Gilles86/Simple gamble decision model.ipynb

## Simple gamble decision model.ipynb
{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "ed1a4b9a",
   "metadata": {},
   "outputs": [],
   "source": [
    "import scipy.stats as ss\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "1a6a68c5",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_posterior(mu1, sd1, mu2, sd2):\n",
    "    \n",
    "    var1, var2 = sd1**2, sd2**2\n",
    "    \n",
    "    return mu1 + (var1/(var1+var2))*(mu2 - mu1), np.sqrt((var1*var2)/(var1+var2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "38a308f8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2.2, 0.22360679774997896)"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "get_posterior(3., .5, 2., .25)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "b38d8800",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_diff_dist(mu1, sd1, mu2, sd2):    \n",
    "    return mu2 - mu1, np.sqrt(sd1**2+sd2**2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "230ebb6d",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_p(diff_dist, thr=0.0):\n",
    "    \n",
    "    return ss.norm(*diff_dist).sf(thr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "7e2253fc",
   "metadata": {},
   "outputs": [],
   "source": [
    "prior = (2., 1.)\n",
    "\n",
    "ll1 = (1., .5)\n",
    "ll2 = (3., 1.)\n",
    "\n",
    "post1 = get_posterior(*prior, *ll1)\n",
    "post2 = get_posterior(*prior, *ll2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "63b16f0c",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_psychm_curve1(base_number=5,\n",
    "                      base_ll_sd=1.,\n",
    "                      risky_ll_sd=1.,\n",
    "                      prior_center=20, \n",
    "                      prior_sd=1.,\n",
    "                      exp=None):\n",
    "    \n",
    "    risky_offers = np.round(2**(np.linspace(-12., 20., 50, True)/4.)* base_number).astype(int)\n",
    "    \n",
    "    ll1 = (np.log(base_number), base_ll_sd)\n",
    "    prior = (np.log(prior_center), prior_sd)\n",
    "    \n",
    "    post1 = get_posterior(*prior, *ll1)\n",
    "    \n",
    "    ps = []\n",
    "        \n",
    "    for r in risky_offers:\n",
    "        post2 = get_posterior(*prior, np.log(r), risky_ll_sd)\n",
    "        diff = get_diff_dist(*post1, *post2)\n",
    "        p = get_p(diff, thr=-np.log(0.55))\n",
    "        ps.append(p)\n",
    "        \n",
    "    return risky_offers, ps"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "5a959537",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.lines.Line2D at 0x7fd5680c4c10>"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(*get_psychm_curve1(risky_ll_sd=1., prior_sd=1., base_ll_sd=1.))\n",
    "plt.plot(*get_psychm_curve1(risky_ll_sd=1., prior_sd=5., base_ll_sd=1.))\n",
    "plt.plot(*get_psychm_curve1(risky_ll_sd=1./5., prior_sd=5., base_ll_sd=1./5.))\n",
    "\n",
    "plt.axvline(5/0.55, c='k')\n",
    "# plt.plot(*get_psychm_curve1(risky_ll_sd=5., prior_sd=1., base_ll_sd=5.))\n",
    "# plt.plot(*get_psychm_curve1(risky_ll_sd=.1, prior_sd=1., base_ll_sd=5., prior_center=5.))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "6931d9d1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(*get_psychm_curve1(risky_ll_sd=1., prior_sd=1., base_ll_sd=1.))\n",
    "plt.plot(*get_psychm_curve1(risky_ll_sd=1., prior_sd=5., base_ll_sd=1.))\n",
    "plt.plot(*get_psychm_curve1(risky_ll_sd=1./5., prior_sd=10., base_ll_sd=1./5.))\n",
    "\n",
    "plt.axvline(5/0.55, c='k')\n",
    "\n",
    "plt.gca().set_xscale('log')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "0fdaf291",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.lines.Line2D at 0x7fd54efa08b0>"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(*get_psychm_curve1(risky_ll_sd=1., prior_sd=1., base_ll_sd=2.))\n",
    "\n",
    "plt.axvline(5/.55)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "07487769",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "02fd4ac5",
   "metadata": {},
   "outputs": [],
   "source": [
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "47103769",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>safe</th>\n",
       "      <th>risky</th>\n",
       "      <th>p</th>\n",
       "      <th>log(risky/safe)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>0.058874</td>\n",
       "      <td>-1.609438</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>0.058874</td>\n",
       "      <td>-1.609438</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>0.058874</td>\n",
       "      <td>-1.609438</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>0.058874</td>\n",
       "      <td>-1.609438</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>0.058874</td>\n",
       "      <td>-1.609438</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45</th>\n",
       "      <td>28</td>\n",
       "      <td>570</td>\n",
       "      <td>0.919798</td>\n",
       "      <td>3.013432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46</th>\n",
       "      <td>28</td>\n",
       "      <td>638</td>\n",
       "      <td>0.931252</td>\n",
       "      <td>3.126134</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47</th>\n",
       "      <td>28</td>\n",
       "      <td>715</td>\n",
       "      <td>0.941506</td>\n",
       "      <td>3.240078</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48</th>\n",
       "      <td>28</td>\n",
       "      <td>800</td>\n",
       "      <td>0.950399</td>\n",
       "      <td>3.352407</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49</th>\n",
       "      <td>28</td>\n",
       "      <td>896</td>\n",
       "      <td>0.958244</td>\n",
       "      <td>3.465736</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>300 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    safe  risky         p  log(risky/safe)\n",
       "0      5      1  0.058874        -1.609438\n",
       "1      5      1  0.058874        -1.609438\n",
       "2      5      1  0.058874        -1.609438\n",
       "3      5      1  0.058874        -1.609438\n",
       "4      5      1  0.058874        -1.609438\n",
       "..   ...    ...       ...              ...\n",
       "45    28    570  0.919798         3.013432\n",
       "46    28    638  0.931252         3.126134\n",
       "47    28    715  0.941506         3.240078\n",
       "48    28    800  0.950399         3.352407\n",
       "49    28    896  0.958244         3.465736\n",
       "\n",
       "[300 rows x 4 columns]"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "4d6751b0",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = []\n",
    "for base_number in [5, 7, 10, 14, 20, 28]:\n",
    "    risky,  p = get_psychm_curve1(risky_ll_sd=1., prior_sd=1.5, base_ll_sd=.5, base_number=base_number, \n",
    "                                  prior_center=14.)\n",
    "    \n",
    "    \n",
    "    d = pd.DataFrame({'safe':base_number, 'risky':risky, 'p':p})\n",
    "    df.append(d)\n",
    "    \n",
    "df = pd.concat(df)\n",
    "\n",
    "df['log(risky/safe)'] = np.log(df['risky'] / df['safe'])\n",
    "\n",
    "# sns.lineplot('log(risky/safe)', 'p', 'safe', data=df)\n",
    "\n",
    "# plt.axvline(np.log(1/.55), c='k', ls='--')\n",
    "# plt.axhline(.5, c='k', ls='--')\n",
    "\n",
    "\n",
    "# sns.despine()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "40b52822",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>safe</th>\n",
       "      <th>risky</th>\n",
       "      <th>p</th>\n",
       "      <th>log(risky/safe)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>0.058874</td>\n",
       "      <td>-1.609438</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>0.058874</td>\n",
       "      <td>-1.609438</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>0.058874</td>\n",
       "      <td>-1.609438</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>0.058874</td>\n",
       "      <td>-1.609438</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>0.058874</td>\n",
       "      <td>-1.609438</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45</th>\n",
       "      <td>28</td>\n",
       "      <td>570</td>\n",
       "      <td>0.919798</td>\n",
       "      <td>3.013432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46</th>\n",
       "      <td>28</td>\n",
       "      <td>638</td>\n",
       "      <td>0.931252</td>\n",
       "      <td>3.126134</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47</th>\n",
       "      <td>28</td>\n",
       "      <td>715</td>\n",
       "      <td>0.941506</td>\n",
       "      <td>3.240078</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48</th>\n",
       "      <td>28</td>\n",
       "      <td>800</td>\n",
       "      <td>0.950399</td>\n",
       "      <td>3.352407</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49</th>\n",
       "      <td>28</td>\n",
       "      <td>896</td>\n",
       "      <td>0.958244</td>\n",
       "      <td>3.465736</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>300 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    safe  risky         p  log(risky/safe)\n",
       "0      5      1  0.058874        -1.609438\n",
       "1      5      1  0.058874        -1.609438\n",
       "2      5      1  0.058874        -1.609438\n",
       "3      5      1  0.058874        -1.609438\n",
       "4      5      1  0.058874        -1.609438\n",
       "..   ...    ...       ...              ...\n",
       "45    28    570  0.919798         3.013432\n",
       "46    28    638  0.931252         3.126134\n",
       "47    28    715  0.941506         3.240078\n",
       "48    28    800  0.950399         3.352407\n",
       "49    28    896  0.958244         3.465736\n",
       "\n",
       "[300 rows x 4 columns]"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "4d3174b9",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/gdehol/miniconda3/envs/pymc3/lib/python3.9/site-packages/seaborn/_decorators.py:36: FutureWarning: Pass the following variables as keyword args: x, y, hue. From version 0.12, the only valid positional argument will be `data`, and passing other arguments without an explicit keyword will result in an error or misinterpretation.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAEGCAYAAABo25JHAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAABHo0lEQVR4nO3dd5xcVf3/8deZfqf3sn2z2Wx6QggJRTAQAoEviAKCKApf9YfK168iIk0UFKQJUpQWRENRiqDUUEIggBQhSKSFFJKQbNq2bJ3dnXZ+f8ySb0gCpOzsZHc/z8djH2TuvXPnfZfd/dx77rnnKK01Qgghhi9TsQMIIYQoLikEQggxzEkhEEKIYU4KgRBCDHNSCIQQYpizFDvArpo9e7Z+6qmnih1DiO3MmDEDgIULFxY1hxCfQn3aikF3RdDU1FTsCEIIMaQMukIghBCifw26piEh9lYXXXRRsSMIsVukEAjRTw4//PBiRxBitwyJQpBOp6mvr6enp6fYUXaLw+GgrKwMq9Va7ChiDyxevBiAyZMnFzWHELtqSBSC+vp6PB4PVVVVKPWpN8b3Slprmpubqa+vp7q6uthxxB4466yzAOk1JAafgt0sVkr9SSnVoJR691PWK6XUjUqpFUqpt5VSU3b3s3p6egiFQoOuCAAopQiFQoP2akYIMfgVstfQXGD2Z6w/Cqjt+zoDuGVPPmwwFoGPDebsQojBr2BNQ1rrF5VSVZ+xyXHAXTo/DvZrSim/Uiqhtd5QqExCCDEQtNb09qZIdibpbO+is62TZEcXXe2ddLUnSbZ10N3ZRU9XD6meXlK9aVK9KTKpFKlUhmwmQzabJZvNkclmyWVzZHM59p06ltMv+Z9+z1vMewSlwNqtXtf3LduuECilziB/1UBFRcWAhNuRDz74gK997WsopXjwwQepqakpWhYhRGElk920NG2maVMzm+ob2LRuE5vqN9G0sYm2zR10dXWR7Oqmu7eXnlSK3nSadDZDJpslk8sVJFOqJ8XpBdhvMQvBjtpDdjhLjtZ6DjAHYOrUqUWbSefhhx/muOOO41e/+lWxIoi92OWXX17sCGInZbNZNq5v4KOV9Sx9ZzlrPlzLhrUbaWxoYnNrG+1dXaQymR2+VwE2sxWr2YzVbMFmseC02Ag4XNitFqwWC3aLBZvVgt1mxWGzYrdasdos2G0W7HYrdrsNh8OGzWHDZrdis9uwGzbsDjt2w47d6cBmt2N12vPbOA3sTjt2n78g349iFoJ6oHyr12XA+oEO0dXVxUknnUR9fT3ZbJZf/OIXLF26lMcee4zu7m4OPPBAbrvtNp588kmuv/56zGYzL774Is8//zz33HMPN954I6lUiunTp3PzzTdjNpsH+hDEXuLAAw8sdgSxjd6eXlau+Igl7yxj8evvsHLZatav20RLWxu5rc7aFWBY7RhWGy67g1jYh9fpwOd05v/rcRLwuQgF3YRCXjwBF4bPid3jxOp1Y3O7sLhdWAwDk82OyWJFmc2D5v5fMQvBo8APlVL3AdOBtmLcH3jqqacoKSnhiSeeAKCtrY1Zs2bxy1/+EoBvfvObPP744xx77LF8//vfx+12c84557BkyRLuv/9+Xn75ZaxWK2eeeSZ/+ctf+Na3vjXQhyD2Eq+88gogBaFY0ukMH7y3nDdfW8ybry5m+dJVbNjYwMfT8ZqVCY/DicduMC5eTtjtJurzUhoOkIj68AWcuP1ODJ8dh8+Fze/B5vVicbmwulyYHQYmqw1lGnoj8xSsECil7gVmAGGlVD1wMWAF0FrfCswDjgZWAEngvwuV5bNMmDCBc845h/POO49jjjmGgw8+mIceeoirr76aZDJJS0sL48aN49hjj/3E+xYsWMCbb77JfvvtB0B3dzfRaLQYhyD2EhdeeCEgzxEMlGRXkrcWvcui1xbz+ktv8sEHK0in8805TqudgOFmfLyChM9PeThIVTxEMOzEF3Ri99hx+JzYPA5sAR82rw+r24PF6cJsdwzJP/afpZC9hk75nPUa6P/b37to1KhRvPnmm8ybN48LLriAI444gptuuolFixZRXl7OJZdcssM+/lprTjvtNK644ooipBZieFr70TpefO5VFjzxAm/9+z2y2SwKRcBwUe2LUR4MMTIWpTzuJxBw4g32neH7DaweA5vXg83nx+r1YXV5MNlsg6b5ppCGxJPFe2L9+vUEg0FOPfVU3G43c+fOBSAcDtPZ2cmDDz7IiSeeuN37Zs6cyXHHHcdPfvITotEoLS0tdHR0UFlZOcBHIMTQpbXm7X+/x7NPvchzT73E2jX524heu5PaYAnV4Qg1sSixkJtIxIUr4MAdcmLzGFjdDqwuJ7ZAEHsgiMXlwWQZ9n/ydmjYf1feeecdfvazn2EymbBardxyyy08/PDDTJgwgaqqqi1NP9saO3Ysl112GUcccQS5XA6r1cpNN90khUCIfrC+fiOP//0ZHn5gHvVrN2BSiqjbz76lIxkVT1AdCxKLufEFDJx+A7vPjs1tYHHbsfv9OMIxrB4vJptdzvh3gvr4RspgMXXqVL1o0aJPLFuyZAljxowpUqL+MRSOYbiTGcr2TLIryfx5L/DI355k0b/+A0DU5aM6FGdCaTnxkJtoxIU3YOAKOnH4Hdg8BmbDjtXnxQhFsfkDmKy2Ih/JXutTK+KwvyIQor9cf/31xY4wKDVsauLeuX/n/rsfprOjC4/DyYR4FXXxMmriQeJRN/6QE6fPseWPv9XtwOxy4ozGsQVCmG32Yh/GoCaFQIh+IsNP75rlS1dy1+3388Q/5pPN5ij3hdl/1CjqEjHiQRfhqBu3344r5MLudWBxO7A47NgCfoxwDKvPj1LDq3dPoUghEKKfPPvss4BMUPN5Fi96l9tuvJOXX3gdq8VCtT/O2EQFI+MhSmJefAEHTr8DZ8DA5nHkz/6dBkY4gj0UxWI4i30IQ44UAiH6yWWXXQZIIfg0q1eu5Yar5rDgqRdxOQwmxqsZnSinIuqjJOrBHzRwh5zYvXZsXgOLYcfidGDEEjgicenxU0DynRVCFFRzYwu33nAnD/71UcwmMxPiVUwoH0F52E087MYbNHAHDBwBA7vXwOK0YzbsOMIxjGhcbv4OACkEQoiC6O7u4a459/Pn2+6lp7uXUdFSxkYqGVEapDTixht04vLZMIIubG47FpcDi2HHHo5ixBJyA3gASSHoR1VVVXg8HsxmMxaLhW27uQoxXLz+yr+55LzfUr9mPSNiJYwpL2NEeZyyoINA2I0nYMcIurG5rVsKgC0QwFVSgdlhFDv+sCOFoJ89//zzhMPhYscQoig6O7q47spb+ds9jxL0ejmsZhI1ZaWUBQ3CYTeevl5ADp8dq8uB2bBhdjlxlVRg8wfl4a8ikUIgRD+57bbbih2hqF56/jUuvfBaGjY2Mr5sBOMi5dRUhIl67HgjblxeC0bYi81lxeI2MNut+fsA8VK5EVxkQ+6737V+DdnuZL/u02zkz1g+j1KKI444AqUU3/ve9zjjjDP6NYfYu9XV1RU7QlG0t3Vw9a//wKMPPkU0FGJmzT7UlJdQGXISiHjw+K04Ay5sXhtWV/5msMXtxl1WhcXlLnZ8wRAsBMX08ssvU1JSQkNDA7NmzWL06NEccsghxY4lBshjjz0GsN2Q5UPZu/9ZwjlnXsKmDY3sW1NHjTtGTXWMmNuKN+bB7bXijPiwGCYsLmffVUAEZ0k5yiSTOO0thlwh2Jkz90IpKSkBIBqN8pWvfIXXX39dCsEwcu211wLDoxBorXngnke4+td/wO10MnPkZCoTcapibvweA2/YgTPgxAgYWAwbZlf+qWBXaSU2f7DY8cU2hlwhKJauri5yuRwej4euri6eeeaZLbOcCTGUJLuSXHrh73ji4fmMLK9goq+ckaPKiBsmPGEvHp8ZI+TF5rZgdTsxO6xY3V5cFdXSJXQvJYWgn2zatImvfOUrAGQyGb7+9a8ze/bsIqcSon+tXL6an/7gYlZ9uIb968ZT7QwzqjZB0GEiUOLHYYAzFsRiB6vHjclmxoiVYMRKpEfQXkwKQT8ZMWIE//nPf4odQ4iCeX7+y5z/o0uxWS3MGrMvJf4QoyqDeJ1W/HEXDrcNZ8SDyaKwet2YbBbc5SOw+fzFji4+hxQCIcTnuv/uh7nilzdQUZpgkreSsoo41UEHbr8Tb9CGM+jB4bNhsucfEDPb7Xiqa2WAuEFCCoEQ/eTuu+8udoR+p7Xmxt/ezh03/YXRI0YwwVVKzdgKYtYcnogPt1dhhHzY3eZ8ryCHBYvTjad6pIwRNIhIIRCin5SXlxc7Qr9Kp9JcfN7VPP73Z9h37FhGWsOMnlRNUKfxlYZxOnIY0QBWQ2HzelFWhc0XwF1ejTJL19DBRAqBEP3k/vvvB+Dkk08ucpI919nRxdnf/wWv/fNNDtlnCiU5D2MnVuInQ7A6is2cxpUIY7bmsPm8KIvCiMQxEmVyU3gQkkIgRD+55ZZbgMFfCBo3NXPm6eeyYtkqZk87gGDawbh9KvHoDOEREawqg7ssismUxerzoczgKqvEEYoWO7rYTVIIhBBbNG5q5junnMWm9Q0cM/UgPBkrE/atxEinCFcFsZlzuMuiKDLYAj6UCVzl1dgDoWJHF3tAJvzsJ0uXLmXy5Mlbvrxer0xmLgaVpoZmvttXBI6csB8+HEyeUoUzkyJcHcJmU3gqIiiVxR7054tAxQgpAkOAXBH0k7q6OhYvXgxANpultLR0ywNmQuztmhtb+H9fP5sN6zYxa+xUAjYXE8eXY+npIVwdxmrReMrDoDPYQwHQOdyVNdh8gWJHF/1ACkEBLFiwgJqaGiorK4sdRYjP1dy0me9+/WzWrd3AkROnEbA4mTShHHNPD5GaMBZTFk9FDLK92EIBIIe7coQUgSFkyBWCtuXLyHR29Os+LW4PvtpRO739fffdxymnnNKvGcTe78EHHyx2hF3W0tzK//v6T6hfs56j99kfT87GxPFlWHp7iYyMYtZpvFVxdLoHW9CPUjp/JeD1Fzu66Edyj6CfpVIpHn30Ub761a8WO4oYYOFweFDNTte6uY0zvnE2a1ev45j9DsKVsTB5SjXWVIpITSRfBEYk8kUgFJAiMIQNuSuCXTlzL4Qnn3ySKVOmEIvFippDDLy5c+cCcPrppxc1x87o6enlf799Aas/XMOXvzADa1uWKQfUYmnvIFqXwJTrxVtThk51YQ+HQGdwlVVJERii5Iqgn917773SLDRMzZ07d0sx2Jtls1ku+NGlvP3W+xxz0CFYWjNMPXgslvYOInUlmLI9eEeUfqIIGPFS7MHBc7Ujdk1BC4FSarZSaqlSaoVS6vwdrPcppR5TSv1HKfWeUuq/C5mn0JLJJPPnz+f4448vdhQhdkhrzdW/+j0Lnn6J2QcdhH1zlv0OnYC5ZTPhUaVYst14qkohncQe7uslFI7iiMSLHV0UUMGahpRSZuAmYBZQD7yhlHpUa/3+Vpv9D/C+1vpYpVQEWKqU+ovWOlWoXIXkdDppbm4udgwhPtWdc+7n3jv/waEHTsffaWHKYRMxNzYTrElgyXbhriyBbBJbMIjOpbD7QzgT5TJsxBBXyCuCacAKrfXKvj/s9wHHbbONBjwq/1PmBlqATAEzCTFsPfnoAn53+S1M328ysS4Ho6eNwtbUhK8yho1uXCVRFL1YPR5QWaweL67yKikCw0AhC0EpsHar1/V9y7b2B2AMsB54B/ix1jq37Y6UUmcopRYppRY1NjYWKq8QQ9Ybr77FRT+9grHjRlGd8VExpgJ/uhtPIozTmsYRDmJ1KCx2ByabGYvhwlM5EmWS24jDQSF7De3oNEJv8/pIYDFwGFADzFdKvaS1bv/Em7SeA8wBmDp16rb7EGKvMG/evGJH2KEVy1Zx1hkXUVISY7KrjGAkSIXPitVmwe1RWF0GRsCBzmUxuw3MVgue6loZSnoYKWS5rwe2HqC9jPyZ/9b+G/i7zlsBrAJGFzCTEAXjdDpxOveuGbnaWtv50XcuxGazcnDlOAzDYOzICCqbJVjmxWwx4y4Nk0untgwi564aiclqLXZ0MYAKWQjeAGqVUtVKKRvwNeDRbbZZA8wEUErFgDpgZQEzCVEwN998MzfffHOxY2yRzWY570eXsnFDA0fusz+qO8u++48i09pOdHQppHrwjighm+zAEQlDLoO7vFqmlxyGClYItNYZ4IfA08AS4AGt9XtKqe8rpb7ft9mlwIFKqXeABcB5WuumQmUSopAeeOABHnjggWLH2OIP19zBKy+8zpdmHoZu7Gb6EfuQ2dBAbFwVJNvxjignl2zHHg6jsymMWImMHzRMFfTJYq31PGDeNstu3erf64EjCplhoHz729/m8ccfJxqN8u677wLQ0tLCySefzOrVq6mqquKBBx4gEJBfNFF4zzzxPHfc/BcO/eKB2DemmPjFCeTWridYU4LqbsVVFkenOrH3dRO1+YM4oolixxZFIl0C+snpp5/OU0899YllV155JTNnzmT58uXMnDmTK6+8skjpxHCyfOlKfnHOVYwdV0uiy0FpXRmernZc0QB2axp7wIfFprE4nSgzWAwnbukmOqxJIegnhxxyCMFg8BPLHnnkEU477TQATjvtNB5++OEiJBPDSXtbB2f9v4twOh1MDdfg8rqojrgwWy34Y05MJoUz5gWtsXpcKJPCXTUSZZIeQsPZkBt0ruH1xfS0tPbrPh1BP9Fpk3f5fZs2bSKRyF9uJxIJGhoa+jWXEFvLZrOc/6NL2bB+E9848igyG7vYZ+ZoMhubSEyuItvZSnDMCDKdrRjxKLl0L+6qkZht9mJHF0U25AqBEMWycOHCon7+bTfexT8X/ouvfeUYUqs6mD57XzLrNxIbX0m2oxV/XTWZzlbs4TC5dC+OSFxGExXAECwEu3PmXiixWIwNGzaQSCTYsGED0Wi02JHEEPXGq28x58a7mDHjAFjVSd30OkybGvBVRFE97bjK4uR6O7C43WidwepyY8S3fdBfDFdyj6CAvvSlL3HnnXcCcOedd3LccdsOtSSGkmuuuYZrrrlmwD93c0srF/z4MkrLEpT1uolWRInbwOZ04PKYsLldWA0zymzG7HJgsphxV4yQm8NiCykE/eSUU07hgAMOYOnSpZSVlXHHHXdw/vnnM3/+fGpra5k/fz7nn7/dSNxiCHn88cd5/PHHB/Qztdb84qdXsnlzGzPHT8GkYcLEclIdXURrY+hUCk9FlFxvN45wCDJp3BUjMFltA5pT7N2GXNNQsdx77707XL5gwYIBTiKGk3vu+BsvPvcqJ3/5v+hd3cGB/7Uf3WvWk9hnJNn25r77Am04olFyqW6MWAlWt7fYscVeRq4IhBik3nv7A6678jYOOHAqpo+S1E4dhV63AX91ArpaMOIRcqlOLG4PuWwKi9sjD42JHZIrAiEGoc6OLs794a8JhQOMdsSwuSwkXGZU2oHba4KsDcNvkO3txuJ2QC6Lu7xa7guIHZIrAiH6iWEYGIZR8M/RWnPphdeyvn4jxx08g97WJFMOrCO9OT+YXLYria+2gkyyE0c0gk6ncJVVyX0B8ankikCIfvLkk08OyOc8+uBTPPnoAk468Rg6ljWy7xFT6F21jsjYSrKtzfhqq8i0t2ALBsn2JnEEIzKYnPhMUgiEGETWrd3AlZfcyOQp43CsSxEZVYarow1L2Ic1l8QeCqBUCmWzYbIolNmBs6T883cshjVpGhKin1x66aVceumlBdt/Lpfjl+fkBy7cr2QUJqUYMypGpidFqDwAaFylIXI9PflRRbMZ3BXVMtOY+FxSCPrJt7/9baLRKOPHj99u3TXXXINSiqYmmWphKFuwYEFBuwv/5c8P8cZri/nykYfTWb+Z6bP3JblmI/GJI8h2tOGvqybd2oIjFiOXSmLESrA43QXLI4YOKQT9ZEfDUAOsXbuW+fPnU1FRUYRUYqhYuXw1N1w1h/0PmEJuZQejDxhDds06vBVR6NqMMx4hm2zr6yrai8Xplq6iYqdJIegnOxqGGuAnP/kJV199tXTbE7stnc7w87Mvx+k0GOcpxRN0U+KxYDKZ8IUNTBYzjqAbpTVWrwulFK4K6Soqdt6Qu1m8fN4rdG5o7td9uhMhao8+cJff9+ijj1JaWsqkSZP6NY8YXv540z289/ZS/vvkE0gubebQEw+ke9lHlE+vI9PcQHDCKDJtTRglCXK9SZyllTK0tNglQ64Q7C2SySS/+c1veOaZZ4odRQyQUCjU7/t87+0PuP33d3HYzIPoXd7CuIPH07uynkBNKemWBjzlJWQ6WrB6fWRT3VjdHuzBcL/nEEPbkCsEu3PmXggffvghq1at2nI1UF9fz5QpU3j99deJx+NFTicK4aGHHurX/fX09PLzn1xOMBRghDmEOaSIO82kei24PAqyDmxeO9muNBa3QS7di6tMppwUu07uERTIhAkTaGhoYPXq1axevZqysjL+/e9/SxEQO+0P19zByhUfcfysw+lqaGe/wyaQ3NBEfEIF2c4u/KMqyXS0YSTi5Hq7cSbKpUlI7BYpBP1kR8NQi+Hlggsu4IILLuiXfb2zeAn33PE3jjr6UNre28SEQyfRs3wNgRElZDY34a4oJdPegtXnJ9PbhUWahMQeGHJNQ8XyacNQf2z16tUDE0QUzauvvtov+0n1pvjlz64iEg1RmnJjiZiJORS9XWZcPguk7djcVrLdKSxuB7mUNAmJPSNXBELsZf5481/4cNkqjp0xg66mDqbNnETXukbiE6vIdnTkm4Q62zHicXI90iQk9pwUAiH2Iss++JA//uFuDp15EMllTUw4dBLJZavwVyfIbm7CVZYg07EZq9cnTUKi30ghEGIvkclkuPhnV+HxehhpjeAOuIk5TCiTCU/AjslqwREw0FpjcRv5B8ekSUj0AykEQvSTsrIyysrKdvv999zxIO+9vZTjj5pF58ZWph46ka51DSQm15BpbyMwegTptlaMeCzfSyheJk1Col/IzWIh+sk999yz2+/9aFU9N117BwcdvB89HzQzalod6dXr8FXFybY24iyJkencjMXtJpfuxeJ0YQ9F+jG9GM7kikCIIsvlclxy3tXYbFYmhiqxGTYq4x60zuELOzGZTDgjPnQ2i83vBZ3DKU1Coh9JIegna9eu5dBDD2XMmDGMGzeOG264AYCWlhZmzZpFbW0ts2bNYvPmzUVOKgrlrLPO4qyzztrl9z3418d481//4cQvH03r6iamHbkvnas3kJhUQ7p1M/66EaQ2N2HEYmS7u3BE4lgchZ8SUwwfBS0ESqnZSqmlSqkVSqnzP2WbGUqpxUqp95RSLxQyTyFZLBauvfZalixZwmuvvcZNN93E+++/z5VXXsnMmTNZvnw5M2fO5Morryx2VFEgixcvZvHixbv0noZNTVx/5W3su99EssvbqBhfBRs24YoFIdmKIxwk19uB2TDQZDDZHRgyvLToZwUrBEopM3ATcBQwFjhFKTV2m238wM3Al7TW44CvFipPoSUSCaZMmQKAx+NhzJgxrFu3jkceeYTTTjsNgNNOO42HH364iCnF3uaqS24k1Zti/xFj0TnNmDElpLt6CI+IkEtncJdFyPX24oiE0ek0rtJKlEku5EX/KuTN4mnACq31SgCl1H3AccD7W23zdeDvWus1AFrrhj390Pl/fJKGVRv3dDefEK2OM+u7R+309qtXr+att95i+vTpbNq0iUQifwaXSCRoaNjjQxRDxAsLXmH+vBc4+aRjaVm6kQO+fADtSz8iNqmGVGMD/lHVpDY34YhGyXS1Yw+Gsbo9xY4thqBCnlqUAmu3el3ft2xro4CAUmqhUupNpdS3drQjpdQZSqlFSqlFjY2NBYrbPzo7OznhhBO4/vrr8Xq9xY4j9lLJriSX/+J6qmsqcG7MEq9J4Ghrw+5zY6UHq8eFyZzFZLGgLAplsWDEd79rqhCfpZBXBDvq0qB38Pn7AjMBA3hVKfWa1nrZJ96k9RxgDsDUqVO33ccn7MqZe39Lp9OccMIJfOMb3+D4448HIBaLsWHDBhKJBBs2bCAajRYtnyisUaNG7fS2N183lw3rNvGDU0+h7f2NTDpmP9rfXUHlgWNINWwksM8YUs2bcJWVkenuwFUxApNFenuLwijkFUE9UL7V6zJg/Q62eUpr3aW1bgJeBAbldF5aa77zne8wZswYzj777C3Lv/SlL3HnnXcCcOedd3LccccVK6IosDlz5jBnzpzP3W7Ju8v4y58eZPZRh9L63kYmHTaZjiUrCdVVkGrahKeqjHRrE1a/n3RPJ1aPD5svMABHIIarQhaCN4BapVS1UsoGfA14dJttHgEOVkpZlFJOYDqwpICZCubll1/m7rvv5rnnnmPy5MlMnjyZefPmcf755zN//nxqa2uZP38+55+/w85TYpjIZrP8+oJr8Pm9VJoCeIJuwjYwWy243CbMVit2rx2tNVZXfhgJZ2mFPDMgCqpg15pa64xS6ofA04AZ+JPW+j2l1Pf71t+qtV6ilHoKeBvIAX/UWr9bqEyF9IUvfAGtd9xqtWDBggFOI4rhjDPOAPjMq4L77voH7729lO9+62Ta397EYScfQteSDymfPpp08yZCk0aT2tyAkUiQ7UnKyKJiQBS00VFrPQ+Yt82yW7d5/Vvgt4XMIcRAWLZs2Weu37i+gd//9o9M238fej9opmbKSNKr6/GURsi2NWPEwmS72jA7neSyKcyGE3tY7imJwpMOyUIMkKsuuZFcNse0ytEopaitiZLpTRGsCEAuh7skTC7Viz0UhGwm/8yANAmJASCFQIgB8MKCV1jw9Esc/+WjaF66gamzp9K29CPiE2tINTbiG1VNb0sjjkiUbLIDeyiKxekqdmwxTEh/NCEKrLu7hyt+eQPVNRUYmzK4KmMYHe3kfC6sugfldmEyZcBiQVkVZK0Y8ZJixxbDiFwRCNFPPu4ttq05v7+L9fUbOeqgL5Bs7WLKQaPpbmolPr6STFcXvtoKMp0dOGIxcr09OEvKMZnlHE0MHPlpE6KfXH/99dstW7l8NXfOuZ9ZRxxC67sbmPDFiXQuXUVwZBnppk24yktItzVj9XrJ9HRidXvlmQEx4OSKoJ982jDUixcvZv/992fy5MlMnTqV119/vchJxUDRWnPZRdfhdBqMNKIYHicJnw1QeEIOUCaMkBudzWL1ulGAS54ZEEUghaCffNow1Oeeey4XX3wxixcv5te//jXnnntusaOKAjn11FM59dRTt7x+7O9Ps+i1xRx/7JG0ftTEfkdMoWPVehL7jCTV3Exg9AhSLf83z4ARTWC2O4p4BGK4kqahfpJIJLaMMrr1MNRKKdrb2wFoa2ujpERuAg5V9fX1W/7d1trOtb+5hfETR8OqTsrGVMDGTTijflR3G/aAD51JYrI70DqLyWbHEYkXMb0YzoZcIbjqV79n6fsr+nWfdWNHct7F/7vT2289DPX111/PkUceyTnnnEMul+OVV17p12xi73Tj1bfTtrmdU4/+L1re38D4SVW0vf8hJQeNpXfTBoJjRtDbtBFXeRmZZAfu6lqZZ0AUjfzk9bNth6G+5ZZbuO6661i7di3XXXcd3/nOd4odURTY22+9z4N/fYxjvzSLpnfWsc+sKbR/sJLI2Cp6Gzbirakk1dKALRAk092JzRfA6vEVO7YYxtSnjY+zt5o6dapetGjRJ5YtWbKEMWPGFCnR/0mn0xxzzDEceeSRW0Yg9fl8tLa2opRCa43P59vSVLS1veUYxO6bMWMGWmtirlqam1o4ft9DIJ1j+rQR9DS1UTKuhFxPN/5R5WTaWzFKYmR7e/DXjcdktRU7vhj6PrUXglwR9JNPG4a6pKSEF17IT8X83HPPUVtbW6yIosAOOOAAEqFqPnhvOV+ZPYv29ZuZevhkOtc2EJ9YTbq1FX/dCNKtLRjxONnuJM54qRQBUXRD7h5BsXw8DPWECRO2PFR0+eWXc/vtt/PjH/+YTCaDw+HYqfHqxeB09lnn8KXDTmXa/vvQ/UEzI6aMJL2qHk9JmFxHC0Y0RDbZisXpJJvtxewwsIdkUDlRfDvVNKSUcgBnAl8gP8vYP4FbtNY9hY23PY/Ho/fdd99PLLv66quZNm0a2WyWFSu2v1EcCoUIh8Ok02lWrly53fpIJEIwGCSVSrFq1art1sdiMfx+Pz09PXz00UfbrU8kEni9XpLJJGvXrt1ufWlpKW63m87OTtatW7fd+vLycj766CPWrVvHZZddtt362267jbq6Oh577DGuvfba7dbffffdlJeXc//993PLLbdst/7BBx8kHA4zd+5c5s6du936efPm4XQ6ufnmm3nggQe2W79w4UIArrnmGh5//PFPrDMMgyeffBKASy+9dLsht0OhEA899BAAF1xwAa+++uon1peVlXHPPfcAcNZZZ7F48eJPrB81atSW4nnGGWdsN8Ln5MmTtzzIdeqpp36i5w7kz9KvuOIKAE444QSam5s/sX7mzJn84he/AOCoo46iu7v7E+uPOeYYzjnnHCDf9LOtk046iTPPPJNkMskhU48i1QX7jxxJhSuBNdjLtPJaqg8cTW9jA6+tW8ER++/H6yuWc/D0qZz/u98z86ijOfnkk1m7di3f/OY3t9v/T3/6U4499liWLl3K9773ve3WX3TRRRx++OEsXryYs846a7v1l19+OQceeCCvvPIKF1544Xbrr7/+eiZPnsyzzz4rP3uD+Gfv6KOP3m796aefzumnn05TUxMnnngiCxcu3OOmobuAccDvgT8AY4C7d/K9Qgx5b7z6FqkuMJtSVLvLaNSNTCsbSY/PTG/DJhwVCQ7bbwpvLlvGgVP34ZmXX2XZ6u1PKoQohp29IviP1nrS5y0bCHvzzeI9MRSOYbhK9aY4Yfa3WbtmLUfX7UdVVSWTxyXIdHYTr42gsxl81TEy3UmMaIhcJoOvbrzMQSwG2h5fEbyllNp/y96Umg68vKep+tNg6/20tcGcXcDcOffz0cq1jIzH8FhdTPniOJIbW4hPqCLd0YF/VBXpjvb8E8S9PTgTZVIExF5lZ38apwPfUkqt6XtdASxRSr0DaK31xIKk20kOh4Pm5mZCodCgG6dFa01zczMOhwwtMBjVr9nA7b+/i0Nm7E9is42GTBPdK9bgq4yTbmnEWRIj3d6MxeMhm0picbmxBULFji3EJ+xsIZhd0BR7qKysjPr6ehobG4sdZbc4HA7KysqKHUPsIq01V158AyaziQnhSjZtXkdl3EsunSFQ4iXV0owz5ifV3IjN7yWb7JJZx8ReaacKgdZ6r76rZbVaqa6uLnYMMcw8/8w/efG5Vzn161+h5f1NuCoMJgeixCeNpLexEf+YkaSaG3FEY2STnTgiccwOo9ixhdiONFQKsRuSXUmuuPgGamqrsK9L4a1JMLYigM7lMGe7MHndmEiBzYYy5VAmG0YsUezYQuyQPFksxG645fo72bShMT/rWHuSSdNH0d3cRmxMGdmuJL6RFWS6OnFEI+TSKZylFSiTudixhdghKQRC7KJlSz7knjv+xlFHH0rL2+uYeOhEOj9YxeqOBpIb1uGuLCXd2oTV5yPT3YXV68fm9Rc7thCfSgqBELsgl8tx2c9/h9fnptIcxOl1EnfbwKQIRtykMhkcAVd+1jGPC6UUzpLyYscW4jNJIRBiF/zj/nksfvNdjv+vI2lb28y0I/elY/V6EpNqqCsr5bXVyz4561isBLPNXuzYQnwmKQRC7KSW5lauu+JWJk8ZR2ZFK5UTq8nWb8AVDUCyldWNm5g2eTxmhwOtM5gdBo6IDCon9n5SCITYSdddcQvJriRfqJtINpNjzNhS0p1JIiNj5FIp1ve0URIJYQ+H0Zl0/gaxkl8xsfeT7qNC7IRFry3mkb89xQknHE3zkg1MP2Y6bUtWEx1fQ2/DJny1Vcwq9bO+vZ3q7k5sgTBWl6fYsYXYKVIIhPgc6VSay37+O0rK4gTazdhLQ7hTSTIuBw5rhhwOzHYTZKzUjB6FzqRwJkqLHVuInSbXrUJ8jjtvv5+VKz7i2ENn0NXYwb6H9A0qN7GKdHs7gbqq/NST8Ti53u6+QeWsxY4txE6TQiDEZ1izup5bb7iTQ744nc4ljYw5aBzdy9cQGFFCurkBV2mcdHsLFreHzY0bWfLhKmyBcLFjC7FLCloIlFKzlVJLlVIrlFLnf8Z2+ymlskqpEwuZR4hdobXm0gt/h81mZVywApvDRnnYidY5vGEj/4xA1IfOZLD5vRiGg1vvf1AGlRODTsEKgVLKDNwEHAWMBU5RSo39lO2uAp4uVBYhdsdjf3+af738JicedxStqxrZ74gpdKxaT3xSDamWFgKjR9Lb3Nj3zEAn/3j2edZu3FTs2ELsskJeEUwDVmitV2qtU8B9wHE72O5/gYeAhgJmEWKXbG5p5ZpLb2bC5DGwqpPS0eWYNjXgjPhR3W3YQwFy6U5MdjuaHCabnYeeWfD5OxZiL1TIQlAKbD2Te33fsi2UUqXAV4BbP2tHSqkzlFKLlFKLBuucA2JwufY3t9DZ0ckhYyeT6U0zflIlqY4kkZoouVQab2WcXHc3jkgYnUnhKqsinckUO7YQu6WQ3Ud31FC67ZyM1wPnaa2zn9WuqrWeA8yB/JzF/RVQiB157Z9v8uiDT3HC8UfT8v4Gph87nfYPVhMdX02qqRHfqGp6mxuwh0JkuzuxB8JY3R5OOumkYkcXYrcUshDUA1uPtlUGrN9mm6nAfX1FIAwcrZTKaK0fLmAuIT5VT08vl/38WsorSvBvNuEsj+Du6SLtdGA3p8HlxGLVZDJmTDYzOgdGIj+73Jlnnlnk9ELsnkI2Db0B1CqlqpVSNuBrwKNbb6C1rtZaV2mtq4AHgTOlCIhimnPjXaxZvY6jDz6YZGsXU74whuSmzcQnVJLp6sI/qnLLRPS5VC/OkootE9Enk0mSyWSRj0CIXVewKwKtdUYp9UPyvYHMwJ+01u8ppb7ft/4z7wsIMdA+eG85f771Xg6fdTDt7zcw6dBJdC5dTbC2jGxLI+6KEtJtzVi9PjK9Xdg8Pmy+wJb3H3300QAsXLiwSEcgxO4p6BATWut5wLxtlu2wAGitTy9kFiE+Szqd4Zc/uwp/wMsISwiCOSIORa/FjMtrIddtxu6xk+5KYfU4yfb29A0qJ88MiMFPniwWAph727188N5yTjj6SDo2tDL1sIl01TcQn1hNpq0N/6gq0h1tOBMJsj1JnHGZZ0AMHVIIxLC3cvnq/DASM/YnuaSJ2ml1pFfX46uIkW1twlkSI9O5GavXSzaVxOJ0YQ/Hih1biH4jhUAMa9lslovPvRqn02CMpwSbw0pVzI3O5YeRMJnNOPxOFGD1ukFrXGXV0iQkhhQZhloMa3+d+3f+8+/3+PY3vkr7e43MOOlgOj9YSck+NWTaWwiOG0mmowVnaSnZ7k6MRBlmh2OH+zr99NMHNrwQ/URpPbiez5o6dapetGhRsWOIIaB+zXqOP+K/mTR5LCPTAWr2qSFmSmP43Hh8Cmc0jMWew+r2YLKbMDsMvDWj5WpADFaf+oMrTUNiWNJac/G5V2M2m5gYrMBu2Kgu8ZFLZ/GXeDCZTDgCBgqF1esCrXGXVX1mEWhqaqKpqWkAj0KI/iGFQAxLD937GG+8+hZfPmoWnetamXZk38iiE6rJtrfjG1lJprMDoyRBtrsLI1aC2WF85j5PPPFETjxRRlIXg48UAjHsrP1oHb+99Gb22Xc8rOpk5NRRZD5ah6c0AsnNGLEw2e42rD4f2d4kZsOFIxIvdmwhCkYKgRhWstksF519BWaTiX3jI7EbNqpibshpAgk3SplwBF0opfp6CeVwl392k5AQg50UAjGs3Dnnft5a9A5f/dJsuta1st/hk+lau4nE5BrSra34ayvIdnXkHxxLdu5Uk5AQg50UAjFsLH1/BX+49g4O/uJ0Uks3M3LfWrJr1uOtiKE7mjGiIbLd7dh8fjK9ScyGU5qExLAgzxGIYSHVm+LCn/wGn8/DaCNBTqeoCLtItaTxhR3kupM4Qi5yPd1YPAbZnm7c5bv24NgPfvCDAh6BEIUjhUAMC3+49g6Wf7CS737tRDo/aOLQr36B7qWrKJ1aS6a1icC4kWQ7WnCWlJDt7sJZWrHLTUInn3xygdILUVjSNCSGvEX/+g93zrmfI444hK4Pmpg4YyK9H67BXxUn29qEqzxBtqsVqy8/vLTV68cejOzy56xdu5a1a9d+/oZC7GXkikAMaZ0dXVx09uWUlsVJJJ14yz0EyZCz23B7zSgMbG4ruZ4MJrsVpcBVVrlbvYS++c1vAjIfgRh85IpADGm//fUf2Li+gcP3mUa2N8PkqSPobWkjPr6cbDKJr6aMbFcn9nAYsmnc5dWYLNZixxZiQEkhEEPW/Hkv8I8H5nH0ETPoXdPGfkdOoevDemKTasg0N+IdUU66rRl7MIjO9OCIxLB6vMWOLcSAk6YhMSTVr9nAJeddzegxI/E2QcWEKiwNjRjxEJZUBxa/F0Uak2GASWN2ODFipcWOLURRyBWBGHLSqTTn/u+v0Fqzf/loHE4HIxI+dC5HoNSD1jlciSA6m8bq9wDgrhiBMsmvgxie5IpADDk3XH077y5ewuknHU/vshYOOW46Pavq/6+r6Ngasp2bMeIxcqkeXGVVmO07nmNgV/z0pz/th/RCDDwpBGJIeWHBK9x1+/0cddSh9C5rYfzB40l/tI5gbRnZ1kY8FaVku1qx+f1kUz3Y/AFsgVC/fPaxxx7bL/sRYqBJIRBDxsYNDVx09hXUjqom3G4lUOEnmEth8rowrGnMdg8WO6CtmOwWlFK4Snevq+iOLF26FIC6urp+2Z8QA0UaRcWQkMlkOP9Hl5JOpzmoehxmrRg3poRMsptIdRidyeCpiJLr7cEWDKAzaVzl1Zgs/Xcu9L3vfY/vfe97/bY/IQaKFAIxJNz8uz/z79ff5rjDDqN3Uyf7zZhA74ZGEhOryHa04autINPRij0cRqd7cJaUY3V7ih1biL2CFAIx6D375Iv88aZ7mHHIAbC2i4mHjCdbv4FwXTm59hbc5Qmy3e1YfT50LoUtEMIeihY7thB7DblHIAa15UtX8vOzL2fM2FpKkk5KRsXxdHXgTISw0YMt6MdsyWKyGiiLwmK4+vW+gBBDgRQCMWi1tbbz4+/+HJfLYFq8Fqs2URV2YdZZfBEHZDMYYTc6k8LiMVAmE56qGnleQIhtSCEQg1Imk+G8//01mzY2curs/6K3vo19DxmLbm0jtk812fbN+EdVkOvuxB4Jgc7irqzFZLUVLNNFF11UsH0LUUhSCMSgdMNVt/PKi29w8peOpmdNK/vOGA8tm0lMGUm2rRlvTXm+CIRCkMvgKqvC6nIXNNPhhx9e0P0LUShyjSwGnScens+dc+7j8EMPwrQmydgDxmBr2UxoVBm59iaciQg63YXV60XrNPZQFHswXPBcixcvZvHixQX/HCH6W0ELgVJqtlJqqVJqhVLq/B2s/4ZS6u2+r1eUUpMKmUcMfu+/s4xLzr2acePriLRZqRhXSSCdxBUNYKMbR8CP1aGwOBwoi8Lq9uIsKRuQbGeddRZnnXXWgHyWEP2pYIVAKWUGbgKOAsYCpyilxm6z2Srgi1rricClwJxC5RGD36aNjZx1xs/x+b1McJUSTISoCjowmc14A1YsDjtG2A1ozC4HZpsNd+UIlJILXyE+SyF/Q6YBK7TWK7XWKeA+4LitN9Bav6K13tz38jVgYE7dxKDT3tbBD771M9rbOpkxchJOm51xo+JkOrsJVwUxmTTu0hC5VA+2gA+lFO6qkTLJjBA7oZCFoBTYegLX+r5ln+Y7wJM7WqGUOkMptUgptaixsbEfI4rBoKenlx9990JWr1zLcQcegrVXs89+I8m0tBKri6MyvXgq4+R6kzjCfT2EyquwGM5iRxdiUChkIdjREzt6hxsqdSj5QnDejtZrredoradqradGIrs+qbgYvLLZLBf86FLeeuMdTpg1Cxp72PfgceimFqKjy1CpJN6qBDqVxB4OoXNp7OEYNn+w2NGFGDQK2X20Hijf6nUZsH7bjZRSE4E/AkdprZsLmEcMMlprfnPRdSx4+iW+fMRMTOu7mXTwWCzNLYRqSzClOnGVxSDbg83vB53BHozgTBSnhfHyyy8vyucKsacKWQjeAGqVUtXAOuBrwNe33kApVQH8Hfim1npZAbOIQeiW6+fy4F8fY9YXD8JoyDLuwDEYba34yqJYc0mc8TAmUxqr1wumHLZABGdpRdGGjzjwwAOL8rlC7KmCFQKtdUYp9UPgacAM/Elr/Z5S6vt9628FfgmEgJv7fnkzWuuphcokBo8H7nmEW6+fy4HTphBqtVA3rRZPZzueeAjDlsYZC2Ox5bC63SiLwh4M4yor7hhCr7zyCiAFQQw+SusdNtvvtaZOnaoXLVpU7BiigBY89SI//cHFTBhXx2hThJGTa4ipNK6wD7cbnLEQVgMshoHZsGIPhHCVVxd9ILkZM2YAsHDhwqLmEOJTfOoviAwxIfYqLz33Guf96FJqRlQwyhShalwVMVMGZ8CTLwKRIDanCZPNitlhxeYP7hVFQIjBTJ60EXuN5+e/zI/P+DnlpQkmOsupqC2j1AEOrwuPz4wR8mHzWlAWMxanDZs/gLtihBQBIfaQXBGIvcKzT77IuT+8hOqqcvZxVxCviFHuMWN3OfAFrdh9LuweKyaTCYvLni8ClTVSBIToB1IIRNE9/fhznP+jy6gZUcEkVznx8ijVfhtOjxNfyIrN7cDutWKy2/JXAj4/7sqRUgSE6CdSCERRPfHwfH7+k8upHVnNRKOE0so4lT4r7qAHr9+05UrA6nZhtluwen24q/bOInD99dcXO4IQu0UKgSiaB+55hMt/cT2jaqqYYC+hoqaUMqcJT8iLx2vK3xNwmbH6vJis+ZFEPVUj99pB5CZPnlzsCELsFikEYsDlcjmuu+I27pxzH2NqRjDeKKWqrpxSO3iiPtxeE85oAKuhsAcCYMphdXvwVNfu1dNMPvvss4BMUCMGHykEYkD19PTy85/8hvnzXmBK3RhqHRFGT6wmSDpfBDwmnPEgFjvY/N6+IuDd64sAwGWXXQZIIRCDjxQCMWBamlv58Xcv5O233uegMROpNEKMn1SFO5vCXxbCYc/hSoQwWzVWrxuTxYQjmsCIleyV9wSEGCqkEIgBsXrlWs487VwaNjYyo24yVaEYY2vjOHSGYFUIm1XjSoQxWzQWtwuL04G7YgRWt7fY0YUY8qQQiIJ76bnXuPAnvyGXzXFo9URqKiuoSbhxWMAf8+BwmnEEXZitGovbiT2Qf0ZAJpURYmBIIRAFk05n+P1vb2fubfeRiEbYNzSCUaOrKHdbMAwrvpAdI+jG7rVhtluxuA2cJeUYsYQ0BQkxgKQQiIJYt3YD5/3vr3n7rfeZUjeakfYINaNKiRsmXEEnHr8VI+zF5rZiMexYvG48lTWDuinotttuK3YEIXaLFALR7557+iV+cc6VZDMZDhs7hRJngBGVQSIeK96IE6fPjhF0YXPZsbgc2ALBvqagwf3jWFdXV+wIQuyWwf2bJ/Yqqd4Uv7viVv7654coi8fZJ1BJWSJKZcggEPHi8Vtx+A2MgIHFZWBx2jESZRjRodEU9NhjjwFw7LHHFjmJELtGCoHoF++/s5Rf/uwqli35kEnVtdR5EoyoiRF3WfHFvbjcZux+J3a/gdVlYHY58VTVYHV5ih2931x77bWAFAIx+EghEHuku7uHW677M3fd/gBup8EhI8YzMlFKddxDwG/gCzlwBuzYfQZ2f75bqNXnw10x+JuChBgq5DdR7BatNS899xpX/epG1n60ntGJCsaFy6kuC1MSdOINOvGEHBhBNza3DavHicluxRkvwxGND4mmICGGCikEYpe9s3gJ111xK4teW0zQ6+PQmkmMKiulIuIiFHLjCdlxRfp6BDlsWFwOzIaBu7IGq8td7PhCiG1IIRA7bc3qem68+naeeWIhLqfB1PJR1EVLqYj5KIm58QQcuMMuHH4Di9OKxeHA6vXgiERxhKIos7nYhyCE2AEpBOJzNTdt5rYb7uRvf30UszIxsaSaukgZ5TE/JSEXvlC+GcgZdGHz5nsD2YIBnLESrF7/sGkGuvvuu4sdQYjdIoVAfKpkV5K7//g3/nzrvfT09DIyUsLYcDkVsSBlETf+kBOX344zYOAIurB5XdgjEZzxUiyGs9jxB1x5eXmxIwixW6QQiO1sXN/AX+c+xN/+8ihdnUnKAxEmVkygIhqkIuYlFHHhDjgwAgZ2n4Ej6MUoLcWIxIf1+ED3338/ACeffHKRkwixa6QQCCDfC+jtt97nnjv+xvx5L6C1ptwX5gt1Y6griZGIuIlEXbiDBkbQmX8mIODDVV6BPRjea2cNG0i33HILIIVADD5SCIa5xk3NPPHwfP5x3xOsWrkGm8XCqFAJYxMV1CRCxEMu/GEn3qgLI+DE6nPiCAZwVVRjdXuGTfu/EEOZFIJhKNmV5KXnX+Ph++fx6j/fJJfLEXZ62a9sFGNLSimP+ihJeHH57HiiLhx+J1a3gRGL4aqoxmyzFfsQhBD9SArBMNHU0MzCBa/w7BMv8Marb5HOZDCsdkaHyxgTL6UmESEedRGMuDB8+fZ/q8fA6rRjlJbhLi1HmaT7pxBDkRSCIaq3p5d3Fi/htRcXsfCZf7J8+So04LI5qPbHqA5FGVteQiLiJhR14Q45MfxGfjRQr4HV68OVKB1W3T+FGK6kEAwRya4k//n3e/zrpTd57aVFLF26kmw2C0DQ8DAhUU1dPEF1LEQk5CIYdeGJuHFHPdi8TqwuFxanE5PDgT0QHpbdP/fUgw8+WOwIQuwWKQSDjNaahk1NfPDuMhb/6x3eW/wBH65YTWPLZgAUEHB6qA0mqAyFGVUSoyweIBJ34Q65cUU8OMIBjFgUm8+PxeFAWaxy1t8PwuFwsSMIsVukEOyFtNa0bm5jzep6Pnx/FSve/5DVy1axft0mNjQ1051KbdnWZXMQMNxMLq2mIhhiZEmUaMhDNO7BE3HjjHhxJhK4KyuwuV3yB7+A5s6dC8Dpp59e1BxC7KqCFgKl1GzgBsAM/FFrfeU261Xf+qOBJHC61vrfhcxUTJlMhtbmVho2bGLjRxvYsGYDm+obaNjYSHNjK5vb2mlp66At2UW6r1nnYw6LFZfNoNQTIubzUR4OUl0SIR7x4QsaGG47dp8TR9CHEY3gLi3B4nLKH/4BJIVADFYFKwRKKTNwEzALqAfeUEo9qrV+f6vNjgJq+76mA7f0/XdA5XI5cuksmVSKbCpDNpUmm8qQSafJptJk0hmy6eyW5R+trOejj+rp6uwi2dVNV0eSrmQ33d29dPf20tObpjed/0plMqQyGXrSaXozKfQOPl8BdosVh8WGx2EQj/oJud2EfR7iIT+liQDhqB9vxIsnHsQZDuHwe7C6XZgdDkxWadoRQuy+Ql4RTANWaK1XAiil7gOOA7YuBMcBd2mtNfCaUsqvlEporTf0d5h3n/oXz929gFw2Ry6n819ao/v+vSve3rCK9zZ9tN1yszJhMZuxmi1YzWZsZgt2ixWPw8CwWXE7DNyGHY/Tic9l4Pe4CPrdRKI+fCEvht+NEfLjikexez1Y3QZm6/AdskEIMTAKWQhKgbVbva5n+7P9HW1TCnyiECilzgDOAKioqNitMK6Am2hpALPZhGnLlxmzZev/mjBbLJgsJkwWM2azGbPVjNlixmSxYLaYMVst7LumnsYNDbicTtweJ/6QH18shCfixxnwYrHbMZlNKIsFsyX/X5PJBEqBQs7ehRB7lUIWgh39tdv21HtntkFrPQeYAzB16tRdO33vUz19HNXTx+3OW7czoV/2IoQQe4dCFoJ6YOtxecuA9buxjRCDwrx584odQYjdUsghI98AapVS1UopG/A14NFttnkU+JbK2x9oK8T9ASEGgtPpxOmUB/HE4FOwKwKtdUYp9UPgafLdR/+ktX5PKfX9vvW3AvPIdx1dQb776H8XKo8QhXbzzTcDcOaZZxY5iRC7RuU77AweU6dO1YsWLSp2DCG2M2PGDAAWLlxY1BxCfIpP7aUis4kIIcQwJ4VACCGGOSkEQggxzEkhEEKIYW7Q3SxWSjUC24/vMDDCQFORPrtQ5JgGh6F2TEPteGDvP6YmrfXsHa0YdIWgmJRSi7TWU4udoz/JMQ0OQ+2YhtrxwOA+JmkaEkKIYU4KgRBCDHNSCHbNnGIHKAA5psFhqB3TUDseGMTHJPcIhBBimJMrAiGEGOakEAghxDAnhWAXKaV+q5T6QCn1tlLqH0opf7Ez7Sml1FeVUu8ppXJKqUHZ/Q1AKTVbKbVUKbVCKXV+sfP0B6XUn5RSDUqpd4udpT8opcqVUs8rpZb0/cz9uNiZ9pRSyqGUel0p9Z++Y/pVsTPtKikEu24+MF5rPRFYBlxQ5Dz94V3geODFYgfZXUopM3ATcBQwFjhFKTW2uKn6xVxghw8BDVIZ4Kda6zHA/sD/DIH/T73AYVrrScBkYHbf/CqDhhSCXaS1fkZrnel7+Rr5WdUGNa31Eq310mLn2EPTgBVa65Va6xRwH3BckTPtMa31i0BLsXP0F631Bq31v/v+3QEsIT9P+aCl8zr7Xlr7vgZVLxwpBHvm28CTxQ4hgPwfk7Vbva5nkP+BGeqUUlXAPsC/ihxljymlzEqpxUADMF9rPaiOqZBzFg9aSqlngfgOVv1ca/1I3zY/J3+Z+5eBzLa7duaYBrkdTboxqM7KhhOllBt4CDhLa91e7Dx7SmudBSb33TP8h1JqvNZ60NzXkUKwA1rrwz9rvVLqNOAYYKYeJA9ifN4xDQH1QPlWr8uA9UXKIj6DUspKvgj8RWv992Ln6U9a61al1ELy93UGTSGQpqFdpJSaDZwHfElrnSx2HrHFG0CtUqpaKWUDvgY8WuRMYhtKKQXcASzRWv+u2Hn6g1Iq8nHvQaWUARwOfFDUULtICsGu+wPgAeYrpRYrpW4tdqA9pZT6ilKqHjgAeEIp9XSxM+2qvhv4PwSeJn8D8gGt9XvFTbXnlFL3Aq8CdUqpeqXUd4qdaQ8dBHwTOKzv92exUuroYofaQwngeaXU2+RPSOZrrR8vcqZdIkNMCCHEMCdXBEIIMcxJIRBCiGFOCoEQQgxzUgiEEGKYk0IghBDDnBQCMegppTo/f6vPfP+DSqkRn7Luj581KJpSauGujNiqlLIqpd7cjYyj+7pavqWUqvmM7Z5VSgV2df9ieJNCIIY1pdQ4wKy1XrmDdWat9Xe11u/340d+AXhlN973ZeARrfU+WusPP2O7u4EzdyeYGL6kEIghQ+X9Vin1rlLqHaXUyX3LTUqpm/vGin9cKTVPKXVi39u+ATyy1T46lVK/Vkr9Czjg4zP+vkHF5m61759s89kmpdSdSqnLlFIvKaUmb7XuZaXUxL6Xs4EnlVIupdQTfWPYv7tV1l8qpd7oWzan75iOBs4CvquUer5vu1P7xsBfrJS6rW8Ybsg/TX1KP39rxRAnhUAMJceTHw9+EvnH/H+rlEr0La8CJgDfJf8E9ccOArZuqnEB72qtp2ut/7nV8slAqdZ6vNZ6AvDnrdZZyA8+uExrfRHwR+B0AKXUKMCutX67b9tDgYXkC8J6rfUkrfV44Km+9X/QWu/Xt8wAjtFazwNuBa7TWh+qlBoDnAwcpLWeDGTJFzS01psBu1IqtAvfNzHMSSEQQ8kXgHu11lmt9SbgBWC/vuV/01rntNYbgee3ek8CaNzqdZb8gGjbWgmMUEr9vm+8qa1HzLyNfPH4Td/rvwHH9A2u9m3yk8uglCoBWvrGqHoHOFwpdZVS6mCtdVvfew9VSv1LKfUOcBgwbgdZZgL7Am/0DX08E9j6HkcDULLjb5EQ25NCIIaSHQ1F/VnLAboBx1ave/qGFP6EvjPtSeTP5v+H/Fn/x14h/wfc0bdtkvxMdscBJwF/7dvuKPJjIaG1Xkb+j/k7wBV9TUIO4GbgxL6rjtu3ybb18dyptZ7c91Wntb5kq/WOvuMSYqdIIRBDyYvAyX3t+RHgEOB14J/ACX3t+DFgxlbvWQKM/LwdK6XCgElr/RDwC2DKVqvvAOYBf1NKfTy0+x+BG4E3tNYfzzA2m76JjPquDpJa63uAa/r29/Ef/aa+8fo/vo+xrQXAiUqpaN++gkqpyr5/K/LzTqz+vGMS4mMyH4EYSv5Bvv3/P+QnpTlXa71RKfUQ+eaTd8nPM/0v4OOmmCfIF4ZnP2ffpcCflVIfnzx9Yq5qrfXvlFI+4G6l1De01m8qpdrpu5fQdzO3Vmv98fDEE8jfw8gBaeAHfWPZ307+KmE1+ZEst6O1fl8pdRHwTF+eNPmrlI/IX2W8ttV0qkJ8Lhl9VAwLSim31rqz7ybq6+RvtG5U+fHjn+97vV2T0B58Xgn5ZqTRWuucUuoLwKla6+/312d8yufeADyqtV5QyM8RQ4tcEYjh4nGVnzzEBlzad9MYrXW3Uupi8mf8a/rjg5RS3wJ+A5yttc71fc4/yTdRFdq7UgTErpIrAiGEGObkZrEQQgxzUgiEEGKYk0IghBDDnBQCIYQY5qQQCCHEMPf/ASnCaZHItttNAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df = []\n",
    "for base_number in [5, 7, 10, 14, 20, 28]:\n",
    "    risky,  p = get_psychm_curve1(risky_ll_sd=.5, prior_sd=1.5, base_ll_sd=.75, base_number=base_number,\n",
    "                                 prior_center=14.)\n",
    "    d = pd.DataFrame({'safe':base_number, 'risky':risky, 'p':p})\n",
    "    df.append(d)\n",
    "    \n",
    "df = pd.concat(df, ignore_index=True)\n",
    "\n",
    "df['log(risky/safe)'] = np.log(df['risky'] / df['safe'])\n",
    "\n",
    "plt.axvline(np.log(1/.55), c='k', ls='--')\n",
    "plt.axhline(.5, c='k', ls='--')\n",
    "\n",
    "\n",
    "sns.lineplot('log(risky/safe)', 'p', 'safe', data=df)\n",
    "\n",
    "sns.despine()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "8a3b7330",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/gdehol/miniconda3/envs/pymc3/lib/python3.9/site-packages/seaborn/_decorators.py:36: FutureWarning: Pass the following variables as keyword args: x, y, hue. From version 0.12, the only valid positional argument will be `data`, and passing other arguments without an explicit keyword will result in an error or misinterpretation.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='log(risky/safe)', ylabel='p'>"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df = []\n",
    "for base_number in [5, 7, 10, 14, 20, 28]:\n",
    "    risky,  p = get_psychm_curve1(risky_ll_sd=.8, prior_sd=2., base_ll_sd=.5, base_number=base_number)\n",
    "    d = pd.DataFrame({'safe':base_number, 'risky':risky, 'p':p})\n",
    "    df.append(d)\n",
    "    \n",
    "df = pd.concat(df, ignore_index=True)\n",
    "\n",
    "df['log(risky/safe)'] = np.log(df['risky'] / df['safe'])\n",
    "\n",
    "sns.lineplot('log(risky/safe)', 'p', 'safe', data=df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "d4b48b27",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "([], [])"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "x_ = [5., 7., 10., 14., 20., 28.]\n",
    "\n",
    "mu_prior = 20.\n",
    "mu_safe = 14.\n",
    "sd_safe = 1.\n",
    "\n",
    "x = np.linspace(np.log(5), np.log(28))\n",
    "\n",
    "\n",
    "evidence_safe = ss.norm(np.log(14.), .5)\n",
    "prior_safe = ss.norm(np.log(20.), .5)\n",
    "\n",
    "plt.plot(x, evidence_safe.pdf(x), c='k', ls='dotted', label='likelihood')\n",
    "# plt.axvline(np.log(mu_safe), c='k', label='prior')\n",
    "\n",
    "plt.plot(x, prior_safe.pdf(x), c='k', ls='--')\n",
    "\n",
    "post = get_posterior(np.log(mu_prior), .5, np.log(mu_safe), .5)\n",
    "posterior_safe = ss.norm(*post)\n",
    "\n",
    "plt.plot(x, posterior_safe.pdf(x), c='k', ls='-')\n",
    "\n",
    "plt.xticks(np.log(x_), x_)\n",
    "\n",
    "sns.despine(left=True)\n",
    "plt.yticks([])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "4c13e12d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(2.04717228111105, 0.3535533905932738)\n",
      "(2.2154083994216567, 0.3535533905932738)\n",
      "(2.3937458713910233, 0.3535533905932738)\n",
      "(2.5619819897016294, 0.3535533905932738)\n",
      "(2.7403194616709956, 0.3535533905932738)\n",
      "(2.908555579981602, 0.3535533905932738)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7fd54edb3730>"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x216 with 6 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "x_ = [5., 7., 10., 14., 20., 28.]\n",
    "\n",
    "mu_prior = 12.\n",
    "sd_risky = .5\n",
    "\n",
    "plt.gcf().set_size_inches(20, 3)\n",
    "\n",
    "df = []\n",
    "\n",
    "for i, x__ in enumerate(x_):\n",
    "    \n",
    "    plt.subplot(1,6,i+1)\n",
    "\n",
    "\n",
    "    x = np.linspace(np.log(5), np.log(28))\n",
    "\n",
    "    evidence_risky = ss.norm(np.log(x__), sd_risky)\n",
    "    prior_safe = ss.norm(np.log(mu_prior), .5)\n",
    "\n",
    "    plt.plot(x, evidence_risky.pdf(x), c='k', ls='dotted', label='likelihood')\n",
    "\n",
    "    plt.plot(x, prior_safe.pdf(x), c='k', ls='--', label='prior')\n",
    "\n",
    "    post = get_posterior(np.log(mu_prior), .5, np.log(x__), sd_risky)\n",
    "    posterior_safe = ss.norm(*post)\n",
    "    \n",
    "    print(post)\n",
    "\n",
    "    plt.plot(x, posterior_safe.pdf(x), c='k', ls='-', label='posterior')\n",
    "\n",
    "    plt.xticks(np.log(x_), x_)\n",
    "\n",
    "    sns.despine(left=True)\n",
    "    plt.yticks([])\n",
    "    \n",
    "    df.append({'x':x__, 'posterior x':np.exp(post[0]), 'likelihood':'precise'})\n",
    "    \n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "3cb58c55",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(2.3098129023172205, 0.4472135954999579)\n",
      "(2.377107349641463, 0.4472135954999579)\n",
      "(2.4484423384292096, 0.4472135954999579)\n",
      "(2.515736785753452, 0.4472135954999579)\n",
      "(2.5870717745411986, 0.4472135954999579)\n",
      "(2.654366221865441, 0.4472135954999579)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7fd54fd34250>"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x216 with 6 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "x_ = [5., 7., 10., 14., 20., 28.]\n",
    "\n",
    "mu_prior = 12.\n",
    "sd_risky = 1.\n",
    "\n",
    "plt.gcf().set_size_inches(20, 3)\n",
    "\n",
    "for i, x__ in enumerate(x_):\n",
    "    \n",
    "    plt.subplot(1,6,i+1)\n",
    "\n",
    "\n",
    "    x = np.linspace(np.log(5), np.log(28))\n",
    "\n",
    "    evidence_risky = ss.norm(np.log(x__), sd_risky)\n",
    "    prior_safe = ss.norm(np.log(mu_prior), .5)\n",
    "\n",
    "    plt.plot(x, evidence_risky.pdf(x), c='k', ls='dotted', label='likelihood')\n",
    "\n",
    "    plt.plot(x, prior_safe.pdf(x), c='k', ls='--', label='prior')\n",
    "\n",
    "    post = get_posterior(np.log(mu_prior), .5, np.log(x__), sd_risky)\n",
    "    print(post)\n",
    "    posterior_safe = ss.norm(*post)\n",
    "\n",
    "    plt.plot(x, posterior_safe.pdf(x), c='k', ls='-', label='posterior')\n",
    "\n",
    "    plt.xticks(np.log(x_), x_)\n",
    "\n",
    "    sns.despine(left=True)\n",
    "    plt.yticks([])\n",
    "    \n",
    "    df.append({'x':x__, 'posterior x':np.exp(post[0]), 'likelihood':'noisy'})\n",
    "    \n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "a2142fe8",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/gdehol/miniconda3/envs/pymc3/lib/python3.9/site-packages/seaborn/_decorators.py:36: FutureWarning: Pass the following variables as keyword args: x, y, hue. From version 0.12, the only valid positional argument will be `data`, and passing other arguments without an explicit keyword will result in an error or misinterpretation.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='x', ylabel='posterior x'>"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df = pd.DataFrame(df)\n",
    "\n",
    "sns.lineplot('x', 'posterior x', 'likelihood', data=df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "0b30406d",
   "metadata": {},
   "outputs": [],
   "source": [
    "def likelihood(prior_mu, \n",
    "               prior_sd, \n",
    "               risky_mu, \n",
    "               risky_sd, \n",
    "               safe_mu, \n",
    "               safe_sd, \n",
    "               choices,\n",
    "               p_risky=.55):\n",
    "    \n",
    "    post_risky = get_posterior(prior_mu, prior_sd, risky_mu, risky_sd)\n",
    "    post_safe = get_posterior(prior_mu, prior_sd, safe_mu, safe_sd)\n",
    "    \n",
    "    diff_dist = get_diff_dist(*post_safe, *post_risky)\n",
    "    \n",
    "    p = get_p(diff_dist, thr=-np.log(0.55))\n",
    "    \n",
    "    ll = np.sum(np.log(ss.bernoulli(p).pmf(choices)))\n",
    "    \n",
    "    return ll"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "f77d018a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-1.4382473059127066"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "prior_mu = np.array([2., 5.])\n",
    "prior_sd = np.array([2., 5.])\n",
    "risky_mu = np.array([1., 2.])\n",
    "risky_sd = np.array([1., 2.])\n",
    "safe_mu = np.array([1., .5])\n",
    "safe_sd = np.array([1., .5])\n",
    "\n",
    "choices = np.array([1., 1.])\n",
    "\n",
    "likelihood(prior_mu, prior_sd, risky_mu, risky_sd, safe_mu, safe_sd, choices)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "c5a87511",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b31e60dbe5564e5795a8408efa0c03a3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/62 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[Errno 2] No such file or directory: '/data/ds-risk/sourcedata/sub-24/behavior/ses-3t2/sub-24_ses-3t2_task-task_run-1_events.tsv'\n",
      "[Errno 2] No such file or directory: '/data/ds-risk/sourcedata/sub-24/behavior/ses-7t2/sub-24_ses-7t2_task-task_run-1_events.tsv'\n"
     ]
    }
   ],
   "source": [
    "from tqdm.contrib.itertools import product\n",
    "df = []\n",
    "\n",
    "for subject, session in product(range(2, 33), ['3t2', '7t2']):\n",
    "        \n",
    "    try:\n",
    "        for run in range(1, 9):\n",
    "            d = pd.read_table(f'/data/ds-risk/sourcedata/sub-{subject:02d}/behavior/ses-{session}/sub-{subject:02d}_ses-{session}_task-task_run-{run}_events.tsv')\n",
    "            d = d[np.in1d(d.phase, [8,9])]\n",
    "            d = d.pivot_table(index=['trial_nr'], values=['choice', 'certainty', 'n1', 'n2', 'prob1', 'prob2'])\n",
    "            d['task'] = 'task'\n",
    "            d['subject'], d['session'], d['scanner'], d['run'] = subject, session, session[:2], run\n",
    "            df.append(d)    \n",
    "    except Exception as e:\n",
    "        print(e)\n",
    "\n",
    "    \n",
    "df = pd.concat(df)\n",
    "\n",
    "df['subject'] = pd.Categorical(df['subject'].apply(lambda x: f'{x:02d}'))\n",
    "\n",
    "df['log(risky/safe)'] = np.log(df['n1'] / df['n2'])\n",
    "ix = df.prob1 == 1.0\n",
    "\n",
    "df.loc[~ix, 'log(risky/safe)'] = np.log(df.loc[~ix, 'n1'] / df.loc[~ix, 'n2'])\n",
    "df.loc[ix, 'log(risky/safe)'] = np.log(df.loc[ix, 'n2'] / df.loc[ix, 'n1'])\n",
    "\n",
    "df['risky/safe'] = np.exp(df['log(risky/safe)'])\n",
    "\n",
    "df.loc[~ix, 'chose risky'] = df.loc[~ix, 'choice'] == 1\n",
    "df.loc[ix, 'chose risky'] = df.loc[ix, 'choice'] == 2\n",
    "df['chose risky'] = df['chose risky'].astype(bool)\n",
    "df = df.reset_index().set_index(['task', 'subject', 'session', 'run', 'trial_nr'])\n",
    "\n",
    "\n",
    "ix_ = pd.IndexSlice\n",
    "\n",
    "n_bins = 6\n",
    "\n",
    "df_ = df.copy()\n",
    "\n",
    "\n",
    "df = df[~df.choice.isnull()]\n",
    "\n",
    "df['risky_first'] = df['prob1'] == 0.55\n",
    "df.loc[df.risky_first, 'base_number'] = df['n2']\n",
    "df.loc[~df.risky_first, 'base_number'] = df['n1']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 151,
   "id": "622976c3",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Two shit subjects\n",
    "\n",
    "df = df.drop(['03', '32'], level='subject')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 152,
   "id": "0d5675c9",
   "metadata": {},
   "outputs": [],
   "source": [
    "from scipy.optimize import minimize"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 153,
   "id": "ae4f19d9",
   "metadata": {},
   "outputs": [],
   "source": [
    "def likelihood_(pars, risky_mu, safe_mu, choices, risky_first):\n",
    "    prior_mu, prior_sd, \\\n",
    "    risky_sd_first, risky_sd_later,  \\\n",
    "    safe_sd_first, safe_sd_later = pars\n",
    "\n",
    "    risky_sd = np.zeros(len(choices))\n",
    "    safe_sd = np.zeros(len(choices))\n",
    "\n",
    "    risky_sd[risky_first] = risky_sd_first\n",
    "    risky_sd[~risky_first] = risky_sd_later\n",
    "\n",
    "    safe_sd[~risky_first] = safe_sd_first\n",
    "    safe_sd[risky_first] = safe_sd_later    \n",
    "\n",
    "    return -likelihood(prior_mu, prior_sd, risky_mu, risky_sd, safe_mu, safe_sd, choices)  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "id": "ff602c4e",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_pars(df):\n",
    "#     print(df)\n",
    "    risky_mu = np.log(np.where(df['prob1'] == 1.0, df['n2'], df['n1']))\n",
    "    safe_mu = np.log(np.where(df['prob1'] == 1.0, df['n1'], df['n2']))\n",
    "    choices = df['chose risky']\n",
    "    risky_first = df['risky_first'].values\n",
    "    \n",
    "    x0 = np.concatenate(([np.log(14.)], np.ones(5)*.5))\n",
    "    bounds = [(np.log(5.), np.log(100.))]+ [(.01, None)] * 5  \n",
    "    \n",
    "    res = minimize(likelihood_, x0, bounds=bounds, args=(risky_mu, safe_mu, choices, risky_first))\n",
    "    \n",
    "    pars = pd.Series(res.x, \n",
    "                     index=['prior_mu', 'prior_sd', 'risky_sd(first)', 'risky_sd (later)', 'safe sd(first)', 'safe sd (later)'])\n",
    "\n",
    "    return pars\n",
    "    \n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 155,
   "id": "6f6c52e9",
   "metadata": {},
   "outputs": [],
   "source": [
    "def likelihood(risky_prior_mu, \n",
    "               risky_prior_sd, \n",
    "               safe_prior_mu,\n",
    "               safe_prior_sd,\n",
    "               risky_mu, \n",
    "               risky_sd, \n",
    "               safe_mu, \n",
    "               safe_sd, \n",
    "               choices,\n",
    "               p_risky=.55):\n",
    "    \n",
    "    post_risky = get_posterior(risky_prior_mu, risky_prior_sd, risky_mu, risky_sd)\n",
    "    post_safe = get_posterior(safe_prior_mu, safe_prior_sd, safe_mu, safe_sd)\n",
    "    \n",
    "    diff_dist = get_diff_dist(*post_safe, *post_risky)\n",
    "    \n",
    "    p = get_p(diff_dist, thr=-np.log(0.55))\n",
    "    \n",
    "    ll = np.sum(np.log(np.clip(ss.bernoulli(p).pmf(choices), 1e-12, None)))\n",
    "    \n",
    "    return ll"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 156,
   "id": "6989e688",
   "metadata": {},
   "outputs": [],
   "source": [
    "def likelihood_(pars, risky_mu, safe_mu, choices, risky_first):\n",
    "    risky_prior_mu, \\\n",
    "    safe_prior_mu, \\\n",
    "    risky_prior_sd, \\\n",
    "    safe_prior_sd, \\\n",
    "    evidence_sd_first, evidence_sd_later = pars\n",
    "\n",
    "    risky_sd = np.zeros(len(choices))\n",
    "    safe_sd = np.zeros(len(choices))\n",
    "\n",
    "    risky_sd[risky_first] = evidence_sd_first\n",
    "    risky_sd[~risky_first] = evidence_sd_later\n",
    "\n",
    "    safe_sd[~risky_first] = evidence_sd_first\n",
    "    safe_sd[risky_first] = evidence_sd_later    \n",
    "\n",
    "    return -likelihood(risky_prior_mu, risky_prior_sd, safe_prior_mu, safe_prior_sd, risky_mu, risky_sd, safe_mu, safe_sd, choices)  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 157,
   "id": "a903ba3e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "parameter\n",
       "risky_prior_mu         2.159452\n",
       "safe_prior_mu          1.702422\n",
       "risky_prior_sd         0.698669\n",
       "safe_prior_sd          1.273685\n",
       "evidence sd (first)    0.441199\n",
       "evidence sd (later)    0.449033\n",
       "dtype: float64"
      ]
     },
     "execution_count": 157,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pars.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 158,
   "id": "bf44c0e8",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_pars(df):\n",
    "    risky_mu = np.log(np.where(df['prob1'] == 1.0, df['n2'], df['n1']))\n",
    "    safe_mu = np.log(np.where(df['prob1'] == 1.0, df['n1'], df['n2']))\n",
    "    choices = df['chose risky']\n",
    "    risky_first = df['risky_first'].values\n",
    "    x0 = [2.34, 2.11, .61, 1.93, .72, .29]  \n",
    "#     x0 = np.concatenate(([np.log(28.), np.log(28.)], np.ones(4)*.5))\n",
    "    bounds = [(0., np.log(28.*4+1)), (0., np.log(28.*4+1))] + [(.01, 10.)] * 4\n",
    "    \n",
    "    print(likelihood_(x0, risky_mu, safe_mu, choices, risky_first))\n",
    "    \n",
    "    res = minimize(likelihood_, x0, bounds=bounds, args=(risky_mu, safe_mu, choices, risky_first), method='L-BFGS-B')\n",
    "    \n",
    "    pars = pd.Series(res.x, \n",
    "                     index=['risky_prior_mu', \n",
    "                            'safe_prior_mu',\n",
    "                            'risky_prior_sd',\n",
    "                            'safe_prior_sd', \n",
    "                            'evidence sd (first)', 'evidence sd (later)'])\n",
    "\n",
    "    return pars\n",
    "    \n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 159,
   "id": "c8613eaf",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "284.4059764137662\n",
      "212.7954144010117\n",
      "244.35108932848001\n",
      "250.94851774168876\n",
      "241.092333959489\n",
      "233.31918373028043\n",
      "293.981762650174\n",
      "223.11327122547675\n",
      "224.76480095517923\n",
      "267.7799200858706\n",
      "234.67794440382286\n",
      "324.41102643531605\n",
      "282.3373742952045\n",
      "234.10139572835126\n",
      "413.20175561098534\n",
      "279.81888282343283\n",
      "260.1140062000761\n",
      "211.70167376459713\n",
      "232.88368408810604\n",
      "238.95450814542085\n",
      "338.1005719118532\n",
      "323.26852047808825\n",
      "299.3734379429487\n",
      "293.9842208045899\n",
      "287.2154892830255\n",
      "289.7133837135036\n",
      "238.24698489536976\n",
      "222.486552837834\n"
     ]
    }
   ],
   "source": [
    "pars = df.groupby('subject').apply(get_pars)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 160,
   "id": "5f3c5267",
   "metadata": {},
   "outputs": [],
   "source": [
    "pars.columns.name = 'parameter'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 161,
   "id": "679c879c",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/gdehol/miniconda3/envs/pymc3/lib/python3.9/site-packages/seaborn/_decorators.py:36: FutureWarning: Pass the following variables as keyword args: x, y. From version 0.12, the only valid positional argument will be `data`, and passing other arguments without an explicit keyword will result in an error or misinterpretation.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='parameter', ylabel='value'>"
      ]
     },
     "execution_count": 161,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.swarmplot('parameter', 'value', data=pars[['risky_prior_mu', 'safe_prior_mu']].stack().to_frame('value').reset_index())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 162,
   "id": "5c62337d",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/gdehol/miniconda3/envs/pymc3/lib/python3.9/site-packages/seaborn/_decorators.py:36: FutureWarning: Pass the following variables as keyword args: x, y. From version 0.12, the only valid positional argument will be `data`, and passing other arguments without an explicit keyword will result in an error or misinterpretation.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='parameter', ylabel='value'>"
      ]
     },
     "execution_count": 162,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.swarmplot('parameter', 'value', data=pars[['risky_prior_sd', 'safe_prior_sd']].stack().to_frame('value').reset_index())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 163,
   "id": "50d6d986",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/gdehol/miniconda3/envs/pymc3/lib/python3.9/site-packages/seaborn/_decorators.py:36: FutureWarning: Pass the following variables as keyword args: x, y. From version 0.12, the only valid positional argument will be `data`, and passing other arguments without an explicit keyword will result in an error or misinterpretation.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='parameter', ylabel='value'>"
      ]
     },
     "execution_count": 163,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.swarmplot('parameter', 'value', data=pars[['evidence sd (first)', 'evidence sd (later)']].stack().to_frame('value').reset_index())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 176,
   "id": "f61799c5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>parameter</th>\n",
       "      <th>risky_prior_mu</th>\n",
       "      <th>safe_prior_mu</th>\n",
       "      <th>risky_prior_sd</th>\n",
       "      <th>safe_prior_sd</th>\n",
       "      <th>evidence sd (first)</th>\n",
       "      <th>evidence sd (later)</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>subject</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>02</th>\n",
       "      <td>0.614323</td>\n",
       "      <td>6.590708e-08</td>\n",
       "      <td>0.085321</td>\n",
       "      <td>0.084811</td>\n",
       "      <td>0.144369</td>\n",
       "      <td>0.140945</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>04</th>\n",
       "      <td>0.656371</td>\n",
       "      <td>3.890125e-04</td>\n",
       "      <td>0.012559</td>\n",
       "      <td>0.015792</td>\n",
       "      <td>0.058559</td>\n",
       "      <td>0.057315</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>05</th>\n",
       "      <td>2.463256</td>\n",
       "      <td>3.245565e+00</td>\n",
       "      <td>0.546210</td>\n",
       "      <td>0.776162</td>\n",
       "      <td>0.558971</td>\n",
       "      <td>0.343371</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>06</th>\n",
       "      <td>2.188224</td>\n",
       "      <td>2.686930e+00</td>\n",
       "      <td>0.896004</td>\n",
       "      <td>1.669682</td>\n",
       "      <td>0.886346</td>\n",
       "      <td>0.409314</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>07</th>\n",
       "      <td>1.628902</td>\n",
       "      <td>2.227192e+00</td>\n",
       "      <td>0.010000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>1.943515</td>\n",
       "      <td>7.769716</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>08</th>\n",
       "      <td>2.467579</td>\n",
       "      <td>3.356345e+00</td>\n",
       "      <td>0.556164</td>\n",
       "      <td>1.120524</td>\n",
       "      <td>0.361737</td>\n",
       "      <td>0.275780</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>09</th>\n",
       "      <td>1.959078</td>\n",
       "      <td>1.603803e+00</td>\n",
       "      <td>1.519479</td>\n",
       "      <td>1.240802</td>\n",
       "      <td>0.353838</td>\n",
       "      <td>0.010043</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>2.452848</td>\n",
       "      <td>1.707468e-06</td>\n",
       "      <td>0.487244</td>\n",
       "      <td>1.627501</td>\n",
       "      <td>0.325344</td>\n",
       "      <td>0.174115</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>2.712099</td>\n",
       "      <td>4.349100e+00</td>\n",
       "      <td>0.276700</td>\n",
       "      <td>0.845281</td>\n",
       "      <td>0.282421</td>\n",
       "      <td>0.191240</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>1.438618</td>\n",
       "      <td>1.262532e+00</td>\n",
       "      <td>2.194407</td>\n",
       "      <td>1.007406</td>\n",
       "      <td>0.447483</td>\n",
       "      <td>0.013820</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>0.872849</td>\n",
       "      <td>1.095048e-03</td>\n",
       "      <td>0.054819</td>\n",
       "      <td>0.073393</td>\n",
       "      <td>0.115816</td>\n",
       "      <td>0.113679</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>4.158962</td>\n",
       "      <td>2.530221e-01</td>\n",
       "      <td>0.828583</td>\n",
       "      <td>5.674717</td>\n",
       "      <td>0.507179</td>\n",
       "      <td>0.259566</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>5.054286e-01</td>\n",
       "      <td>0.564738</td>\n",
       "      <td>0.537364</td>\n",
       "      <td>0.254850</td>\n",
       "      <td>0.272163</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>2.694181</td>\n",
       "      <td>3.241917e+00</td>\n",
       "      <td>0.603323</td>\n",
       "      <td>1.836281</td>\n",
       "      <td>0.715422</td>\n",
       "      <td>0.101879</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>2.950818</td>\n",
       "      <td>1.419511e+00</td>\n",
       "      <td>2.954196</td>\n",
       "      <td>0.340890</td>\n",
       "      <td>0.333159</td>\n",
       "      <td>0.255373</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>3.096833</td>\n",
       "      <td>7.484693e-01</td>\n",
       "      <td>0.777710</td>\n",
       "      <td>0.751475</td>\n",
       "      <td>0.389965</td>\n",
       "      <td>0.010099</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>2.918215</td>\n",
       "      <td>4.610642e-01</td>\n",
       "      <td>0.375331</td>\n",
       "      <td>1.267624</td>\n",
       "      <td>0.643245</td>\n",
       "      <td>0.385441</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>2.510348</td>\n",
       "      <td>3.615390e+00</td>\n",
       "      <td>0.314517</td>\n",
       "      <td>1.530625</td>\n",
       "      <td>0.341811</td>\n",
       "      <td>0.242941</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>0.619435</td>\n",
       "      <td>2.823532e-06</td>\n",
       "      <td>0.020845</td>\n",
       "      <td>0.026403</td>\n",
       "      <td>0.091509</td>\n",
       "      <td>0.088985</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>2.341648</td>\n",
       "      <td>2.894896e+00</td>\n",
       "      <td>0.649364</td>\n",
       "      <td>0.630642</td>\n",
       "      <td>0.378547</td>\n",
       "      <td>0.030655</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>4.727030</td>\n",
       "      <td>2.096046e+00</td>\n",
       "      <td>1.127142</td>\n",
       "      <td>0.432238</td>\n",
       "      <td>0.351611</td>\n",
       "      <td>0.176159</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>0.611512</td>\n",
       "      <td>3.864846e-07</td>\n",
       "      <td>0.014676</td>\n",
       "      <td>0.015631</td>\n",
       "      <td>0.068897</td>\n",
       "      <td>0.070481</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>3.484213</td>\n",
       "      <td>4.727388e+00</td>\n",
       "      <td>0.440334</td>\n",
       "      <td>1.152562</td>\n",
       "      <td>0.192274</td>\n",
       "      <td>0.234571</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>3.036075</td>\n",
       "      <td>1.517136e+00</td>\n",
       "      <td>1.169033</td>\n",
       "      <td>0.683681</td>\n",
       "      <td>0.430668</td>\n",
       "      <td>0.010000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>3.474404</td>\n",
       "      <td>2.348890e+00</td>\n",
       "      <td>0.875028</td>\n",
       "      <td>0.572336</td>\n",
       "      <td>0.189723</td>\n",
       "      <td>0.134137</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>3.630747</td>\n",
       "      <td>1.660103e+00</td>\n",
       "      <td>1.486692</td>\n",
       "      <td>1.241606</td>\n",
       "      <td>0.624690</td>\n",
       "      <td>0.010000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>2.226123</td>\n",
       "      <td>2.014686e+00</td>\n",
       "      <td>0.995737</td>\n",
       "      <td>1.126423</td>\n",
       "      <td>0.718496</td>\n",
       "      <td>0.010000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>2.848859</td>\n",
       "      <td>4.727388e+00</td>\n",
       "      <td>0.242299</td>\n",
       "      <td>0.910832</td>\n",
       "      <td>0.220902</td>\n",
       "      <td>0.152603</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "parameter  risky_prior_mu  safe_prior_mu  risky_prior_sd  safe_prior_sd  \\\n",
       "subject                                                                   \n",
       "02               0.614323   6.590708e-08        0.085321       0.084811   \n",
       "04               0.656371   3.890125e-04        0.012559       0.015792   \n",
       "05               2.463256   3.245565e+00        0.546210       0.776162   \n",
       "06               2.188224   2.686930e+00        0.896004       1.669682   \n",
       "07               1.628902   2.227192e+00        0.010000      10.000000   \n",
       "08               2.467579   3.356345e+00        0.556164       1.120524   \n",
       "09               1.959078   1.603803e+00        1.519479       1.240802   \n",
       "10               2.452848   1.707468e-06        0.487244       1.627501   \n",
       "11               2.712099   4.349100e+00        0.276700       0.845281   \n",
       "12               1.438618   1.262532e+00        2.194407       1.007406   \n",
       "13               0.872849   1.095048e-03        0.054819       0.073393   \n",
       "14               4.158962   2.530221e-01        0.828583       5.674717   \n",
       "15               0.000000   5.054286e-01        0.564738       0.537364   \n",
       "16               2.694181   3.241917e+00        0.603323       1.836281   \n",
       "17               2.950818   1.419511e+00        2.954196       0.340890   \n",
       "18               3.096833   7.484693e-01        0.777710       0.751475   \n",
       "19               2.918215   4.610642e-01        0.375331       1.267624   \n",
       "20               2.510348   3.615390e+00        0.314517       1.530625   \n",
       "21               0.619435   2.823532e-06        0.020845       0.026403   \n",
       "22               2.341648   2.894896e+00        0.649364       0.630642   \n",
       "23               4.727030   2.096046e+00        1.127142       0.432238   \n",
       "25               0.611512   3.864846e-07        0.014676       0.015631   \n",
       "26               3.484213   4.727388e+00        0.440334       1.152562   \n",
       "27               3.036075   1.517136e+00        1.169033       0.683681   \n",
       "28               3.474404   2.348890e+00        0.875028       0.572336   \n",
       "29               3.630747   1.660103e+00        1.486692       1.241606   \n",
       "30               2.226123   2.014686e+00        0.995737       1.126423   \n",
       "31               2.848859   4.727388e+00        0.242299       0.910832   \n",
       "\n",
       "parameter  evidence sd (first)  evidence sd (later)  \n",
       "subject                                              \n",
       "02                    0.144369             0.140945  \n",
       "04                    0.058559             0.057315  \n",
       "05                    0.558971             0.343371  \n",
       "06                    0.886346             0.409314  \n",
       "07                    1.943515             7.769716  \n",
       "08                    0.361737             0.275780  \n",
       "09                    0.353838             0.010043  \n",
       "10                    0.325344             0.174115  \n",
       "11                    0.282421             0.191240  \n",
       "12                    0.447483             0.013820  \n",
       "13                    0.115816             0.113679  \n",
       "14                    0.507179             0.259566  \n",
       "15                    0.254850             0.272163  \n",
       "16                    0.715422             0.101879  \n",
       "17                    0.333159             0.255373  \n",
       "18                    0.389965             0.010099  \n",
       "19                    0.643245             0.385441  \n",
       "20                    0.341811             0.242941  \n",
       "21                    0.091509             0.088985  \n",
       "22                    0.378547             0.030655  \n",
       "23                    0.351611             0.176159  \n",
       "25                    0.068897             0.070481  \n",
       "26                    0.192274             0.234571  \n",
       "27                    0.430668             0.010000  \n",
       "28                    0.189723             0.134137  \n",
       "29                    0.624690             0.010000  \n",
       "30                    0.718496             0.010000  \n",
       "31                    0.220902             0.152603  "
      ]
     },
     "execution_count": 176,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pars"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 164,
   "id": "6ee7d55d",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/gdehol/miniconda3/envs/pymc3/lib/python3.9/site-packages/seaborn/distributions.py:2619: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
      "  warnings.warn(msg, FutureWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:ylabel='Density'>"
      ]
     },
     "execution_count": 164,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "sns.distplot(pars['evidence sd (first)'] - pars['evidence sd (later)'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 165,
   "id": "c1b66058",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/gdehol/miniconda3/envs/pymc3/lib/python3.9/site-packages/seaborn/distributions.py:2619: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
      "  warnings.warn(msg, FutureWarning)\n",
      "/Users/gdehol/miniconda3/envs/pymc3/lib/python3.9/site-packages/seaborn/distributions.py:2619: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
      "  warnings.warn(msg, FutureWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='safe_prior_mu', ylabel='Density'>"
      ]
     },
     "execution_count": 165,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.distplot(np.exp(pars['risky_prior_mu']))\n",
    "sns.distplot(np.exp(pars['safe_prior_mu']))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 166,
   "id": "21f178b6",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/gdehol/miniconda3/envs/pymc3/lib/python3.9/site-packages/seaborn/distributions.py:2619: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
      "  warnings.warn(msg, FutureWarning)\n",
      "/Users/gdehol/miniconda3/envs/pymc3/lib/python3.9/site-packages/seaborn/distributions.py:2619: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
      "  warnings.warn(msg, FutureWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='safe_prior_sd', ylabel='Density'>"
      ]
     },
     "execution_count": 166,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.distplot(pars['risky_prior_sd'])\n",
    "sns.distplot(pars['safe_prior_sd'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 167,
   "id": "8d23eeb7",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pingouin"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 168,
   "id": "77b6fbd2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>T</th>\n",
       "      <th>dof</th>\n",
       "      <th>alternative</th>\n",
       "      <th>p-val</th>\n",
       "      <th>CI95%</th>\n",
       "      <th>cohen-d</th>\n",
       "      <th>BF10</th>\n",
       "      <th>power</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>T-test</th>\n",
       "      <td>1.79582</td>\n",
       "      <td>27</td>\n",
       "      <td>two-sided</td>\n",
       "      <td>0.083723</td>\n",
       "      <td>[-0.07, 1.06]</td>\n",
       "      <td>0.363431</td>\n",
       "      <td>0.822</td>\n",
       "      <td>0.458105</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              T  dof alternative     p-val          CI95%   cohen-d   BF10  \\\n",
       "T-test  1.79582   27   two-sided  0.083723  [-0.07, 1.06]  0.363431  0.822   \n",
       "\n",
       "           power  \n",
       "T-test  0.458105  "
      ]
     },
     "execution_count": 168,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pingouin.ttest(pars['risky_prior_mu'], pars['safe_prior_mu'], True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 169,
   "id": "d1bfb4c3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>T</th>\n",
       "      <th>dof</th>\n",
       "      <th>alternative</th>\n",
       "      <th>p-val</th>\n",
       "      <th>CI95%</th>\n",
       "      <th>cohen-d</th>\n",
       "      <th>BF10</th>\n",
       "      <th>power</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>T-test</th>\n",
       "      <td>-1.477538</td>\n",
       "      <td>27</td>\n",
       "      <td>two-sided</td>\n",
       "      <td>0.151102</td>\n",
       "      <td>[-1.46, 0.24]</td>\n",
       "      <td>0.408834</td>\n",
       "      <td>0.53</td>\n",
       "      <td>0.550198</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               T  dof alternative     p-val          CI95%   cohen-d  BF10  \\\n",
       "T-test -1.477538   27   two-sided  0.151102  [-1.46, 0.24]  0.408834  0.53   \n",
       "\n",
       "           power  \n",
       "T-test  0.550198  "
      ]
     },
     "execution_count": 169,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pingouin.ttest(pars['risky_prior_sd'], pars['safe_prior_sd'], True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 170,
   "id": "b8c824d2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>T</th>\n",
       "      <th>dof</th>\n",
       "      <th>alternative</th>\n",
       "      <th>p-val</th>\n",
       "      <th>CI95%</th>\n",
       "      <th>cohen-d</th>\n",
       "      <th>BF10</th>\n",
       "      <th>power</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>T-test</th>\n",
       "      <td>-0.002122</td>\n",
       "      <td>27</td>\n",
       "      <td>two-sided</td>\n",
       "      <td>0.998323</td>\n",
       "      <td>[-0.45, 0.45]</td>\n",
       "      <td>0.000442</td>\n",
       "      <td>0.201</td>\n",
       "      <td>0.050001</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               T  dof alternative     p-val          CI95%   cohen-d   BF10  \\\n",
       "T-test -0.002122   27   two-sided  0.998323  [-0.45, 0.45]  0.000442  0.201   \n",
       "\n",
       "           power  \n",
       "T-test  0.050001  "
      ]
     },
     "execution_count": 170,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pingouin.ttest(pars['evidence sd (first)'], pars['evidence sd (later)'], True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 171,
   "id": "fc5c7ecf",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/gdehol/miniconda3/envs/pymc3/lib/python3.9/site-packages/seaborn/_decorators.py:36: FutureWarning: Pass the following variables as keyword args: x, y. From version 0.12, the only valid positional argument will be `data`, and passing other arguments without an explicit keyword will result in an error or misinterpretation.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7fd554174b20>]"
      ]
     },
     "execution_count": 171,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.scatterplot('evidence sd (first)', 'evidence sd (later)', data=pars)\n",
    "plt.plot([0., 1.], [0., 1.])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 172,
   "id": "de02d0a7",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/gdehol/miniconda3/envs/pymc3/lib/python3.9/site-packages/seaborn/_decorators.py:36: FutureWarning: Pass the following variables as keyword args: x, y. From version 0.12, the only valid positional argument will be `data`, and passing other arguments without an explicit keyword will result in an error or misinterpretation.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7fd5549550a0>]"
      ]
     },
     "execution_count": 172,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.scatterplot('risky_prior_mu', 'safe_prior_mu', data=pars)\n",
    "plt.plot([1.5, 4.,], [1.5, 4.])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 173,
   "id": "0c65cbeb",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/gdehol/miniconda3/envs/pymc3/lib/python3.9/site-packages/seaborn/_decorators.py:36: FutureWarning: Pass the following variables as keyword args: x, y. From version 0.12, the only valid positional argument will be `data`, and passing other arguments without an explicit keyword will result in an error or misinterpretation.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7fd558a1df10>]"
      ]
     },
     "execution_count": 173,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.scatterplot('risky_prior_sd', 'safe_prior_sd', data=pars)\n",
    "plt.plot([0., 4.,], [0., 4.])"
   ]
  }
 ],
 "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": "3.9.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}