEEmS4lEIpFIrCCTpUQikUgkVpDJUiKRSCQSK8hkKZFIJBKJFWSylEgkEonECv8fHGXsp1MrMOsAAAAASUVORK5CYII=\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "fig, axes = plt.subplots(10, 10, figsize=(8, 8),\n", " subplot_kw={'xticks':[], 'yticks':[]},\n", " gridspec_kw=dict(hspace=0.1, wspace=0.1))\n", "\n", "test_images = Xtest.reshape(-1, 8, 8)\n", "\n", "for i, ax in enumerate(axes.flat):\n", " ax.imshow(test_images[i], cmap='binary', interpolation='nearest')\n", " ax.text(0.05, 0.05, str(y_model[i]),\n", " transform=ax.transAxes,\n", " color='green' if (ytest[i] == y_model[i]) else 'red')" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "Examining this subset of the data, we can gain insight regarding where the algorithm might be not performing optimally.\n", "\n", "To go beyond our 80% classification rate, we might move to a more sophisticated algorithm such as \n", "- support vector machines (see [In-Depth: Support Vector Machines](05.07-Support-Vector-Machines.ipynb)), \n", "- random forests (see [In-Depth: Decision Trees and Random Forests](05.08-Random-Forests.ipynb)) \n", "- the other classification approaches." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "## Summary" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "In this section we have covered the essential features of the Scikit-Learn \n", "- data representation\n", "- the estimator API.\n", "\n", "Regardless of the type of estimator, the same import/instantiate/fit/predict pattern holds.\n", "\n", "Armed with this information about the estimator API, you can explore the Scikit-Learn documentation and begin trying out various models on your data.\n", "\n", "In the next section, we will explore perhaps the most important topic in machine learning: how to select and validate your model." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "\n", "< [What Is Machine Learning?](05.01-What-Is-Machine-Learning.ipynb) | [Contents](Index.ipynb) | [Hyperparameters and Model Validation](05.03-Hyperparameters-and-Model-Validation.ipynb) >" ] } ], "metadata": { "anaconda-cloud": {}, "celltoolbar": "Slideshow", "kernelspec": { "display_name": "Python [default]", "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": "3.5.4" }, "toc": { "base_numbering": 1, "nav_menu": {}, "number_sections": false, "sideBar": true, "skip_h1_title": false, "title_cell": "Table of Contents", "title_sidebar": "Contents", "toc_cell": false, "toc_position": {}, "toc_section_display": true, "toc_window_display": false } }, "nbformat": 4, "nbformat_minor": 1 }