{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Neural Networks #\n",
    "\n",
    "As outlined in [Carreau and Bengio (2009)](references.ipynb), the parameters of the Phat distribution can also be fit utilizing a simple neural network. For a univariate model, the need for such a structure may not be obvious, but the structure can be built upon to add additional free paramters (such as the mixture weights between the Carbens) and also conditional models with exogeneous variables.\n",
    "\n",
    "First, we will demonstrate the technique simply on a Gaussian distribution. \n",
    "\n",
    "**Tensorflow** is required."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Fitting a Standard Gaussian ##\n",
    "\n",
    "A conditional density model is estimated by providing one or many independent variables, $X$, and a dependent variable, $Y$. In our case, we are looking to fit a univariate independent variable. In Tensorflow, we must provide both $X$ and $Y$ input tensors, so to accomplish this we can simply set $X=0$ for every sample of $Y$:\n",
    "\n",
    "$$\n",
    "X_i = 0; i = 1 ... n\n",
    "\\\\Y_i = \\text{independent variable}\n",
    "$$\n",
    "\n",
    "In this example, we generate 100,000 samples from a standard Gaussian and fit the via the negative log-likelihood. `phat-tails` has a custom `DataSplit` class we can use to split the data for training purposes."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "nbsphinx": "hidden"
   },
   "outputs": [],
   "source": [
    "%load_ext autoreload\n",
    "%autoreload 2\n",
    "\n",
    "import seaborn as sns; sns.set(style = 'whitegrid')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-01-09 05:21:37.355360: I tensorflow/core/platform/cpu_feature_guard.cc:142] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  AVX2 FMA\n",
      "To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import scipy.stats as scist\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import phat as ph\n",
    "\n",
    "n = 100000\n",
    "y_data = scist.norm(0, 1).rvs(size=n)\n",
    "data = ph.DataSplit(y_data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Below we can see the kernel density of our samples looking clearly like the PDF of the Gaussian"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(y_data, bins=100)\n",
    "plt.rcParams['patch.edgecolor'] = 'C0'\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We have built a very simple neural network of `DN` class that takes in both $X$ and $Y$ variables, passes $X$ through 1 hidden layer (utilizing a `tanh` activation), then to an intermediate layer with two nodes, $\\mu$ and $\\sigma$, the parameters of the Normal distribution. $\\sigma$ is then passed through a customized `nnelu` activation, which is simply the `relu` with a restriciton to only positive numbers.\n",
    "\n",
    "The loss function is the Gaussian negative log-likelihood."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model\"\n",
      "__________________________________________________________________________________________________\n",
      "Layer (type)                    Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      "input_1 (InputLayer)            [(None, 1)]          0                                            \n",
      "__________________________________________________________________________________________________\n",
      "h1 (Dense)                      (None, 200)          400         input_1[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "mu (Dense)                      (None, 1)            201         h1[0][0]                         \n",
      "__________________________________________________________________________________________________\n",
      "sigma (Dense)                   (None, 1)            201         h1[0][0]                         \n",
      "__________________________________________________________________________________________________\n",
      "pvec (Concatenate)              (None, 2)            0           mu[0][0]                         \n",
      "                                                                 sigma[0][0]                      \n",
      "==================================================================================================\n",
      "Total params: 802\n",
      "Trainable params: 802\n",
      "Non-trainable params: 0\n",
      "__________________________________________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "from phat.learn.normnet import DN, gnll_loss\n",
    "\n",
    "dn = DN(neurons=200)\n",
    "lr = tf.keras.optimizers.schedules.ExponentialDecay(\n",
    "    initial_learning_rate=1e-3,\n",
    "    decay_steps=250,\n",
    "    decay_rate=0.8\n",
    ")\n",
    "dn.compile(\n",
    "    loss=gnll_loss, \n",
    "    optimizer=tf.keras.optimizers.Adam(learning_rate=lr),\n",
    "    metrics=['mean', 'std']\n",
    ")\n",
    "dn.build_graph().summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see the graph visually via `plot_model`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dn.plot_model()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-01-09 05:21:38.797366: I tensorflow/compiler/mlir/mlir_graph_optimization_pass.cc:176] None of the MLIR Optimization Passes are enabled (registered 2)\n"
     ]
    }
   ],
   "source": [
    "stop_loss = tf.keras.callbacks.EarlyStopping(monitor='val_loss', patience=2, verbose=0)\n",
    "history = dn.fit(\n",
    "    data.train, epochs=3, \n",
    "    validation_data=data.test, \n",
    "    callbacks=[stop_loss], batch_size=32, verbose=0\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The model minimized the loss almost immediately, resulting in the parameters below. They are shown next to return values from `scipy`'s fit function utilizing the Maximum Likelihood Estimate (MLE):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "</style>\n",
       "<table id=\"T_f878b_\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th class=\"col_heading level0 col0\" >mean</th>\n",
       "      <th class=\"col_heading level0 col1\" >std</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_f878b_level0_row0\" class=\"row_heading level0 row0\" >ANN</th>\n",
       "      <td id=\"T_f878b_row0_col0\" class=\"data row0 col0\" >0.0007</td>\n",
       "      <td id=\"T_f878b_row0_col1\" class=\"data row0 col1\" >1.0004</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_f878b_level0_row1\" class=\"row_heading level0 row1\" >MLE</th>\n",
       "      <td id=\"T_f878b_row1_col0\" class=\"data row1 col0\" >-0.0006</td>\n",
       "      <td id=\"T_f878b_row1_col1\" class=\"data row1 col1\" >0.9979</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x191eec3d0>"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "mean, std = dn.predict(np.zeros(1))[0]\n",
    "m, s = scist.norm.fit(data.raw.y)\n",
    "\n",
    "df = pd.DataFrame([[mean, std], [m, s]], index=['ANN', 'MLE'], columns=['mean', 'std'])\n",
    "df.style.format({'mean': '{:.4f}', 'std': '{:.4f}'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "tags": [
     "hide_input"
    ]
   },
   "outputs": [
    {
     "data": {
      "text/markdown": [
       "The fit for both mean and standard deviation is fairly close, though we should be cognizant that, in terms of daily returns, the delta of 0.0013  still translates to a 0.32% CAGR."
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from IPython.core.display import Markdown as md\n",
    "\n",
    "text = \"The fit for both mean and standard deviation is fairly close,\"\n",
    "text += \" though we should be cognizant that, in terms of daily returns,\"\n",
    "text += f' the delta of {mean - m:.4f} '\n",
    "text += f' still translates to a {(1 + (mean-m)/100)**252-1:.2%} CAGR.'\n",
    "\n",
    "md(text)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Fitting S&P 500 Daily Returns ##\n",
    "\n",
    "We will repeat the same process now for S&P 500 daily returns."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[*********************100%***********************]  1 of 1 completed\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import yfinance as yf\n",
    "\n",
    "sp = yf.download('^GSPC')\n",
    "sp_rets = sp.Close.pct_change()[1:]\n",
    "sp_rets.plot()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 00064: early stopping\n"
     ]
    }
   ],
   "source": [
    "data = ph.DataSplit(sp_rets.values)\n",
    "\n",
    "dn = DN()\n",
    "lr = tf.keras.optimizers.schedules.ExponentialDecay(\n",
    "    initial_learning_rate=1e-3,\n",
    "    decay_steps=100,\n",
    "    decay_rate=0.9\n",
    ")\n",
    "dn.compile(\n",
    "    loss=gnll_loss, \n",
    "    optimizer=tf.keras.optimizers.Adam(learning_rate=lr),\n",
    "    metrics=['mean', 'std']\n",
    ")\n",
    "stop_loss = tf.keras.callbacks.EarlyStopping(\n",
    "    monitor='val_loss', min_delta=0, \n",
    "    patience=20, verbose=1, mode='auto'\n",
    ")\n",
    "history = dn.fit(\n",
    "    data.train, epochs=100, \n",
    "    validation_data=data.test, \n",
    "    callbacks=[stop_loss], batch_size=32, verbose=0\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "</style>\n",
       "<table id=\"T_7d23d_\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th class=\"col_heading level0 col0\" >mean</th>\n",
       "      <th class=\"col_heading level0 col1\" >std</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_7d23d_level0_row0\" class=\"row_heading level0 row0\" >ANN</th>\n",
       "      <td id=\"T_7d23d_row0_col0\" class=\"data row0 col0\" >0.00035</td>\n",
       "      <td id=\"T_7d23d_row0_col1\" class=\"data row0 col1\" >0.0133</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_7d23d_level0_row1\" class=\"row_heading level0 row1\" >MLE</th>\n",
       "      <td id=\"T_7d23d_row1_col0\" class=\"data row1 col0\" >0.00037</td>\n",
       "      <td id=\"T_7d23d_row1_col1\" class=\"data row1 col1\" >0.0131</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x191a964c0>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mean, std = dn.predict(np.zeros(1))[0]\n",
    "m, s = scist.norm.fit(sp_rets)\n",
    "\n",
    "df = pd.DataFrame([[mean, std], [m, s]], index=['ANN', 'MLE'], columns=['mean', 'std'])\n",
    "df.style.format({'mean': '{:.5f}', 'std': '{:.4f}'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "tags": [
     "hide_input"
    ]
   },
   "outputs": [
    {
     "data": {
      "text/markdown": [
       " In this instance, the delta between the estimates accounts for just 0.004% CAGR."
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from IPython.core.display import Markdown as md\n",
    "\n",
    "diff = df['mean'].diff().iloc[1]\n",
    "text = ' In this instance, the delta between the estimates'\n",
    "text += f' accounts for just {(1 + diff/100)**252-1:.3%} CAGR.'\n",
    "\n",
    "md(text)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A visualation of the gradient descent (towargs the mean) is avaible via `loss_progress`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<video width=\"432\" height=\"288\" controls autoplay>\n",
       "  <source type=\"video/mp4\" src=\"data:video/mp4;base64,AAAAIGZ0eXBNNFYgAAACAE00ViBpc29taXNvMmF2YzEAAAAIZnJlZQAAPQdtZGF0AAACrQYF//+p\n",
       "3EXpvebZSLeWLNgg2SPu73gyNjQgLSBjb3JlIDE2MyByMzA2MCA1ZGI2YWE2IC0gSC4yNjQvTVBF\n",
       "Ry00IEFWQyBjb2RlYyAtIENvcHlsZWZ0IDIwMDMtMjAyMSAtIGh0dHA6Ly93d3cudmlkZW9sYW4u\n",
       "b3JnL3gyNjQuaHRtbCAtIG9wdGlvbnM6IGNhYmFjPTEgcmVmPTMgZGVibG9jaz0xOjA6MCBhbmFs\n",
       "eXNlPTB4MzoweDExMyBtZT1oZXggc3VibWU9NyBwc3k9MSBwc3lfcmQ9MS4wMDowLjAwIG1peGVk\n",
       "X3JlZj0xIG1lX3JhbmdlPTE2IGNocm9tYV9tZT0xIHRyZWxsaXM9MSA4eDhkY3Q9MSBjcW09MCBk\n",
       "ZWFkem9uZT0yMSwxMSBmYXN0X3Bza2lwPTEgY2hyb21hX3FwX29mZnNldD0tMiB0aHJlYWRzPTkg\n",
       "bG9va2FoZWFkX3RocmVhZHM9MSBzbGljZWRfdGhyZWFkcz0wIG5yPTAgZGVjaW1hdGU9MSBpbnRl\n",
       "cmxhY2VkPTAgYmx1cmF5X2NvbXBhdD0wIGNvbnN0cmFpbmVkX2ludHJhPTAgYmZyYW1lcz0zIGJf\n",
       "cHlyYW1pZD0yIGJfYWRhcHQ9MSBiX2JpYXM9MCBkaXJlY3Q9MSB3ZWlnaHRiPTEgb3Blbl9nb3A9\n",
       "MCB3ZWlnaHRwPTIga2V5aW50PTI1MCBrZXlpbnRfbWluPTUgc2NlbmVjdXQ9NDAgaW50cmFfcmVm\n",
       "cmVzaD0wIHJjX2xvb2thaGVhZD00MCByYz1jcmYgbWJ0cmVlPTEgY3JmPTIzLjAgcWNvbXA9MC42\n",
       "MCBxcG1pbj0wIHFwbWF4PTY5IHFwc3RlcD00IGlwX3JhdGlvPTEuNDAgYXE9MToxLjAwAIAAABT8\n",
       "ZYiEABT//vfHT8Cm6Plmy51E1H/Gd2uRAJnXn97KhCdFlzVoBkpM6urGucDCiNuOcQFEW4+msupc\n",
       "swwz8Uab2X2TPGmVrvK+fc6fXtEsEn0T1JxM51DVAa7qHUMJEmYk9oiwf4klMJnvayMO3yZV6Yy7\n",
       "npr3uZHc6F14b3XNce0sDIFUBHgANOzUii3DGFJ6niHmQp9Rf7fqh3aoZPor2i3J3l6XAMhImafj\n",
       "vCtkwb8kk7hCpQX9pFYeBrMQwtu/uuJk7KfNf/bte2CYHObASme3n82dp9FNg+B7H0XDDGEwibYt\n",
       "ej7GCs0J0+7vYd5ZPRKS93bA5X3vCbXAKByezSBdNZWF0gllNHqmahWWFed5R66Z6OsgaRIpDPvr\n",
       "iwFy+hYZQ4FRhaOSGm1wsbKCqEOstuTYiH2kFXU5lfd5oCxAqMJurwiPpiVI2+L1WZqULLPLqhqc\n",
       "SXt25NrOX4J8iVTbawTOPkd/p1tuIBSNWVqiWbwsn7Jfup11JhAJGxvAAwA3s8givN66O/EFcKMS\n",
       "3QVcExxcgibEYPFB2zfogIdVRuuxIU21oU7A3tY1JBZovH/BIxDWyWcYek2HBvSvXqW+DP9KCHRh\n",
       "v/rQnz/IJgC0bP6rom9z2DnOJn/HyHAEHtPK5nqshnQTx8ZUXjXLdUNwW3EF7wXWn1jAelM5GSw/\n",
       "6AfhQl8aWO2IL5T04SgeNfGMXShLesz4Bvn/SYxcHz5n5uei39NRQM3IaKRN5AsW0G0GvxCxGC8/\n",
       "+jGeiIo085a/HuD0O15V4ZL9Fkr3ZorsH90bb96mGSlshyvxejtjYQ0U7UkGs0D37VUXVWV8UWYw\n",
       "w6Qn81V3I2VoTsb4qEVn0NErYgX/nerxRO56gboM8yqBNIrV6vsCtphx61MvIPIKRCnyu5L2G4AB\n",
       "qfKlSxtoH7H//ASiM5JpWOdVm9KBVGGvj+5dIl7meMHA40rim+ghMj840ndqVAR6qI1G+DrpE8JZ\n",
       "PtUQFuDAfWSzgPZDsHqme3sxP87qLXPsExBv6OawU/xWYYn5gHrQl2gj8HZCnhc21EWYQ/1ISnQZ\n",
       "oRImcQgmyHiOwiiXOdHedZvKQ6sNMZOqksw7JgEIWwPQ671L23nx0kf6WV1SSyzLtJgfDr13+dqB\n",
       "GvODOmFuqFZBBwlSvwzqPGscDDWTrIXSeZoGzn0eR/Ys/PvkJXWFPGDV0bne48rNsXwqy4oV4qak\n",
       "eJMJ0sngEaiNImk3ds8lMyfulZZikYZQSMXvIBNVA7BC2tndWy9lU5LL1wq5Z5sjqkL7OYwJk1Nd\n",
       "Xp1fCt4k9flwm+/aoD/1wCQRYy2XDuvnNSj/XsImGlrVoQ4oFyfa1Jw78LnsATibGmiCIPy8aWA+\n",
       "wwh4etqrTu6gg5B/aWf+Ih8248F2KiKFt3Lfg3jE3QnF2l++JOGHQfQKCyUReNdelm1Fsg+wHqWj\n",
       "TqBMa8QVDJQNaLIVRaGOa8TCynmer9Y3QSsxy0uVdvANKKGvDu/pGtUF2HsF1YBMyEGu0hXBVXD1\n",
       "IIZjWaDLKx14F9gLsZkxqkpVK9x9jkbTWQoHACpujM526g5jYrhTUJ0sNci640j5SyerKpyw/Wp7\n",
       "Xo8H/R2r5irSIIOf5Y2jYyjRbaC2e6GLylMaFr/479b+323Dd2N5M3qzENgmJlnhpc1HA5jyWmEg\n",
       "bDVaAgxhU76NdSt7R9QOk5zDfkpvuPZWpkn8/zM1BJzdq2/UaMjbbtyMc2ctzQpcZ9pFD8P8qGpV\n",
       "nN86X3TVywS1UhYlZobpt3VLMCVIZPZUHNLJ0nM37SuQraSRCDo3+6EQbsC8nN9WeAstntSBG3ap\n",
       "Fc6gRJoUTal87E7IVcusK8+cb5KTe3FATbCB5aH1ZbrF5uQ0gdpTmuBK1IO2nSoinTYQmie7Qcm6\n",
       "HU8dvVopp4VTDh6pgolfINcEyayWDTbfCAtf76q4NyevdnfZbSgxcWwhl1Kt7nYbimlOqPz1R5w0\n",
       "Owq0NrD0AAScmdrNiKORq6yH0hNyyFhceluLcp0pS4CuRRVw4+0dSdPQMPMteau8LkwT1gEI2LAv\n",
       "VWAEWORJn6ryd+HViRPvYfgI5EWQJGtOXSt0DWO10NtSY2I08srm/Bl71WgnACSrK9vUX8SBCgWd\n",
       "JY1nJNezDTu3G7C8fYVpV/3dRm3liA0kMwt8pJ9JuH6C4W+rrUbGyHBDkTbKqUftf+VDU3krgi3L\n",
       "BsaJic9I0k9n40wRbCsm+sW/h/aHt/hUCx4DA+rbDv5nb5xioGCkYTxk/vkkox5/Ja/YGjPVNNhF\n",
       "R5n9vNVE81xZRe5ul2WZPzPCMpNrDIe/K6PtSzH+xdSjMZvEz2MtGinjJuqTf/gdpGqDrGjaQ5F2\n",
       "N6NKRFsqxtNoz1RJpd15g6jC8TBIZSFX6YqOXhzRyIRvoIdACtyrYIITI0vvPXryy6qd1gynhNl6\n",
       "0KYd3L2oJihalpqXhJNSGbTvkrWZUt+jPjP16sfvl17Tvb6nBUfM3x76kkVbwfSiBtEyzYf44uda\n",
       "ABzcAgl+kQVvB1ogGAf7gStDX+4UKxuWgrCbvmzNxFoUdxBrG4OSrit7Gxuw6B4nHUwMORwW/LDn\n",
       "2J5DKhiXtRDImNb598zLDmJvibM2hFF0t6UWz4XNODSJj2uhIgN58U+hgsSvzKBf1t21jBBfkfHF\n",
       "PGc7WZ33STPnc2czEJLtAuJVdg0tVasw2jfjgTk2PGuqPA5Zai49JTH1Wj79CZEbiNpUfwn5pkIf\n",
       "L9Uxrts6IIfMLkOeSkJdsGKCif5Dk5sjTB6lnPg9j6EXImvglti/MJ4dTm9sQuhAKGji4pBzWx14\n",
       "CjQWK1U5BrXwd9oZDQ0h0rI7fIMyThhLCJLqoELi3R7KO/3JDua2cmaff6XPXwOspTIe1TcnsaiX\n",
       "xEi6LgEGeDDP2SO2i079GDI4yg9yaaAo1BHqsQjONwsNbBF5W5ejGkiFiJrEfgSu7jIiBDPfrSfB\n",
       "78QpEZFtipodpwnbthlurWbbtUiudQI4u1TqqUK/jkOTq225+GEYn87P+XOtUN+/DPjs8e0mIENx\n",
       "DiVSfJkpTy2WRgnqa0F1SCdRRumTfYFST1FlRfH0GLmJ0W3AdfyGbRDyesouKxstOjgOueqhOAhl\n",
       "j740rmlTn5LsMw1Y044JVG6kaMg3Cdeu/bPUAAj6tnMihDQjrJ1a9x0TZ9ooUs2RC8FDk3tvs6Of\n",
       "WpwqhMv8rYfum9/QEniLgNjUPnlkas8uND2/V00vuRG9VAWVDG5cEVj+XKQl69Ayhjgu1N7erb8T\n",
       "PoN7yPjgXX1HQkhsnYxpvIFLggdm6/85YjKb2DPt/n3vCE6xGJqNdLWLJhXmUIoXzUdqwRKP241u\n",
       "O3EZCNXHxsuHa89lwlsdcMV5iQJk0kWX0UTJ9Duh0CE5BM0NwEJpjch/Do3zda4PNC5Oc+GAztou\n",
       "9g/XL7Yc7viySWRggJOycK9BzIxLW1B+QTGsAZ1T4gCQ9p70p3ciUcK5Hhc8bYCPeu6Q2PNJi7up\n",
       "j2+T40mhl89QfhVFMBDXRjT4CCR/dTwHm6Fc2yMLnM6fQiPG4iurWrcRj2vmC856slvsyq2zc2PW\n",
       "G8xt3OnqYD+DhO/1NvfBv7QycOem5tmXKln6Y4ipIg+8+hWh1/5Hbl+4JW9gOlQUBvr1wh+ws/++\n",
       "sydqbJDkTaiaxE5pZrsQ8zuS5C+b1N5K8561gCGTH77S+u7pgN81/Gc2AmwL2+j1TD0xcyjR8Dsv\n",
       "Uo4Dl/Zafyf2HMVOyiAQLUhxv66PZNzsMux5k4Enh0nR74rgYuYZICbVs9W2HMT/8vJOxWRqC5+G\n",
       "lL/oNpe9q8yIOcHyBxOUd8OtQstlLpTKbr50ZUUYKx2NV9+JEURcT4t22QBtVgPSa6sPTmwrH9wR\n",
       "+3nPkkONcPoi/fSFaT/Od41DT23wk4BrxX1X0UPrOoCSgPqxHup+cD5ePDRB5rgZ5pPYBDtq78QZ\n",
       "bDuoRKAMr4+w8QKKoNdsfiCqKUvivmHh5eddP7cQeaMwMYdn+fSMn5qKb023QDOA3Fnt/j/gOiSV\n",
       "r4fjGV4iKCF9mKx9ZQh2Q7ajKJfEcjxdPKtc60N7rsHk6FXyBr0d7X1R75GZH4/502THvrSHv2oZ\n",
       "Y+gND+AmE8zxpXHtJUh21WDm04+wYVKs1FUETR2upr7zLpoX1PYn2znehx4tVY6pVcUbzXBiLlpQ\n",
       "Xg3kNYct37CESMWYh1lTkv91ge99vkWJDqJsVsAvUUOVFeT20C78x1DEG4vHolJLtwFNq6Qwu9Oe\n",
       "G0dqOV38AepWK9532Do34nl8Adn5ixLBs9u1OLbPaTSsFyEqOxEDw4e5j0CFKXiZQQSaP5XY+10+\n",
       "K7Aa9J6fPrFeboDp/a+cIy4gUjF4YcSTp6DDuKsUY2tSBCfInjKiXV9UrpHeDUYmeCnpv4AYa6GJ\n",
       "O75NaZIB561lQ0cY5L86CV3Jamy4nF3VE91OcL7yRWnCRN7sV+g+9gCCUrUcVidk3tjaj5FXLHI7\n",
       "Y9kHs+TqJEP2I4Kr0r9GCH55wUwSP/qwEVCrwlghkoWPpWacdx0zV19TUGf1jwiNXaq5dJxTFbF+\n",
       "0/K7u2xjWnIWsNa6C/VzlsLI1HLSMm4HMMQUNHgFnXk/TSXX3vWYwTBhu5QmdIUki2CtMQ+EhOyY\n",
       "VUK9Rmn/MgAxJuBJhNz/BuwNuQAwIHW7QC18xP/dAw+2sud3dCbx9tq0gAAbQAAOerJx/27TpDPz\n",
       "Y5uJU9eq8fAsENMihCQXe1bQTiUzYg5/NcBz4B0rf57Dy+JBydeet55cdBU+eJ9kuNX8VdUZWCI8\n",
       "wBdbkSMRKhaPIKW+FAEn7VOB2SWw2xZDAWnly1QjZT6mWj8/WxqRc2roU3YMSKcXwP32ec52PMNM\n",
       "sX/eS2eWWFmSOg0SDoQZpUpHhEjDDfZWd5CC4rdecuCcTVjVnKe0pZp84n7o+YGF041kTjbS3apt\n",
       "KNLYWw9XkJDB4A5SJkH/lshC1+jTldCE/xn+UON3bYNOsO0GauEbA9rgDafsbRqamTi1wLzwh4nE\n",
       "V9s648djOeNeZZQ43XkWpz4ZF+tJ7s7iQ11Duabp/9DG8mnnCB6kiSneWue2wCr5obT8gNtfhU6P\n",
       "M3lYoPJavk9Pu9xud1GyTVHs5iH66qD/op9FVnHWm1M9hVinFf5IbaCC1IvwJ3giRE0bxbzmdyYe\n",
       "jd6SW6bcfOeqUj9KXap00FetKvH6HCEljeiktXFgfkZSa2w8rznrAB8WAneO8OOgCskujTw905T8\n",
       "DdfShcX3s/KeeVK/QZaL/1SRXac7SCblgixHB9fPGqQJl8fHgZtt7VR3ExfLyhV3fgWCdd5TqlqZ\n",
       "W0I3eDEqCp2x2AFNrq2cpTo0VGDP84GWNKvlulW1GO7BegPRFGSiIvtTKDLvB38RnOqVvZU4fZZS\n",
       "deW/UuGc06wDj9PDsjvef2314qQ3kGwvVJFEJ0AZXzszV4TAwEt1Ox2aHs5xNZuwWDIvcMmeT1Jv\n",
       "gyoeAhYTDwA05iYDbfsaiqHI7vnhp6Pd0VdthxYjI9+KOVLU0t4o/wnZCVmNCKenhxOL9pFJ1q7U\n",
       "iylAtHnCgXRhRpaGTQaS6Q0D5GHuZADnrvF+swnJ6/kdyh5sAEbm5rBvtbwOU0WnT5ublqxvIt9B\n",
       "qIo76hDuZrcnRn+AVRddv6VvgkmCyhyP0ZGLNpnyIlYB8wisDonYGXWIVKD4LldrrO9ueRv1zhDE\n",
       "ylpOpfCkTAhq3jaEqDj0EWPZieSXpLlgXpNIPP8pRchuCyOG3aLSRx9Y+zmdqMmbtJ1Ku/1YPiHt\n",
       "kOzxIqd+yU6PkUDANJCxSMhcotaB67Fua0wbdKjbewZV0Bz63vZee5c5mjngdReWusHv9od7cDAV\n",
       "c5toxT1uWSB1cpO6eeADK5RkLNVlVwWFxs4lDA2DpGLXfiOVeynwZxK+SMUCsZ6Ww7gm0HIibJKS\n",
       "U8sqKYadRL2FWqxZtMUyj8a84geSZYlFa3P53w4PUsFHGBC1Mbops5CQ/U/OPxQ+c6CW2YHmJB1x\n",
       "HaPxnF6SIQRboTERgGGNiAWAkuDSgStzJcpKo30VJY/Hn7k98CWPXHL7JZNMKqEPY/TGb6zzD5Rs\n",
       "AXJCGBS1vBTk3w0wiI5lYQFs7E+1ExHO8QyMwwrcAO+pk6IXb2bAaPEhOKn9UEIsmGT+s0+euAz0\n",
       "phUvAo0g8RP15LJ9JBnpe1qIYXmoMuuSpU+ZPGq/p8GhrM5RjbsZE8IcuOgrpSRMLI8pHi4HLe20\n",
       "UlYUXopSAiQFmjTeVOXgzVKg10z+OMjwUXv8N0/u15M+5UivtlnbHJ9gscnDB9MrHCDaa69yLPAQ\n",
       "dtRWazkwspoO4wcvHFNHnkZaEjDfeELD/nKIlnaBjs3Fy9ugnrJRreSZQu/qXFrAZxQwPhNrXnad\n",
       "wp85n6/8AU/dXfbsL7U51/UXzKdLA18qkV9hRv/k17BV4WIN81pAGSc6s0jnZ1tR6qWY6MJyHXcn\n",
       "OXknzqDvXnZ8xQWr4rpoP7vK5H/IK6TZ3ahdp+7s9uGsP3UiF+W60yJ8oNHd2ZBgUcFV91J59OzR\n",
       "wrJaNypG8ELzN8/5miLDqGY7Ad755hMYJjDqkshC4vBG8iYPbrEkkcECtJzHWikPaNyy1JpvFN3F\n",
       "f3d9LvCyOLyAnogdjUaxrqH6U2F8TV78grJs/tVjKrbZdg3Ab+FjoTwE/PkDv0C50NuiPwJdmlbD\n",
       "6gCQ2t/zH8SDg6ecXMjNzV1ZbLQ6jbputq4G6JEyg6gQAGLj57GkE0fqoAmTgav0QTWkt+XKCmFm\n",
       "pDrd+Za/yRIdEZFSpMRws2N5CPlgfWCvwoV129qDuKBeqbUPg2JBueosQZ7wE6ayN3bb0omDv1WY\n",
       "QQEv1ZMxgvFs3+eSLkZvI44As5BGw1jEvyEoyJt48Cc0pAgr4DTyeW+VZsdYgxDaXx1qr9O/X3VD\n",
       "PSh75k2u+4HMPJIvS6PXkSvMXD9ov03bZmhcw53hWadjEOM2xxPSbBl3lsmCA9psLMA4mEZ28Wer\n",
       "+1QrXK1179T/Z2VX4rmc70MjOIMcgegCb7ZSWBpdlVHXIQjs3lHx69eKhhcaYE8BxBxyEcs8DHx0\n",
       "OX1UNZv6Ho67reewNMhBmImF8a76zFSFyoX9DWLb41yAtTzm97+sZbQf7ZTtuFNuc+cH1NaeCFBV\n",
       "bE60YnVFAB9FtjE/uokAAAIOQZokbEFP/mjt4oBCDsGUiMf32P/6Rr/HIY33orPzzYNynun328TU\n",
       "MtvMWQTffv3wEKUlYleMOnDcabdIoB4NHBAfv+tue1gzQ+uvdTLUJ0hOwW1ojBWlr5LZIVp9FuRg\n",
       "KUb4BGd6DgCkz5s6H0hlm38qZ6DhWtHtOJNNDyBllSrKJxrlQgyNUBb+6S/GDMDwzzhYvxR1kuPy\n",
       "vpfcwA/teUdWacNEMVOKzaZ7DEu8ivacRmEGWycqZtzpDxhYSE87OlzfkIW0rYeBuN87MYrQJ30W\n",
       "Tlg1XHJMqzAcejdUsmMz6LeDxydN7uWu2v4OBIflqsf7WUC1Npmgz0AurrYua+ckjvcky1gL+ted\n",
       "TJc10XQudtEKEirse7M1UPWJXrf7ynjXmNhoFY0uGKq2x1KsJgamVRKugsWFbadyQE15hZ9MNQ4x\n",
       "r7T7bI1EIYoYNd8JPhdgbXhy5+Kp2CCQktBiVDJVAqc7jHR5jkENe+1ZpxpO+NJ6t+fOirD8R0yN\n",
       "llFjgO47Dql9jUB53r2J1TSiEpq/W0fMj9sCjR3ZOgRWba1UdDzukfmMPrAOoqBEuJAwGkavAbtP\n",
       "ePbZ1QJwC0qljYMjxmaCEqbuXlmVw44WRW0uqStfZrl5YBKpVmMmXz5mhKTldui+x9PAKDWj5rii\n",
       "QXERNgsqXL9UUxk30a+VK1IxTD5pTM4TCGcfbB2uEAAAAOZBnkJ4gh9EwlMAkIABGVwot1art9qL\n",
       "lrgCBMP7k8i18MuJ32Mi0lI6L2pwKgTaN1HEX8QpjcqE8My4YZmQkcVvU3147i7bfDXHeATdTCG0\n",
       "qCgp6IkMIhKyo/jYgideiCenHIBW/4vx6C8UQ18XVV3+ESG8clM7GK8+RrMlNDvmPwFMEJlmx/sE\n",
       "wU5RLbYCZcEb0bFlSoDL5ahAPTbB1xIL3XClu65lZcKEadBBwfb9OwaQrmAMQXM3wjKh61BOai9k\n",
       "On/ofdSv31ko8VO6qrJJabBXeVKdLPQykpKkPI9EJBpwwZoO6QAAALsBnmF0Q/9WoQ1Ov0vO+IAO\n",
       "PSQ460GmOP9ASw4qC3krPppzqiOOKV1XthzMu24SNCtpdOCeR1ADOZeMzGajkJZ56FR1dI3mdKr2\n",
       "1NecRAYCXdEyQT0NuYiae8LWGjlgTb1gnZVi7XFm1zXms0a+stnkv06sPoSLNOvJp6hBHL5ZfFBC\n",
       "+bdtj/EQBfN2hwOK31jnFxJbQAX9X4VbNTJEqr61Nb2zXv9nEe50jXSonT+yJe7NasiwddQGvsWA\n",
       "AAAA1gGeY2pD/x5/nneF60ALXZHPeNFxG4a+yxMtF4K2YiidwARSF2MMLYaU8fu9/n067V0F4UWM\n",
       "tV2VmdWJTfLCC2DqPrZvjm6DHUuJwrPo3ODvX0JMptd61jyKR832N/qfRvqjI4q2F9Q5Hq3902sR\n",
       "6uVJAVf6jz//34RUoreOIZf9mdfnnwe4FN/rnPPbI1CcfW8alsGRN4olaGaplzpHy/2a5ItHTBve\n",
       "KzwTxVrhGiRrN68XEWkvL1KRTUU0mhjRBV1cjshPW64AnX0oj3A/8Gc4wroSD+kAAAHlQZpoSahB\n",
       "aJlMCCf//rDRtv4AHMGqyL2YR7DbiCXDJRDFz0u3UZTYYpDuB+ncrYnKD0pT3tS3EnyrVTu62G+4\n",
       "N1/VGM1j4HjF0RqKSl7+WN4ZRY3HqWjD5M0VBY7sO+j3GdT6lpZPv9n+tUGbnKuL+Uqbd+AlUAey\n",
       "/pVugl+Oku/uyUhrjnCbMkwcaMpLYfeFkUQ6gdahm9osV0owNtwX/AgpS33DrAf2K4mPa2YD5i99\n",
       "XeACmCh7XYU8mQ4w+gksZqkFKSUDP1FbXeYYhyf/YaqgHynd7DWOZ6wLJHzvVgfxWiuP8wzGTYzx\n",
       "MbcLLqko0mpjdq/fYIohs/EUrWSkFgy3RWVNDhI8dEf27BItgs6JmIu7PpRyjomw2fWYCDTPLQhQ\n",
       "eipueC1zKO/TClevzk0WAnC7x1HvPb71+GGvFwgbPPHxt5retPdYmkLwLKOh6OhsS2Uu6jFnMS5k\n",
       "VebnSxCN8TAC7b0ZRJuosDg1hu3GSn8kcIk5zR0pfsOxcJk28+tlkscIznbNtzLgQOfcjwsbBCpk\n",
       "JNeTGqyoNZUVk238rVETzKyDsbDLtdVc0XVtqUKKQ/FhHb3WZdKQXk2fXyDFNoDFz9GAAQTIQgMa\n",
       "k+ft3bMN3IDl0Dvs/lk8RjKGnx9qJlEAAADFQZ6GRREsEP8WJfFAABbNJiboXKuRPkeeQhhPT5dA\n",
       "TmtPAv3aYeUQt0hmsDB3xuGAeb5rkDN55+2WVZRLqEQkLot2tTfi2mNhMZ6VYw6C/x84G1OkEXMl\n",
       "CreseqKy3pbV/wcHMccjWQrwAy44i+j9ApDcilzv3ZRUuL0jnBEGvm1eecSRHFsHLN+eK20MBbJR\n",
       "EyWp3i8nDv7Zau7jcPeozDTT6ypSDwfyqY2Kcdcw923b5pU6qj0Wptmddr7uB6AlmHJ+ZwkAAAB+\n",
       "AZ6ldEP/IsBmeADdkPmIovy9/ISoHx1UcngXNXGaH2m0Fzypm9TweS+u/pRtc6K/mFqhqnkiruqg\n",
       "zUbaAcvsHCTiTb36TvnNfd1LdSKHWwTst0R+HLxTsmX4ko2oKeVjLw1MTgfbJz48OaCsfLFw53+A\n",
       "/OZGCLLa8logMcmVAAAAvgGep2pD/yLZTPABus2nNdtMJ9DoPF50ORdqiMrLepE/mxv79DEP+ocU\n",
       "ANJA6Tlrc7M5aD4g+r1QOYUZEf/xtvL58Ea/Z5SMwQJX9scNRB3LZivjArU5k6Nb+x+QHHJ8c0om\n",
       "4vdnJVWOgtiKseXYdtuxUEcXVeQ4DjrqmhuuDcOX0Hl1cPA9MjwBe/qR05EJb7c7Llu1i9Lp71jN\n",
       "ntFoDSviRhEY4Kahi8PixkEBZ/JaC8P2sZ2K+3oFCGwCO4AAAAISQZqsSahBbJlMCCf//rDQNv++\n",
       "KANxnila4VejlDbWjLv1/62n/O1p45SlYa/9Ed44qbLRXALXP/ke86Xmh458qWzh1BzxVo0k3Bp/\n",
       "qI8jaowUzsxvV5vv7cQSrVs0lREN3SP49cMjVops6y+Zbve5VNt0iSRdxlBmQ4mGEddHTDHdSNsZ\n",
       "Cj1WTzkTkvoCP5tiJT1onajpxDS76/DpBG4dioQKaA2AU7qCQYnaGPLWvSev4PbSON5tHy/ayTn7\n",
       "kpp/sBIbb8KkyGtIlNtgu99QsrjwzBIX4tBTn9I+TTFFRsirlKmvgHm701W1DB04yabrfBQ+gGOP\n",
       "yqv3b/mB/AUnjmvI3QfOyKbEvRrHk9SJA9FvtH0BPGSwKaTIYzE9nO/Q2jMk+GeizQhn8OshHZsA\n",
       "ESUzUU9HlZ02iBSaID/s/4qjRlOZuuRbV6qTdmvw7damaAuNExNxjMdIxhuwelyLH1g7a5x0Bzyq\n",
       "UFlDSRxAX20vn4xoszBFd3hul2bw+lJvgRz0p91Qs0ETEPr6ULnl9rKA+09q61abr3vcFHryZVkm\n",
       "k5LlzVnEaLU3ftoGxCFSd2jOakTjzxF2OjQC+ov0MN4mDQWGbmKBV+2jlNq/wjYY0uQKMcAdvErX\n",
       "1DBc5qsiNhcrG6PtSZk2ONPVj7jloPPjFks8F6cwnafBn9nj3exe+UGp3OuTiLLCLxopdE61VUAA\n",
       "AAC0QZ7KRRUsEP8aY0XO/dz/AC3qgPYnPn8nNtZThpfa+5kFQIEboSXDajKFarit4Lex6mQUKHcT\n",
       "ETKKIHskUj/h+Ixvwhiqxv2vKH0jSVxrwr0l+e38QkFq8eIi1/CctLRelEGUcuGJA4aLiOhzh8Mr\n",
       "AxbOw89Dn5w4BtFIiHwDyEcnXI1MyW+ZihbGTEk8ibg8xipMSOMsCz2OpypJK+rZeyvxcr9IwLrQ\n",
       "tWhX7hyurkx9ZPnxAAAAfQGe6XRD/yMI66gA4zZnAReUtrzhYgoNqBr39GzEv8tpK9Xk0CqZsmbw\n",
       "3G0TcDxLQx0zrbT928i6R7xsSa7OBXrLcjCoAIdmby1tmRDIRDXipifB98luzfW4fEwhSYuIy7i2\n",
       "IO9V4hBHlX/AVBpRUyqWZ7T+9IHg5NUwRPhAAAAAowGe62pD/xq/foDuRgBLVzS1MtGsjWFiDcoM\n",
       "iWU/ax2OWaMHp4ookmOOKT2K21dZ5rolQW3eeHCYYTHEtxvl2olHBTB3zOEpxI1St8ikLWvgvpbI\n",
       "Cr8xUWSM0EzmmqkXsxTGP9fAJYtLIMWE0ZSasmKTuX1n9Ml9xFHD3DGiQ73ZieUgCoob+XMuek7q\n",
       "/10kUD9S/hZ7tCrxss4dpr9RwQAJQ+UAAAE3QZrwSahBbJlMCCf//rNr0mKACJt5zzEzTn3shp5Y\n",
       "XIDbs48QrsqDn0jYXdgKcMBeGM7BfaEA6ehFX1+wHKy8j/j8oEBwX0NcU4NWTrqMoFXo+ZU9NWQX\n",
       "DvinMTnWNCwO97sfT5dCdLuVzA04Qec8eYgrgLVt69vK9CV3aEQC0bd7+xRJARaQ4G7MsDVSwUqw\n",
       "XwaxujvONV49tyo4+ngyws1KSK8QV0DCY8wQ6H7001HTMoqtYO4rCNQHNJi4ipJ34wBHCfVyoXUD\n",
       "rRrJkR2FLDS6+fbb48/z0Tt4DvctlJa0/X7AD4+a+L78wRcqkYp2KMJ7wOu7GjSRkFmUyJpNERkk\n",
       "kXklRS5qVS8J3VA7tyGhKXR0bCqif5HhL9+TUjZfbjKD5KTzWlUIdy6OfD/S7fk9nE6phK0AAACW\n",
       "QZ8ORRUsEP8Sz7QrlN3YAE0k/zF7OAk5a84lv8sZWClkieOxDCXgmh2iyWgjRCstrhqrjWZeNeGj\n",
       "SX0cCma6hBxaADAQy+pBxUsHvP8nwtasgX94gXoRWrJ+6G1716+MBwABn3s6FuxRtGDIAHIq7iW6\n",
       "FzFeNgHaQ3O/vDR5Ja5LvECm65SEhSYul0uQF8IRonE21sqBAAAAUQGfLXRD/xqOqc+wAfyXb9Hn\n",
       "Wat1IItZeyQPF/e6VsY/vXRTUzvflXnO6ZVK5g0BJQy5lzV5NTzDoFN5EMYfquP3ZIuf8ylXTW5h\n",
       "gyiJr4ve4QAAAFIBny9qQ/8aqEhyMAH9IhKfmD4ZH6XfGvsOFkH+egQruBS/sq5vYikIlELL0X0L\n",
       "cDIjqb7/K2boQzq3951Md4pawkSOV4E3y3m/QS3HzypPrgHAAAAAxUGbMkmoQWyZTBRME//+s048\n",
       "XgAImZsUJlXUa/QlF8rCAllspdNlst3+AZk2p3wm9E1ieZD43U7FCTOusy9A4pRb/j2YCKq5pzl5\n",
       "dtGDrIIT7WJSmVYVMSmOWvR6nY/fpdGop+P7/7BYxTbfjeyXBinUGRp6g+L9qEB+SDgAH2f1MSLo\n",
       "tqX38ULW/5Rh6wb9fP4PdsFhg9nrQNQ2G8vk8SPPBxErK68o7RetQfCzoWCd43K8VPvDpnSMb0wu\n",
       "hM99v8/iGBqQAAAAYAGfUWpD/xsKB1ABNUnTmrZ5LN+QfcSKDK3Jbh5/pPKYAPga6PcXZDxWv4SF\n",
       "+6bHogqoTTX0gaFu0OmpWnVVOvejNrEgwhS9ndbfD70ScL8xaH1cHmQJ6rxJa3FbKSd6vwAAAYdB\n",
       "m1ZJ4QpSZTAgn//+sCaPo3b98UAUHd+05fzp7HGFf/j+/Dc8vwhalw03yp25wczJuqsdRXQ8qBPo\n",
       "CrbFP8hDjP6F4Kql6H+3EE2zSTK4Qr8IOPK2ciP32AUjTFZWmfEAjQvd2r916c7DfYOhNMsu4M72\n",
       "jDa3/W5pfsnCJ4ag58mCbwHEoaMvROkYX3INhDvjEz7r8rRAfcisiBEf9kmjmuv4S9Rhl5PIoycE\n",
       "coZnOKhIwW0HMlwav2IRH8y2Q3QSrwZOQ81tFprsNhkgt8Ug+Cmdi0WvD/yyRHcAwQ2hujcWEEyb\n",
       "qP0hWo3uvZjFKi/9G/bE4lS1UmdNNQnogyxE0SvtNOVs/QLdmNAFM6155qIt9d3aLZuvHrP1HJSb\n",
       "32ULwAZIWEZ5xEEluEjISCl1T4yW0drw7wYabszCz7gWBSZntpjK2XtxF8o+zA9nH41cbhzwcaLn\n",
       "NJahhf9YMHN8uORxoJrs65rdpjSQWnTiqlUDccmAiTOwqlhkB41ktEFlZIyhVhHzAAAAcUGfdEU0\n",
       "TBD/GmNF32g4AD+kB1JZ2r95WtFV5FDlkv3LOwsgXC7ajQ85m1A3QT51w+U7IwMBuI2Be2s52JsU\n",
       "4hJZYn//w4GZ54ZXokdpwu0V6LRu5MAx5PF7Am984Pm2OnGeMM52qNlIP2A0fMdD7a4sAAAAWQGf\n",
       "k3RD/yMIz9juiAA/laP2grcJpyUM3zW6frp8k0O5euSoTgOAAP8JqO5CmybCgiwh566g/u6sjEyt\n",
       "pnbCZhT8W16jUkeks4TRwPQDbIHuu/p1YlalfF3BAAAAZgGflWpD/xye+wAlq5o5ltW+CQzcF5Dt\n",
       "xooQFUsJg31S2YXhRukCwuqnm6o38nUC83nhFcKe6pO3+iwvgg7pUDkeBdffxriMhkfRUpjH0+2r\n",
       "reqFKa1cyNkdTLddzqmUZFc/NMiGBAAAAJpBm5pJqEFomUwIJ//+s2ODVNACMr7AgUys4+MRU8sS\n",
       "/h4p4LCDLQUEHXgKIqZsMTSR3hq4I0ofWhyni+zKg3aftBwEwuAtYae1SAqXUOdsX/CjtbQ9jdqS\n",
       "we4hl7kz/bpnqe2IxDyNac9wsId4JkFejSgT1rD8XSAIQXt6uSy3+0u7LgNjuB3/zZNeOk2Iox6N\n",
       "iDBfK+zlr3dBAAAAcEGfuEURLBD/Euf5TzcAE7fMNFtPLns8G7Cf08vgXMqJLTjowGEoZjAQGGFa\n",
       "Yga7CX60d2rSxYa09dmBgzeVHqJlP9VZQmWN+oeQbWw7nTW9eeUFFW1nQZYqLtGZCvuyd012Odti\n",
       "MpbsddSPhLtgC8kAAABAAZ/XdEP/GsRHe9ABOylxjEWs15lkiYQ7F+qCc3zwP6l+kZ+RkKdgvhGA\n",
       "XK75/j1uphmZiW5W1TuJmv+Lox9I8AAAAEMBn9lqQ/8cnvsAH9IhJ4b0Ej2SJT5RuI5i1d2SJN7i\n",
       "aEiW+U4nkKQtchqTh/qsKuLCr+Vy4w5zKAqzD+AcEERB0AKbAAAAe0Gb3EmoQWyZTBRMFP/+1YhW\n",
       "L4AETDx+rZ0g/8dUJ8cuwg+yfxTTy1izKWo7sJaZpS10WLfZY8tT+hXEUa2vV+9KEfp/kyqutWiv\n",
       "ncw09+2ugyziXuJAVhf+cVc8yb+HK8YIMgSwMP0SA5ir5zh1q7S4j78RFDMl3/e7oAAAAGABn/tq\n",
       "Q/8bCgdQATVJ04q7G1zbL6GCykRvidSnBqdE+Et+BigRpLbvluE+y0AFrrQbhTCnyKkY7lOnGPXM\n",
       "4ft/c1T69O1BqngSaUCG9y4tIYEsmnDkIQb30xNrENQ5/B0AAAEbQZvgSeEKUmUwIJ///rBTt4oA\n",
       "WG0sSFxtXNXcUxeMhyBevb9MlP5ZLJyzzxpD1YQBKLTEWuwhBXl21EmRCeGebayGqUSxqGxXWNDs\n",
       "h7QlYF9nf9GOWiowURQ9tSc0IVItNxcQqSg2SwOcjuIryjZ2P0NLdqOWdaJIv5M/gtF6WF+z6HsR\n",
       "3ZAwu9TX6J0rkC4YMGh8SyOEwdPPSxtUnGyfASh2LVhOwoXS1r9QKF0MF6v7YWh9TrJ6Vhxy6NnF\n",
       "p+BSAWk6Q6gCWGRDq5Dd+fYvsp8evU62yM9wJplO6RtouD14kizjy/FehxvJJThWE/H/eP5WvxoT\n",
       "kBG/3vcyzJWN/PodvnG5UDiYlVRszP7etBiFAR3NoQj50vA3oQAAAF1Bnh5FNEwQ/xpjRZtn+AE1\n",
       "awpgQCriErMlnETPZ9fO+TcJpbp5ga76a/5QnXED0vIH4XwC5JHlZRBGhx1iEHcf8cPyGcWc/QXK\n",
       "YLO5VQ1Vj97Xkv6UJwi6q4R062gAAABZAZ49dEP/GhL2b7AB/K0fup9RnQbuN7aRi6JIdESlK76b\n",
       "NrqpuW/R1YZkYePafCVLBkoggC1/DSZ5bqOxMxcqcTUST9XwZeV6vJeX3n7kYrbkA4c6GqxObFwA\n",
       "AABcAZ4/akP/HJ77ACWrsrzLxUVLSJQLYFLbbawBMYSr3RnuNdjqS3dnVg0pJ++A3Y/7R6P0QAeq\n",
       "i7erN9fRsE9qXEQBfKRfyz6fXpSq4MP2hJ6QH2DUhWXPNms4/f8AAACJQZokSahBaJlMCCf//rNr\n",
       "0mKACJqLlrmbybG0Ve4EoImX8xc2EGTzEtJxcv8BAUIKi4awkj+jRomH+9P6y1EDhFnK2T3WslUS\n",
       "ldkM6BR0Q1meNOQ+VaJSgAq9PAe9ye0kF7x82TqkKUH4V7z+d7w615lXMmiDKMpqg3cZR1Ey9nkE\n",
       "dd6pVo/qx4AAAAB1QZ5CRREsEP8P+STC0XVAATt8mgtp5ctO7Mr75EYog7iTi6f24PyQ/RdaNfOm\n",
       "gIqVgmk9SIDDSGiCZepOBsmD8sv75mkgpoT8cYPlo0pg2loXYcmO/mpDL4UjVzvJirWhTZT3u8Lf\n",
       "7Rn6QjS7+37UuF8Jg5uVAAAARAGeYXRD/xqOqWOZeQ0n70V9b3oAIcS0m458PrJg6pa2jMbQpeaV\n",
       "3u1shik9E6Otg40+dWj1vff9GA6IT96ibw3GCsi4AAAATQGeY2pD/xye+wAf0iEndxJn2Mg4G43P\n",
       "yLuWTsLdVucT9O2+CWXQN3bl2ysr9093pQVhBKOi3L06Fxov378qEJhqMgiIZ7/z1t+fiCjLAAAA\n",
       "ikGaZkmoQWyZTBRMFP/+1YhWL4ADdeU6aTr6NCTPZlX+5UEXlqurwfndqNjR23mAOLOgUT8xJXx6\n",
       "aLFoKjf4e4Tz1rrHfHxKLIWnmnDw8s3rLwSlnSjTJSqhQIKbS1A1guN1bUNGrRWrIIFQP4sM974v\n",
       "QONzMLZg07LYwSS4W/iIEmhNGZFk9uQLgQAAAGYBnoVqQ/8bCgdQATVJ05q2eSaIMn3EigytyVM/\n",
       "qt7CYAPggZoNLmavRrH8JDw+LIha3IvoF0xVoKTBL/qzyy1xZReAvxG6ciTX1F2PEaA3d8BCHlga\n",
       "ZAPokzPQEDMkyBK/9Aqc8m8AAAEAQZqKSeEKUmUwIKf//tQ2TNnfFAQrfg0qQeh9fKlsubxhoqkS\n",
       "gBJmUxyW/uwCxPLQ38T3moeeqhXAagi/abSvd4gvKmWzAZIg6OFzKAKYZpX+X8IDVsX+gQE2bBmh\n",
       "F9rcAAssgNXvakJONuYnmRXjFODf2haisfYtDOFUZZTKMHng4dq0tVUGJh/Dzix07cC91ZpuA/nQ\n",
       "EBR8zMUEzPpNuxfj+F/hPxcQg9iTSGsrDt8RoJAsBdgO7Me+jDV18/FRFN+AvfF6+f9hjBdkHjw3\n",
       "ed5M60jjFEH4bFFnRDzXUgGQ1UKaNM4khwjquDsM0Z6QUikzlBN+QN9GX/8Z62TlMQAAAGhBnqhF\n",
       "NEwQ/xpjRd9oOABLV5xSWdsdDla0h3kUOWS/U/kXDrjGb9pKZkcYzPML6KcDupUjTai0U2Ox5Rt1\n",
       "Wc/KWJ4ce6rzid1LoG23rT2qnzbEEcFMHjSXGu1DHEuWzCHOzLRYWR/2YAAAAFQBnsd0Q/8jCM/Y\n",
       "7ogAP5WnjmpYxvn+TqepTvVqBeW99ynOrJJr3dG3k59C/wclBvt9co5Qk3nricIN8Xg8UmQ1ZLTD\n",
       "mFml6pP2u3JTVjY2ofZBncAAAABfAZ7JakP/Go5e2sSxGAEsoeWEiegvz7Py2GtlKRX4fMrzVFoM\n",
       "4XHimeDdNdy8mqeOxYL4nbtGMRtmcXaCiM/YaJWgZYUlQntUxQsFSUbFpITeh8YtZ9pNillRFUd7\n",
       "7aEAAACPQZrOSahBaJlMCCf//rNjg1+KAC6qLdGsw3QSiFun9sIOyPz9p4pgE92t+QmaygRwWxaH\n",
       "u0Czpzkr5fMuiVdKfxwLiDg9HqirYGJ6+UzxEPfrmovcL2v+z3utOjO0ERB1jRP6qCpJMNh1PR4H\n",
       "iEv9qotbiQTiNPaEG3DyioB1yH/A1vyFhuQubnMnBEd+UZUAAABwQZ7sRREsEP8S5/dHdgATV609\n",
       "K08uez9a2/GqDO2gzGT6ZAv1hhXDdvLqSVzxYiVDiBxmXvjBKVarLUKDMVq1usMEGHBBwN81AjIt\n",
       "2BNtRXJ+nyBRB4ZMNa7gyt6L041H6XPNB/4bG1MpotfY5ADFgAAAADsBnwt0Q/8axEWMcmUrH9LK\n",
       "oR0AEN512DWpuOu22afcb8t1GUoEbrqjDIYJ/U1IGDK9yCKc7LLjy2AlYQAAAEUBnw1qQ/8cnvsA\n",
       "H9IhJ4cC0Y30iU+UbiOYtSdkiTe+ZCM/nFuALJ/IWLIKSTzDwjZlKUxE/BV1sNT5ki66QkNFIHQA\n",
       "poEAAAD/QZsSSahBbJlMCCf//rEUjCEFADpbzsTj7vqxfsEM2SrqDT2dp/FKb1Sk1fyTSt8aVdCM\n",
       "3USN84ZkWifHNFeLNFS0+d9T+EZdAAChKr8k5MVDZ7pEtC6xNTGMtwG2DaH2Fjjgaat46IlPMXKt\n",
       "Rw53soWwFhC2ZenPHA3vgtRKjsUTBoAwGWwTbgtahDXQSr2WqDUR2FtcCNE2G2w2Kqb9E3m+FnDY\n",
       "tq6iGeW1od30ACgwOtNK0phwfvu2qZYVTZVkxOFqQTFjAC4wluZk1qe34V0Iaj7TrcnDM4aUq5xE\n",
       "ErHtGPJpwuz+C1e5cU3waRfmIajL8qruxKMMUepjbmVxAAAAQUGfMEUVLBD/GnxnjMp/NXhFcxaQ\n",
       "AAQ4mqY/RZ9FI97NYPBooa6lWi3owz04wuunqXN11ARtXb3c8K/y8uIv0oTMAAAAYAGfT3RD/xye\n",
       "+wAf0iErvrnvYs0d7egekLbHHgxHtfU9KhfFXTE9BnULAFh3KfFfWAzLm4EUk8YkCEtiaW9X631Z\n",
       "Das1avvL1Nuu9dw3aSwvjtSA41WtjNXDuSCq/j8LaAAAAFYBn1FqQ/8i9mfsd0QAH8vEKQiQxD3a\n",
       "WnfKcavUReE1nnjyE0X1BSzH3IHF/oJEIGlCa7ygW8YLe+lPQK0CZb40n224qkT2r5a3vqgCHqoP\n",
       "L6NwhCNiwQAAAN9Bm1ZJqEFsmUwIJ//+s3Bw+0uZwmgAnY2H9m6mt30gCMoRr3UIGERVI6EzmHKc\n",
       "f/KdJMEaOa09TAYmdyOoAAFMluQcpCicKUNSTDEl4Dz9MZcSDz382EKA1KG01lnUSZxpYgp+DS+J\n",
       "Vrc6xmPbxUXoShFBWimTw/rbLHrZYXRT81vryMFylX/RJnB/fe7apljIEw97ImrLAa1toEGvFqJh\n",
       "JqdgECT+cYmcUio3z5+zgXnp3TSFrjnJs+8VW/EXk70RlxVDqSj/HMSvpYMwPLrmjUuCjlKg9+KM\n",
       "hyZrWMC4AAAAakGfdEUVLBD/EvyEn2AAS1Ykqxl9VBugJqk/T24/WeWyL4kyv5pY2nNkaiJz35D7\n",
       "wXbWDDocUoffjmAqCXZ3z22SjeI47N0ocfMZlF/QgPbLuB8ygi7nwOJI2rfdlBTdbDVlo+DncdtQ\n",
       "J8AAAABMAZ+TdEP/GqtxgZGAEsiA4OSAl7rKq9hEjKas/4bmBbbZujzE3V8luAZek4qSACFNEW/H\n",
       "7P4eqAfUbnRHo8WIjex+cMNIAFJjzG2hMwAAADsBn5VqQ/8avzgOm1JyfQANkT9ux/wbO/SkTmRB\n",
       "TfF1YtsdadQmmmKevMuC8ZSMreMOdrPmZYDj8MADegAAAGhBm5pJqEFsmUwIJf/+s0pgRgALe2lm\n",
       "XBlPYQb+C3WKWNpn+FacPb0PvHXCp0KsJw6dML897EAvtLnSB8ORNfyM7VN/tQm4CGcrrx+zkwxV\n",
       "EtLKI21LFtY5fk8/4iqvXWVSEatmnqyQIwAAAE5Bn7hFFSwQ/xKJFtqgALqe6e7dezZ0fXVhiWwc\n",
       "qk3bXKnImg4eD9uzpS6ekFA7uqG5XHsmCJgjDvgjS3w4286dpe9tV6MARZj3vC/DcCEAAABGAZ/X\n",
       "dEP/HJ77AB/SISeG9BI9+nDH3PDOYza7ZIk3ueKxE+Rxw9gpFdkFL5ur7GOwayOah6N96+97EX7Y\n",
       "BzOrfffpD/gBEwAAAF8Bn9lqQ/8cnvsAH9IhK766elI/Lvb0D0hbY48GH6D3poJb4pvfF3Wr0NDY\n",
       "KQSKy47i52nC1RxjZLMBMXFSwWd8AAxms6fdXCe3THd7mRfMocnyI2XxdE0BSXS53/ACJwAAAKBB\n",
       "m95JqEFsmUwIIf/+oQYMAAAiZGZJE5rU80+7Jg7QsqIXwcPCKZrQ+4mvXi//djcQUgjFtKX7X1fm\n",
       "Jyv04tJhIzaMnCgr6KKnydtkRc0pFSem888sTuSORYg7aEIRFQ/1RKjaLbdCte5DCobW4eQ4/ngw\n",
       "DLl1dscmF6XQKW/BIYTbVEKeetCCFIWL+l8XBIal0lDDLm6TDSOPjC3ezsMoAAAAWEGf/EUVLBD/\n",
       "GmNFm2f4ATV4vtgA4F1mLseMlIVjr3o6xhccfjAk30n05Ep3j3r2mnKurQuCVrN+Vk0a9AS6ZQj3\n",
       "U4TsTSlhgShnB8quOLSiSsVPXKQAF3EAAABRAZ4bdEP/IwjP2O6IAD+jLCkIjbqx4PX7JLycIBrB\n",
       "HPqysW0ZOkOblqfAk7qztNbBjl6yTU6qHBGFrys9aK5hfCZ1j7fnpt9IYv7trzCdABDxAAAAYQGe\n",
       "HWpD/xye+wAlq5o5l0thdLS2hycaVumi3bZGtTT+5us1o5ICkDMo+aEmtxMwS3z0YypySVXW2Dqt\n",
       "XTjSW6p5z7NYRjtN8f5B0urv6ntmK/1fhRiXc8UXcX1LCyoAG9AAAAA8QZofSahBbJlMCH///qVu\n",
       "luLWDA8xTQAbInj5/MB88gED5JXvSsr9e1fkuHPGzqn/mMPY4yy3sIJzNz1gAAAGMm1vb3YAAABs\n",
       "bXZoZAAAAAAAAAAAAAAAAAAAA+gAADIAAAEAAAEAAAAAAAAAAAAAAAABAAAAAAAAAAAAAAAAAAAA\n",
       "AQAAAAAAAAAAAAAAAAAAQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAIAAAVcdHJhawAA\n",
       "AFx0a2hkAAAAAwAAAAAAAAAAAAAAAQAAAAAAADIAAAAAAAAAAAAAAAAAAAAAAAABAAAAAAAAAAAA\n",
       "AAAAAAAAAQAAAAAAAAAAAAAAAAAAQAAAAAGwAAABIAAAAAAAJGVkdHMAAAAcZWxzdAAAAAAAAAAB\n",
       "AAAyAAAAEAAAAQAAAAAE1G1kaWEAAAAgbWRoZAAAAAAAAAAAAAAAAAAAKAAAAgAAVcQAAAAAAC1o\n",
       "ZGxyAAAAAAAAAAB2aWRlAAAAAAAAAAAAAAAAVmlkZW9IYW5kbGVyAAAABH9taW5mAAAAFHZtaGQA\n",
       "AAABAAAAAAAAAAAAAAAkZGluZgAAABxkcmVmAAAAAAAAAAEAAAAMdXJsIAAAAAEAAAQ/c3RibAAA\n",
       "ALdzdHNkAAAAAAAAAAEAAACnYXZjMQAAAAAAAAABAAAAAAAAAAAAAAAAAAAAAAGwASAASAAAAEgA\n",
       "AAAAAAAAAQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABj//wAAADVhdmNDAWQAFf/h\n",
       "ABhnZAAVrNlBsJaEAAADAAQAAAMAKDxYtlgBAAZo6+PLIsD9+PgAAAAAHHV1aWRraEDyXyRPxbo5\n",
       "pRvPAyPzAAAAAAAAABhzdHRzAAAAAAAAAAEAAABAAAAIAAAAABRzdHNzAAAAAAAAAAEAAAABAAAC\n",
       "EGN0dHMAAAAAAAAAQAAAAAEAABAAAAAAAQAAKAAAAAABAAAQAAAAAAEAAAAAAAAAAQAACAAAAAAB\n",
       "AAAoAAAAAAEAABAAAAAAAQAAAAAAAAABAAAIAAAAAAEAACgAAAAAAQAAEAAAAAABAAAAAAAAAAEA\n",
       "AAgAAAAAAQAAKAAAAAABAAAQAAAAAAEAAAAAAAAAAQAACAAAAAABAAAYAAAAAAEAAAgAAAAAAQAA\n",
       "KAAAAAABAAAQAAAAAAEAAAAAAAAAAQAACAAAAAABAAAoAAAAAAEAABAAAAAAAQAAAAAAAAABAAAI\n",
       "AAAAAAEAABgAAAAAAQAACAAAAAABAAAoAAAAAAEAABAAAAAAAQAAAAAAAAABAAAIAAAAAAEAACgA\n",
       "AAAAAQAAEAAAAAABAAAAAAAAAAEAAAgAAAAAAQAAGAAAAAABAAAIAAAAAAEAACgAAAAAAQAAEAAA\n",
       "AAABAAAAAAAAAAEAAAgAAAAAAQAAKAAAAAABAAAQAAAAAAEAAAAAAAAAAQAACAAAAAABAAAoAAAA\n",
       "AAEAABAAAAAAAQAAAAAAAAABAAAIAAAAAAEAACgAAAAAAQAAEAAAAAABAAAAAAAAAAEAAAgAAAAA\n",
       "AQAAKAAAAAABAAAQAAAAAAEAAAAAAAAAAQAACAAAAAABAAAoAAAAAAEAABAAAAAAAQAAAAAAAAAB\n",
       "AAAIAAAAAAEAABAAAAAAHHN0c2MAAAAAAAAAAQAAAAEAAABAAAAAAQAAARRzdHN6AAAAAAAAAAAA\n",
       "AABAAAAXsQAAAhIAAADqAAAAvwAAANoAAAHpAAAAyQAAAIIAAADCAAACFgAAALgAAACBAAAApwAA\n",
       "ATsAAACaAAAAVQAAAFYAAADJAAAAZAAAAYsAAAB1AAAAXQAAAGoAAACeAAAAdAAAAEQAAABHAAAA\n",
       "fwAAAGQAAAEfAAAAYQAAAF0AAABgAAAAjQAAAHkAAABIAAAAUQAAAI4AAABqAAABBAAAAGwAAABY\n",
       "AAAAYwAAAJMAAAB0AAAAPwAAAEkAAAEDAAAARQAAAGQAAABaAAAA4wAAAG4AAABQAAAAPwAAAGwA\n",
       "AABSAAAASgAAAGMAAACkAAAAXAAAAFUAAABlAAAAQAAAABRzdGNvAAAAAAAAAAEAAAAwAAAAYnVk\n",
       "dGEAAABabWV0YQAAAAAAAAAhaGRscgAAAAAAAAAAbWRpcmFwcGwAAAAAAAAAAAAAAAAtaWxzdAAA\n",
       "ACWpdG9vAAAAHWRhdGEAAAABAAAAAExhdmY1OC43Ni4xMDA=\n",
       "\">\n",
       "  Your browser does not support the video tag.\n",
       "</video>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib\n",
    "from IPython.core.display import HTML\n",
    "\n",
    "matplotlib.use(\"Agg\")\n",
    "\n",
    "Writer = matplotlib.animation.writers['ffmpeg']\n",
    "writer = Writer(fps=15, metadata=dict(artist='rskene'), bitrate=1800)\n",
    "\n",
    "anime = dn.loss_progress(history)\n",
    "anime.save('nnet_norm_fit_sp.mp4', writer=writer)\n",
    "\n",
    "HTML(anime.to_html5_video())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Fitting the Phat #\n",
    "\n",
    "## Failure of a Standard Loss Function ##\n",
    "\n",
    "Of course, daily returns on the S&P 500 are not Guassian ... or, at best, if they are Gaussian it means we are living in a one-in-$10^{100+}$ universe where a dozen or more six sigma events have occured in the past 100 years. Fans of the Many Worlds interpretation would agree this is entirely possible.\n",
    "\n",
    "Nevertheless, we will explore the fit of the Phat distribution, utilizing a network similar to that employed in [Carreau and Bengio (2009)](references.ipynb). We will first test our model against a generative version of Phat, with parameters chosen to mirror that of daily S&P 500 returns.\n",
    "\n",
    "We will use the negative log-likelihood of the entire Phat distribution, a standard loss function used for most probabilitiy distributions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "genmod = ph.Phat(.0003, .0032, .17, .19)\n",
    "\n",
    "n = 60000\n",
    "y = genmod.rvs(size=n, seed=16433)\n",
    "data = ph.DataSplit(y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "tags": [
     "hide_input"
    ]
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/by/3p7tzvtd0_9cn60snk5cv8g40000gn/T/ipykernel_73140/2489721303.py:7: UserWarning: Matplotlib is currently using agg, which is a non-GUI backend, so cannot show the figure.\n",
      "  plt.show()\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x432 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, (ax1, ax2) = plt.subplots(1,2,figsize=(18,6))\n",
    "ax1.plot(data.train_raw.y)\n",
    "ax2.plot(data.test_raw.y)\n",
    "\n",
    "ax1.set_title('Training Set')\n",
    "ax2.set_title('Test Set')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We developed a second custom network for this process, `PhatNetBeta`, containing a few key changes.\n",
    "\n",
    "1. we reduced the nodes in the hidden layer to just one. the input values are all zero and so their weighting is valueless. the bias from the hidden layer is the only meaningful input to the parameter layer.\n",
    "2. the hidden layer has *no activation function*. the `x` values provided are all zero. They cannot be \"activated\".\n",
    "2. we have added two additional parameters, `shape_l` and `shape_r`, representing the tail indices of the left and right-tailed Pareto distributions incorporated in the Phat distribution.\n",
    "3. the loss function is now the negative log-likelihood of the Phat distribution (available as `BodyLoss`, referencing the fact that it over-fits the body of the distribution).\n",
    "4. the `PhatMetric` class instantiates a metric for any one of the Phat parameters in the body and in both tails (which tail must be specified).\n",
    "4. a number of operations were pushed lower-level for convenience"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from phat.learn.phatnet import PhatNetBeta, PhatMetric, BodyLoss\n",
    "\n",
    "dn = PhatNetBeta(neurons=1)\n",
    "dn.plot_model()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 00017: early stopping\n"
     ]
    }
   ],
   "source": [
    "metrics = [\n",
    "    PhatMetric('mean_left'),  PhatMetric('std_left'), \n",
    "    PhatMetric('shape_left'), PhatMetric('shape_right'),\n",
    "]\n",
    "dn.compile(loss=BodyLoss(), optimizer='adam', metrics=metrics)\n",
    "history = dn.fit(\n",
    "    data.train, \n",
    "    validation_data=data.test,\n",
    "    epochs=100, \n",
    "    batch_size=32, \n",
    "    verbose=0\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can compare results with the generative model:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "nbsphinx": "hidden"
   },
   "source": [
    "```python\n",
    "pd.DataFrame(\n",
    "    (genmod.args, phat_fit.args), \n",
    "    index=['Gen Model', 'Neural Net Fit'],\n",
    "    columns=['mean', 'std', \n",
    "        r'$\\xi_l$', r'$\\xi_r$', r'$a_l$', \n",
    "        r'$a_r$', r'$b_l$', r'$b_r$',\n",
    "    ]\n",
    ").T.to_csv('phat_fit_no_tails_comp.csv', index=False)\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "</style>\n",
       "<table id=\"T_d45ce_\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th class=\"col_heading level0 col0\" >Gen Model</th>\n",
       "      <th class=\"col_heading level0 col1\" >Neural Net Fit</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_d45ce_level0_row0\" class=\"row_heading level0 row0\" >0</th>\n",
       "      <td id=\"T_d45ce_row0_col0\" class=\"data row0 col0\" >0.0003</td>\n",
       "      <td id=\"T_d45ce_row0_col1\" class=\"data row0 col1\" >-0.0003001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d45ce_level0_row1\" class=\"row_heading level0 row1\" >1</th>\n",
       "      <td id=\"T_d45ce_row1_col0\" class=\"data row1 col0\" >0.0032</td>\n",
       "      <td id=\"T_d45ce_row1_col1\" class=\"data row1 col1\" >0.003883</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d45ce_level0_row2\" class=\"row_heading level0 row2\" >2</th>\n",
       "      <td id=\"T_d45ce_row2_col0\" class=\"data row2 col0\" >0.17</td>\n",
       "      <td id=\"T_d45ce_row2_col1\" class=\"data row2 col1\" >3.899e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d45ce_level0_row3\" class=\"row_heading level0 row3\" >3</th>\n",
       "      <td id=\"T_d45ce_row3_col0\" class=\"data row3 col0\" >0.19</td>\n",
       "      <td id=\"T_d45ce_row3_col1\" class=\"data row3 col1\" >4.822e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d45ce_level0_row4\" class=\"row_heading level0 row4\" >4</th>\n",
       "      <td id=\"T_d45ce_row4_col0\" class=\"data row4 col0\" >-0.001064</td>\n",
       "      <td id=\"T_d45ce_row4_col1\" class=\"data row4 col1\" >-0.001746</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d45ce_level0_row5\" class=\"row_heading level0 row5\" >5</th>\n",
       "      <td id=\"T_d45ce_row5_col0\" class=\"data row5 col0\" >0.001684</td>\n",
       "      <td id=\"T_d45ce_row5_col1\" class=\"data row5 col1\" >0.001145</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d45ce_level0_row6\" class=\"row_heading level0 row6\" >6</th>\n",
       "      <td id=\"T_d45ce_row6_col0\" class=\"data row6 col0\" >0.008784</td>\n",
       "      <td id=\"T_d45ce_row6_col1\" class=\"data row6 col1\" >0.01043</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_d45ce_level0_row7\" class=\"row_heading level0 row7\" >7</th>\n",
       "      <td id=\"T_d45ce_row7_col0\" class=\"data row7 col0\" >0.008807</td>\n",
       "      <td id=\"T_d45ce_row7_col1\" class=\"data row7 col1\" >0.01043</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x1929de5b0>"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mean_, std_, shl_, shr_ = dn.predict([0])[0]\n",
    "phat_fit = ph.Phat(*dn.predict([0])[0])\n",
    "\n",
    "df = pd.read_csv('phat_fit_no_tails_comp.csv')\n",
    "df.style.format('{:.4}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As with the [MLE](#mle_fit.ipynb) fit, the neural net approach leads to significant underestimation of the tail index, driving them to near zero."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see why below. We compare the change in the log-likelihood for changes in the different parameters. Changes in the mean have a clear absolute minimum, changes in standard deviation are actually asymptotic to declining loss, and changes in the tail are linear across a very narrow range.\n",
    "\n",
    "So $\\xi$ can be turned all the way down to zero to the benefit of the loss function."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "tags": [
     "hide_input"
    ]
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/by/3p7tzvtd0_9cn60snk5cv8g40000gn/T/ipykernel_73140/2497768408.py:26: UserWarning: Matplotlib is currently using agg, which is a non-GUI backend, so cannot show the figure.\n",
      "  plt.show()\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x864 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "\n",
    "y = np.linspace(-.1, .1, 1000)\n",
    "siglin = np.linspace(0.001, 0.01, 1000)\n",
    "shlin = np.linspace(0, .5, 1000)\n",
    "\n",
    "shl_pdf = [ph.Phat(mean_, std_, sh, shr_).pdf(mean_) for sh in shlin]\n",
    "shr_pdf = [ph.Phat(mean_, std_, shl_, sh).pdf(mean_) for sh in shlin]\n",
    "\n",
    "shl_nll = [ph.Phat(mean_, std_, sh, shr_).nll(mean_) for sh in shlin]\n",
    "shr_nll = [ph.Phat(mean_, std_, shl_, sh).nll(mean_) for sh in shlin]\n",
    "\n",
    "\n",
    "fig, ((ax1, ax2), (ax3, ax4)) = plt.subplots(2,2,figsize=(18,12))\n",
    "\n",
    "ax1.plot(y.reshape(-1,1), phat_fit.nll(np.tile(y, 1)).T)\n",
    "ax2.plot(siglin.reshape(-1,1), ph.Phat(mean_, siglin, shl_, shr_).nll(mean_).T)\n",
    "ax3.plot(shlin, shl_nll)\n",
    "ax4.plot(shlin, shr_nll)\n",
    "\n",
    "ax1.set_title('NLL Relative to Log Returns')\n",
    "ax2.set_title('NLL Relative to Deviation')\n",
    "ax3.set_title('NLL Relative to Changes in the Left Tail Shape')\n",
    "ax4.set_title('NLL Relative to Changes in the Right Tail Shape')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Below we can see just how sinister the lack of a tail index can be. The fitted model, with no tail indices, appears to be as good a fit, if not better, for the random samples than the actual model that created them!\n",
    "\n",
    "It is only deep in the tails that the larger tail index plays a role."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "tags": [
     "hide_input"
    ]
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/by/3p7tzvtd0_9cn60snk5cv8g40000gn/T/ipykernel_73140/1042267465.py:14: UserWarning: Matplotlib is currently using agg, which is a non-GUI backend, so cannot show the figure.\n",
      "  plt.show()\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlIAAAFoCAYAAACc1hUlAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8rg+JYAAAACXBIWXMAAAsTAAALEwEAmpwYAABpdUlEQVR4nO3deXxU9b3/8dc5syaZJJOdJCTsO7JIFXEBsVZABBS1tVq36/Vq69LV1vqjaq2t1PYWW7Xa3nptr9Xa2qogKu5aFWQTwSA72fd9mX3OOb8/hgyE7CGZyfJ5Ph55JDlzzplPcjLDm+/3e75fxTAMAyGEEEII0WtqtAsQQgghhBiqJEgJIYQQQvSRBCkhhBBCiD6SICWEEEII0UcSpIQQQggh+sgc6SfUdR2Xy4XFYkFRlEg/vRBCCCFEjxmGQSAQIC4uDlVt3/4U8SDlcrk4ePBgpJ9WCCGEEKLPJk+eTHx8fLvtEQ9SFoslXJDVao3000dEXl4eM2fOjHYZog/k2g1tcv2GNrl+Q9dwvnZ+v5+DBw+G88vJIh6kWrvzrFYrNpst0k8fMcP5Zxvu5NoNbXL9hja5fkPXcL92nQ1HksHmQgghhBB9JEFKCCGEEKKPJEgJIYQQQvRRxMdICSGEEIOZruvU1NTQ0NCApmnRLmdIMJvN7Nu3L9plnBKTyYTT6SQ1NbXDaQ46I0FKCCGEOEFJSQmKojB27FiZ87CHXC4XcXFx0S6jz1rniqqsrKSkpITc3NweHytde0IIIcQJXC4X2dnZWK1WCVEjhKIoWK1WsrOzcblcvTpWgpQQQghxkt507Yjhoy/XXf5ShBBCiEEuGAzyxBNPsGzZMi6++GKWLFnCk08+iWEYUamnuLiYe+65B4DPP/+cBx544JTPOWXKFG666aY22+rq6pgxYwaPPvpoj89TUlLCBRdc0OU+jz76aK/O2RUZIyWEEEIMcj/96U+pqanh73//OwkJCbS0tHDbbbcRHx/PNddcE/F6ysrKKC4uBuC0007j3nvv7ZfzFhQU0NjYSGJiIgBvvvkmCQkJ/XLugSItUkIIIcQgVlFRwYYNG1i7dm04VDgcDu69915SU1MBqKmp4Vvf+harV6/m8ssvZ/PmzUCo5WXNmjVce+21XHDBBTzxxBMAaJrGQw89xGWXXcbKlSv585//DMDWrVu54oorWL16NT/60Y+orKzkpptu4qtf/SqLFy/m17/+NQAPPvggeXl5/PSnP2Xr1q3cfPPN7N+/n0suuSRc93vvvcett94KwB//+Mfwcz388MOdtqRdcMEFvP322+Hv33jjDb7yla+Ev//ss8+48sorWblyJddffz2FhYUAfPHFF1x22WVcdtllPP744+H9O/u99CdpkRJCCCG60LznfZp3vzsg546ffQHxs87vcp89e/YwYcKEcCtNqwkTJjBhwgQAfv7zn3P55Zfz5S9/maqqKq6++mpefvllAA4cOMCzzz5Lc3MzF154Iddccw2vvvoqAC+99BJ+v5+bbropvFZeQUEB7733HvHx8Tz11FNccsklXHbZZTQ3N7No0SL+4z/+gzVr1vDYY49x3333sXXrVgCmTp2KqqocPHiQyZMns3HjRlauXMm///1v8vLy+Oc//4miKNx1111s2LCBVatWtftZly1bxpNPPsnll19OdXU1hmGQlpYGhNa8+973vscjjzzCrFmzeP311/ne977Hv/71L370ox/x4x//mLPPPpvHH388XFNXv5f+IkFKCDGs6EE/ismMokiDuxg+Trx7cNOmTTzxxBPouo7VauVf//oXmzdv5ujRo/zud78DQmOqWrve5s+fj9VqJSUlBafTSXNzM1u2bGHfvn188sknALjdbg4cOMDEiRMZN24c8fHxANx000188sknPPXUUxw6dIhAIIDH4+m0zlWrVvHqq6+Sk5PDtm3b+MUvfsEjjzzCnj17WL16NQBer5esrKwOj587dy75+fk0NzfzxhtvsGTJEmpqaoBQwEtISGDWrFlAKHTde++9lJaWUlVVxdlnnw3A6tWr+de//gXQ5e+lv0iQEkIMC4ahE7PvLQreWIs5MZWMy+/CNmp8tMsSw0D8rPO7bTUaSDNmzODIkSO0tLTgcDhYunQpS5cupaSkhOuuuw4ITSL6l7/8BafTCUBlZSWpqam8/fbbbRYTVhQFwzDQNI277rqLiy66CAgN6o6NjWX37t3Y7fbw/mvXrqW4uJhLLrmECy+8kM2bN3c5wP2SSy7h+uuvZ+rUqZx77rnYbDY0TeP666/nxhtvBKCpqQmTydTh8YqisHjxYt555x3efPNNHnnkEZ599tnwz3gywzCIjY1tU9OJ5+7q99Jf5L9sQohhoXHbRuyF23HMOBdD16n4x1o0b+/mgxFiMMrOzmblypX86Ec/oqmpCQiNcXr//ffDt+ufddZZPPfccwAcPnyYlStXdtlydNZZZ/GPf/yDQCCAy+Xi6quvZvfu3e32+/jjj7nppptYtmwZ5eXlVFZWous6JpOJYDDYbv+MjAwyMzP54x//yMqVK8PPtX79elwuF8FgkNtuu4033nij09qWLVvGc889h8ViITk5Obx9/PjxNDQ0sGfPHgBee+01srKySEpKIisri/fffx+AjRs3tvk5e/N76QtpkRJCDHmap4WGD1/AnzaRtJV34is7TNmf76Zx20aSF34t2uUJccruv/9+nn76aa677joMw8Dv9zNnzhz+53/+B4A1a9Zw7733smLFCgAefvhhHA5Hp+e76qqrKCws5LLLLiMYDLJ69Wrmz58fHlvU6pZbbuGHP/whCQkJpKSkMHPmTEpKSpg2bRrNzc3cddddXHHFFW2OWbVqFevWrWP+/PlAaAD5/v37+epXv4qmaZx33nlcdtllndY2Z84cqqurufLKK9tst1qtrFu3jp/97Gd4PB4SExNZt24dAL/61a/48Y9/zCOPPMKcOXPCx/T299IXihHhSSh8Ph95eXnMnDmzTXPjcLJz507mzZsX7TJEH8i1G5oaNr9E3Xt/pemc/2TO+csAqPjnw3gL88j99p9QzdYoVyh6YrC8/vbt28e0adOiXcaQMtSXiDnRyde/u9wiXXtCiCHNMAyaP38f2+gpaPHp4e0J85age124D+2IXnFCiGFPgpQQYkjzV+YTqCkhfuaiNttjxszE5EimJe/fUapMCDESSJASQgxpoRYnhbipZ7XZrqgm4qbOx5O/Bz3oj05xQohhT4KUEGJIcx/djS1zPKa4xHaPxY6fixHw4S3eF4XKhBAjgQQpIcSQpXtd+EoPEjN+ToeP28fMAJMZz5FdkS1MCDFiSJASQgxZnsK9YOjEjJ/d4eOq1Y599BS8RV9EuDIhxEghQUoIMWR5Sw+AasaWNanTfezZU/BV5KP7vRGsTAgxUkiQEkIMWb7SQ9hGjetynih7zlQwdHzlhyNYmRD9p6SkhJkzZ7Jq1SouvfRSli9fzo033khFRQUXXHABJSUlPT7Xu+++y9NPPw3Aiy++yPz588Nr2bU+1wUXXNDlOYqLi7nnnnsA2Lp1K3PnzuWqq65i5cqVLFu2jCeeeKLbOv7+97+3mYF8KJMgJYQYkgxdw1d+GFt2561RALbsyQB4Sw5EoiwhBkR6ejrr16/n5Zdf5tVXX2XmzJn87Gc/6/V59u7dS0tLS/h7l8vFfffd16tzlJWVtVn4d+bMmTz//PNs2LCBf/7znzz//PMcPtz1f1x27dqF3z887qaVJWKEEEOSv6oII+DDfiwodcYUE48lJRtf6cEIVSbEwPvSl77Eu+++C8Djjz/Ovn378Hg8PPzww8yePZtt27axbt06vF4vjY2N3HXXXUyaNInnn38egKysLBRFYcmSJRw4cIBXXnklvIxKK5fLxQMPPMChQ4fQNI2bb76ZSy65hAcffJCSkhJ++tOfsnTp0jbHeL1eTCYT8fHxAOzZs4eHHnoIr9dLUlISP/3pTykuLubdd9/lk08+IS0tjYyMDH72s5/hdrupq6vjxhtvDC/GPBRIkBJCDEm+skMAXY6PamXLnBAamC5EH3yQ/wnv5W8ekHMvHnc2i8ad1f2OJwgEArz++uucfvrpfPzxx0ycOJGHHnqIv/71rzz11FP87ne/469//SsPPvggEyZMYMuWLfziF7/glVde4aqrrgLg8ssv58UXX8RisfDQQw9xyy23sGDBgjbP88QTTzBjxgx++ctf0tLSwlVXXcXs2bNZs2YNjz32GPfddx9bt24lLy8vfN6ioiKWLVtGeno6fr+fNWvW8OSTT5KVlcWHH37IT37yE/785z9zwQUXcOaZZ3Leeefx85//nG9961ssWLCA4uJiVq5cKUFKCCEGmr+yANUWi9mZ0e2+1oyxtOT9G83dhCk2IQLVCdG/qqqqWLVqFQB+v59Zs2bx/e9/n48//pgLL7wQgIkTJ/LGG28AoUV833vvPTZt2sTu3btxuVydnvu0007jyiuv5L777uPHP/5xePvmzZvxer3861//AsDtdnPo0KF2a+rNnDmTJ598kri4OFwuF7feeit//OMfWbx4McXFxXzzm98M73tit2Kru+++mw8//JA//OEPHDhwALfb3cffUnRIkBJCDEm+qkKs6WNQFKXbfa0ZY4FQ+IoZN2uAKxPDzaJxZ/W61ai/tY6R6ojJZAJo81q4+uqrmT9/PvPnz2fBggX84Ac/6PL8t912G6tXr24zAFzXdX71q18xY8YMAGpqakhMTOTTTz/t9DxxcXFceOGFbN68mUWLFjF69Ohw3ZqmtRnY3uo73/kOCQkJLF68mIsvvphXX321y1oHm24Hm7/wwgusWrUq/DFv3jweeOABNm/ezIoVK7joootYt25dJGoVQgggtFCx/1iQ6glbxjgAfJUFA1iVEINDQ0MDBQUFfPvb32bRokV8/PHHaJoGhEJXMBhsd4zVauWhhx7iySefDG8766yz+Nvf/gaEWsRWrlxJeXl5p+eAUFjatm0b06dPZ/z48TQ2NrJjR2jh8H/961/hQGcymcI1ffzxx9x5551ceOGFbN++PXyeoaLbFqkrr7ySK6+8EoBDhw5x2223cfPNN/P1r3+dZ555hszMTG655RY++OADFi1a1M3ZhBDi1AUbqzH8nh4HKVNsAqb4FPyV+QNcmRDR53Q6ufLKK1m+fDkOh4M5c+bg9Xpxu92cccYZ/OhHPyI1NbVdF91pp53G9ddfzyuvvALA7bffzv33388ll1yCpmncdddd5ObmEh8fT3NzM3fddRdXXHFFeIyUqqp4PB5OO+00br75ZqxWK7/97W/5+c9/js/nw+Fw8Mtf/hKAs88+m9/85jfEx8dzxx13cPXVV5OQkMC4cePIzs6mpKSEMWN69vqONsUwDKOnO19zzTV84xvfICUlhccff5y//OUvALz88sts3bqVhx56qNtz+Hw+8vLymDlzJjabre+VD2I7d+5k3rx50S5D9IFcu6HBdXA7lS+sJev6X2AfPSW8vavrV/H3XxBsqmb0zdKCPlgNltffvn37mDZtWrTLGFJcLle7YDZUnXz9u8stPZ5HqnXQ2bJly6iqqiItLS38WHp6OpWVladYuhBC9Iy/qhAAa1puj4+xpGYTqC3H0IdOl4EQYvDr8WDz559/nhtvvBEIjU84WU8GfJ4oLy+vV/sPNTt37ox2CaKP5NoNfnEHPsMU42RXXvs19Dq7ftYWjTgtwGcfv4cemzTQJYo+GgyvP7PZ3OVdbqJjw+V35vf7e/V32KMg5ff72b59O2vXrgUgIyOjzcj7qqoq0tPTe1WodO2JwUiu3dBQ8ulfMWdPYNJJ16qr6+fNcFCW9yqTRzmJmyTXeDAaLK+/ffv2DZtuqkgZTl17VquV2bOPL4Te2rXXmR517R04cICxY8cSGxsLwOzZs8nPz6ewsBBN09i4cSMLFy48xdKFEKJ7hqETqKvAkpLdq+Na9w/U9HxdMiGE6E6PWqSKi4sZNWpU+HubzcbatWu544478Pl8LFq0qN008UIIMRC0plqMoB9LcmavjjPFODDFOfHXlA5QZUKIkahHQeriiy/m4osvbrNtwYIFbNiwYUCKEkKIzvjrygCwpGT1+tjQgHNpkRJC9B+Z2VwIMaQEassBsCT3PkhZU0bTsvdDDMPo9Q0yYmS79v5NNDT7+v28zngbz9zfdY9OSUkJS5cuZcKECUBoxnGXy8Wll17KnXfeeco1vPjii2zbti08Dro/GIbBo48+yltvvYWiKFitVu68884BHQZ09913c+aZZ7J69eoBe46OSJASQgwpgboyFIsdk6P3d95ZUkej+9xorgbMfThejFwDEaJ6c96Tl4iprKxkyZIlLF++PBywBpPXX3+dvXv38tJLL2E2m8nPz+frX/86r776KikpKdEur19JkBJCDCmBujIsyZl9alFqHVcVrK+QICWGtOrqagzDIC4ujmAwyP3338+hQ4eoqalh3LhxPPbYY9TU1HD77bczadIk9u3bR0pKCr/97W9xOp28/PLLPPHEEzgcDrKzs8M3k3322WfhmciTkpJ44IEHGDNmDNdeey3Tpk1jy5YteL1e1qxZwzPPPMPhw4e54YYbwiugnFifpmn4/X7MZjPjxo3jd7/7HWZzKHasW7eOLVu20NjYSFJSEo8++ihpaWmcc845LF68mB07dpCWlsbVV1/NM888Q0VFBWvXruXMM8/k2muvZfz48ezZswefz8c999zDueee2+b5X375Zf7yl7+g6zozZszgvvvuQ1VV7rnnHg4dOgSE1iP86le/esrXoscTcgohxGAQqCvv0/goAEvSqPA5hBhKqqqqWLVqFUuXLmX+/Pk88sgjPPbYY4waNYpdu3ZhsVj4+9//zltvvYXP5+ODDz4AYP/+/dx4441s3LiRhIQEXnnlFSorK/n1r3/Ns88+y9///vfw/E9+v5/vfe97/OQnP2HDhg1cddVVfO9732tTxyuvvMKqVat48MEHefTRR3n22Wd5/PHH29V76aWX0tLSwoIFC7jpppv44x//yLhx40hMTKSwsJCjR4/y/PPP88Ybb5Cbmxtelqampobzzz+fTZs2AfD222/z3HPPcccdd4RXU2mt9aWXXuK///u/ufvuu/H7/eHHDh06xD/+8Q+ef/551q9fT0pKCk899RS7du2isbGRl19+maeffrrLxZd7Q1qkhBBDhqEFCDZU4Zhxbvc7d8CcmAaKSqC+op8rE2JgtXbt6brO2rVrOXDgAGeddRYAZ5xxBk6nk2effZajR49SUFCA2+0GICUlhenTpwMwadIkGhsb2bVrF3PnziU1NRWAFStW8Mknn1BQUEBCQgKzZs0CYNmyZdx77700NzcDhMc3ZWVlMXv2bGJiYsjOzqapqaldvYmJiTz//PMcOHCAzZs38+677/LUU0/xz3/+kzFjxvCjH/2IF154gfz8fD777DNyc4+vUtD6PNnZ2eF5xbKysto8T2tL0rRp00hLS+PAgQPhx7Zu3UphYWF4n0AgwPTp0/n6179Ofn4+N910EwsXLgwvoHyqpEVKCDFkBOorwdD7NNAcQDGZMTvTJUiJIUtVVX74wx9SW1vL//7v/wLwzjvv8IMf/AC73c7q1as544wzwiuQnDjxtaIo4RstdF0Pb2/tbjtxWyvDMNC00LJKFoul3TGdefrpp9m/fz9Tpkzhxhtv5JlnnuHcc8/ljTfeIC8vj5tuugld11myZAkXXnhhmxVTrFZr+GuTydTh+U/crut6m3o0TWPZsmWsX7+e9evX88ILL3DvvfeSlJTEq6++yje+8Q3y8/O57LLLOgyBvSVBSggxZLR2yfU1SEGoey9QJ0FKDF1ms5kf/vCHPPnkk1RXV7NlyxaWLVvG5ZdfTmpqKtu3bw+Hn47MmzeP3bt3U1lZia7rvPbaawCMHz+ehoYG9uzZA8Brr71GVlYWTqez1zU2NzfzyCOPhLsNPR4PpaWlTJs2je3bt3PmmWfy9a9/nYkTJ/Lxxx93WW9HWmv+/PPPaWpqYvLkyeHH5s+fz1tvvUVtbS2GYXD//ffzl7/8JRw4zz//fNasWUNsbCzl5afezS9de0KIISPQOodULyfjPJElaRS+0oMyBYIY0hYuXMicOXN45JFHuO666/jBD37Apk2bsFqtzJkzh5KSzudLS01NZc2aNdxwww3ExMQwceJEINQStG7dOn72s5/h8XhITExk3bp1farvW9/6FuvWrWPlypXYbDZUVeWaa67hnHPOYeLEidx+++2sWLECi8XClClTuqy3I8XFxVx22WVAaOD6iS1UU6dO5fbbb+f6669H13WmTZvGf/3Xf6GqKm+88QbLly/HZrNx0UUXMWXKlD79fCdSjI5WIB5ArWvWyFp7YjCSaze4Vb/6BK6D2xj73ac7fLwn169x20Zq33qaMd99GlNswkCUKfposLz+9u3bx7Rp09psi+Y8UkNBJNfau/baa7n99tuZP3/+gJz/5OvfXW6RFikhxJARqCvH2ss19k5mbr1zr75CgpToseEQdsTAkCAlhBgyAnXlxIyf3f2OXThxCgR79uRu9hZCDDbPPPNMtEtoQwabCyGGBD3gQ2upw+LMOKXzhI5X5M49IUS/kCAlhBgSgo3VAJid6ad0HsVswZyYSlCClOhCR1MBiOGvL9ddgpQQYkgINlQBYEk6tRap0DlGyezmolNxcXGUlpbi9/uJ8P1YIkoMw8Dv91NaWtrrQfMyRkoIMSQEjgUpc+KpBylzUiau/VtO+TxieBo9ejQ1NTUUFhYSDAajXc6Q4Pf720ykORSZzWYSExPDM773+LgBqkcIIfpVsKESxWzF5HCe8rksyaPQPc1onhZMMY5TL04MK6qqkp6eTnr6qXUjjyQ7d+5k9uxTuxFkqJKuPSHEkBBoqMScmNYvk2i23rkn46SEEKdKgpQQYkgINlSd8kDzVpak0MzocueeEOJUSZASQgwJwcaqU576oJX52IB1CVJCiFMlQUoIMehpnhZ0rwtzPwUp1WLD5EgmUF/ZL+cTQoxcEqSEEINeeOqDfurag9B8VMEGCVJCiFMjQUoIMegFGkOBp7/GSEFoPioJUkKIUyVBSggx6LW2SPVX1x6AOTGdYFMthhbot3MKIUYeCVJCiEEv2FCFandgsvduxuGuhGZINwg21vTbOYUQI48EKSHEoBeor+y0W6+vS3i0ni8g3XtCiFMgM5sLIQa9YGMl1rTcNtuqXbX8aefz7K74grS4FL4x+7JevaFZnMcm5TzWbSiEEH0hLVJCiEHNMHSCDdVtWqTqPA3c+85/s7/mMEsnLiLWbOe/P/4je5sP9fi8pvgkMJmlRUoIcUqkRUoIMahpLQ0YWiA8GadhGDy+9S+0BNz87ILvMzYph4AW4Bf/foxNVR9xUcuXGeVI6/a8iqJiSZQpEIQQp0ZapIQQg1pr0Gm9Y++Tkk/5vHI/18y6lLFJOQBYTBZun38DqqLyf7v+2eNzm53pBOqla08I0XcSpIQQg1qg4fgcUoZh8K+9rzM6IZOLJixss19KbBLznbPYUbaHo3WFPTq3xZlBsFFapIQQfSdBSggxqIXnkEpMY1d5HkWNpVw6bQmq2v7ta55zBnGWGF458HaPzm12pqMfW35GCCH6QoKUEGJQCzRUYXIko5qtvH3kI5LsiZyd+6UO97WpVhaOPYutJZ/R5Gvp9tzhxYvlzj0hRB/1KEi9++67rF69mqVLl/Lggw8CsHnzZlasWMFFF13EunXrBrRIIcTIFWyowpKUQZO3mV3leZw39kzMqqnT/S+ccC5BPcgH+Z90e25LYsax55DuPSFE33QbpIqLi7nvvvv4/e9/zyuvvMIXX3zBBx98wD333MPvf/97XnvtNfLy8vjggw8iUa8QYoQJNIQm4/yoaDuaobNwzPwu989JzGJC0hg2F+/o9tzSIiWEOFXdBqm33nqLiy++mFGjRmGxWFi3bh0xMTGMGTOGnJwczGYzK1asYNOmTZGoVwgxghhaAK2pFnNiOpuLdjLWOZpcZ3a3x83PmcuRukKqXLVd7meyx6Ha46RFSgjRZ90GqcLCQjRN46abbmLlypU899xzVFVVkZZ2fJ6W9PR0KivljUgI0b9C6+AZeBwJHKrN54zs2T067qyc0wH4pPjTbvc1OzNkUk4hRJ91OyGnpmns2LGDZ555htjYWL71rW8RExPTbj9FUXr1xHl5eb3af6jZuXNntEsQfSTXbvAw1+QTD7xTfhQDg9hGS7fXp/XxdGsyHxzYTLYrucv94wwLpopCue6DhFyHoWukXrtug1RqaioLFiwgOTn0ZvTlL3+ZTZs2YTIdH+xZVVVFenrHC4p2ZubMmdhstl6WOzTs3LmTefPmRbsM0Qdy7QaXpk9rqQEqHQbJupOLz76oy/+0nXj99puL2HjgbabPmkGMxd7pMbUNe2nafpTTT5+LosiNzNEkr7+hazhfO5/P12XjT7fvGosXL+ajjz6iqakJTdP48MMPWbp0Kfn5+eFuv40bN7Jw4cLuTiWEEL0SaKhEU818XnuE07NO61XL9+xR09EMnb1VB7rcz+JMD43Faq4/1XKFECNQty1Ss2fP5j//8z+5+uqrCQQCnHPOOXz9619n/Pjx3HHHHfh8PhYtWsTSpUsjUa8QYgQJNlRRnpKCN+hjVsbUXh07JXU8NpOVzyq+4EtdjK1qXXom2FiFOSHllOoVQow8PVq0+IorruCKK65os23BggVs2LBhQIoSQggIBamCeAfQzPS0Sb061mKyMCN9Mrsr9nW5X2uQCtRXYs+Z1tdShRAjlAwIEEIMWoHGKo5YQnNDJdjje338zIypVLZUU+du6HQfS2IaoISXohFCiN6QICWEGJR0n4eAu4mjupsZaZP7dI5paRMB2FdzqNN9FLMFU3yyTIEghOgTCVJCiEEp0FBJqc2Mz9CYnt67br1WY52jsZtt7Ks+3OV+Fme6TMophOgTCVJCiEEp2FDF0RgrQK/HR7UyqSYmp4xnf/WRLveTSTmFEH0lQUoIMSgFG6vIj7GQ7Ujv0/ioVlPTJlLcWEaL39XpPmZnOlpzPUYw0OfnEUKMTBKkhBCDkq+ugqIYK1P72K3XalraRAwMDtQc7XQfizMdMAg0Vp/ScwkhRh4JUkKIQamssRSPqjA5ZfwpnWdi8lhMqqnLcVJmZ2hlBhknJYToLQlSQohB6bC7BoDJqacWpGxmK+OdORys6XyclKV1Uk6ZAkEI0UsSpIQQg45hGORrbmIVE5nxvVvHsyOTUsdztL6IoK51+LjJkQSqmUCjBCkhRO9IkBJCDDq6u4kiq8p4ezJqPywkPDllHH4tQFFDaYePK6oJc2KqdO0JIXpNgpQQYtBprCmiympikjOnX843KWUcAIdq8zvdx+LMkK49IUSvSZASQgw6B8v3YSgKUzL6NqP5yVJjk3HaE7oMUmZnOgEJUkKIXpIgJYQYdA7WF6IYBpNHz+6X8ymKwsSUcd20SKWje5rRfZ5+eU4hxMggQUoIMegccVWRETRwxDn77ZyTU8ZR3lJFs6+lw8fNx+7ckxnOhRC9IUFKCDGo6IZOftDFOOz9et7j46QKOnzcnNg6l5R07wkhek6ClBBiUClrqsSrGIyPSerX805IykVRlE679yytk3LKFAhCiF4wR7sAIYQ4UevEmZMSsrvd99r7N9HQ7Gu78bmSdvspChgG2GY4+MeWrfz1z3oHZzN4OMnMq5u2cfWZl/SldCHECCQtUkKIQeVo1WGsuk5Wythu920XojphGKHPusuJGtcIGB3spVCrOYjTGntcqxBCSJASQgwq+XWFZPmC2JIy+v3ceksiijmIYnd1+Hit7iBF7XgwuhBCdESClBBi0NB1nSJXNVm+IOaBCFKuRIBjrVLt1ekOUkwtGEZHLVZCCNGeBCkhxKBR0VKFz9DI8muYE1L7/fyGx4GhmVDjmjp8vFZzYFOC6O6OHxdCiJNJkBJCDBpH64sByDXHo6imAXgGBd2dgNJJi1St7gCQGc6FED0mQUoIMWjk1xdhMiA7Pm3AnsNwJaDGNgHt79yr0+MBmQJBCNFzEqSEEINGQUMxmQEdu7P/x0e10l2JKCYdJab9gPNaLdQiFZTZzYUQPSRBSggxKBiGwdG6YrI8vvByLQNBdyUAHQ8492GhRbdJ154QosckSAkhBoUadx2ugJssXwDLAAYpwxvX5YDzOt0hy8QIIXpMgpQQYlDIPzbQPMsXxHxsuZaBoaC7EjqdAqFWc8jCxUKIHpMgJYQYFI7WF6GikOkf6CB1bJxUbDMo7Qec1+oOgo01GLo2oDUIIYYHCVJCiEGhoL6YUSY7VpMVU5xzQJ/LcCWgqDqKvf0s5nW6A/QgWkv9gNYghBgeJEgJIQaF/PpisjUVszMdRVEG9LmOz3DefpzU8bmkpHtPCNE9CVJCiKhr8DRS720ky+sf0IHmrQxfLEbQ3OE4qVrt2FxSMuBcCNEDEqSEEFGX3xAaaD6qqXHAx0eFtA44b98iVafHAYpMgSCE6BEJUkKIqGu9Yy+z2RWhINU64Lyp3YBzDROm+CRpkRJC9Ii5Jztdd9111NbWYjaHdn/ggQcoKiriiSeeIBAIcMMNN3DNNdcMaKFCiOErv76YDLsTu1EVka49AMOdgKIaKDHNGO7ENo9ZnBkyu7kQoke6DVKGYXD06FHef//9cJCqrKzku9/9Li+++CJWq5WrrrqK+fPnM3HixAEvWAgx/OTXF5FrC4WZgZzV/EThGc5jm9FOClJmZwaegs8jUocQYmjrtmvv6NGjKIrCzTffzMqVK/nrX//K5s2bOeuss3A6ncTGxrJkyRI2bdoUiXqFEMNMi99FlauWHKwAWCLUtWf4Yjud4dzsTEdrrsMIBiJSixBi6Oo2SDU1NbFgwQIef/xx/vznP/P8889TVlZGWtrx1dnT09OprJRmcCFE7xXUlwCQHdBQYxyottgIPXNowLkS2z5IhcKcQbCpOkK1CCGGqm679ubOncvcuXMBiI2N5YorruChhx7i1ltvbbNfb+d9ycvL69X+Q83OnTujXYLoI7l2kbWtfg8AjrIK/BYHO3fu5FcvluHytp91vL8Z7gRMaSWAARx/Dzta2UA88MX2jwmmjh/wOsRx8vobukbqtes2SO3YsYNAIMCCBQuA0Jip7OxsampqwvtUVVWRnt675viZM2dis9l6We7QsHPnTubNmxftMkQfyLWLvI8/2UOKO4m0uias2eOYPG8erudKIvLcujses0lDsbswvI7w9hnzz6No218Zl5ZIwuny9xAp8vobuobztfP5fF02/nTbtdfc3MzDDz+Mz+ejpaWFl156iV/96lds2bKFuro6PB4Pb775JgsXLuzXwoUQI0N+fRHjkkYTaKyO2EDzVicOOD+RyZEEqllmNxdCdKvbFqnFixeze/duLr30UnRd5+qrr2bevHl897vf5brrriMQCHDFFVcwa9asSNQrhBhGvEEfZU2VzE+fBloQS2JkBpq3MrwODF0JjZOqywxvV1QT5sRUmUtKCNGtHs0j9Z3vfIfvfOc7bbatWLGCFStWDERNQogRoqihFAODHFNogHmkJuMMM1QMT3yHd+6F5pKSICWE6JrMbC6EiJqj9UUAjNZCA70j3bUHoXFSoa49o812szOdQKMEKSFE1yRICSGipqC+mHibA0dLM6BgSUzr9pj+prsSUCx+sPjabLc409HdTeh+T8RrEkIMHT3q2hNCiIGQX1/M+KQctMZqTPHJKGZLxGsw3McGnMc1oTfYAVjx/fXMtRZxgwNuXfN3yrWkdsc54208c//SiNYqhBh8pEVKCBEVQS1IUVMZY505BBsqsSRFvlsPQl17hgHqSRNz1mrxAKSoLR0e19Ds63C7EGJkkSAlhIiK4qZyNF1jXFIugfrKqIyPAkA3Y3hj202BUKuH5pXqLEgJIQRIkBJCREn+sYHmY+Iz0FrqIrbGXkd0dwLKSXfuuQwbPsNMskmClBCicxKkhBBRcbS+iBiznZRg6G65qLVIERonpdo8YDpxkWKFWs1Bitrc6XFCCCFBSggRFQX1JYxNykE/NsVAtMZIQWicFHQwTkp3SNeeEKJLEqSEEBGn6zqFDSWMc44mUB9ahiWaLVLhpWJO6t6r0x3HuvaMDo4SQggJUkKIKChrqcSn+RmXlEuwsQrFbMUU54xeQUEbht+GcvKAc82BXQkSp8gdekKIjkmQEkJEXH5dMQDjknKO3bGXjqIoUa1Jdyd02LUHcueeEKJzEqSEEBGXX1+ExWQhO2FUaA6pKHbrtdJdCSgxLlC08LY6PTR2Su7cE0J0RoKUECLi8huKGZOYjaqoBBqqojo+qpXujkdRjDbde7WatEgJIbomQUoIEVGGYZBfXxy6Y8/TjOH3RPWOvXBdrQPOTwhSPiy06DYJUkKITkmQEkJEVJWrBnfAw/hj46MgunfstTL8MRhBc7txUqE792QuKSFExyRICSEiKr++daB5LsGGUJAaDGOkQAkNOI87ec09mUtKCNE5CVJCiIjKry9GVVRyErMINIQm4zQ706JcVYjhjkeJaebEeaNqdQfJqgtF5pISQnRAgpQQIqIKGorJScjEarIQbKjEFJeIao2JdlnAsTv3TDqK3RXeVqc7MCs6CYo7ipUJIQYrc7QLEEKMHN+4/3V8Ew+hNaSy4vvruS0+D4ti4fbvr492aUBoLikIzXCueUN37IXnkjK10BiMi1ptQojBSVqkhBAR0+htQrH4w4ElWW2h7tgUA4OB4YnD0FWUEwac12qhuaRknJQQoiMSpIQQEdN6R5zhSkRFJ0l1UXNs0svBQcVwx7cZcF6nx6EbEqSEEB2TICWEiBg1rgnDCE1+maS6MClGeNLLwUJ3xx8LfKHB5RommoxYmd1cCNEhCVJCiIhRY5sxvLGgm0lRQ3Mz1Q6qFqljA87NQRSrJ7xNpkAQQnRGgpQQImKUuMbw+KiUYy08rYO5B4vW+pS4E5aK0R3h4CeEECeSICWEiIhmXwuqzRteiiVFbUEzFBr02ChX1pbhiccwaDPDeZ3uIFF1Y0Lr4kghxEgkQUoIEREFDSXA8RafFLWZOt2BMdjehnQThsfRJkjVag5UBZJUVxcHCiFGokH2DiaEGK7y64uA40Eq1dQy6Lr1Wukn3bkXnktKxkkJIU4iQUoIERFH64vRfXYIWoFQi1TrHE2DjeFOQLH6wOwHQl17gNy5J4RoR4KUECIiCuqLw+OjYhUfcaqf6kF2x14r/Vidrd17DXosQUOVFikhRDsSpIQQA84b8FLeXHW8W+/YHXDVg7RFSneH6mrt3jNQqdfjJEgJIdqRICWEGHAFDaUYGCeMjwoFqRo9IZpldU6zovvsbZaKqdMd0rUnhGhHgpQQYsCFB5of6zJLU0MBZbDNan4iw5XQ7s49aZESQpysx0Hql7/8JXfffTcA+/bt4/LLL2fJkiX8v//3/wgGgwNWoBBi6MuvLybRFg8BGxBqkarXYglgjnJlndPdCSh2N6ih97da3UG86sVGIMqVCSEGkx4FqS1btvDSSy+Fv7/rrrv4yU9+whtvvIFhGPzjH/8YsAKFEENffkMx45JyAAWANLWZ6sHarXeM7k5AUULL2gDUHBvP1dotKYQQ0IMg1dDQwLp167j11lsBKC0txev1MmfOHABWr17Npk2bBrRIIcTQ5dcClDSWMTYpJ7wt1dQcDiaDVWs3ZOs4qdbg19otKYQQ0IMgde+99/Ld736XhITQm0hVVRVpaWnhx9PS0qisrBy4CoUQQ1pxYxmaoTM+KRcAu+InXvUO2qkPwgI2jIAlfOde6x2GadIiJYQ4QZcDFF544QUyMzNZsGABL774IgCGYbTbT1GUXj9xXl5er48ZSnbu3BntEkQfybXrX5817gfAVdoIHJ/6YLC3SIGC7k4Id+35sdCox5BmOt4iJX8r/U9+p0PXSL12XQap1157jerqalatWkVjYyNutxtFUaipqQnvU11dTXp6eq+feObMmdhstt5XPATs3LmTefPmRbsM0Qdy7frfpzsOENsQwwXzz+eRFzaEg8ignfrgBLorAfOoAlB0MFSqtYQ2XXvyt9K/5PU3dA3na+fz+bps/OkySD399NPhr1988UW2bdvGQw89xCWXXBL+pb388sssXLiw/yoWQgwr+fWhgeatLdfHW6QG79QHrQx3AopqoMS0YLgTqNbimWktiXZZQohBpE/zSP3617/moYceYtmyZXg8Hq677rr+rksIMQwEdY3ChhLGHRsfBaExRg16DH4sUaysZ8IznJ8w4Dw0BYI/mmUJIQaRHk/isnr1alavXg3A1KlT+ec//zlgRQkhhofSpnICepDxJ96xpzZTow3+bj0AwxuHoZlQY5vQOD6uK83UTImWEt3ihBCDgsxsLoQYMPn1xQAntUg1UTPY79gLU9Dd8Shxx9YGbJ0CwSRTIAghQiRICSEGzNG6IuxmG5nxoRtSbARIUL2DdrHijhju1qVijONTIKgyBYIQIkSClBBiwOTXFzHWORpVCb3VtM4KXj1EuvYgdOeeYtJQbG4CmGnQY6VFSggRJkFKCDEgdF2noKEkPBEnnHDH3pDp2jthwPkJE3NKi5QQopUEKSHEgChrrsSn+duNj4KhMBnncYYnHkNXji8VoyVIi5QQIkyClBBiQBytLwJgfHLbqQ+adDu+ITD1QZihYngc4RnOq/V4HKqPGEWmQBBCSJASQgyQo/VFWE0WsuIzwtvS1UaqtMQoVtU3ujvhWNeeER7flSqLFwshkCAlhBggoYHmOZhUU3hbuqmJqiE00LyV7o5HsfjB4jthCgQZJyWEkCAlhBgAuqGHl4Zppbmbcag+qobAGnsnM1yhmtXYJmq0eHSDNmvuCSFGLglSQoh+V9FchTfoazPQPFBXCkDlEO3aM4zQnXtBTDTocdIiJYQAJEgJIQbA0WMzmp849UGgtgxgSHbtoZsxvHGocY1AaPoGuXNPCAESpIQQA+BofREW1czoxMzwNn9tKUFDpU53RLGyvtNdieG5pKq0BNLUJgzDiHJVQohokyAlhOh3+fVF5DqzMZ8w0DxQW0q1Fo8+RN92dFcCitUHFi+VWiJxqh/dLa1SQox0Q/MdTQgxaBmGcWygeW6b7YHaMqr0oTc+qpXhCtWuxjWGx3n5a0qiWZIQYhCQICWE6FeVrhrcAU+b8VGGFiRQXzE0x0cdExpwroSC1LFAGKgtjXJVQohokyAlhOhXR+sKARh/wtQHgYYq0LUhecdemG7CcDtQHY006rH4DLO0SAkhJEgJIfrX4bpCLKqZ3MTs8LbWlpuh3CIFxwecG4R+FmmREkJIkBJC9KsjdQWMTcrBbDKHt4WD1BCcjPNEuisBxRxAsXmo1BIJSIuUECOeBCkhRL/RdI2jdUVMTB7bZnugtgxTXCIewxadwvqJftKA82BTDbrfG+WqhBDRZO5+FyGE6JmSpnJ8mr99kKorw5KcBcXRqau/GJ54DF1FiWuksjkUqm5b8xwlWkqnxzjjbTxz/9JIlSiEiDBpkRJC9JvDtQUATEwZ22a7v7YUS0p2+wOGGkPFcMe3uXMvw9TY5SENzb5IVCaEiBIJUkKIfnOkrpA4SwyjHGnhbZq7Gd3dNDyCFMcHnFdrDjRD6TZICSGGNwlSQoh+c7iugAnJY1EUJbzNX1MEgDUtp7PDhhTdlYhi0tDtXmp1B+my5p4QI5oEKSFEv/AF/RQ1ljExZUyb7YHq0MCo4ROkQncetg44H6VKi5QQI5kEKSFEv8ivL0Y39HYDzf3VxSi2WEzxnQ/IHkoMjwNDM4WDVJqpCRU92mUJIaJEgpQQol8crisAYEIHQcqamtOmu29oU9BdCaiOUJAyKzopaku0ixJCRIlMfyCEOGXX3r8JV/pWVIed69a83+axnzsPs8efw9+/vz46xQ0A3ZWIOaOICn08ABmmBqqH+GSjQoi+kRYpIcQpa2j2hZZOcbVdS8+heHCoPio0Z3QKGyB6ixNF1am0mQDINDVEtyAhRNRIkBJCnDqzH9XuRm9pG6RGHZsaYDgGKYCgo4VazSFBSogRTIKUEOKUqXGhwKSf1CKVaaoHoHyYBSkCdnSfHdXRQLnmJEuClBAjlgQpIcQpUx0NGEZHQaoRl26lyYiJUmUDR29xhoNUuqkRE1q0SxJCRIEEKSHEKVMdDRjuBNDb3r8yytRwrFtvuNyxd5ze4kS1eSlUHZgUgwyZmFOIEUmClBDilGi6hupoCI8bOs44IUgNP60/b5k99Dba2o0phBhZehSkfvvb33LxxRezfPlynn76aQA2b97MihUruOiii1i3bt2AFimEGLyKG8tCS6acFKQSFA9xqp9yLbHjA4c4w52Aoas0x/rRDEUGnAsxQnU7j9S2bdv45JNP2LBhA8FgkIsvvpgFCxZwzz338Mwzz5CZmcktt9zCBx98wKJFiyJRsxBiEDlQcxQAvTmpzfbWYDFcW6Qw1NByMY4mKhsTyZIWKSFGpG5bpM4880z+7//+D7PZTG1tLZqm0dTUxJgxY8jJycFsNrNixQo2bdoUiXqFEIPMgdqjGH4bht/eZnuWORQsSrWkjg4bFvQWJ2pcIyV6orRICTFC9ahrz2Kx8Lvf/Y7ly5ezYMECqqqqSEtLCz+enp5OZWXlgBUphBi8DtQcOdat13ZA+WhTHfVaLG7D3uFxw0FoYk6Dw1YHySYXNvzRLkkIEWE9XiLmzjvv5Oabb+bWW2+loKCg3eO9XUcrLy+vV/sPNTt37ox2CaKP5Nr1XHPQRbWrFq15arvHsk31lGrJUagqclrHhZXYVfBCprmBgmB6u/3kb6rn5Hc1dI3Ua9dtkDpy5Ah+v59p06YRExPDRRddxKZNmzCZTOF9qqqqSE9v/+bRlZkzZ2Kz2Xpf8RCwc+dO5s2bF+0yRB/IteudT4o/hQLaDTS3ECTd1MieQE5U6oqYYxNzNsZq4IUsU8dBSv6mekZef0PXcL52Pp+vy8afbrv2SkpKWLNmDX6/H7/fzzvvvMNVV11Ffn4+hYWFaJrGxo0bWbhwYb8WLoQY/A7WHMVisoTmkDrBKFMDJsWgNDi8W6QA9JYkgo5mPIZFpkAQYgTqtkVq0aJF7N69m0svvRSTycRFF13E8uXLSU5O5o477sDn87Fo0SKWLl0aiXqFEIPIgdqjTEwew6dG2/+TZR8baF4yzLv2APSmJMwp5exXkhhtqot2OUKICOvRGKk777yTO++8s822BQsWsGHDhgEpSggx+PmDfo7WF7F88pf59KTHsk11eA0LdbojKrVFkt4cCotf2OO4wihGQceQuY6FGDHk1S6E6JNDdQVousb0tIntHhttqqM0mIQxDJeGOZnhjcMIWCiOVbEpQdLU5miXJISIIAlSQog+2Vt1EEVRmJraNkgpGGSZG4b1/FFtKejNyTTFhaY+yDHXRrkeIUQkSZASQvTJvupDjHWOJtYa02Z7stqCXQkM+6kPTqQ1J6HbfNSYzDJOSogRRoKUEKLX/FqAg7X5TE+b3O6x1iBRGhwpLVLHx0ntsjkZbZYgJcRIIkFKCNFrh2sLCGgBZqRPavdYjrkWzVAoHzFde2C44zGCZg7Z7ceCpBHtkoQQESJBSgjRa19UH0JBYWoHA81zzTWUaUkEMXVw5HCloLckURlrEKv6SVZbol2QECJCJEgJIXrti6qD5DqzcVjj2mxXMMgx1VIUTI1SZdGjNyXhswdoMSnkyDgpIUaMHq+1J4QQAEEtyMHao1w4/tx2j6WpTcSqAQqDKVGoLLq05mQswBG7ldHmOnYHxoQfW/H99d0e74y38cz9MrGxEEONtEgJIXrlcF0hfi3A9PT2A81zj936X6SNvBYpw52AoZnIszvIMfV+CoSGZt8AVCWEGGjSIiWE6NC192/q8B93c+YRLDnws0cPQbCwzWO55hp8hplKLTFSZQ4ehorenMzR2FouNdcQGnA+/CckFWKkkxYpIUSHOmshURNr0d3xELS2eyzXVEtJMBl9hL61aI0puGw6fqvMcC7ESDEy3+2EEH2jBlEd9WiN7cdAqehkm+so0kbe+KhWelPoZz8cY2WsuTrK1QghIkGClBCix9T4ehTVQG9sPwYq01SPVdEoHIF37LUyPA4Mv5UDMXYJUkKMEBKkhBA9ZkqswdBV9Ob2k22GB5qP4CAFClpTKodjreSaa6JdjBAiAiRICSF6TE2oDYUoo/1km+PNVTTrdmp1RxQqGzz0phS8ZlBjmrAQjHY5QogBJkFKCNEzFi9qbAt6B+OjIBSkjgbTGel3qrWOHzsaZyHH3PtpEIQQQ4sEKSFEj5gSQqFAa2rfdReveEg1tZAfTIt0WYNPwA6eWBlwLsQIIUFKCNEjamItRsCK4Y5v99h4cxUARwPpkS5rUAo2pnI0xspoiwQpIYY7CVJCiB4wMCXUoDWl0FHX3XhLFX7DRImWHPnSBiGtMY2gqqDG1xOamFMIMVxJkBJCdEuJa0Sx+tEbOu66G2euoiiYikb7Qegjkd6UjKopFDkgXW2KdjlCiAEkQUoI0S2TsxrDAK2D+aOsBBhtqjs20FwAYJhQmpzsj7MxwVIR7WqEEANIgpQQolsmZ1Vo2oOOloUx12JSDBlofhJPQyb1FhNpcRKkhBjOJEgJIbpm8aLGNXfarTfJUoFuKORLi1QbWkPo9xFIbEDGSQkxfEmQEkJ0yeQM3ZHXGgxONtlcTpGWgsdo31o1ogXsxHisFDgUUtSWaFcjhBggEqSEEF0yJVWje2MwvHHtHrMSYIy5hkOBUVGobPDTG9IoslsYayuLdilCiAEiQUoI0TlVCy0L09DxjOUTLFWYFIODEqQ61Fg3Gl1RiE0qj3YpQogBIkFKCNEpNaEGRdXROhkfNdlcTtBQZXxUJwyXE3sAmhJbkHFSQgxPEqSEEJ0yJVdgBC3ozR1PtDnZUk5+MI0A5ghXNlQoOBoTyI8zkW6ui3YxQogBIEFKCNExRceUVIVWnw5G+7eKWMXLaHM9BwOZUShu6GiqzSWgKmQk50e7FCHEAJAgJYTokJpYg2LS0Oo6Hv80xRIa93MwKEGqK42N2cQEDYJJtdEuRQgxACRICSE6FOrWM6M3pXT4+HRLKS26jcJgx4+LVgqpTXbK4nXMij/axQgh+pkEKSFEO0EtiMnZebeegs50Syn7A1kY8jbSrUBdJn5VYVTykWiXIoToZz16B3zsscdYvnw5y5cv5+GHHwZg8+bNrFixgosuuoh169YNaJFCiMjaU7kfxRxEq++4Wy/XVItD9bE3MDrClQ1NpfXjiNF0LCmyXIwQw023QWrz5s189NFHvPTSS7z88svs3buXjRs3cs899/D73/+e1157jby8PD744INI1CuEiICPiraHuvU6WKQYYIa1BN1Q2B/IinBlQ5PfsJHTpFKT4AU1GO1yhBD9qNsglZaWxt13343VasVisTBhwgQKCgoYM2YMOTk5mM1mVqxYwaZNmyJRrxBigHkCXraXfIZWl9lhtx6ExkflB9NwG7YIVzd0qbUZBFUFZ3JBtEsRQvSjbid/mTRpUvjrgoICXnvtNa699lrS0o5P0Jeenk5lZeXAVCiEiKitJbvwaX60mo5bmxIVNznmOl5xz41wZUNbfsNEkgNlaGnFUDOxw31WfH99t+dxxtt45v6l/V2eEKKPejyL3qFDh7jlllv40Y9+hNlsJj+/7ZwoitJ++Yiu5OXl9Wr/oWbnzp3RLkH00Ui/dq+Wvo3TkkB5i7PDx2dZiwD43J8TwaqGvgY9ntMaFXakeMHihYC9b+dp9g3rv9Hh/LMNdyP12vUoSO3cuZM777yTe+65h+XLl7Nt2zZqamrCj1dVVZGe3rslImbOnInNNjy7BXbu3Mm8efOiXYbog5F+7Wrd9RQdLueKGRfzfx9rHe4zx1pIedBJpe6MbHHDgFKTAanVxKUW4Sqf3OfzDNe/0ZH++hvKhvO18/l8XTb+dDtGqry8nNtuu41f//rXLF++HIDZs2eTn59PYWEhmqaxceNGFi5c2H9VCyGi4sPCbRgYnDd2foePJyhuxpsr+SyQG+HKhofDrvHkeANYU0uQtfeEGB66bZF66qmn8Pl8rF27NrztqquuYu3atdxxxx34fD4WLVrE0qXSZy/EUHDt/ZtoaPZ18IiBbdaHGIEkbr5vc4fHzrIWoSrwmX/sgNY4XJVpTlY2GBSP8qPENWK4nNEuSQhxiroNUmvWrGHNmjUdPrZhw4Z+L0gIMbA6DlGgJtSh2t34SzseCA2hbr0KLZEKzTlA1Q13Cv7aLKzpVSgZ+biPyoB9IYY6mZJYCAGAOb0II2BBq8vo8PEExc0EcxWf+cdEuLLh5TPvROY2eyG5CkyyZIwQQ50EKSEEWLyoSVUEa7LBMHW4yzxbPqpisNM3LsLFDS8VmpNxdRYM1cCUWhbtcoQQp0iClBACc1opimKgVXU2pYHBfOsR8gNpVOmJEa1tOCpqHkeuN4AtvRAZdC7E0CZBSoiRTtExpxehNaZg+OI63CXHVEumuYGt/gkRLm54+tQ/lvmNHowYD2pCbbTLEUKcAglSQoxwpuRyFKuPYMXYTvc503aEgKHK3Xr9pF53EFufSFzQwDwqv/sDhBCDlgQpIUY0A3NmAbrb0ekCxWY05lnz2ePPxWNYI1zf8LXVO5lzG12YnLUoMc3RLkcI0UcSpIQYwdSEWtTY5mOtUR0v8zTXWkCc6ucT36QOHxd9s8efy8x6HbOOtEoJMYRJkBJiBDNn5mP4bWi1HS9QDAYL7fup0BI5GBwV0dqGOw0TezzjOaPRjTmlPLT+nhBiyJEgJcQIpcQ1YEqsJVg5BoyO3wrGmGrINdfyoXcKnbVYib7b4pvMeY1uUAwsmdIqJcRQJEFKiBHKkn0YI2AhWNn5unkL7fvx6Ba2+eRuvYFQrSdQ4R7F7CY/pvRiaZUSYgiSICXECKQ66jE5awiWjwO945WinKqLudYCtvkn4McS4QpHjve8M1hS34Si6Fiyjka7HCFEL0mQEmIEMmcfxghYCVZ13hp1gX0vAO95p0eqrBHpYHAUbm8iMxs1TGnFKFZPtEsSQvRCt4sWCyGGFzW+FlNiLYGiKZ22RjkUDwtsh9juH0+97ohwhSONwrveGaxs2ExeQhrmrCMECmZ2ecSK76/v0Zmd8TaeuX9pfxQphOiEtEgJMYLouo4ldz+6z97l2Kjz7fswo/GOp+t/0EX/2OUfi+aLYXaDhimtBCW2qV/O29Ds65fzCCE6J0FKiBHk/YItqHHNBIundLo4sUPxcJ59P5/5x8i6ehGio/KGZxaXNtRi0kxYcvcja/AJMTRIkBJihPAEvPzt8w1ozU60us7nhFoSswcLGq975kSuOMF2/wRaAg7Oq/FjSqhDdVZFuyQhRA/IGCkhholr79/UZVeOOWc/lswmAkVn0dmcUClqM+fYDvKJb5K0RkWYjsrrntlcY/qILZ7R6Ln78TWldDqOTQgxOEiLlBDDRFchSoltxDyqgGBVDobL2el+l8TsIoiJTZ5ZA1Ch6M6n/rFUBJNYXd2Eavdgzj4c7ZKEEN2QICXEsKdjHbcXAjYCxZM73WuiuYLTbQW8551OkxEbwfpEKwOVF91nMMffxKjaWMyjClBiG6NdlhCiCxKkhBjmzKMKUeOa8BdOA63jiTVNaFwZt5UazcHbcqdeVB0OjmKXfww3NJSiBCyhEKzo0S5LCNEJCVJCDGNKTBPm0QfR6jLQ6zM63e98+z5GmRp50X0mARk6GXXr3fOI1XXOqQQ1rglz1pFolySE6IQEKSGGK0XDOmEPBK34C2bQ2QDzNLWJpTG72ePPYW9gdGRrFB2q1x284ZnFCl8pCbWJmLOOoDrqol2WEKIDEqSEGKYsuQdQY1vwHz0NgtYO91HRucbxEQHDxD9d8yNcoejKu94ZFAVT+K+6YhRfDJYJe8AUiHZZQoiTSJASYhgyJZdjzigiWDEGvSm10/2+bN/LOHMNL7jn0ygDzAcVHZXnXGfjxM+FpTqKxYd1/OfIRJ1CDC4SpIQYZpSYJizjPkdrdhIontLpfrmmGpbG7OZT31h2+cdFsELRU+VaEps8s7nQKCa7LBlTUpWMlxJikJEgJcRwYvJjnbQLNAv+w3PB6PglHqt4udHxAY16DC+4pUtvMHvbO4MDgUxu8RzEUpuKZfRhmfVciEFEgpQQw4WiYZ20C8XqxXdoLgRsHe+GznVxH5Ggeni6ZRFuo+P9xOBgoPJMy7n4DAu31JZitMRjnbC73xY2FkKcGglSQgwDuqFjGf85poR6AvmndTl7+bKY3UyzlvFP95kUa52PnxKDR7MRw/+1nEem0sTqMi8Ezdgm70SxuqNdmhAjngQpIYY4wzD4y65/Yk6pIFA0Ba02q9N951sPsyTmc7Z4J7LFNymCVYpTdSiYyUvuM5ivlnJ2YSyoOtYpO8Dsj3ZpQoxoEqSEGMIMw+DveRt4/dB7BCvGEKwY2+m+k81lfC1uC/sDmfzD3fnCxWLw+tA3lQ+9U1ilHmBSfhqKzYttyg4wSZgSIlokSAkxRLWGqBe/2MSXx59LoGgqnYWjbFMt/+H4gEotkf9tXoQuL/0h60X3GezzZ/Ef7CY7PxslphnbVGmZEiJa5N1UiCHIMAye//x4iLr5S1+nsxCVZarntvi38BhW/tDyZXx0PDmnGBp0VP63ZRGFwVS+pe8i/ehYlJgWbFO2S5gSIgokSAkxxGi6xv/seI6X9h0PUarS8Us5Q23gW/FvEcDMY80X0aDHRbhaMRD8WPhDy5ep0BK5w9hBypGxKDEubNO2olg90S5PiBGlx6uTtrS0cNVVV/Hkk08yevRoNm/ezEMPPYTP52PZsmV897vfHcg6hRCAL+jnt1ueYkfZHi6btpSrTluJonTcEpVpqueb8W+jo/BY00XU6vERrlYMJI9h5cnmC7k94U2+o27n8cNnUDG+CNv0T/AdmIfhSQBgxffXd3suZ7yNZ+5fOtAlCzEs9ShI7d69mzVr1lBQUACA1+vlnnvu4ZlnniEzM5NbbrmFDz74gEWLFg1krUKMSNfev4mGZh+YfVgn7UJ1NBAonMZz2+A5NnR4zFhzFbc43sWPid83fYVqPSHCVYtIaDZieLTpIr4Z/zZ3mLbxh0NnUji+DNv0rfgPze1yeaATNTT7BrhSIYavHnXt/eMf/+C+++4jPT0dgD179jBmzBhycnIwm82sWLGCTZs2DWihQoxUDc0+lNhGbDO2oMY14T88B61qTKf7T7eUcFv8W7gMG79tWkal7oxcsSLiWowYHmu+iBItmVvNW5l+OAPDF4N1yg7Mo/KRtfmEGFg9ClI///nP+dKXvhT+vqqqirS0tPD36enpVFZW9n91QghMKaXYpm8FwPfFfPT6UZ3uu8B2kP90vEellsgjTUup0x2RKlNEkcew8fumr3AoOIobbNtZnG9Hr8/AknsAy4Q9oGrRLlGIYavHY6ROZBjt/4fT2TiNzuTl5fXlqYeMnTt3RrsE0UeD5doF9CDv1W7FOmEfWlMy/sNzINjxHXcqOpfG7mCRfT/7/Fk83bJQ7s4bYXxY+EPzl7kidivL7F+QXp7L864JmEYfQY1pwX94Noa382A9WP7uB0sdovdG6rXrU5DKyMigpqYm/H1VVVW426+nZs6cic02PNf42rlzJ/PmzYt2GaIPBsu1K2oo5bef/C/FjWUEyscSLJnc6QLEMYqf6x0fMM1SznveaWxwz5N5okYoHZV/uM+iWk/g0tidpLlc/OnQLHzjDmKbsZlA0TS06tF0NFXGYPi7HyyvP9F7w/na+Xy+Lht/+vRuO3v2bPLz8yksLETTNDZu3MjChQv7XKQQIsQwDN449AE/fvuXNPlauGfhHQSLp3YaokabavlBwkYmmSv5W8sCXnafISFqxFN4zzuD/2leTKraxI/4mNH7J6C3JGEdtxfrxM9kJnQh+lGfWqRsNhtr167ljjvuwOfzsWjRIpYulVtnhTgV1a5a/rD9WfZU7mPOqOl8a/71OO0JwKEO9jY413aAy2J30KzbebT5IgqCvWsVFsNbXiCH/25azn84PuA2+795o+g03k6chHn0YeynNeAvmI7ekBHtMoUY8noVpN59993w1wsWLGDDho5vvRZCdC88rQEGpvQiLDkHAQgUT2fLthy2bHivw+NiFD9fi9vCXGshe/3Z/NV1Dm7DHsHKxVBRoyewrmkZV8ZtZVnM50xxpfGX/XPwjD2KbfIugrWjCBRO73TsnRCie31qkRJCnLqGZh+KvQXL2L2YEurRGlIJFMzA8Md0eswUcxlXOzYTr3hY7z6d97wzMGTxYdGFAGaec53NgUAmV8Zu5cfK+/zj0BnsTsnAnHUEU0ItgaJprPj+y3S3kLVM3ClEexKkhIgCT8CLOecA5owC0E34j85Eq8mms3/IbARYFbuTc+wHqdAS+VPz+RRrPZtsUQhQ2Okfz9FgOt+I+5jr4jbzWVMuLzTMIzD2ENYJe9DSigkUTsfwdD4DvkzcKUR7EqSEiCDDMPiocDt/3f0ilsxGglWjCZRMgmDnd7BONpfxtbhPSFZbeMczndc8cwliimDVYrio1x081vwVFtu/4OKYz5hsVPDy4dPZkZiNJecQtpmb0SpzCZROBM0S7XKFGBIkSAkRIQdrjvLX3S+yv+YIE5LGUL5zGobL2en+8YqHy2K3M89WQLUWz++al5IvA8rFKTJQedc7kzx/Dl+L+4SrHZ9whjeD5z//Eo1ZFZgyCjGllBMonRiaKqGTO0aFECESpIQYYGVNFTz3+Xq2lXxGoj2B//rSNVww/mxWvfFKh/sr6JxrO8jymF1YFI3XPbN423OatEKJflWlJ/JY80XMtx1mVcxOfux4nfeqpvNW9RkYuYexjv0CfVQBgeLJ6PUZdDd+SoiRSoKUEAOk3tPIC3tf5d2jH2M1WfjqzBVcMvkC7JbO77CbZC5nVexOcsx17A9k8k/XfFlwWAwYA4VPfJPY6x/NytidfCUmjzP1I7xydA6fxo7FPPoQtkmfobckhgJVc0q0SxZi0JEgJUQ/q3M3sH7/m7x99CN0Q+eiiQu5fPoyEu2dB6IMtYGVsZ8y01pCnRbHn1vOY5d/LNIKICKh2YjhWde5fOSdwmVx2/mGYwsLgym8tH8ehc4g5uzD2KZtR2tKIq9yKjPSJ/d6WTAhhisJUkL0k2pXLev3vcm7+ZsxDJ3zxs5n9fRljHKkdXpMguJmScweFtgO4TfMrHefzr+906QbT0RFoZbGb5uWcbo1n5Wxn/LthDfZ58tk497ZlKe4sWTm88D7jzA1dQJXzFjOaRlTJVCJEU+ClBCnqKypgg373+KDgk9AUVg87mwunbaE9LjOu0GCzfVcFruds20HMaHzsW8ymzyzccnEmiLKjGNTJezx53Ku/QAX2vO4K2ETu125vJo3myv+cxwv73uDBz/4HZNSxrFq6kV8KWsWqiqD0sXIJEFKiD4wDIO9VQfYeOAdPi3Pw6Ka+cqEhbz7upVXPlF5hY86PC5BcfPlmL2cbTvIeTad7f4JvOk5jVq987l7hIiGAGbe885gs3cS59v3sdj+BafFF/H5X3NRvdPxJ3k54DvKr2v/gO6NJVgxJjQXmm6WiTvFiCJBSoheCGpBPi7awcaD71DYUEKCzcGVM5Zz0cSFJNoTeOm59R0el642cr59H2fYjmBCZ7t/PG96ZkmAEoOeDytveGfzoW8K59v3cZ7tALMTijjoHcVbe2dwJEHFPKoA69h9GKMPE6zKoaEyN9plCxExEqSE6IHKlmreOfox7+VvodHbxOiETG494xucO+ZMrKbOJi40mGCuYrF9L6dZSwgYKtt9E3jHO4MauRNPDDFuw85rnrm845nJ2faDnG//gtsS3qE4mMz7h6ey2zoZZVQR5syjmEfl898fN/KVCecxM2MKqiLdfmL4kiAlRCeCusbOsj28feRDdlfswzBAb0gjWDWPQ42prHu7hnW81u44C0HmWgs4z36AXHMtLbqNTZ5ZfOidQovR+Tp6QgwFPiy8553Bv71TOcN6lMUxX3CtYzOX6Ta2lE7iw9J5uFLr+MJ+kK0lu8hwpPGVCedy/tgFJNilBVYMPxKkhDhJSWM5/y7cygf5n1DvbSQ5xkmgdAJa9eguFxTOUBs4x36QM6xHiVX9VGiJ/MM1n22+CQTkpSaGGQ0Tn/gn8Yl/IpPMFZxnP8CX7Xv5MnvJqx/Nx5UTqIs3UZZWwl9bXuKZXS+j1Weg1WahN6aCocpYKjEsyLu7EECDt4mPC7ezqfg9Kg/Xoioqc0ZN5+YJX2du5kwuvWtjh8fZCDDLWsRZtkNMtFQRNFR2+3P52DeFI8F0ZB4oMfwpHApmcqglkyS1hXNsBznLdphZ1mIatRh2FI5nszKLxtRGTCllmFMqMAJWtNpMGmuyMAxDplAQQ5oEKTFiuf0edpZ9zkdF29hdsQ/d0MmwpXD9nCs4Z8wZODuZQFNFZ6qljC9Zj3KatRirolGtxbPefTpbfRNlCgMxYtXrDjZ6Tud1z2ymW0qZbzvM+fYv+LKyl4KGVLZVjuOzGAeB1BpM6UWYRxXy/U0FnDvmDM4aPTfa5QvRJxKkxIjS4nOxvXQ3W0t2sadyP0E9SEpsEiunfoWFY+ZTebiMeVPmtTtOQWecuZo51kJOtxYQr3px6Va2+Saw3T+BgmAq0vokRIiGic8DuXweyCVe8TDPls9862G+GredKww4UpnB9rIc8hw24r6k8/znG3j+8w2kWpPIt1dw1ujTGZ2YGe0fQ4gekSAlhr06dwM7yz5na8ku9lYdQDN0UmOTWTJxEfNHz2Vy6rjwXUWVlIWPM7QAnoI8XAe28jPnh8SrXgKGSl4ghx2+cewLZKPJDORCdKnZiOF973Te905nlKmBudYC5loLuNqyAz2gEFsxE9/4C8mLtfBewS5eyHuVf+RtJDthFPNHz2Ve1mlMSB4jd/6JQUuClBh2dF3nYG0+u8rz2FWeR0FDCQCjHGmsmPoV5o+ey/ik3A7HZSh+Ny17P8J9eCfuQzvQfW4Uq51DgVHsDuSyz5+Nj86mOxBCdKVCc/K6Zw6ve2aTeSxUXdJUi/Huc8wApjrSMM04gy8SHexoLuGlfZt48YvXSbA5mJs5k9OzZjI7YzqxVrn7VQweEqTEsFDvaeTzyv3sKs9jd8U+WvwuVEVlSuoErpl1GXMzZ5CTmNUuPF13/2skuMuYZi1lmqWMXFMNVQq06DbyAqPZ4x/DgUCmrH0nRL9SKNeSKPck8c1vrsJfW4b78A4qP30f0/Y3mK5rzIyJRx93GkdSktmLhx1le/ig4BNMisrUtInMGTWD0zKmMNaZI8vTiKiSICWGpGZfC3urDpJXdYC9lQcpba4AINEWz7ys0zg9ayazMqYRZ41tc5xh6ASqS/AU7cVbtJe7TDuJS/SjG1CopfKGZzZfBLIp1pIxkDdnIQbaiu+fuBrAImIUP1MsZcz0lTAl71MmqF4mAAuMBPRZMzjgsPO5p5Zn97wEQJw1lhnpk5mZPoXTMqaSFZ8hdwGKiJIgJYaEBk8jB2vz2Vd9mLyqAxQe666zm21MS5vI4vFnMzN9MmOTctqMpTAMHX9VEd6ivXgK9+It3ofubgLAlJBKXiCHfYEsDgQyccvddkJEncew8pl/LJ/5xwIGo0yNTDaXM9lSzqyDu8nwuVkIeFKzKMzI5LDdxP6afLaVfAZAUkwiM9ImMzVtAlNSJ5CTkCUtVmJASZASg46maxQ3lnGg5igHao9yqOYola4aACyqmSmpE7jqtJXMTJ/C+OQxmNXj3W7BlgY8ZYfwlR7EV3YIb/kRDJ8bAHNiOrETTycmdwb2MTMwJ6Zz+w82ROVnFEL0hEKF5qRCc/Jv3zQ23HcJvvIjeAo+J6Z4P3FH9jPV62I50BCfSMGoTI4oVj4v38tHRdsBiLHYmZwynimpE5iaOp6JKeOwm23R/bHEsCJBSkSVbuhUttSQX1/E0fpijtYVcriuAG/QB4DTnsCU1AlcNHERU1LHMy4pB8uxte00TzP+oi9oqczHV3YYX+khgo1VoRMrKtb0MThmnIt99BRicmdgTkyL1o8phOgHimrCnj0Ze/Zk4FhXfU0J3pIDxJfsJ614P3PrKzCAOrNKSWoaRQlW8muL2FPxBQagKiqjEzIZn5zL+KRcJiSPYYxzdBdrZgrRNQlSImJ0XaespZL8umKO1heRX19EfkMxnoAXALNqJjcxi0Vjz2JK6ngmp04gLTYZMAjWV+CrLKQ5byv+ynx8VYVoTTXhc5sSUrFnTyLhS8uwZ0/COmo8qkX+1ynEcKYoKta0XKxpuSTM/QoAmqsRX8VRksuPkFN+hNMrjqI11eBRFYrsFoqdTsqMWnY2VfF+/hYATIrK6MQsJiTlMj45l9zE0eQmZsndgaJHJEiJfqcbOtWuWoobyyluLKO4KfS5rKmCgB4EwGKyMDYxm/PGnMn4pFzGJeUy2pGK0VhLoKYEf0kxgc+2UFZbir+mFONY2EJRsaRmE5MzDWvG2NBH+ljMDmebGq69fxMNzb4I/+RCiIHUdmB6x1rX74udcHym9NZwlVV+hNMrC/DXFOOvLafRBCU2M6V2G2X+MrY2lPFu/ubwcakxSeQ4s8lNzCI3MZucxCyyEzLCreJCgAQpcQq8QR8VzdVUtFRR3lxFWXMlJY3llDSV49P84f1SYpPISchkVsZUchJGkWOJJyOgozdWE6ivIHD0QwI1pRTXV4Chh48zJaRiTckmfs4FWNPHYssYiyUtB9Vs7bY2CVFCjEwdvfZNcYnETpjbJlwZwQD+2lImVxfjry7CX12Er7qI2pYaKqzm0IfNS3lTHXvMeWjHbgQ0oZARm0xWYmboIz6DzPh0MuMzSLTFyx2DI5AEKdElb8BLtbuO8uaqY4HpeHCq8zS02ddpT2B0QiaLc75EljmOUYaJdK8PS1MdwdJKAnn7CDZWg6FT1XqQasaSPAprei5x0xZgSR2NNSUbS0oWagfN6tLSJIToTm9armwZY9tsz/F7Q//BqysjUFdOoK4Mb10ZZU2VlOteKqxmqlo8FNdXssv6OdoJwSlGNTPKnkRWfDpZSaPJSMwkPS6FtLgUkmISZXb2YUqC1AhmGAYtfhfVrjpq3HVUu2qpPva5xlVHtbuOFr+rzTHx1lgybIlMtaeQZs8gNaiT4vWR3NKMuaaeYHMBaMHw/l7AH5uAxZmBLXsSjhnnYnZmYEnKwOLMwBSfzHUPvEXDwdZwVH/sIy9CvwUhxEjU0OzrQeAyoyi5GEYuADGKjzS1mTRTE5PUFs5Um7HaXeg2D35bgDqrSrXVzd7GCj6u3HvymUgy2UizxpMWm0x6fDoZziwykrNJiUvBaU9scweyGDokSA1TvqCfem8j9Z4G6j1Noc/eRuo8rdsaqfM0hO+Oa2VTzKSY7SRjZrRuxamrOH1+klwukpobidGq2j6RyYw5PgVzQirm0VMxxScf+z4lFJic6ai2tpNinkxamIQQg5VhHP/aY9go0mwUaanHN4ZmV0HB4MX7FxFsrCLYWI2noZzK+jKqXXVU+xqpCbipw0uduZnClkpctQfaPI9iQLyikqhaSTbHkmR1kByTSEpcCinxaaQ4M0lJyibOLt2Hg40EqSFC13Va/C6afC00+ZrbfvaGvm70NtHgaaTe24g72D6cmFFIxESCDmlBnfEBP06PF2cgSFJQxxnQiNUNFEAxWTDFJWKKc2KKS8eU6gx93xqSElIwx6egxiZ0+qIOdcPtHuDfjBBCRJ+Bgjk+CXN8EoyeggM4ecIVQ9fQXI0Em+twNVRS1VBKVXMVte566n1NNAS81OseKpQWDppq8LSoUN32HGbdwGEoJCgm4lUL8SYbCZYYEixxOGMSSLQnkBibTFJ8CgmONCyxCZhi4lBkgPyAkSAVYf6gH1fAg8vvxhVw4/K7afG7afE04vI00eJrocXbgtvvojngpingoTnoxa0HMDo5Z4xuEKfpxAU1kjSd3KBOgqaTENRJCGqhrxUzcVYH5pgY1BgHptj40OdwWHKGvzbHJaLYYrsJSGUD90sSQoghqCdjs+D4+KwUYFoHjxuGgeH34G6qobahnJrGCmpdtdR56kP/cQ64ada8NGh+SjQvLcEGNK8Cze3PZdd0YnWDGB1iUYlTTcSpFuJUK3FmG3FmOw5rLPHWOOJsDhz2eOJiErDb4zHZYlFtMSjWGFRbDKo1BsUkseFkp/QbeeWVV3jiiScIBALccMMNXHPNNf1V16BhGAYBvwe3twWPz4XbH/rs8bvx+ty4/S68fg+egBdv0IMn6KOuqYE3i1/CqwXw6kHcegAPGm5DJ9hNi6xFN4jRdWI0g1hdJ1XTGaPpxGkGDkL/A3GYbSSYY4i3xpJgjcNiiwuFopj49p/tDtQYR7/OqSRdcUII0Xc9G591PHCN6WY/wzBo8bVQ31hFQ0s1DS21NLjrafI00uJ30eL34Ap6cGl+yvQALsOPGx+G1gwa0MFbumoY2HQDu25g0/Vjnw3shoJdUYnBhF01YVctxKhmDL9Gy9GN2Mw27GY7dqsdm9mO3RaL3RqLyWpHtdhQLHYUixXVYkcxW1HMltCH6YSvzRaUITQwv89BqrKyknXr1vHiiy9itVq56qqrmD9/PhMnTuz1uQzDAEPH0IKgBTFaP3QttE0Pfa8F/QSDAfxBH8GgP/RZCxAIBghqfgJagKAWwB/049MD+LUAAS2IT/Pj1wL49QB+LYhfDxLQg/gMjYChETB0/IZOAIMABn4MAopBAPCrSpu7MrrS+odn1Q1s/tAfnBWFDEUlBjOxqpU41UKMKfQ/gVizHYclhjhrDHH2eBz2BKwx8ajWGFRb7LGP0NeK1Y5ywkDEzu9ecx/7qAxvaX0xdkfuiBNCiMGjp4FLUdqO5QqJP/Zx3In/FuiGjjfgOxa0XDT7mml2N9LsbsTlbcbjd+H2u3H7PXiCXjxBHx7NT70WwGsE8RgaAXRCKcwHFkCvAD+hD3fbakxG6N9Gq25gaf3aOOFrndDXhoFFNzCjYFZMWFQVi2LCopgwm8xYVRMW1YJVNWM2W7CaLFhMVqwmKxZz6MNqtmBWrZjMFhSTOdSKpppRTKbj35vMKKr5hMePP3Z8/9CHpnXWHxTS5yC1efNmzjrrLJxOJwBLlixh06ZN3H777T06/rf//DHNgSY0IIiBpigEFdAUJbRNUdpuU0DvhwF2ZsPAYihYACsqFhQsqopVsRCnmLAqJqyqCatqxmIyY1Mt2E02YsxWYo4l7RiLDbsllhhrDE+9cpjGZoWAZsOnW/BgodEwo9H27ouO/9BP5sMZD8/cf063P0dvAk9PX4xCCCGGnu7/bQk58d8NVVGJtcYQa40hndQujupcUNe44cFXafK4wKShmAJg0kDVUNTQZ5MpgEkNYLMbnDMnDV/Aiy/owxv0hRs5GvQAPi2I39Dw6UECho7ebjCLAQSOfXiPb9aOfXTAZBiYDQOzEWroMBmhDKAaYKKD7eFjTtiOQawlnjNm3djp76HPQaqqqoq0tOND6dLT09mzZ0+3xxnHrnijxYnbsKEYAAqKoWDWFcyGgmKoYCjHPlQMXcUwVNCPfW+YMAwFXTdhGCpWi5UrLpiKxWTBZLJiNVuxWuxYzTasFhtWsw2Lyczdj26m2dUmQ3dOgU4HJYV5gdGhJG4BFbAd++gzPci1927sdjdnnNwmK4QQond68u9Lz/79a91VJdFyrOXLAIId76c1wr/f6P58CXFWfvOdRWi6RtDQCGqhHqSgHiSoBfFrQYJGgKCmhbcHtGOfj+0TaP066Mev+dG00HbNCBLUNDQ9SFDX0I99bn0uXdcJGBpeXUczdIJo6IaOXY3hDI7nl5P1OUh1dMKe3JIZCAQAWD1heV+fumPHmhF1wEsALwFOHnl304V9S91CCCGEiIy8vJ7PI2gCTKiA9djHSRRCSacfxsgHAgHsdnu77X0+dUZGBjt27Ah/X1VVRXp6erfHxcXFMXnyZCwWi8yFIYQQQohBzTAMAoEAcXFxHT7e5yB19tln8+ijj1JXV0dMTAxvvvkmP/vZz7o9TlVV4uPju91PCCGEEGIw6KglqtUptUh997vf5brrriMQCHDFFVcwa9asvp5OCCGEEGLIUYzORk8JIYQQQoguDZ0Zr4QQQgghBhkJUkIIIYQQfSRBSgghhBCijyRICSGEEEL0kQQpIYQQQog+kiDVR2VlZVxzzTUsXbqUb37zm7hcrnb7+P1+7rrrLpYtW8Zll13GkSNH2jweDAb52te+xosvvhipsgWndu1cLhff/va3WbFiBStWrODVV1+NdPkj1iuvvMLFF1/MV77yFZ599tl2j+/bt4/LL7+cJUuW8P/+3/8jGAytVdGT6y0GVl+v3c6dO7n88stZtWoV119/PaWlpZEuXdD369fqiy++YObMmZEqN/IM0Sf/9V//ZWzcuNEwDMN47LHHjIcffrjdPn/605+Mn/zkJ4ZhGMa2bduMK664os3jjzzyiHHmmWca//rXvwa+YBF2KtfuN7/5jbF27VrDMAyjpqbGOOecc4zq6uoIVT5yVVRUGIsXLzbq6+sNl8tlrFixwjh06FCbfZYvX27s2rXLMAzD+PGPf2w8++yzhmH07HqLgXMq127x4sXGvn37DMMwjBdeeMG49dZbI1q7OLXrZxiG4Xa7ja997WvG5MmTI1l2REmLVB8EAgG2b9/OkiVLAFi9ejWbNm1qt9/777/PypUrATjjjDOor6+nrKwMCP1P68CBAyxevDhyhYtTvnZnnnkm1157LQApKSk4nU5qamoi9wOMUJs3b+ass87C6XQSGxvLkiVL2ly30tJSvF4vc+bMAY5f155ebzFw+nrt/H4/3/72t5k6dSoAU6ZMoby8PBo/wojW1+vXau3atdxwww0RrjqyJEj1QX19PQ6HA7M5NDF8WloalZWV7farqqoiLS0t/H1aWhoVFRW0tLSwdu1aHnjggYjVLEJO9dqdc845ZGVlAfDaa6/h9/uZOHFiZIofwU6+Hunp6W2uW0fXq7KyssfXWwycvl47q9XKqlWrANB1nccee4wLL7wwcoULoO/XD+Cdd97B6/WydOnSyBUcBf2wHvLw9vrrr/PQQw+12TZ27Nh2+/V0AWZVVfnpT3/KrbfeSmpqan+UKDoxENfuxHP/4he/4E9/+lP4H2kxcIwOFmA48bp19nh3x4mB19dr18rv93P33XcTDAa55ZZbBqZI0am+Xr/q6mqeeOIJ/vznPw9keYOC/AvQjWXLlrFs2bI22wKBAPPnz0fTNEwmE9XV1aSnp7c7Nj09nerqasaMGQNAdXU1aWlpbNmyhYMHD/K73/2O8vJyPvnkE8xmc7grSfSP/r52rfs988wzPPXUUzz11FNMmTJl4H8QQUZGBjt27Ah/X1VV1ea6ZWRktOlibb1eycnJtLS0dHu9xcDp67WD0M0d3/zmN3E6nTzxxBNYLJbIFS6Avl+/999/n4aGBq655prwY6tWreLZZ5/F4XBEpvgIka69PrBYLHzpS1/itddeA+Dll19m4cKF7fZbtGgR69evB2DHjh3YbDays7P56KOPWL9+PevXr+eCCy7gzjvvlBAVIady7bKysnj77bf585//zN/+9jcJURF09tlns2XLFurq6vB4PLz55pttrlt2djY2m42dO3cCx69rT6+3GDh9vXYAd911F2PGjOG3v/0tVqs1KvWPdH29fldeeSVvv/12+N86gPXr1w+7EAWyaHGflZaWcvfdd1NbW0tmZia/+c1vSExM5G9/+xtVVVV8+9vfxufzce+995KXl4fVauXBBx9kxowZbc5z9913c+aZZ7J69eoo/SQjz6lcu5UrV1JXV0dKSkr4fA8++CCnnXZaFH+ikeGVV17hD3/4A4FAgCuuuIKbb76Zm2++mTvvvJPTTjuN/fv3s2bNGlwuF9OnT+ehhx7CarV2er1F5PTl2h0+fJjLLruMiRMnhrvP09PT+Z//+Z8o/zQjT19feyeaMmUKBw4ciNJPMLAkSAkhhBBC9JF07QkhhBBC9JEEKSGEEEKIPpIgJYQQQgjRRxKkhBBCCCH6SIKUEEIIIUQfSZASQgghhOgjCVJCCCGEEH0kQUoIIYQQoo/+PzIACgPmw6U1AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1,1,figsize=(10,6))\n",
    "\n",
    "y = genmod.rvs(size=10000)\n",
    "y_ = np.linspace(y.min(), y.max(), 10000)\n",
    "\n",
    "bins = np.linspace(y.min(), y.max(), 100)\n",
    "ax.hist(y, bins=bins, density=True, label='Random Samples')\n",
    "ax.plot(y_, genmod.pdf(y_), label='Generative Model')\n",
    "ax.plot(y_, phat_fit.pdf(y_), label='PhatNetBeta')\n",
    "\n",
    "ax.set_xlim((-.05, .05))\n",
    "\n",
    "ax.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## PhatLoss: A Custom Loss Function ##\n",
    "\n",
    "As with our MLE, we can incorporate a tail estimation method into our process in order to alleviate underfitting in the tails. To do so, we have to again amend the network model, including the use of a new custom loss function, `PhatLoss`. The updated network is available as `PhatNet`.\n",
    "\n",
    "First, we will generate some data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "genmod = ph.Phat(.0003, .0032, .25, .29)\n",
    "\n",
    "n = 60000\n",
    "y = genmod.rvs(size=n)\n",
    "data = ph.DataSplit(y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Then, we will find our estimate of each tail index."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "29a8d056e8944e39807989a6722ecc4b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "xi_l, xi_r = ph.two_tailed_hill_double_bootstrap(data.raw.y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For our neural net, we want to train the shape parameters against our bootstrapped estimates and the remaining parameters against our Phat distribution. This means we must employ two different loss functions:\n",
    "\n",
    "1. For each of the shape parameters, we use the asymptotic error discussed earlier, however, we take the log of both values to create a more valuable gradient (same concept as log-likelihood): \n",
    "\n",
    "    $$\\text{AMLSE} = E[(log\\hat{\\xi} - log\\xi)^2]$$\n",
    "\n",
    "    where: $\\xi$ is now the estimate derived from the double bootstrap method and $\\hat{\\xi}$ is the value resulting from the gradient descent.\n",
    "\n",
    "    The AMSE of the left and right tails is then averaged. This is the `TailLoss`.\n",
    "\n",
    "2. For the body params, $\\mu$ and $\\sigma$, we will continue to use the negative log-likelihood of the resulting Phat distribution. Note this means the ongoing shape parameters must also be provided to the $mu$ and $sigma$ loss calculation. This is the `BodyLoss` used earlier.\n",
    "\n",
    "We then combine these two losses, at each step, into a single loss driver according to the formula:\n",
    "\n",
    "$$ \\text{Loss} = \\frac{\\text{Loss}_{\\textit{body}}}{\\text{Loss}_{\\textit{tail}}+1}$$\n",
    "\n",
    "The above relationship was established empirically. As we'll see, it produces a loss curve that scales the relative importance of both the Body and Tail losses and allows for asymptotic behavior in either without negating it in the other. This seems to produce good convergence, although it does have [scale drawbacks](#Caution-On-Scaling).\n",
    "\n",
    "Below we can see the surface of the resulting loss function in terms of its constituents, the log-likelihood of the Phat distribution and the average of the AMLSE of each tail index."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "tags": [
     "hide_input"
    ]
   },
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "/* Put everything inside the global mpl namespace */\n",
       "/* global mpl */\n",
       "window.mpl = {};\n",
       "\n",
       "mpl.get_websocket_type = function () {\n",
       "    if (typeof WebSocket !== 'undefined') {\n",
       "        return WebSocket;\n",
       "    } else if (typeof MozWebSocket !== 'undefined') {\n",
       "        return MozWebSocket;\n",
       "    } else {\n",
       "        alert(\n",
       "            'Your browser does not have WebSocket support. ' +\n",
       "                'Please try Chrome, Safari or Firefox ≥ 6. ' +\n",
       "                'Firefox 4 and 5 are also supported but you ' +\n",
       "                'have to enable WebSockets in about:config.'\n",
       "        );\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure = function (figure_id, websocket, ondownload, parent_element) {\n",
       "    this.id = figure_id;\n",
       "\n",
       "    this.ws = websocket;\n",
       "\n",
       "    this.supports_binary = this.ws.binaryType !== undefined;\n",
       "\n",
       "    if (!this.supports_binary) {\n",
       "        var warnings = document.getElementById('mpl-warnings');\n",
       "        if (warnings) {\n",
       "            warnings.style.display = 'block';\n",
       "            warnings.textContent =\n",
       "                'This browser does not support binary websocket messages. ' +\n",
       "                'Performance may be slow.';\n",
       "        }\n",
       "    }\n",
       "\n",
       "    this.imageObj = new Image();\n",
       "\n",
       "    this.context = undefined;\n",
       "    this.message = undefined;\n",
       "    this.canvas = undefined;\n",
       "    this.rubberband_canvas = undefined;\n",
       "    this.rubberband_context = undefined;\n",
       "    this.format_dropdown = undefined;\n",
       "\n",
       "    this.image_mode = 'full';\n",
       "\n",
       "    this.root = document.createElement('div');\n",
       "    this.root.setAttribute('style', 'display: inline-block');\n",
       "    this._root_extra_style(this.root);\n",
       "\n",
       "    parent_element.appendChild(this.root);\n",
       "\n",
       "    this._init_header(this);\n",
       "    this._init_canvas(this);\n",
       "    this._init_toolbar(this);\n",
       "\n",
       "    var fig = this;\n",
       "\n",
       "    this.waiting = false;\n",
       "\n",
       "    this.ws.onopen = function () {\n",
       "        fig.send_message('supports_binary', { value: fig.supports_binary });\n",
       "        fig.send_message('send_image_mode', {});\n",
       "        if (fig.ratio !== 1) {\n",
       "            fig.send_message('set_dpi_ratio', { dpi_ratio: fig.ratio });\n",
       "        }\n",
       "        fig.send_message('refresh', {});\n",
       "    };\n",
       "\n",
       "    this.imageObj.onload = function () {\n",
       "        if (fig.image_mode === 'full') {\n",
       "            // Full images could contain transparency (where diff images\n",
       "            // almost always do), so we need to clear the canvas so that\n",
       "            // there is no ghosting.\n",
       "            fig.context.clearRect(0, 0, fig.canvas.width, fig.canvas.height);\n",
       "        }\n",
       "        fig.context.drawImage(fig.imageObj, 0, 0);\n",
       "    };\n",
       "\n",
       "    this.imageObj.onunload = function () {\n",
       "        fig.ws.close();\n",
       "    };\n",
       "\n",
       "    this.ws.onmessage = this._make_on_message_function(this);\n",
       "\n",
       "    this.ondownload = ondownload;\n",
       "};\n",
       "\n",
       "mpl.figure.prototype._init_header = function () {\n",
       "    var titlebar = document.createElement('div');\n",
       "    titlebar.classList =\n",
       "        'ui-dialog-titlebar ui-widget-header ui-corner-all ui-helper-clearfix';\n",
       "    var titletext = document.createElement('div');\n",
       "    titletext.classList = 'ui-dialog-title';\n",
       "    titletext.setAttribute(\n",
       "        'style',\n",
       "        'width: 100%; text-align: center; padding: 3px;'\n",
       "    );\n",
       "    titlebar.appendChild(titletext);\n",
       "    this.root.appendChild(titlebar);\n",
       "    this.header = titletext;\n",
       "};\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function (_canvas_div) {};\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function (_canvas_div) {};\n",
       "\n",
       "mpl.figure.prototype._init_canvas = function () {\n",
       "    var fig = this;\n",
       "\n",
       "    var canvas_div = (this.canvas_div = document.createElement('div'));\n",
       "    canvas_div.setAttribute(\n",
       "        'style',\n",
       "        'border: 1px solid #ddd;' +\n",
       "            'box-sizing: content-box;' +\n",
       "            'clear: both;' +\n",
       "            'min-height: 1px;' +\n",
       "            'min-width: 1px;' +\n",
       "            'outline: 0;' +\n",
       "            'overflow: hidden;' +\n",
       "            'position: relative;' +\n",
       "            'resize: both;'\n",
       "    );\n",
       "\n",
       "    function on_keyboard_event_closure(name) {\n",
       "        return function (event) {\n",
       "            return fig.key_event(event, name);\n",
       "        };\n",
       "    }\n",
       "\n",
       "    canvas_div.addEventListener(\n",
       "        'keydown',\n",
       "        on_keyboard_event_closure('key_press')\n",
       "    );\n",
       "    canvas_div.addEventListener(\n",
       "        'keyup',\n",
       "        on_keyboard_event_closure('key_release')\n",
       "    );\n",
       "\n",
       "    this._canvas_extra_style(canvas_div);\n",
       "    this.root.appendChild(canvas_div);\n",
       "\n",
       "    var canvas = (this.canvas = document.createElement('canvas'));\n",
       "    canvas.classList.add('mpl-canvas');\n",
       "    canvas.setAttribute('style', 'box-sizing: content-box;');\n",
       "\n",
       "    this.context = canvas.getContext('2d');\n",
       "\n",
       "    var backingStore =\n",
       "        this.context.backingStorePixelRatio ||\n",
       "        this.context.webkitBackingStorePixelRatio ||\n",
       "        this.context.mozBackingStorePixelRatio ||\n",
       "        this.context.msBackingStorePixelRatio ||\n",
       "        this.context.oBackingStorePixelRatio ||\n",
       "        this.context.backingStorePixelRatio ||\n",
       "        1;\n",
       "\n",
       "    this.ratio = (window.devicePixelRatio || 1) / backingStore;\n",
       "\n",
       "    var rubberband_canvas = (this.rubberband_canvas = document.createElement(\n",
       "        'canvas'\n",
       "    ));\n",
       "    rubberband_canvas.setAttribute(\n",
       "        'style',\n",
       "        'box-sizing: content-box; position: absolute; left: 0; top: 0; z-index: 1;'\n",
       "    );\n",
       "\n",
       "    // Apply a ponyfill if ResizeObserver is not implemented by browser.\n",
       "    if (this.ResizeObserver === undefined) {\n",
       "        if (window.ResizeObserver !== undefined) {\n",
       "            this.ResizeObserver = window.ResizeObserver;\n",
       "        } else {\n",
       "            var obs = _JSXTOOLS_RESIZE_OBSERVER({});\n",
       "            this.ResizeObserver = obs.ResizeObserver;\n",
       "        }\n",
       "    }\n",
       "\n",
       "    this.resizeObserverInstance = new this.ResizeObserver(function (entries) {\n",
       "        var nentries = entries.length;\n",
       "        for (var i = 0; i < nentries; i++) {\n",
       "            var entry = entries[i];\n",
       "            var width, height;\n",
       "            if (entry.contentBoxSize) {\n",
       "                if (entry.contentBoxSize instanceof Array) {\n",
       "                    // Chrome 84 implements new version of spec.\n",
       "                    width = entry.contentBoxSize[0].inlineSize;\n",
       "                    height = entry.contentBoxSize[0].blockSize;\n",
       "                } else {\n",
       "                    // Firefox implements old version of spec.\n",
       "                    width = entry.contentBoxSize.inlineSize;\n",
       "                    height = entry.contentBoxSize.blockSize;\n",
       "                }\n",
       "            } else {\n",
       "                // Chrome <84 implements even older version of spec.\n",
       "                width = entry.contentRect.width;\n",
       "                height = entry.contentRect.height;\n",
       "            }\n",
       "\n",
       "            // Keep the size of the canvas and rubber band canvas in sync with\n",
       "            // the canvas container.\n",
       "            if (entry.devicePixelContentBoxSize) {\n",
       "                // Chrome 84 implements new version of spec.\n",
       "                canvas.setAttribute(\n",
       "                    'width',\n",
       "                    entry.devicePixelContentBoxSize[0].inlineSize\n",
       "                );\n",
       "                canvas.setAttribute(\n",
       "                    'height',\n",
       "                    entry.devicePixelContentBoxSize[0].blockSize\n",
       "                );\n",
       "            } else {\n",
       "                canvas.setAttribute('width', width * fig.ratio);\n",
       "                canvas.setAttribute('height', height * fig.ratio);\n",
       "            }\n",
       "            canvas.setAttribute(\n",
       "                'style',\n",
       "                'width: ' + width + 'px; height: ' + height + 'px;'\n",
       "            );\n",
       "\n",
       "            rubberband_canvas.setAttribute('width', width);\n",
       "            rubberband_canvas.setAttribute('height', height);\n",
       "\n",
       "            // And update the size in Python. We ignore the initial 0/0 size\n",
       "            // that occurs as the element is placed into the DOM, which should\n",
       "            // otherwise not happen due to the minimum size styling.\n",
       "            if (fig.ws.readyState == 1 && width != 0 && height != 0) {\n",
       "                fig.request_resize(width, height);\n",
       "            }\n",
       "        }\n",
       "    });\n",
       "    this.resizeObserverInstance.observe(canvas_div);\n",
       "\n",
       "    function on_mouse_event_closure(name) {\n",
       "        return function (event) {\n",
       "            return fig.mouse_event(event, name);\n",
       "        };\n",
       "    }\n",
       "\n",
       "    rubberband_canvas.addEventListener(\n",
       "        'mousedown',\n",
       "        on_mouse_event_closure('button_press')\n",
       "    );\n",
       "    rubberband_canvas.addEventListener(\n",
       "        'mouseup',\n",
       "        on_mouse_event_closure('button_release')\n",
       "    );\n",
       "    rubberband_canvas.addEventListener(\n",
       "        'dblclick',\n",
       "        on_mouse_event_closure('dblclick')\n",
       "    );\n",
       "    // Throttle sequential mouse events to 1 every 20ms.\n",
       "    rubberband_canvas.addEventListener(\n",
       "        'mousemove',\n",
       "        on_mouse_event_closure('motion_notify')\n",
       "    );\n",
       "\n",
       "    rubberband_canvas.addEventListener(\n",
       "        'mouseenter',\n",
       "        on_mouse_event_closure('figure_enter')\n",
       "    );\n",
       "    rubberband_canvas.addEventListener(\n",
       "        'mouseleave',\n",
       "        on_mouse_event_closure('figure_leave')\n",
       "    );\n",
       "\n",
       "    canvas_div.addEventListener('wheel', function (event) {\n",
       "        if (event.deltaY < 0) {\n",
       "            event.step = 1;\n",
       "        } else {\n",
       "            event.step = -1;\n",
       "        }\n",
       "        on_mouse_event_closure('scroll')(event);\n",
       "    });\n",
       "\n",
       "    canvas_div.appendChild(canvas);\n",
       "    canvas_div.appendChild(rubberband_canvas);\n",
       "\n",
       "    this.rubberband_context = rubberband_canvas.getContext('2d');\n",
       "    this.rubberband_context.strokeStyle = '#000000';\n",
       "\n",
       "    this._resize_canvas = function (width, height, forward) {\n",
       "        if (forward) {\n",
       "            canvas_div.style.width = width + 'px';\n",
       "            canvas_div.style.height = height + 'px';\n",
       "        }\n",
       "    };\n",
       "\n",
       "    // Disable right mouse context menu.\n",
       "    this.rubberband_canvas.addEventListener('contextmenu', function (_e) {\n",
       "        event.preventDefault();\n",
       "        return false;\n",
       "    });\n",
       "\n",
       "    function set_focus() {\n",
       "        canvas.focus();\n",
       "        canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    window.setTimeout(set_focus, 100);\n",
       "};\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function () {\n",
       "    var fig = this;\n",
       "\n",
       "    var toolbar = document.createElement('div');\n",
       "    toolbar.classList = 'mpl-toolbar';\n",
       "    this.root.appendChild(toolbar);\n",
       "\n",
       "    function on_click_closure(name) {\n",
       "        return function (_event) {\n",
       "            return fig.toolbar_button_onclick(name);\n",
       "        };\n",
       "    }\n",
       "\n",
       "    function on_mouseover_closure(tooltip) {\n",
       "        return function (event) {\n",
       "            if (!event.currentTarget.disabled) {\n",
       "                return fig.toolbar_button_onmouseover(tooltip);\n",
       "            }\n",
       "        };\n",
       "    }\n",
       "\n",
       "    fig.buttons = {};\n",
       "    var buttonGroup = document.createElement('div');\n",
       "    buttonGroup.classList = 'mpl-button-group';\n",
       "    for (var toolbar_ind in mpl.toolbar_items) {\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) {\n",
       "            /* Instead of a spacer, we start a new button group. */\n",
       "            if (buttonGroup.hasChildNodes()) {\n",
       "                toolbar.appendChild(buttonGroup);\n",
       "            }\n",
       "            buttonGroup = document.createElement('div');\n",
       "            buttonGroup.classList = 'mpl-button-group';\n",
       "            continue;\n",
       "        }\n",
       "\n",
       "        var button = (fig.buttons[name] = document.createElement('button'));\n",
       "        button.classList = 'mpl-widget';\n",
       "        button.setAttribute('role', 'button');\n",
       "        button.setAttribute('aria-disabled', 'false');\n",
       "        button.addEventListener('click', on_click_closure(method_name));\n",
       "        button.addEventListener('mouseover', on_mouseover_closure(tooltip));\n",
       "\n",
       "        var icon_img = document.createElement('img');\n",
       "        icon_img.src = '_images/' + image + '.png';\n",
       "        icon_img.srcset = '_images/' + image + '_large.png 2x';\n",
       "        icon_img.alt = tooltip;\n",
       "        button.appendChild(icon_img);\n",
       "\n",
       "        buttonGroup.appendChild(button);\n",
       "    }\n",
       "\n",
       "    if (buttonGroup.hasChildNodes()) {\n",
       "        toolbar.appendChild(buttonGroup);\n",
       "    }\n",
       "\n",
       "    var fmt_picker = document.createElement('select');\n",
       "    fmt_picker.classList = 'mpl-widget';\n",
       "    toolbar.appendChild(fmt_picker);\n",
       "    this.format_dropdown = fmt_picker;\n",
       "\n",
       "    for (var ind in mpl.extensions) {\n",
       "        var fmt = mpl.extensions[ind];\n",
       "        var option = document.createElement('option');\n",
       "        option.selected = fmt === mpl.default_extension;\n",
       "        option.innerHTML = fmt;\n",
       "        fmt_picker.appendChild(option);\n",
       "    }\n",
       "\n",
       "    var status_bar = document.createElement('span');\n",
       "    status_bar.classList = 'mpl-message';\n",
       "    toolbar.appendChild(status_bar);\n",
       "    this.message = status_bar;\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.request_resize = function (x_pixels, y_pixels) {\n",
       "    // Request matplotlib to resize the figure. Matplotlib will then trigger a resize in the client,\n",
       "    // which will in turn request a refresh of the image.\n",
       "    this.send_message('resize', { width: x_pixels, height: y_pixels });\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.send_message = function (type, properties) {\n",
       "    properties['type'] = type;\n",
       "    properties['figure_id'] = this.id;\n",
       "    this.ws.send(JSON.stringify(properties));\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.send_draw_message = function () {\n",
       "    if (!this.waiting) {\n",
       "        this.waiting = true;\n",
       "        this.ws.send(JSON.stringify({ type: 'draw', figure_id: this.id }));\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_save = function (fig, _msg) {\n",
       "    var format_dropdown = fig.format_dropdown;\n",
       "    var format = format_dropdown.options[format_dropdown.selectedIndex].value;\n",
       "    fig.ondownload(fig, format);\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_resize = function (fig, msg) {\n",
       "    var size = msg['size'];\n",
       "    if (size[0] !== fig.canvas.width || size[1] !== fig.canvas.height) {\n",
       "        fig._resize_canvas(size[0], size[1], msg['forward']);\n",
       "        fig.send_message('refresh', {});\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_rubberband = function (fig, msg) {\n",
       "    var x0 = msg['x0'] / fig.ratio;\n",
       "    var y0 = (fig.canvas.height - msg['y0']) / fig.ratio;\n",
       "    var x1 = msg['x1'] / fig.ratio;\n",
       "    var y1 = (fig.canvas.height - msg['y1']) / fig.ratio;\n",
       "    x0 = Math.floor(x0) + 0.5;\n",
       "    y0 = Math.floor(y0) + 0.5;\n",
       "    x1 = Math.floor(x1) + 0.5;\n",
       "    y1 = Math.floor(y1) + 0.5;\n",
       "    var min_x = Math.min(x0, x1);\n",
       "    var min_y = Math.min(y0, y1);\n",
       "    var width = Math.abs(x1 - x0);\n",
       "    var height = Math.abs(y1 - y0);\n",
       "\n",
       "    fig.rubberband_context.clearRect(\n",
       "        0,\n",
       "        0,\n",
       "        fig.canvas.width / fig.ratio,\n",
       "        fig.canvas.height / fig.ratio\n",
       "    );\n",
       "\n",
       "    fig.rubberband_context.strokeRect(min_x, min_y, width, height);\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_figure_label = function (fig, msg) {\n",
       "    // Updates the figure title.\n",
       "    fig.header.textContent = msg['label'];\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_cursor = function (fig, msg) {\n",
       "    var cursor = msg['cursor'];\n",
       "    switch (cursor) {\n",
       "        case 0:\n",
       "            cursor = 'pointer';\n",
       "            break;\n",
       "        case 1:\n",
       "            cursor = 'default';\n",
       "            break;\n",
       "        case 2:\n",
       "            cursor = 'crosshair';\n",
       "            break;\n",
       "        case 3:\n",
       "            cursor = 'move';\n",
       "            break;\n",
       "    }\n",
       "    fig.rubberband_canvas.style.cursor = cursor;\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_message = function (fig, msg) {\n",
       "    fig.message.textContent = msg['message'];\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_draw = function (fig, _msg) {\n",
       "    // Request the server to send over a new figure.\n",
       "    fig.send_draw_message();\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_image_mode = function (fig, msg) {\n",
       "    fig.image_mode = msg['mode'];\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_history_buttons = function (fig, msg) {\n",
       "    for (var key in msg) {\n",
       "        if (!(key in fig.buttons)) {\n",
       "            continue;\n",
       "        }\n",
       "        fig.buttons[key].disabled = !msg[key];\n",
       "        fig.buttons[key].setAttribute('aria-disabled', !msg[key]);\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_navigate_mode = function (fig, msg) {\n",
       "    if (msg['mode'] === 'PAN') {\n",
       "        fig.buttons['Pan'].classList.add('active');\n",
       "        fig.buttons['Zoom'].classList.remove('active');\n",
       "    } else if (msg['mode'] === 'ZOOM') {\n",
       "        fig.buttons['Pan'].classList.remove('active');\n",
       "        fig.buttons['Zoom'].classList.add('active');\n",
       "    } else {\n",
       "        fig.buttons['Pan'].classList.remove('active');\n",
       "        fig.buttons['Zoom'].classList.remove('active');\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function () {\n",
       "    // Called whenever the canvas gets updated.\n",
       "    this.send_message('ack', {});\n",
       "};\n",
       "\n",
       "// A function to construct a web socket function for onmessage handling.\n",
       "// Called in the figure constructor.\n",
       "mpl.figure.prototype._make_on_message_function = function (fig) {\n",
       "    return function socket_on_message(evt) {\n",
       "        if (evt.data instanceof Blob) {\n",
       "            var img = evt.data;\n",
       "            if (img.type !== 'image/png') {\n",
       "                /* FIXME: We get \"Resource interpreted as Image but\n",
       "                 * transferred with MIME type text/plain:\" errors on\n",
       "                 * Chrome.  But how to set the MIME type?  It doesn't seem\n",
       "                 * to be part of the websocket stream */\n",
       "                img.type = 'image/png';\n",
       "            }\n",
       "\n",
       "            /* Free the memory for the previous frames */\n",
       "            if (fig.imageObj.src) {\n",
       "                (window.URL || window.webkitURL).revokeObjectURL(\n",
       "                    fig.imageObj.src\n",
       "                );\n",
       "            }\n",
       "\n",
       "            fig.imageObj.src = (window.URL || window.webkitURL).createObjectURL(\n",
       "                img\n",
       "            );\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        } else if (\n",
       "            typeof evt.data === 'string' &&\n",
       "            evt.data.slice(0, 21) === 'data:image/png;base64'\n",
       "        ) {\n",
       "            fig.imageObj.src = evt.data;\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        var msg = JSON.parse(evt.data);\n",
       "        var msg_type = msg['type'];\n",
       "\n",
       "        // Call the  \"handle_{type}\" callback, which takes\n",
       "        // the figure and JSON message as its only arguments.\n",
       "        try {\n",
       "            var callback = fig['handle_' + msg_type];\n",
       "        } catch (e) {\n",
       "            console.log(\n",
       "                \"No handler for the '\" + msg_type + \"' message type: \",\n",
       "                msg\n",
       "            );\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        if (callback) {\n",
       "            try {\n",
       "                // console.log(\"Handling '\" + msg_type + \"' message: \", msg);\n",
       "                callback(fig, msg);\n",
       "            } catch (e) {\n",
       "                console.log(\n",
       "                    \"Exception inside the 'handler_\" + msg_type + \"' callback:\",\n",
       "                    e,\n",
       "                    e.stack,\n",
       "                    msg\n",
       "                );\n",
       "            }\n",
       "        }\n",
       "    };\n",
       "};\n",
       "\n",
       "// from http://stackoverflow.com/questions/1114465/getting-mouse-location-in-canvas\n",
       "mpl.findpos = function (e) {\n",
       "    //this section is from http://www.quirksmode.org/js/events_properties.html\n",
       "    var targ;\n",
       "    if (!e) {\n",
       "        e = window.event;\n",
       "    }\n",
       "    if (e.target) {\n",
       "        targ = e.target;\n",
       "    } else if (e.srcElement) {\n",
       "        targ = e.srcElement;\n",
       "    }\n",
       "    if (targ.nodeType === 3) {\n",
       "        // defeat Safari bug\n",
       "        targ = targ.parentNode;\n",
       "    }\n",
       "\n",
       "    // pageX,Y are the mouse positions relative to the document\n",
       "    var boundingRect = targ.getBoundingClientRect();\n",
       "    var x = e.pageX - (boundingRect.left + document.body.scrollLeft);\n",
       "    var y = e.pageY - (boundingRect.top + document.body.scrollTop);\n",
       "\n",
       "    return { x: x, y: y };\n",
       "};\n",
       "\n",
       "/*\n",
       " * return a copy of an object with only non-object keys\n",
       " * we need this to avoid circular references\n",
       " * http://stackoverflow.com/a/24161582/3208463\n",
       " */\n",
       "function simpleKeys(original) {\n",
       "    return Object.keys(original).reduce(function (obj, key) {\n",
       "        if (typeof original[key] !== 'object') {\n",
       "            obj[key] = original[key];\n",
       "        }\n",
       "        return obj;\n",
       "    }, {});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.mouse_event = function (event, name) {\n",
       "    var canvas_pos = mpl.findpos(event);\n",
       "\n",
       "    if (name === 'button_press') {\n",
       "        this.canvas.focus();\n",
       "        this.canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    var x = canvas_pos.x * this.ratio;\n",
       "    var y = canvas_pos.y * this.ratio;\n",
       "\n",
       "    this.send_message(name, {\n",
       "        x: x,\n",
       "        y: y,\n",
       "        button: event.button,\n",
       "        step: event.step,\n",
       "        guiEvent: simpleKeys(event),\n",
       "    });\n",
       "\n",
       "    /* This prevents the web browser from automatically changing to\n",
       "     * the text insertion cursor when the button is pressed.  We want\n",
       "     * to control all of the cursor setting manually through the\n",
       "     * 'cursor' event from matplotlib */\n",
       "    event.preventDefault();\n",
       "    return false;\n",
       "};\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function (_event, _name) {\n",
       "    // Handle any extra behaviour associated with a key event\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.key_event = function (event, name) {\n",
       "    // Prevent repeat events\n",
       "    if (name === 'key_press') {\n",
       "        if (event.key === this._key) {\n",
       "            return;\n",
       "        } else {\n",
       "            this._key = event.key;\n",
       "        }\n",
       "    }\n",
       "    if (name === 'key_release') {\n",
       "        this._key = null;\n",
       "    }\n",
       "\n",
       "    var value = '';\n",
       "    if (event.ctrlKey && event.key !== 'Control') {\n",
       "        value += 'ctrl+';\n",
       "    }\n",
       "    else if (event.altKey && event.key !== 'Alt') {\n",
       "        value += 'alt+';\n",
       "    }\n",
       "    else if (event.shiftKey && event.key !== 'Shift') {\n",
       "        value += 'shift+';\n",
       "    }\n",
       "\n",
       "    value += 'k' + event.key;\n",
       "\n",
       "    this._key_event_extra(event, name);\n",
       "\n",
       "    this.send_message(name, { key: value, guiEvent: simpleKeys(event) });\n",
       "    return false;\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onclick = function (name) {\n",
       "    if (name === 'download') {\n",
       "        this.handle_save(this, null);\n",
       "    } else {\n",
       "        this.send_message('toolbar_button', { name: name });\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onmouseover = function (tooltip) {\n",
       "    this.message.textContent = tooltip;\n",
       "};\n",
       "\n",
       "///////////////// REMAINING CONTENT GENERATED BY embed_js.py /////////////////\n",
       "// prettier-ignore\n",
       "var _JSXTOOLS_RESIZE_OBSERVER=function(A){var t,i=new WeakMap,n=new WeakMap,a=new WeakMap,r=new WeakMap,o=new Set;function s(e){if(!(this instanceof s))throw new TypeError(\"Constructor requires 'new' operator\");i.set(this,e)}function h(){throw new TypeError(\"Function is not a constructor\")}function c(e,t,i,n){e=0 in arguments?Number(arguments[0]):0,t=1 in arguments?Number(arguments[1]):0,i=2 in arguments?Number(arguments[2]):0,n=3 in arguments?Number(arguments[3]):0,this.right=(this.x=this.left=e)+(this.width=i),this.bottom=(this.y=this.top=t)+(this.height=n),Object.freeze(this)}function d(){t=requestAnimationFrame(d);var s=new WeakMap,p=new Set;o.forEach((function(t){r.get(t).forEach((function(i){var r=t instanceof window.SVGElement,o=a.get(t),d=r?0:parseFloat(o.paddingTop),f=r?0:parseFloat(o.paddingRight),l=r?0:parseFloat(o.paddingBottom),u=r?0:parseFloat(o.paddingLeft),g=r?0:parseFloat(o.borderTopWidth),m=r?0:parseFloat(o.borderRightWidth),w=r?0:parseFloat(o.borderBottomWidth),b=u+f,F=d+l,v=(r?0:parseFloat(o.borderLeftWidth))+m,W=g+w,y=r?0:t.offsetHeight-W-t.clientHeight,E=r?0:t.offsetWidth-v-t.clientWidth,R=b+v,z=F+W,M=r?t.width:parseFloat(o.width)-R-E,O=r?t.height:parseFloat(o.height)-z-y;if(n.has(t)){var k=n.get(t);if(k[0]===M&&k[1]===O)return}n.set(t,[M,O]);var S=Object.create(h.prototype);S.target=t,S.contentRect=new c(u,d,M,O),s.has(i)||(s.set(i,[]),p.add(i)),s.get(i).push(S)}))})),p.forEach((function(e){i.get(e).call(e,s.get(e),e)}))}return s.prototype.observe=function(i){if(i instanceof window.Element){r.has(i)||(r.set(i,new Set),o.add(i),a.set(i,window.getComputedStyle(i)));var n=r.get(i);n.has(this)||n.add(this),cancelAnimationFrame(t),t=requestAnimationFrame(d)}},s.prototype.unobserve=function(i){if(i instanceof window.Element&&r.has(i)){var n=r.get(i);n.has(this)&&(n.delete(this),n.size||(r.delete(i),o.delete(i))),n.size||r.delete(i),o.size||cancelAnimationFrame(t)}},A.DOMRectReadOnly=c,A.ResizeObserver=s,A.ResizeObserverEntry=h,A}; // eslint-disable-line\n",
       "mpl.toolbar_items = [[\"Home\", \"Reset original view\", \"fa fa-home icon-home\", \"home\"], [\"Back\", \"Back to previous view\", \"fa fa-arrow-left icon-arrow-left\", \"back\"], [\"Forward\", \"Forward to next view\", \"fa fa-arrow-right icon-arrow-right\", \"forward\"], [\"\", \"\", \"\", \"\"], [\"Pan\", \"Left button pans, Right button zooms\\nx/y fixes axis, CTRL fixes aspect\", \"fa fa-arrows icon-move\", \"pan\"], [\"Zoom\", \"Zoom to rectangle\\nx/y fixes axis, CTRL fixes aspect\", \"fa fa-square-o icon-check-empty\", \"zoom\"], [\"\", \"\", \"\", \"\"], [\"Download\", \"Download plot\", \"fa fa-floppy-o icon-save\", \"download\"]];\n",
       "\n",
       "mpl.extensions = [\"eps\", \"jpeg\", \"pgf\", \"pdf\", \"png\", \"ps\", \"raw\", \"svg\", \"tif\"];\n",
       "\n",
       "mpl.default_extension = \"png\";/* global mpl */\n",
       "\n",
       "var comm_websocket_adapter = function (comm) {\n",
       "    // Create a \"websocket\"-like object which calls the given IPython comm\n",
       "    // object with the appropriate methods. Currently this is a non binary\n",
       "    // socket, so there is still some room for performance tuning.\n",
       "    var ws = {};\n",
       "\n",
       "    ws.binaryType = comm.kernel.ws.binaryType;\n",
       "    ws.readyState = comm.kernel.ws.readyState;\n",
       "    function updateReadyState(_event) {\n",
       "        if (comm.kernel.ws) {\n",
       "            ws.readyState = comm.kernel.ws.readyState;\n",
       "        } else {\n",
       "            ws.readyState = 3; // Closed state.\n",
       "        }\n",
       "    }\n",
       "    comm.kernel.ws.addEventListener('open', updateReadyState);\n",
       "    comm.kernel.ws.addEventListener('close', updateReadyState);\n",
       "    comm.kernel.ws.addEventListener('error', updateReadyState);\n",
       "\n",
       "    ws.close = function () {\n",
       "        comm.close();\n",
       "    };\n",
       "    ws.send = function (m) {\n",
       "        //console.log('sending', m);\n",
       "        comm.send(m);\n",
       "    };\n",
       "    // Register the callback with on_msg.\n",
       "    comm.on_msg(function (msg) {\n",
       "        //console.log('receiving', msg['content']['data'], msg);\n",
       "        var data = msg['content']['data'];\n",
       "        if (data['blob'] !== undefined) {\n",
       "            data = {\n",
       "                data: new Blob(msg['buffers'], { type: data['blob'] }),\n",
       "            };\n",
       "        }\n",
       "        // Pass the mpl event to the overridden (by mpl) onmessage function.\n",
       "        ws.onmessage(data);\n",
       "    });\n",
       "    return ws;\n",
       "};\n",
       "\n",
       "mpl.mpl_figure_comm = function (comm, msg) {\n",
       "    // This is the function which gets called when the mpl process\n",
       "    // starts-up an IPython Comm through the \"matplotlib\" channel.\n",
       "\n",
       "    var id = msg.content.data.id;\n",
       "    // Get hold of the div created by the display call when the Comm\n",
       "    // socket was opened in Python.\n",
       "    var element = document.getElementById(id);\n",
       "    var ws_proxy = comm_websocket_adapter(comm);\n",
       "\n",
       "    function ondownload(figure, _format) {\n",
       "        window.open(figure.canvas.toDataURL());\n",
       "    }\n",
       "\n",
       "    var fig = new mpl.figure(id, ws_proxy, ondownload, element);\n",
       "\n",
       "    // Call onopen now - mpl needs it, as it is assuming we've passed it a real\n",
       "    // web socket which is closed, not our websocket->open comm proxy.\n",
       "    ws_proxy.onopen();\n",
       "\n",
       "    fig.parent_element = element;\n",
       "    fig.cell_info = mpl.find_output_cell(\"<div id='\" + id + \"'></div>\");\n",
       "    if (!fig.cell_info) {\n",
       "        console.error('Failed to find cell for figure', id, fig);\n",
       "        return;\n",
       "    }\n",
       "    fig.cell_info[0].output_area.element.on(\n",
       "        'cleared',\n",
       "        { fig: fig },\n",
       "        fig._remove_fig_handler\n",
       "    );\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_close = function (fig, msg) {\n",
       "    var width = fig.canvas.width / fig.ratio;\n",
       "    fig.cell_info[0].output_area.element.off(\n",
       "        'cleared',\n",
       "        fig._remove_fig_handler\n",
       "    );\n",
       "    fig.resizeObserverInstance.unobserve(fig.canvas_div);\n",
       "\n",
       "    // Update the output cell to use the data from the current canvas.\n",
       "    fig.push_to_output();\n",
       "    var dataURL = fig.canvas.toDataURL();\n",
       "    // Re-enable the keyboard manager in IPython - without this line, in FF,\n",
       "    // the notebook keyboard shortcuts fail.\n",
       "    IPython.keyboard_manager.enable();\n",
       "    fig.parent_element.innerHTML =\n",
       "        '<img src=\"' + dataURL + '\" width=\"' + width + '\">';\n",
       "    fig.close_ws(fig, msg);\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.close_ws = function (fig, msg) {\n",
       "    fig.send_message('closing', msg);\n",
       "    // fig.ws.close()\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.push_to_output = function (_remove_interactive) {\n",
       "    // Turn the data on the canvas into data in the output cell.\n",
       "    var width = this.canvas.width / this.ratio;\n",
       "    var dataURL = this.canvas.toDataURL();\n",
       "    this.cell_info[1]['text/html'] =\n",
       "        '<img src=\"' + dataURL + '\" width=\"' + width + '\">';\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function () {\n",
       "    // Tell IPython that the notebook contents must change.\n",
       "    IPython.notebook.set_dirty(true);\n",
       "    this.send_message('ack', {});\n",
       "    var fig = this;\n",
       "    // Wait a second, then push the new image to the DOM so\n",
       "    // that it is saved nicely (might be nice to debounce this).\n",
       "    setTimeout(function () {\n",
       "        fig.push_to_output();\n",
       "    }, 1000);\n",
       "};\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function () {\n",
       "    var fig = this;\n",
       "\n",
       "    var toolbar = document.createElement('div');\n",
       "    toolbar.classList = 'btn-toolbar';\n",
       "    this.root.appendChild(toolbar);\n",
       "\n",
       "    function on_click_closure(name) {\n",
       "        return function (_event) {\n",
       "            return fig.toolbar_button_onclick(name);\n",
       "        };\n",
       "    }\n",
       "\n",
       "    function on_mouseover_closure(tooltip) {\n",
       "        return function (event) {\n",
       "            if (!event.currentTarget.disabled) {\n",
       "                return fig.toolbar_button_onmouseover(tooltip);\n",
       "            }\n",
       "        };\n",
       "    }\n",
       "\n",
       "    fig.buttons = {};\n",
       "    var buttonGroup = document.createElement('div');\n",
       "    buttonGroup.classList = 'btn-group';\n",
       "    var button;\n",
       "    for (var toolbar_ind in mpl.toolbar_items) {\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) {\n",
       "            /* Instead of a spacer, we start a new button group. */\n",
       "            if (buttonGroup.hasChildNodes()) {\n",
       "                toolbar.appendChild(buttonGroup);\n",
       "            }\n",
       "            buttonGroup = document.createElement('div');\n",
       "            buttonGroup.classList = 'btn-group';\n",
       "            continue;\n",
       "        }\n",
       "\n",
       "        button = fig.buttons[name] = document.createElement('button');\n",
       "        button.classList = 'btn btn-default';\n",
       "        button.href = '#';\n",
       "        button.title = name;\n",
       "        button.innerHTML = '<i class=\"fa ' + image + ' fa-lg\"></i>';\n",
       "        button.addEventListener('click', on_click_closure(method_name));\n",
       "        button.addEventListener('mouseover', on_mouseover_closure(tooltip));\n",
       "        buttonGroup.appendChild(button);\n",
       "    }\n",
       "\n",
       "    if (buttonGroup.hasChildNodes()) {\n",
       "        toolbar.appendChild(buttonGroup);\n",
       "    }\n",
       "\n",
       "    // Add the status bar.\n",
       "    var status_bar = document.createElement('span');\n",
       "    status_bar.classList = 'mpl-message pull-right';\n",
       "    toolbar.appendChild(status_bar);\n",
       "    this.message = status_bar;\n",
       "\n",
       "    // Add the close button to the window.\n",
       "    var buttongrp = document.createElement('div');\n",
       "    buttongrp.classList = 'btn-group inline pull-right';\n",
       "    button = document.createElement('button');\n",
       "    button.classList = 'btn btn-mini btn-primary';\n",
       "    button.href = '#';\n",
       "    button.title = 'Stop Interaction';\n",
       "    button.innerHTML = '<i class=\"fa fa-power-off icon-remove icon-large\"></i>';\n",
       "    button.addEventListener('click', function (_evt) {\n",
       "        fig.handle_close(fig, {});\n",
       "    });\n",
       "    button.addEventListener(\n",
       "        'mouseover',\n",
       "        on_mouseover_closure('Stop Interaction')\n",
       "    );\n",
       "    buttongrp.appendChild(button);\n",
       "    var titlebar = this.root.querySelector('.ui-dialog-titlebar');\n",
       "    titlebar.insertBefore(buttongrp, titlebar.firstChild);\n",
       "};\n",
       "\n",
       "mpl.figure.prototype._remove_fig_handler = function (event) {\n",
       "    var fig = event.data.fig;\n",
       "    if (event.target !== this) {\n",
       "        // Ignore bubbled events from children.\n",
       "        return;\n",
       "    }\n",
       "    fig.close_ws(fig, {});\n",
       "};\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function (el) {\n",
       "    el.style.boxSizing = 'content-box'; // override notebook setting of border-box.\n",
       "};\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function (el) {\n",
       "    // this is important to make the div 'focusable\n",
       "    el.setAttribute('tabindex', 0);\n",
       "    // reach out to IPython and tell the keyboard manager to turn it's self\n",
       "    // off when our div gets focus\n",
       "\n",
       "    // location in version 3\n",
       "    if (IPython.notebook.keyboard_manager) {\n",
       "        IPython.notebook.keyboard_manager.register_events(el);\n",
       "    } else {\n",
       "        // location in version 2\n",
       "        IPython.keyboard_manager.register_events(el);\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function (event, _name) {\n",
       "    var manager = IPython.notebook.keyboard_manager;\n",
       "    if (!manager) {\n",
       "        manager = IPython.keyboard_manager;\n",
       "    }\n",
       "\n",
       "    // Check for shift+enter\n",
       "    if (event.shiftKey && event.which === 13) {\n",
       "        this.canvas_div.blur();\n",
       "        // select the cell after this one\n",
       "        var index = IPython.notebook.find_cell_index(this.cell_info[0]);\n",
       "        IPython.notebook.select(index + 1);\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_save = function (fig, _msg) {\n",
       "    fig.ondownload(fig, null);\n",
       "};\n",
       "\n",
       "mpl.find_output_cell = function (html_output) {\n",
       "    // Return the cell and output element which can be found *uniquely* in the notebook.\n",
       "    // Note - this is a bit hacky, but it is done because the \"notebook_saving.Notebook\"\n",
       "    // IPython event is triggered only after the cells have been serialised, which for\n",
       "    // our purposes (turning an active figure into a static one), is too late.\n",
       "    var cells = IPython.notebook.get_cells();\n",
       "    var ncells = cells.length;\n",
       "    for (var i = 0; i < ncells; i++) {\n",
       "        var cell = cells[i];\n",
       "        if (cell.cell_type === 'code') {\n",
       "            for (var j = 0; j < cell.output_area.outputs.length; j++) {\n",
       "                var data = cell.output_area.outputs[j];\n",
       "                if (data.data) {\n",
       "                    // IPython >= 3 moved mimebundle to data attribute of output\n",
       "                    data = data.data;\n",
       "                }\n",
       "                if (data['text/html'] === html_output) {\n",
       "                    return [cell, data, j];\n",
       "                }\n",
       "            }\n",
       "        }\n",
       "    }\n",
       "};\n",
       "\n",
       "// Register the function which deals with the matplotlib target/channel.\n",
       "// The kernel may be null if the page has been refreshed.\n",
       "if (IPython.notebook.kernel !== null) {\n",
       "    IPython.notebook.kernel.comm_manager.register_target(\n",
       "        'matplotlib',\n",
       "        mpl.mpl_figure_comm\n",
       "    );\n",
       "}\n"
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<img src=\"data:image/png;base64,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\" width=\"800\">"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib notebook\n",
    "\n",
    "from matplotlib import cm\n",
    "from mpl_toolkits.mplot3d import Axes3D\n",
    "from mpl_toolkits.axes_grid1 import make_axes_locatable\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(8,5),subplot_kw={\"projection\": \"3d\"})\n",
    "\n",
    "history = pd.read_csv('history_phat_learning.csv')\n",
    "bodyloss = history.nll\n",
    "tailloss = history.two_tailed_amlse\n",
    "X, Y = np.meshgrid(bodyloss, tailloss)\n",
    "Z = X / (Y + 1)\n",
    "\n",
    "surf = ax.plot_surface(X, Y, Z, cmap=cm.coolwarm,\n",
    "                       linewidth=0, antialiased=False)\n",
    "\n",
    "plt.gca().invert_xaxis()\n",
    "\n",
    "ax.set_xlabel('Log Likelihood', labelpad=10)\n",
    "ax.set_ylabel('AMLSE', labelpad=10)\n",
    "ax.set_zlabel('Total Loss')\n",
    "\n",
    "ax.get_proj = lambda: np.dot(Axes3D.get_proj(ax), np.diag([1, 1.5, 1.25, 1.1]))\n",
    "\n",
    "ax.set_title('Phat Loss', y=.95)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## PhatNet ##"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here we see the model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "from datetime import datetime as dt\n",
    "import tensorflow as tf\n",
    "\n",
    "nnet = ph.PhatNet()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To compile the model, we pass our custom loss function `PhatLoss` as well as a number of custom metrics to monitor results, mainly through the `PhatMetric` class. Both `PhatLoss` and `TwoTailedAMLSE` take the estimated shape parameters as arguments."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "from phat.learn.phatnet import NLL, TwoTailedAMLSE\n",
    "\n",
    "metrics = [NLL(), TwoTailedAMLSE(xi_l, xi_r)]\n",
    "optimizer = tf.keras.optimizers.Adam(learning_rate=5*10**-5)\n",
    "nnet.compile(loss=ph.PhatLoss(xi_l, xi_r), optimizer=optimizer, metrics=metrics)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We have added some customization to the `fit` method including a number of callbacks. For instance, a TensorBoard log is created simply by passing the `logdir` keyword."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 00028: early stopping\n"
     ]
    }
   ],
   "source": [
    "history = nnet.fit(\n",
    "    data.train, \n",
    "    epochs=100,\n",
    "    validation_data=data.test,\n",
    "    verbose=0\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "nbsphinx": "hidden"
   },
   "source": [
    "```python\n",
    "pd.DataFrame(history.history).to_csv('history_phat_learning.csv', index=False)\n",
    "results = pd.DataFrame(\n",
    "    list(zip(nnet.predicted_params().values[:, 0], genmod.learnable_params)),\n",
    "    index=genmod.PARAM_NAMES[:4],\n",
    "    columns=['Trained', 'Actual']\n",
    ").to_csv('phat_w_tails_results.csv', index=False)\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "</style>\n",
       "<table id=\"T_4def7_\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th class=\"col_heading level0 col0\" >Trained</th>\n",
       "      <th class=\"col_heading level0 col1\" >Actual</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_4def7_level0_row0\" class=\"row_heading level0 row0\" >0</th>\n",
       "      <td id=\"T_4def7_row0_col0\" class=\"data row0 col0\" >0.0001879</td>\n",
       "      <td id=\"T_4def7_row0_col1\" class=\"data row0 col1\" >0.0003</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4def7_level0_row1\" class=\"row_heading level0 row1\" >1</th>\n",
       "      <td id=\"T_4def7_row1_col0\" class=\"data row1 col0\" >0.003583</td>\n",
       "      <td id=\"T_4def7_row1_col1\" class=\"data row1 col1\" >0.0032</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4def7_level0_row2\" class=\"row_heading level0 row2\" >2</th>\n",
       "      <td id=\"T_4def7_row2_col0\" class=\"data row2 col0\" >0.3316</td>\n",
       "      <td id=\"T_4def7_row2_col1\" class=\"data row2 col1\" >0.25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4def7_level0_row3\" class=\"row_heading level0 row3\" >3</th>\n",
       "      <td id=\"T_4def7_row3_col0\" class=\"data row3 col0\" >0.3574</td>\n",
       "      <td id=\"T_4def7_row3_col1\" class=\"data row3 col1\" >0.29</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x1940ef730>"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results = pd.read_csv('phat_w_tails_results.csv')\n",
    "results.style.format('{:.4}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Above we see a much improved fit of the tail indices while not sacrificing accuracy in the body parameters."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Caution on Scaling ##\n",
    "\n",
    "The model inputs are all `0` so the usual concerns regarding normalization/standardization/activation in the hidden layer do not apply. Still, the scale of the target `y` values do impact performance in an important way.\n",
    "\n",
    "If the scale of the `y` values is too large, our custom loss function will work in the exact opposite fashion we expect. To see why, recall our loss function:\n",
    "\n",
    "$$ \\text{Loss} = \\frac{\\text{Loss}_{\\textit{body}}}{\\text{Loss}_{\\textit{tail}}+1}$$\n",
    "\n",
    "We can see that if $\\Delta \\text{Loss}_{\\textit{tail}} >>> \\Delta\\text{Loss}_{\\textit{body}}$, then an *increase* in both will lead to a decreasing loss value.\n",
    "\n",
    "This can result if the scaling of `y` is too large. \n",
    "\n",
    "To demonstrate, we'll repeat the prior experiment and simply increase the standard deviation of our Phat distribution by a factor of 100."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0e0cfcab58ff4c87961dd5f600fbc498",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/10 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "\n",
    "genmod = ph.Phat(.0003, .32, .25, .29)\n",
    "\n",
    "n = 60000\n",
    "y = genmod.rvs(size=n)\n",
    "data = ph.DataSplit(y)\n",
    "\n",
    "xi_l, xi_r = ph.two_tailed_hill_double_bootstrap(data.raw.y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Note that the tail estimates are essentially unchanged. The location of each tail will be impacted, but not the index."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "from phat.learn.phatnet import NLL, TwoTailedAMLSE\n",
    "\n",
    "nnet = ph.PhatNet()\n",
    "metrics = [NLL(), TwoTailedAMLSE(xi_l, xi_r)]\n",
    "optimizer = tf.keras.optimizers.Adam(learning_rate=5*10**-5)\n",
    "nnet.compile(loss=ph.PhatLoss(xi_l, xi_r), optimizer=optimizer, metrics=metrics)\n",
    "history = nnet.fit(\n",
    "    data.train, \n",
    "    epochs=100,\n",
    "    validation_data=data.test,\n",
    "    verbose=0,\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As we can see below, this results in a markedly different loss region where the Total Loss improves as both the log-likelihood and the AMSE increases!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "tags": [
     "hide_input"
    ]
   },
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "/* Put everything inside the global mpl namespace */\n",
       "/* global mpl */\n",
       "window.mpl = {};\n",
       "\n",
       "mpl.get_websocket_type = function () {\n",
       "    if (typeof WebSocket !== 'undefined') {\n",
       "        return WebSocket;\n",
       "    } else if (typeof MozWebSocket !== 'undefined') {\n",
       "        return MozWebSocket;\n",
       "    } else {\n",
       "        alert(\n",
       "            'Your browser does not have WebSocket support. ' +\n",
       "                'Please try Chrome, Safari or Firefox ≥ 6. ' +\n",
       "                'Firefox 4 and 5 are also supported but you ' +\n",
       "                'have to enable WebSockets in about:config.'\n",
       "        );\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure = function (figure_id, websocket, ondownload, parent_element) {\n",
       "    this.id = figure_id;\n",
       "\n",
       "    this.ws = websocket;\n",
       "\n",
       "    this.supports_binary = this.ws.binaryType !== undefined;\n",
       "\n",
       "    if (!this.supports_binary) {\n",
       "        var warnings = document.getElementById('mpl-warnings');\n",
       "        if (warnings) {\n",
       "            warnings.style.display = 'block';\n",
       "            warnings.textContent =\n",
       "                'This browser does not support binary websocket messages. ' +\n",
       "                'Performance may be slow.';\n",
       "        }\n",
       "    }\n",
       "\n",
       "    this.imageObj = new Image();\n",
       "\n",
       "    this.context = undefined;\n",
       "    this.message = undefined;\n",
       "    this.canvas = undefined;\n",
       "    this.rubberband_canvas = undefined;\n",
       "    this.rubberband_context = undefined;\n",
       "    this.format_dropdown = undefined;\n",
       "\n",
       "    this.image_mode = 'full';\n",
       "\n",
       "    this.root = document.createElement('div');\n",
       "    this.root.setAttribute('style', 'display: inline-block');\n",
       "    this._root_extra_style(this.root);\n",
       "\n",
       "    parent_element.appendChild(this.root);\n",
       "\n",
       "    this._init_header(this);\n",
       "    this._init_canvas(this);\n",
       "    this._init_toolbar(this);\n",
       "\n",
       "    var fig = this;\n",
       "\n",
       "    this.waiting = false;\n",
       "\n",
       "    this.ws.onopen = function () {\n",
       "        fig.send_message('supports_binary', { value: fig.supports_binary });\n",
       "        fig.send_message('send_image_mode', {});\n",
       "        if (fig.ratio !== 1) {\n",
       "            fig.send_message('set_dpi_ratio', { dpi_ratio: fig.ratio });\n",
       "        }\n",
       "        fig.send_message('refresh', {});\n",
       "    };\n",
       "\n",
       "    this.imageObj.onload = function () {\n",
       "        if (fig.image_mode === 'full') {\n",
       "            // Full images could contain transparency (where diff images\n",
       "            // almost always do), so we need to clear the canvas so that\n",
       "            // there is no ghosting.\n",
       "            fig.context.clearRect(0, 0, fig.canvas.width, fig.canvas.height);\n",
       "        }\n",
       "        fig.context.drawImage(fig.imageObj, 0, 0);\n",
       "    };\n",
       "\n",
       "    this.imageObj.onunload = function () {\n",
       "        fig.ws.close();\n",
       "    };\n",
       "\n",
       "    this.ws.onmessage = this._make_on_message_function(this);\n",
       "\n",
       "    this.ondownload = ondownload;\n",
       "};\n",
       "\n",
       "mpl.figure.prototype._init_header = function () {\n",
       "    var titlebar = document.createElement('div');\n",
       "    titlebar.classList =\n",
       "        'ui-dialog-titlebar ui-widget-header ui-corner-all ui-helper-clearfix';\n",
       "    var titletext = document.createElement('div');\n",
       "    titletext.classList = 'ui-dialog-title';\n",
       "    titletext.setAttribute(\n",
       "        'style',\n",
       "        'width: 100%; text-align: center; padding: 3px;'\n",
       "    );\n",
       "    titlebar.appendChild(titletext);\n",
       "    this.root.appendChild(titlebar);\n",
       "    this.header = titletext;\n",
       "};\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function (_canvas_div) {};\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function (_canvas_div) {};\n",
       "\n",
       "mpl.figure.prototype._init_canvas = function () {\n",
       "    var fig = this;\n",
       "\n",
       "    var canvas_div = (this.canvas_div = document.createElement('div'));\n",
       "    canvas_div.setAttribute(\n",
       "        'style',\n",
       "        'border: 1px solid #ddd;' +\n",
       "            'box-sizing: content-box;' +\n",
       "            'clear: both;' +\n",
       "            'min-height: 1px;' +\n",
       "            'min-width: 1px;' +\n",
       "            'outline: 0;' +\n",
       "            'overflow: hidden;' +\n",
       "            'position: relative;' +\n",
       "            'resize: both;'\n",
       "    );\n",
       "\n",
       "    function on_keyboard_event_closure(name) {\n",
       "        return function (event) {\n",
       "            return fig.key_event(event, name);\n",
       "        };\n",
       "    }\n",
       "\n",
       "    canvas_div.addEventListener(\n",
       "        'keydown',\n",
       "        on_keyboard_event_closure('key_press')\n",
       "    );\n",
       "    canvas_div.addEventListener(\n",
       "        'keyup',\n",
       "        on_keyboard_event_closure('key_release')\n",
       "    );\n",
       "\n",
       "    this._canvas_extra_style(canvas_div);\n",
       "    this.root.appendChild(canvas_div);\n",
       "\n",
       "    var canvas = (this.canvas = document.createElement('canvas'));\n",
       "    canvas.classList.add('mpl-canvas');\n",
       "    canvas.setAttribute('style', 'box-sizing: content-box;');\n",
       "\n",
       "    this.context = canvas.getContext('2d');\n",
       "\n",
       "    var backingStore =\n",
       "        this.context.backingStorePixelRatio ||\n",
       "        this.context.webkitBackingStorePixelRatio ||\n",
       "        this.context.mozBackingStorePixelRatio ||\n",
       "        this.context.msBackingStorePixelRatio ||\n",
       "        this.context.oBackingStorePixelRatio ||\n",
       "        this.context.backingStorePixelRatio ||\n",
       "        1;\n",
       "\n",
       "    this.ratio = (window.devicePixelRatio || 1) / backingStore;\n",
       "\n",
       "    var rubberband_canvas = (this.rubberband_canvas = document.createElement(\n",
       "        'canvas'\n",
       "    ));\n",
       "    rubberband_canvas.setAttribute(\n",
       "        'style',\n",
       "        'box-sizing: content-box; position: absolute; left: 0; top: 0; z-index: 1;'\n",
       "    );\n",
       "\n",
       "    // Apply a ponyfill if ResizeObserver is not implemented by browser.\n",
       "    if (this.ResizeObserver === undefined) {\n",
       "        if (window.ResizeObserver !== undefined) {\n",
       "            this.ResizeObserver = window.ResizeObserver;\n",
       "        } else {\n",
       "            var obs = _JSXTOOLS_RESIZE_OBSERVER({});\n",
       "            this.ResizeObserver = obs.ResizeObserver;\n",
       "        }\n",
       "    }\n",
       "\n",
       "    this.resizeObserverInstance = new this.ResizeObserver(function (entries) {\n",
       "        var nentries = entries.length;\n",
       "        for (var i = 0; i < nentries; i++) {\n",
       "            var entry = entries[i];\n",
       "            var width, height;\n",
       "            if (entry.contentBoxSize) {\n",
       "                if (entry.contentBoxSize instanceof Array) {\n",
       "                    // Chrome 84 implements new version of spec.\n",
       "                    width = entry.contentBoxSize[0].inlineSize;\n",
       "                    height = entry.contentBoxSize[0].blockSize;\n",
       "                } else {\n",
       "                    // Firefox implements old version of spec.\n",
       "                    width = entry.contentBoxSize.inlineSize;\n",
       "                    height = entry.contentBoxSize.blockSize;\n",
       "                }\n",
       "            } else {\n",
       "                // Chrome <84 implements even older version of spec.\n",
       "                width = entry.contentRect.width;\n",
       "                height = entry.contentRect.height;\n",
       "            }\n",
       "\n",
       "            // Keep the size of the canvas and rubber band canvas in sync with\n",
       "            // the canvas container.\n",
       "            if (entry.devicePixelContentBoxSize) {\n",
       "                // Chrome 84 implements new version of spec.\n",
       "                canvas.setAttribute(\n",
       "                    'width',\n",
       "                    entry.devicePixelContentBoxSize[0].inlineSize\n",
       "                );\n",
       "                canvas.setAttribute(\n",
       "                    'height',\n",
       "                    entry.devicePixelContentBoxSize[0].blockSize\n",
       "                );\n",
       "            } else {\n",
       "                canvas.setAttribute('width', width * fig.ratio);\n",
       "                canvas.setAttribute('height', height * fig.ratio);\n",
       "            }\n",
       "            canvas.setAttribute(\n",
       "                'style',\n",
       "                'width: ' + width + 'px; height: ' + height + 'px;'\n",
       "            );\n",
       "\n",
       "            rubberband_canvas.setAttribute('width', width);\n",
       "            rubberband_canvas.setAttribute('height', height);\n",
       "\n",
       "            // And update the size in Python. We ignore the initial 0/0 size\n",
       "            // that occurs as the element is placed into the DOM, which should\n",
       "            // otherwise not happen due to the minimum size styling.\n",
       "            if (fig.ws.readyState == 1 && width != 0 && height != 0) {\n",
       "                fig.request_resize(width, height);\n",
       "            }\n",
       "        }\n",
       "    });\n",
       "    this.resizeObserverInstance.observe(canvas_div);\n",
       "\n",
       "    function on_mouse_event_closure(name) {\n",
       "        return function (event) {\n",
       "            return fig.mouse_event(event, name);\n",
       "        };\n",
       "    }\n",
       "\n",
       "    rubberband_canvas.addEventListener(\n",
       "        'mousedown',\n",
       "        on_mouse_event_closure('button_press')\n",
       "    );\n",
       "    rubberband_canvas.addEventListener(\n",
       "        'mouseup',\n",
       "        on_mouse_event_closure('button_release')\n",
       "    );\n",
       "    rubberband_canvas.addEventListener(\n",
       "        'dblclick',\n",
       "        on_mouse_event_closure('dblclick')\n",
       "    );\n",
       "    // Throttle sequential mouse events to 1 every 20ms.\n",
       "    rubberband_canvas.addEventListener(\n",
       "        'mousemove',\n",
       "        on_mouse_event_closure('motion_notify')\n",
       "    );\n",
       "\n",
       "    rubberband_canvas.addEventListener(\n",
       "        'mouseenter',\n",
       "        on_mouse_event_closure('figure_enter')\n",
       "    );\n",
       "    rubberband_canvas.addEventListener(\n",
       "        'mouseleave',\n",
       "        on_mouse_event_closure('figure_leave')\n",
       "    );\n",
       "\n",
       "    canvas_div.addEventListener('wheel', function (event) {\n",
       "        if (event.deltaY < 0) {\n",
       "            event.step = 1;\n",
       "        } else {\n",
       "            event.step = -1;\n",
       "        }\n",
       "        on_mouse_event_closure('scroll')(event);\n",
       "    });\n",
       "\n",
       "    canvas_div.appendChild(canvas);\n",
       "    canvas_div.appendChild(rubberband_canvas);\n",
       "\n",
       "    this.rubberband_context = rubberband_canvas.getContext('2d');\n",
       "    this.rubberband_context.strokeStyle = '#000000';\n",
       "\n",
       "    this._resize_canvas = function (width, height, forward) {\n",
       "        if (forward) {\n",
       "            canvas_div.style.width = width + 'px';\n",
       "            canvas_div.style.height = height + 'px';\n",
       "        }\n",
       "    };\n",
       "\n",
       "    // Disable right mouse context menu.\n",
       "    this.rubberband_canvas.addEventListener('contextmenu', function (_e) {\n",
       "        event.preventDefault();\n",
       "        return false;\n",
       "    });\n",
       "\n",
       "    function set_focus() {\n",
       "        canvas.focus();\n",
       "        canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    window.setTimeout(set_focus, 100);\n",
       "};\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function () {\n",
       "    var fig = this;\n",
       "\n",
       "    var toolbar = document.createElement('div');\n",
       "    toolbar.classList = 'mpl-toolbar';\n",
       "    this.root.appendChild(toolbar);\n",
       "\n",
       "    function on_click_closure(name) {\n",
       "        return function (_event) {\n",
       "            return fig.toolbar_button_onclick(name);\n",
       "        };\n",
       "    }\n",
       "\n",
       "    function on_mouseover_closure(tooltip) {\n",
       "        return function (event) {\n",
       "            if (!event.currentTarget.disabled) {\n",
       "                return fig.toolbar_button_onmouseover(tooltip);\n",
       "            }\n",
       "        };\n",
       "    }\n",
       "\n",
       "    fig.buttons = {};\n",
       "    var buttonGroup = document.createElement('div');\n",
       "    buttonGroup.classList = 'mpl-button-group';\n",
       "    for (var toolbar_ind in mpl.toolbar_items) {\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) {\n",
       "            /* Instead of a spacer, we start a new button group. */\n",
       "            if (buttonGroup.hasChildNodes()) {\n",
       "                toolbar.appendChild(buttonGroup);\n",
       "            }\n",
       "            buttonGroup = document.createElement('div');\n",
       "            buttonGroup.classList = 'mpl-button-group';\n",
       "            continue;\n",
       "        }\n",
       "\n",
       "        var button = (fig.buttons[name] = document.createElement('button'));\n",
       "        button.classList = 'mpl-widget';\n",
       "        button.setAttribute('role', 'button');\n",
       "        button.setAttribute('aria-disabled', 'false');\n",
       "        button.addEventListener('click', on_click_closure(method_name));\n",
       "        button.addEventListener('mouseover', on_mouseover_closure(tooltip));\n",
       "\n",
       "        var icon_img = document.createElement('img');\n",
       "        icon_img.src = '_images/' + image + '.png';\n",
       "        icon_img.srcset = '_images/' + image + '_large.png 2x';\n",
       "        icon_img.alt = tooltip;\n",
       "        button.appendChild(icon_img);\n",
       "\n",
       "        buttonGroup.appendChild(button);\n",
       "    }\n",
       "\n",
       "    if (buttonGroup.hasChildNodes()) {\n",
       "        toolbar.appendChild(buttonGroup);\n",
       "    }\n",
       "\n",
       "    var fmt_picker = document.createElement('select');\n",
       "    fmt_picker.classList = 'mpl-widget';\n",
       "    toolbar.appendChild(fmt_picker);\n",
       "    this.format_dropdown = fmt_picker;\n",
       "\n",
       "    for (var ind in mpl.extensions) {\n",
       "        var fmt = mpl.extensions[ind];\n",
       "        var option = document.createElement('option');\n",
       "        option.selected = fmt === mpl.default_extension;\n",
       "        option.innerHTML = fmt;\n",
       "        fmt_picker.appendChild(option);\n",
       "    }\n",
       "\n",
       "    var status_bar = document.createElement('span');\n",
       "    status_bar.classList = 'mpl-message';\n",
       "    toolbar.appendChild(status_bar);\n",
       "    this.message = status_bar;\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.request_resize = function (x_pixels, y_pixels) {\n",
       "    // Request matplotlib to resize the figure. Matplotlib will then trigger a resize in the client,\n",
       "    // which will in turn request a refresh of the image.\n",
       "    this.send_message('resize', { width: x_pixels, height: y_pixels });\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.send_message = function (type, properties) {\n",
       "    properties['type'] = type;\n",
       "    properties['figure_id'] = this.id;\n",
       "    this.ws.send(JSON.stringify(properties));\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.send_draw_message = function () {\n",
       "    if (!this.waiting) {\n",
       "        this.waiting = true;\n",
       "        this.ws.send(JSON.stringify({ type: 'draw', figure_id: this.id }));\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_save = function (fig, _msg) {\n",
       "    var format_dropdown = fig.format_dropdown;\n",
       "    var format = format_dropdown.options[format_dropdown.selectedIndex].value;\n",
       "    fig.ondownload(fig, format);\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_resize = function (fig, msg) {\n",
       "    var size = msg['size'];\n",
       "    if (size[0] !== fig.canvas.width || size[1] !== fig.canvas.height) {\n",
       "        fig._resize_canvas(size[0], size[1], msg['forward']);\n",
       "        fig.send_message('refresh', {});\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_rubberband = function (fig, msg) {\n",
       "    var x0 = msg['x0'] / fig.ratio;\n",
       "    var y0 = (fig.canvas.height - msg['y0']) / fig.ratio;\n",
       "    var x1 = msg['x1'] / fig.ratio;\n",
       "    var y1 = (fig.canvas.height - msg['y1']) / fig.ratio;\n",
       "    x0 = Math.floor(x0) + 0.5;\n",
       "    y0 = Math.floor(y0) + 0.5;\n",
       "    x1 = Math.floor(x1) + 0.5;\n",
       "    y1 = Math.floor(y1) + 0.5;\n",
       "    var min_x = Math.min(x0, x1);\n",
       "    var min_y = Math.min(y0, y1);\n",
       "    var width = Math.abs(x1 - x0);\n",
       "    var height = Math.abs(y1 - y0);\n",
       "\n",
       "    fig.rubberband_context.clearRect(\n",
       "        0,\n",
       "        0,\n",
       "        fig.canvas.width / fig.ratio,\n",
       "        fig.canvas.height / fig.ratio\n",
       "    );\n",
       "\n",
       "    fig.rubberband_context.strokeRect(min_x, min_y, width, height);\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_figure_label = function (fig, msg) {\n",
       "    // Updates the figure title.\n",
       "    fig.header.textContent = msg['label'];\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_cursor = function (fig, msg) {\n",
       "    var cursor = msg['cursor'];\n",
       "    switch (cursor) {\n",
       "        case 0:\n",
       "            cursor = 'pointer';\n",
       "            break;\n",
       "        case 1:\n",
       "            cursor = 'default';\n",
       "            break;\n",
       "        case 2:\n",
       "            cursor = 'crosshair';\n",
       "            break;\n",
       "        case 3:\n",
       "            cursor = 'move';\n",
       "            break;\n",
       "    }\n",
       "    fig.rubberband_canvas.style.cursor = cursor;\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_message = function (fig, msg) {\n",
       "    fig.message.textContent = msg['message'];\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_draw = function (fig, _msg) {\n",
       "    // Request the server to send over a new figure.\n",
       "    fig.send_draw_message();\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_image_mode = function (fig, msg) {\n",
       "    fig.image_mode = msg['mode'];\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_history_buttons = function (fig, msg) {\n",
       "    for (var key in msg) {\n",
       "        if (!(key in fig.buttons)) {\n",
       "            continue;\n",
       "        }\n",
       "        fig.buttons[key].disabled = !msg[key];\n",
       "        fig.buttons[key].setAttribute('aria-disabled', !msg[key]);\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_navigate_mode = function (fig, msg) {\n",
       "    if (msg['mode'] === 'PAN') {\n",
       "        fig.buttons['Pan'].classList.add('active');\n",
       "        fig.buttons['Zoom'].classList.remove('active');\n",
       "    } else if (msg['mode'] === 'ZOOM') {\n",
       "        fig.buttons['Pan'].classList.remove('active');\n",
       "        fig.buttons['Zoom'].classList.add('active');\n",
       "    } else {\n",
       "        fig.buttons['Pan'].classList.remove('active');\n",
       "        fig.buttons['Zoom'].classList.remove('active');\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function () {\n",
       "    // Called whenever the canvas gets updated.\n",
       "    this.send_message('ack', {});\n",
       "};\n",
       "\n",
       "// A function to construct a web socket function for onmessage handling.\n",
       "// Called in the figure constructor.\n",
       "mpl.figure.prototype._make_on_message_function = function (fig) {\n",
       "    return function socket_on_message(evt) {\n",
       "        if (evt.data instanceof Blob) {\n",
       "            var img = evt.data;\n",
       "            if (img.type !== 'image/png') {\n",
       "                /* FIXME: We get \"Resource interpreted as Image but\n",
       "                 * transferred with MIME type text/plain:\" errors on\n",
       "                 * Chrome.  But how to set the MIME type?  It doesn't seem\n",
       "                 * to be part of the websocket stream */\n",
       "                img.type = 'image/png';\n",
       "            }\n",
       "\n",
       "            /* Free the memory for the previous frames */\n",
       "            if (fig.imageObj.src) {\n",
       "                (window.URL || window.webkitURL).revokeObjectURL(\n",
       "                    fig.imageObj.src\n",
       "                );\n",
       "            }\n",
       "\n",
       "            fig.imageObj.src = (window.URL || window.webkitURL).createObjectURL(\n",
       "                img\n",
       "            );\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        } else if (\n",
       "            typeof evt.data === 'string' &&\n",
       "            evt.data.slice(0, 21) === 'data:image/png;base64'\n",
       "        ) {\n",
       "            fig.imageObj.src = evt.data;\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        var msg = JSON.parse(evt.data);\n",
       "        var msg_type = msg['type'];\n",
       "\n",
       "        // Call the  \"handle_{type}\" callback, which takes\n",
       "        // the figure and JSON message as its only arguments.\n",
       "        try {\n",
       "            var callback = fig['handle_' + msg_type];\n",
       "        } catch (e) {\n",
       "            console.log(\n",
       "                \"No handler for the '\" + msg_type + \"' message type: \",\n",
       "                msg\n",
       "            );\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        if (callback) {\n",
       "            try {\n",
       "                // console.log(\"Handling '\" + msg_type + \"' message: \", msg);\n",
       "                callback(fig, msg);\n",
       "            } catch (e) {\n",
       "                console.log(\n",
       "                    \"Exception inside the 'handler_\" + msg_type + \"' callback:\",\n",
       "                    e,\n",
       "                    e.stack,\n",
       "                    msg\n",
       "                );\n",
       "            }\n",
       "        }\n",
       "    };\n",
       "};\n",
       "\n",
       "// from http://stackoverflow.com/questions/1114465/getting-mouse-location-in-canvas\n",
       "mpl.findpos = function (e) {\n",
       "    //this section is from http://www.quirksmode.org/js/events_properties.html\n",
       "    var targ;\n",
       "    if (!e) {\n",
       "        e = window.event;\n",
       "    }\n",
       "    if (e.target) {\n",
       "        targ = e.target;\n",
       "    } else if (e.srcElement) {\n",
       "        targ = e.srcElement;\n",
       "    }\n",
       "    if (targ.nodeType === 3) {\n",
       "        // defeat Safari bug\n",
       "        targ = targ.parentNode;\n",
       "    }\n",
       "\n",
       "    // pageX,Y are the mouse positions relative to the document\n",
       "    var boundingRect = targ.getBoundingClientRect();\n",
       "    var x = e.pageX - (boundingRect.left + document.body.scrollLeft);\n",
       "    var y = e.pageY - (boundingRect.top + document.body.scrollTop);\n",
       "\n",
       "    return { x: x, y: y };\n",
       "};\n",
       "\n",
       "/*\n",
       " * return a copy of an object with only non-object keys\n",
       " * we need this to avoid circular references\n",
       " * http://stackoverflow.com/a/24161582/3208463\n",
       " */\n",
       "function simpleKeys(original) {\n",
       "    return Object.keys(original).reduce(function (obj, key) {\n",
       "        if (typeof original[key] !== 'object') {\n",
       "            obj[key] = original[key];\n",
       "        }\n",
       "        return obj;\n",
       "    }, {});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.mouse_event = function (event, name) {\n",
       "    var canvas_pos = mpl.findpos(event);\n",
       "\n",
       "    if (name === 'button_press') {\n",
       "        this.canvas.focus();\n",
       "        this.canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    var x = canvas_pos.x * this.ratio;\n",
       "    var y = canvas_pos.y * this.ratio;\n",
       "\n",
       "    this.send_message(name, {\n",
       "        x: x,\n",
       "        y: y,\n",
       "        button: event.button,\n",
       "        step: event.step,\n",
       "        guiEvent: simpleKeys(event),\n",
       "    });\n",
       "\n",
       "    /* This prevents the web browser from automatically changing to\n",
       "     * the text insertion cursor when the button is pressed.  We want\n",
       "     * to control all of the cursor setting manually through the\n",
       "     * 'cursor' event from matplotlib */\n",
       "    event.preventDefault();\n",
       "    return false;\n",
       "};\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function (_event, _name) {\n",
       "    // Handle any extra behaviour associated with a key event\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.key_event = function (event, name) {\n",
       "    // Prevent repeat events\n",
       "    if (name === 'key_press') {\n",
       "        if (event.key === this._key) {\n",
       "            return;\n",
       "        } else {\n",
       "            this._key = event.key;\n",
       "        }\n",
       "    }\n",
       "    if (name === 'key_release') {\n",
       "        this._key = null;\n",
       "    }\n",
       "\n",
       "    var value = '';\n",
       "    if (event.ctrlKey && event.key !== 'Control') {\n",
       "        value += 'ctrl+';\n",
       "    }\n",
       "    else if (event.altKey && event.key !== 'Alt') {\n",
       "        value += 'alt+';\n",
       "    }\n",
       "    else if (event.shiftKey && event.key !== 'Shift') {\n",
       "        value += 'shift+';\n",
       "    }\n",
       "\n",
       "    value += 'k' + event.key;\n",
       "\n",
       "    this._key_event_extra(event, name);\n",
       "\n",
       "    this.send_message(name, { key: value, guiEvent: simpleKeys(event) });\n",
       "    return false;\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onclick = function (name) {\n",
       "    if (name === 'download') {\n",
       "        this.handle_save(this, null);\n",
       "    } else {\n",
       "        this.send_message('toolbar_button', { name: name });\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onmouseover = function (tooltip) {\n",
       "    this.message.textContent = tooltip;\n",
       "};\n",
       "\n",
       "///////////////// REMAINING CONTENT GENERATED BY embed_js.py /////////////////\n",
       "// prettier-ignore\n",
       "var _JSXTOOLS_RESIZE_OBSERVER=function(A){var t,i=new WeakMap,n=new WeakMap,a=new WeakMap,r=new WeakMap,o=new Set;function s(e){if(!(this instanceof s))throw new TypeError(\"Constructor requires 'new' operator\");i.set(this,e)}function h(){throw new TypeError(\"Function is not a constructor\")}function c(e,t,i,n){e=0 in arguments?Number(arguments[0]):0,t=1 in arguments?Number(arguments[1]):0,i=2 in arguments?Number(arguments[2]):0,n=3 in arguments?Number(arguments[3]):0,this.right=(this.x=this.left=e)+(this.width=i),this.bottom=(this.y=this.top=t)+(this.height=n),Object.freeze(this)}function d(){t=requestAnimationFrame(d);var s=new WeakMap,p=new Set;o.forEach((function(t){r.get(t).forEach((function(i){var r=t instanceof window.SVGElement,o=a.get(t),d=r?0:parseFloat(o.paddingTop),f=r?0:parseFloat(o.paddingRight),l=r?0:parseFloat(o.paddingBottom),u=r?0:parseFloat(o.paddingLeft),g=r?0:parseFloat(o.borderTopWidth),m=r?0:parseFloat(o.borderRightWidth),w=r?0:parseFloat(o.borderBottomWidth),b=u+f,F=d+l,v=(r?0:parseFloat(o.borderLeftWidth))+m,W=g+w,y=r?0:t.offsetHeight-W-t.clientHeight,E=r?0:t.offsetWidth-v-t.clientWidth,R=b+v,z=F+W,M=r?t.width:parseFloat(o.width)-R-E,O=r?t.height:parseFloat(o.height)-z-y;if(n.has(t)){var k=n.get(t);if(k[0]===M&&k[1]===O)return}n.set(t,[M,O]);var S=Object.create(h.prototype);S.target=t,S.contentRect=new c(u,d,M,O),s.has(i)||(s.set(i,[]),p.add(i)),s.get(i).push(S)}))})),p.forEach((function(e){i.get(e).call(e,s.get(e),e)}))}return s.prototype.observe=function(i){if(i instanceof window.Element){r.has(i)||(r.set(i,new Set),o.add(i),a.set(i,window.getComputedStyle(i)));var n=r.get(i);n.has(this)||n.add(this),cancelAnimationFrame(t),t=requestAnimationFrame(d)}},s.prototype.unobserve=function(i){if(i instanceof window.Element&&r.has(i)){var n=r.get(i);n.has(this)&&(n.delete(this),n.size||(r.delete(i),o.delete(i))),n.size||r.delete(i),o.size||cancelAnimationFrame(t)}},A.DOMRectReadOnly=c,A.ResizeObserver=s,A.ResizeObserverEntry=h,A}; // eslint-disable-line\n",
       "mpl.toolbar_items = [[\"Home\", \"Reset original view\", \"fa fa-home icon-home\", \"home\"], [\"Back\", \"Back to previous view\", \"fa fa-arrow-left icon-arrow-left\", \"back\"], [\"Forward\", \"Forward to next view\", \"fa fa-arrow-right icon-arrow-right\", \"forward\"], [\"\", \"\", \"\", \"\"], [\"Pan\", \"Left button pans, Right button zooms\\nx/y fixes axis, CTRL fixes aspect\", \"fa fa-arrows icon-move\", \"pan\"], [\"Zoom\", \"Zoom to rectangle\\nx/y fixes axis, CTRL fixes aspect\", \"fa fa-square-o icon-check-empty\", \"zoom\"], [\"\", \"\", \"\", \"\"], [\"Download\", \"Download plot\", \"fa fa-floppy-o icon-save\", \"download\"]];\n",
       "\n",
       "mpl.extensions = [\"eps\", \"jpeg\", \"pgf\", \"pdf\", \"png\", \"ps\", \"raw\", \"svg\", \"tif\"];\n",
       "\n",
       "mpl.default_extension = \"png\";/* global mpl */\n",
       "\n",
       "var comm_websocket_adapter = function (comm) {\n",
       "    // Create a \"websocket\"-like object which calls the given IPython comm\n",
       "    // object with the appropriate methods. Currently this is a non binary\n",
       "    // socket, so there is still some room for performance tuning.\n",
       "    var ws = {};\n",
       "\n",
       "    ws.binaryType = comm.kernel.ws.binaryType;\n",
       "    ws.readyState = comm.kernel.ws.readyState;\n",
       "    function updateReadyState(_event) {\n",
       "        if (comm.kernel.ws) {\n",
       "            ws.readyState = comm.kernel.ws.readyState;\n",
       "        } else {\n",
       "            ws.readyState = 3; // Closed state.\n",
       "        }\n",
       "    }\n",
       "    comm.kernel.ws.addEventListener('open', updateReadyState);\n",
       "    comm.kernel.ws.addEventListener('close', updateReadyState);\n",
       "    comm.kernel.ws.addEventListener('error', updateReadyState);\n",
       "\n",
       "    ws.close = function () {\n",
       "        comm.close();\n",
       "    };\n",
       "    ws.send = function (m) {\n",
       "        //console.log('sending', m);\n",
       "        comm.send(m);\n",
       "    };\n",
       "    // Register the callback with on_msg.\n",
       "    comm.on_msg(function (msg) {\n",
       "        //console.log('receiving', msg['content']['data'], msg);\n",
       "        var data = msg['content']['data'];\n",
       "        if (data['blob'] !== undefined) {\n",
       "            data = {\n",
       "                data: new Blob(msg['buffers'], { type: data['blob'] }),\n",
       "            };\n",
       "        }\n",
       "        // Pass the mpl event to the overridden (by mpl) onmessage function.\n",
       "        ws.onmessage(data);\n",
       "    });\n",
       "    return ws;\n",
       "};\n",
       "\n",
       "mpl.mpl_figure_comm = function (comm, msg) {\n",
       "    // This is the function which gets called when the mpl process\n",
       "    // starts-up an IPython Comm through the \"matplotlib\" channel.\n",
       "\n",
       "    var id = msg.content.data.id;\n",
       "    // Get hold of the div created by the display call when the Comm\n",
       "    // socket was opened in Python.\n",
       "    var element = document.getElementById(id);\n",
       "    var ws_proxy = comm_websocket_adapter(comm);\n",
       "\n",
       "    function ondownload(figure, _format) {\n",
       "        window.open(figure.canvas.toDataURL());\n",
       "    }\n",
       "\n",
       "    var fig = new mpl.figure(id, ws_proxy, ondownload, element);\n",
       "\n",
       "    // Call onopen now - mpl needs it, as it is assuming we've passed it a real\n",
       "    // web socket which is closed, not our websocket->open comm proxy.\n",
       "    ws_proxy.onopen();\n",
       "\n",
       "    fig.parent_element = element;\n",
       "    fig.cell_info = mpl.find_output_cell(\"<div id='\" + id + \"'></div>\");\n",
       "    if (!fig.cell_info) {\n",
       "        console.error('Failed to find cell for figure', id, fig);\n",
       "        return;\n",
       "    }\n",
       "    fig.cell_info[0].output_area.element.on(\n",
       "        'cleared',\n",
       "        { fig: fig },\n",
       "        fig._remove_fig_handler\n",
       "    );\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_close = function (fig, msg) {\n",
       "    var width = fig.canvas.width / fig.ratio;\n",
       "    fig.cell_info[0].output_area.element.off(\n",
       "        'cleared',\n",
       "        fig._remove_fig_handler\n",
       "    );\n",
       "    fig.resizeObserverInstance.unobserve(fig.canvas_div);\n",
       "\n",
       "    // Update the output cell to use the data from the current canvas.\n",
       "    fig.push_to_output();\n",
       "    var dataURL = fig.canvas.toDataURL();\n",
       "    // Re-enable the keyboard manager in IPython - without this line, in FF,\n",
       "    // the notebook keyboard shortcuts fail.\n",
       "    IPython.keyboard_manager.enable();\n",
       "    fig.parent_element.innerHTML =\n",
       "        '<img src=\"' + dataURL + '\" width=\"' + width + '\">';\n",
       "    fig.close_ws(fig, msg);\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.close_ws = function (fig, msg) {\n",
       "    fig.send_message('closing', msg);\n",
       "    // fig.ws.close()\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.push_to_output = function (_remove_interactive) {\n",
       "    // Turn the data on the canvas into data in the output cell.\n",
       "    var width = this.canvas.width / this.ratio;\n",
       "    var dataURL = this.canvas.toDataURL();\n",
       "    this.cell_info[1]['text/html'] =\n",
       "        '<img src=\"' + dataURL + '\" width=\"' + width + '\">';\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function () {\n",
       "    // Tell IPython that the notebook contents must change.\n",
       "    IPython.notebook.set_dirty(true);\n",
       "    this.send_message('ack', {});\n",
       "    var fig = this;\n",
       "    // Wait a second, then push the new image to the DOM so\n",
       "    // that it is saved nicely (might be nice to debounce this).\n",
       "    setTimeout(function () {\n",
       "        fig.push_to_output();\n",
       "    }, 1000);\n",
       "};\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function () {\n",
       "    var fig = this;\n",
       "\n",
       "    var toolbar = document.createElement('div');\n",
       "    toolbar.classList = 'btn-toolbar';\n",
       "    this.root.appendChild(toolbar);\n",
       "\n",
       "    function on_click_closure(name) {\n",
       "        return function (_event) {\n",
       "            return fig.toolbar_button_onclick(name);\n",
       "        };\n",
       "    }\n",
       "\n",
       "    function on_mouseover_closure(tooltip) {\n",
       "        return function (event) {\n",
       "            if (!event.currentTarget.disabled) {\n",
       "                return fig.toolbar_button_onmouseover(tooltip);\n",
       "            }\n",
       "        };\n",
       "    }\n",
       "\n",
       "    fig.buttons = {};\n",
       "    var buttonGroup = document.createElement('div');\n",
       "    buttonGroup.classList = 'btn-group';\n",
       "    var button;\n",
       "    for (var toolbar_ind in mpl.toolbar_items) {\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) {\n",
       "            /* Instead of a spacer, we start a new button group. */\n",
       "            if (buttonGroup.hasChildNodes()) {\n",
       "                toolbar.appendChild(buttonGroup);\n",
       "            }\n",
       "            buttonGroup = document.createElement('div');\n",
       "            buttonGroup.classList = 'btn-group';\n",
       "            continue;\n",
       "        }\n",
       "\n",
       "        button = fig.buttons[name] = document.createElement('button');\n",
       "        button.classList = 'btn btn-default';\n",
       "        button.href = '#';\n",
       "        button.title = name;\n",
       "        button.innerHTML = '<i class=\"fa ' + image + ' fa-lg\"></i>';\n",
       "        button.addEventListener('click', on_click_closure(method_name));\n",
       "        button.addEventListener('mouseover', on_mouseover_closure(tooltip));\n",
       "        buttonGroup.appendChild(button);\n",
       "    }\n",
       "\n",
       "    if (buttonGroup.hasChildNodes()) {\n",
       "        toolbar.appendChild(buttonGroup);\n",
       "    }\n",
       "\n",
       "    // Add the status bar.\n",
       "    var status_bar = document.createElement('span');\n",
       "    status_bar.classList = 'mpl-message pull-right';\n",
       "    toolbar.appendChild(status_bar);\n",
       "    this.message = status_bar;\n",
       "\n",
       "    // Add the close button to the window.\n",
       "    var buttongrp = document.createElement('div');\n",
       "    buttongrp.classList = 'btn-group inline pull-right';\n",
       "    button = document.createElement('button');\n",
       "    button.classList = 'btn btn-mini btn-primary';\n",
       "    button.href = '#';\n",
       "    button.title = 'Stop Interaction';\n",
       "    button.innerHTML = '<i class=\"fa fa-power-off icon-remove icon-large\"></i>';\n",
       "    button.addEventListener('click', function (_evt) {\n",
       "        fig.handle_close(fig, {});\n",
       "    });\n",
       "    button.addEventListener(\n",
       "        'mouseover',\n",
       "        on_mouseover_closure('Stop Interaction')\n",
       "    );\n",
       "    buttongrp.appendChild(button);\n",
       "    var titlebar = this.root.querySelector('.ui-dialog-titlebar');\n",
       "    titlebar.insertBefore(buttongrp, titlebar.firstChild);\n",
       "};\n",
       "\n",
       "mpl.figure.prototype._remove_fig_handler = function (event) {\n",
       "    var fig = event.data.fig;\n",
       "    if (event.target !== this) {\n",
       "        // Ignore bubbled events from children.\n",
       "        return;\n",
       "    }\n",
       "    fig.close_ws(fig, {});\n",
       "};\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function (el) {\n",
       "    el.style.boxSizing = 'content-box'; // override notebook setting of border-box.\n",
       "};\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function (el) {\n",
       "    // this is important to make the div 'focusable\n",
       "    el.setAttribute('tabindex', 0);\n",
       "    // reach out to IPython and tell the keyboard manager to turn it's self\n",
       "    // off when our div gets focus\n",
       "\n",
       "    // location in version 3\n",
       "    if (IPython.notebook.keyboard_manager) {\n",
       "        IPython.notebook.keyboard_manager.register_events(el);\n",
       "    } else {\n",
       "        // location in version 2\n",
       "        IPython.keyboard_manager.register_events(el);\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function (event, _name) {\n",
       "    var manager = IPython.notebook.keyboard_manager;\n",
       "    if (!manager) {\n",
       "        manager = IPython.keyboard_manager;\n",
       "    }\n",
       "\n",
       "    // Check for shift+enter\n",
       "    if (event.shiftKey && event.which === 13) {\n",
       "        this.canvas_div.blur();\n",
       "        // select the cell after this one\n",
       "        var index = IPython.notebook.find_cell_index(this.cell_info[0]);\n",
       "        IPython.notebook.select(index + 1);\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_save = function (fig, _msg) {\n",
       "    fig.ondownload(fig, null);\n",
       "};\n",
       "\n",
       "mpl.find_output_cell = function (html_output) {\n",
       "    // Return the cell and output element which can be found *uniquely* in the notebook.\n",
       "    // Note - this is a bit hacky, but it is done because the \"notebook_saving.Notebook\"\n",
       "    // IPython event is triggered only after the cells have been serialised, which for\n",
       "    // our purposes (turning an active figure into a static one), is too late.\n",
       "    var cells = IPython.notebook.get_cells();\n",
       "    var ncells = cells.length;\n",
       "    for (var i = 0; i < ncells; i++) {\n",
       "        var cell = cells[i];\n",
       "        if (cell.cell_type === 'code') {\n",
       "            for (var j = 0; j < cell.output_area.outputs.length; j++) {\n",
       "                var data = cell.output_area.outputs[j];\n",
       "                if (data.data) {\n",
       "                    // IPython >= 3 moved mimebundle to data attribute of output\n",
       "                    data = data.data;\n",
       "                }\n",
       "                if (data['text/html'] === html_output) {\n",
       "                    return [cell, data, j];\n",
       "                }\n",
       "            }\n",
       "        }\n",
       "    }\n",
       "};\n",
       "\n",
       "// Register the function which deals with the matplotlib target/channel.\n",
       "// The kernel may be null if the page has been refreshed.\n",
       "if (IPython.notebook.kernel !== null) {\n",
       "    IPython.notebook.kernel.comm_manager.register_target(\n",
       "        'matplotlib',\n",
       "        mpl.mpl_figure_comm\n",
       "    );\n",
       "}\n"
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<img src=\"data:image/png;base64,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\" width=\"600\">"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib notebook\n",
    "from matplotlib import cm\n",
    "from mpl_toolkits.mplot3d import Axes3D\n",
    "from mpl_toolkits.axes_grid1 import make_axes_locatable\n",
    "\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(6,4),subplot_kw={\"projection\": \"3d\"})\n",
    "\n",
    "# pd.DataFrame(history.history).to_csv('history_phat_learning_on_100x_scale.csv', index=False)\n",
    "histy = pd.read_csv('history_phat_learning_on_100x_scale.csv')\n",
    "bodyloss = histy.nll\n",
    "tailloss = histy.two_tailed_amlse\n",
    "X, Y = np.meshgrid(bodyloss, tailloss)\n",
    "Z = X / (Y + 1)\n",
    "\n",
    "surf = ax.plot_surface(X, Y, Z, cmap=cm.coolwarm,\n",
    "                       linewidth=0, antialiased=False)\n",
    "\n",
    "plt.gca().invert_xaxis()\n",
    "\n",
    "ax.set_xlabel('Log Likelihood', labelpad=10)\n",
    "ax.set_ylabel('AMLSE', labelpad=10)\n",
    "ax.set_zlabel('Total Loss')\n",
    "\n",
    "ax.set_title('Phat Loss with Incorrect Scale')\n",
    "\n",
    "ax.get_proj = lambda: np.dot(Axes3D.get_proj(ax), np.diag([1, 1.5, 1.25, 1.1]))\n",
    "\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "celltoolbar": "Edit Metadata",
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}