From 33173175d7bc133e70dfe68b2da05eae75664c96 Mon Sep 17 00:00:00 2001
From: Matt Barlin <33430774+matttyb80@users.noreply.github.com>
Date: Tue, 21 Jul 2020 11:27:51 -0400
Subject: [PATCH] added viz for states, agents h and v

---
 .../mutliscale_ABM_main-checkpoint.ipynb      | 1663 +++++++++++
 __pycache__/run2.cpython-36.pyc               |  Bin 872 -> 886 bytes
 mutliscale_ABM_main.ipynb                     | 1663 +++++++++++
 run2.py                                       |    1 +
 src/sim/config.py                             |    4 +-
 src/sim/model/parts/private_beliefs.py        |    2 +-
 src/sim/run.py                                |  106 +-
 tests/mutliscale_ABM.ipynb                    | 2497 +++--------------
 8 files changed, 3800 insertions(+), 2136 deletions(-)
 create mode 100644 .ipynb_checkpoints/mutliscale_ABM_main-checkpoint.ipynb
 create mode 100644 mutliscale_ABM_main.ipynb

diff --git a/.ipynb_checkpoints/mutliscale_ABM_main-checkpoint.ipynb b/.ipynb_checkpoints/mutliscale_ABM_main-checkpoint.ipynb
new file mode 100644
index 0000000..437289c
--- /dev/null
+++ b/.ipynb_checkpoints/mutliscale_ABM_main-checkpoint.ipynb
@@ -0,0 +1,1663 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# NOTES\n",
+    "## USING run.py and NOT run2.py\n",
+    "## sys path (or run2.py)  may interfere with overwriting configs, NEED TO LOOK INTO MORE\n",
+    "## WILL ONLY WORK FOR SIMULATIONS WHERE N>1. CAN STILL VISUALIZE SINGLE RUN DATA AS SEEN BELOW, BUT cadCAD RUN MUST BE N>1."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {
+    "scrolled": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\n",
+      "<class 'list'>\n",
+      "[1000]\n",
+      "\n",
+      "CHECKPOINT 1\n",
+      "CHECKPOINT 2\n",
+      "Initial Conditions (config.py) :  {'reserve': 300, 'price': 1, 'realized_price': 0, 'spot_price': 1, 'private_price': 0, 'private_alpha': 0, 'kappa': 0, 'supply': 600.0, 'alpha': [0.5], 'supply_0': 200, 'supply_1': 200, 'supply_free': 600.0, 'attestations': 40, 'attestations_0': 20, 'attestations_1': 20, 'invariant_V': 1200.0, 'invariant_I': 650.0, 'agents':    agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply_1  \\\n",
+      "0                     0                     0             50              10   \n",
+      "1                     0                     0             50              10   \n",
+      "2                     0                     0             50              10   \n",
+      "3                     0                     0             50              10   \n",
+      "4                     0                     0             50              10   \n",
+      "\n",
+      "   agent_supply_0  agent_supply_free  \n",
+      "0              10                 30  \n",
+      "1              10                 30  \n",
+      "2              10                 30  \n",
+      "3              10                 30  \n",
+      "4              10                 30  , 'chosen_agent': 0}\n",
+      "Params (config.py) :  {'starting_kappa': 2, 'starting_alpha': 0.5, 'money_raised': 1000, 'C': 700, 'f': 0.03, 'm': 0.15, 'period': 2000, 'rules_price': 'martin'}\n",
+      "Params (config.py) :  {'starting_kappa': 2, 'starting_alpha': 0.5, 'money_raised': 1000, 'C': 700, 'f': 0.03, 'm': 0.15, 'period': 2000, 'rules_price': 'martin'}\n",
+      "Params (config.py) :  {'starting_kappa': 2, 'starting_alpha': 0.5, 'money_raised': 1000, 'C': 700, 'f': 0.03, 'm': 0.15, 'period': 2000, 'rules_price': 'martin'}\n",
+      "Params (config.py) :  {'starting_kappa': 2, 'starting_alpha': 0.5, 'money_raised': 1000, 'C': 700, 'f': 0.03, 'm': 0.15, 'period': 2000, 'rules_price': 'step'}\n",
+      "Params (config.py) :  {'starting_kappa': 2, 'starting_alpha': 0.5, 'money_raised': 1000, 'C': 700, 'f': 0.03, 'm': 0.15, 'period': 2000, 'rules_price': 'step'}\n",
+      "Params (config.py) :  {'starting_kappa': 2, 'starting_alpha': 0.5, 'money_raised': 1000, 'C': 700, 'f': 0.03, 'm': 0.15, 'period': 2000, 'rules_price': 'step'}\n",
+      "\n",
+      "                  ___________    ____\n",
+      "  ________ __ ___/ / ____/   |  / __ \\\n",
+      " / ___/ __` / __  / /   / /| | / / / /\n",
+      "/ /__/ /_/ / /_/ / /___/ ___ |/ /_/ /\n",
+      "\\___/\\__,_/\\__,_/\\____/_/  |_/_____/\n",
+      "by cadCAD\n",
+      "\n",
+      "Execution Mode: multi_proc\n",
+      "Configuration Count: 6\n",
+      "Dimensions of the first simulation: (Timesteps, Params, Runs, Vars) = (100, 8, 1, 19)\n",
+      "Execution Method: parallelize_simulations\n",
+      "Execution Mode: parallelized\n",
+      "Total execution time: 9.21s\n"
+     ]
+    }
+   ],
+   "source": [
+    "# import sys\n",
+    "# sys.path.append('../')\n",
+    "import pandas as pd\n",
+    "\n",
+    "import matplotlib.pyplot as plt\n",
+    "# import run2\n",
+    "\n",
+    "from src.sim import run\n",
+    "\n",
+    "import seaborn as sns\n",
+    "\n",
+    "# For analysis\n",
+    "import numpy as np\n",
+    "import pandas as pd\n",
+    "\n",
+    "sns.set_style(\"whitegrid\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "cadCAD                        0.4.17             \n"
+     ]
+    }
+   ],
+   "source": [
+    "!pip list | grep cadCAD"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {
+    "scrolled": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "config_ids =  [{'M': {'starting_kappa': 2, 'starting_alpha': 0.5, 'money_raised': 1000, 'C': 700, 'f': 0.03, 'm': 0.15, 'period': 2000, 'rules_price': 'martin'}, 'simulation_id': 0, 'run_id': 0}, {'M': {'starting_kappa': 2, 'starting_alpha': 0.5, 'money_raised': 1000, 'C': 700, 'f': 0.03, 'm': 0.15, 'period': 2000, 'rules_price': 'martin'}, 'simulation_id': 0, 'run_id': 1}, {'M': {'starting_kappa': 2, 'starting_alpha': 0.5, 'money_raised': 1000, 'C': 700, 'f': 0.03, 'm': 0.15, 'period': 2000, 'rules_price': 'martin'}, 'simulation_id': 0, 'run_id': 2}, {'M': {'starting_kappa': 2, 'starting_alpha': 0.5, 'money_raised': 1000, 'C': 700, 'f': 0.03, 'm': 0.15, 'period': 2000, 'rules_price': 'step'}, 'simulation_id': 1, 'run_id': 0}, {'M': {'starting_kappa': 2, 'starting_alpha': 0.5, 'money_raised': 1000, 'C': 700, 'f': 0.03, 'm': 0.15, 'period': 2000, 'rules_price': 'step'}, 'simulation_id': 1, 'run_id': 1}, {'M': {'starting_kappa': 2, 'starting_alpha': 0.5, 'money_raised': 1000, 'C': 700, 'f': 0.03, 'm': 0.15, 'period': 2000, 'rules_price': 'step'}, 'simulation_id': 1, 'run_id': 2}]\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "C:\\Users\\mbarl\\Projects\\ICF_Internal\\src\\sim\\run.py:78: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
+      "  sub_df = df[df.simulation == config_id['simulation_id']][df.run == config_id['run_id'] + 1]\n",
+      "C:\\Users\\mbarl\\Projects\\ICF_Internal\\src\\sim\\run.py:78: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
+      "  sub_df = df[df.simulation == config_id['simulation_id']][df.run == config_id['run_id'] + 1]\n",
+      "C:\\Users\\mbarl\\Projects\\ICF_Internal\\src\\sim\\run.py:78: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
+      "  sub_df = df[df.simulation == config_id['simulation_id']][df.run == config_id['run_id'] + 1]\n",
+      "C:\\Users\\mbarl\\Projects\\ICF_Internal\\src\\sim\\run.py:78: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
+      "  sub_df = df[df.simulation == config_id['simulation_id']][df.run == config_id['run_id'] + 1]\n",
+      "C:\\Users\\mbarl\\Projects\\ICF_Internal\\src\\sim\\run.py:78: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
+      "  sub_df = df[df.simulation == config_id['simulation_id']][df.run == config_id['run_id'] + 1]\n",
+      "C:\\Users\\mbarl\\Projects\\ICF_Internal\\src\\sim\\run.py:78: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
+      "  sub_df = df[df.simulation == config_id['simulation_id']][df.run == config_id['run_id'] + 1]\n"
+     ]
+    }
+   ],
+   "source": [
+    "experiments = run.run()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "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>index</th>\n",
+       "      <th>starting_kappa</th>\n",
+       "      <th>starting_alpha</th>\n",
+       "      <th>money_raised</th>\n",
+       "      <th>C</th>\n",
+       "      <th>f</th>\n",
+       "      <th>m</th>\n",
+       "      <th>period</th>\n",
+       "      <th>rules_price</th>\n",
+       "      <th>dataset</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>0</td>\n",
+       "      <td>2</td>\n",
+       "      <td>0.5</td>\n",
+       "      <td>1000</td>\n",
+       "      <td>700</td>\n",
+       "      <td>0.03</td>\n",
+       "      <td>0.15</td>\n",
+       "      <td>2000</td>\n",
+       "      <td>martin</td>\n",
+       "      <td>reserve     price  realized_price  sp...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>0</td>\n",
+       "      <td>2</td>\n",
+       "      <td>0.5</td>\n",
+       "      <td>1000</td>\n",
+       "      <td>700</td>\n",
+       "      <td>0.03</td>\n",
+       "      <td>0.15</td>\n",
+       "      <td>2000</td>\n",
+       "      <td>step</td>\n",
+       "      <td>reserve     price  realized_price  sp...</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   index  starting_kappa  starting_alpha  money_raised    C     f     m  \\\n",
+       "0      0               2             0.5          1000  700  0.03  0.15   \n",
+       "1      0               2             0.5          1000  700  0.03  0.15   \n",
+       "\n",
+       "   period rules_price                                            dataset  \n",
+       "0    2000      martin           reserve     price  realized_price  sp...  \n",
+       "1    2000        step           reserve     price  realized_price  sp...  "
+      ]
+     },
+     "execution_count": 4,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "experiments"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "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>reserve</th>\n",
+       "      <th>price</th>\n",
+       "      <th>realized_price</th>\n",
+       "      <th>spot_price</th>\n",
+       "      <th>private_price</th>\n",
+       "      <th>private_alpha</th>\n",
+       "      <th>kappa</th>\n",
+       "      <th>supply</th>\n",
+       "      <th>alpha</th>\n",
+       "      <th>supply_0</th>\n",
+       "      <th>...</th>\n",
+       "      <th>attestations_0</th>\n",
+       "      <th>attestations_1</th>\n",
+       "      <th>invariant_V</th>\n",
+       "      <th>invariant_I</th>\n",
+       "      <th>agents</th>\n",
+       "      <th>chosen_agent</th>\n",
+       "      <th>simulation</th>\n",
+       "      <th>run</th>\n",
+       "      <th>substep</th>\n",
+       "      <th>timestep</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>300.000000</td>\n",
+       "      <td>1.000000</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.000000</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0.00</td>\n",
+       "      <td>2.000000</td>\n",
+       "      <td>600.000000</td>\n",
+       "      <td>0.500000</td>\n",
+       "      <td>200.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>20.0</td>\n",
+       "      <td>20.000000</td>\n",
+       "      <td>1200.000000</td>\n",
+       "      <td>650.000000</td>\n",
+       "      <td>agent_attestations_1  agent_attestations_0 ...</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>5</th>\n",
+       "      <td>300.026889</td>\n",
+       "      <td>1.000022</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.084917</td>\n",
+       "      <td>1.001076</td>\n",
+       "      <td>0.74</td>\n",
+       "      <td>2.169738</td>\n",
+       "      <td>600.026888</td>\n",
+       "      <td>0.501361</td>\n",
+       "      <td>200.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>20.0</td>\n",
+       "      <td>20.153041</td>\n",
+       "      <td>3554.198921</td>\n",
+       "      <td>650.979671</td>\n",
+       "      <td>agent_attestations_1  agent_attestations_0 ...</td>\n",
+       "      <td>{'agent_attestations_1': 0.15304119675024097, ...</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>5</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>10</th>\n",
+       "      <td>301.512069</td>\n",
+       "      <td>1.083468</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.086216</td>\n",
+       "      <td>0.993533</td>\n",
+       "      <td>0.71</td>\n",
+       "      <td>2.166573</td>\n",
+       "      <td>601.397654</td>\n",
+       "      <td>0.502480</td>\n",
+       "      <td>200.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>20.0</td>\n",
+       "      <td>20.272443</td>\n",
+       "      <td>3482.997889</td>\n",
+       "      <td>653.248033</td>\n",
+       "      <td>agent_attestations_1  agent_attestations_0 ...</td>\n",
+       "      <td>{'agent_attestations_1': 0.11940162751985284, ...</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>5</td>\n",
+       "      <td>2</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>15</th>\n",
+       "      <td>301.564639</td>\n",
+       "      <td>1.086165</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.087566</td>\n",
+       "      <td>1.082990</td>\n",
+       "      <td>0.73</td>\n",
+       "      <td>2.169062</td>\n",
+       "      <td>601.446053</td>\n",
+       "      <td>0.503640</td>\n",
+       "      <td>200.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>20.0</td>\n",
+       "      <td>20.391865</td>\n",
+       "      <td>3538.916623</td>\n",
+       "      <td>654.112491</td>\n",
+       "      <td>agent_attestations_1  agent_attestations_0 ...</td>\n",
+       "      <td>{'agent_attestations_1': 0.1194224597725313, '...</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>5</td>\n",
+       "      <td>3</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>20</th>\n",
+       "      <td>301.717442</td>\n",
+       "      <td>1.087418</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.089257</td>\n",
+       "      <td>1.078198</td>\n",
+       "      <td>0.71</td>\n",
+       "      <td>2.171841</td>\n",
+       "      <td>601.586572</td>\n",
+       "      <td>0.505093</td>\n",
+       "      <td>200.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>20.0</td>\n",
+       "      <td>20.533524</td>\n",
+       "      <td>3602.415418</td>\n",
+       "      <td>655.282373</td>\n",
+       "      <td>agent_attestations_1  agent_attestations_0 ...</td>\n",
+       "      <td>{'agent_attestations_1': 0.14165863612184193, ...</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>5</td>\n",
+       "      <td>4</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>5 rows × 23 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "       reserve     price  realized_price  spot_price  private_price  \\\n",
+       "0   300.000000  1.000000               0    1.000000       0.000000   \n",
+       "5   300.026889  1.000022               0    1.084917       1.001076   \n",
+       "10  301.512069  1.083468               0    1.086216       0.993533   \n",
+       "15  301.564639  1.086165               0    1.087566       1.082990   \n",
+       "20  301.717442  1.087418               0    1.089257       1.078198   \n",
+       "\n",
+       "    private_alpha     kappa      supply     alpha  supply_0  ...  \\\n",
+       "0            0.00  2.000000  600.000000  0.500000     200.0  ...   \n",
+       "5            0.74  2.169738  600.026888  0.501361     200.0  ...   \n",
+       "10           0.71  2.166573  601.397654  0.502480     200.0  ...   \n",
+       "15           0.73  2.169062  601.446053  0.503640     200.0  ...   \n",
+       "20           0.71  2.171841  601.586572  0.505093     200.0  ...   \n",
+       "\n",
+       "    attestations_0  attestations_1  invariant_V  invariant_I  \\\n",
+       "0             20.0       20.000000  1200.000000   650.000000   \n",
+       "5             20.0       20.153041  3554.198921   650.979671   \n",
+       "10            20.0       20.272443  3482.997889   653.248033   \n",
+       "15            20.0       20.391865  3538.916623   654.112491   \n",
+       "20            20.0       20.533524  3602.415418   655.282373   \n",
+       "\n",
+       "                                               agents  \\\n",
+       "0      agent_attestations_1  agent_attestations_0 ...   \n",
+       "5      agent_attestations_1  agent_attestations_0 ...   \n",
+       "10     agent_attestations_1  agent_attestations_0 ...   \n",
+       "15     agent_attestations_1  agent_attestations_0 ...   \n",
+       "20     agent_attestations_1  agent_attestations_0 ...   \n",
+       "\n",
+       "                                         chosen_agent  simulation run substep  \\\n",
+       "0                                                   0           0   1       0   \n",
+       "5   {'agent_attestations_1': 0.15304119675024097, ...           0   1       5   \n",
+       "10  {'agent_attestations_1': 0.11940162751985284, ...           0   1       5   \n",
+       "15  {'agent_attestations_1': 0.1194224597725313, '...           0   1       5   \n",
+       "20  {'agent_attestations_1': 0.14165863612184193, ...           0   1       5   \n",
+       "\n",
+       "    timestep  \n",
+       "0          0  \n",
+       "5          1  \n",
+       "10         2  \n",
+       "15         3  \n",
+       "20         4  \n",
+       "\n",
+       "[5 rows x 23 columns]"
+      ]
+     },
+     "execution_count": 5,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "experiments.dataset[0].head()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "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>reserve</th>\n",
+       "      <th>price</th>\n",
+       "      <th>realized_price</th>\n",
+       "      <th>spot_price</th>\n",
+       "      <th>private_price</th>\n",
+       "      <th>private_alpha</th>\n",
+       "      <th>kappa</th>\n",
+       "      <th>supply</th>\n",
+       "      <th>alpha</th>\n",
+       "      <th>supply_0</th>\n",
+       "      <th>...</th>\n",
+       "      <th>attestations_0</th>\n",
+       "      <th>attestations_1</th>\n",
+       "      <th>invariant_V</th>\n",
+       "      <th>invariant_I</th>\n",
+       "      <th>agents</th>\n",
+       "      <th>chosen_agent</th>\n",
+       "      <th>simulation</th>\n",
+       "      <th>run</th>\n",
+       "      <th>substep</th>\n",
+       "      <th>timestep</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>1482</th>\n",
+       "      <td>306.664756</td>\n",
+       "      <td>1.119910</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.119913</td>\n",
+       "      <td>1.118950</td>\n",
+       "      <td>0.68</td>\n",
+       "      <td>2.213216</td>\n",
+       "      <td>606.042960</td>\n",
+       "      <td>0.531501</td>\n",
+       "      <td>200.095493</td>\n",
+       "      <td>...</td>\n",
+       "      <td>20.00158</td>\n",
+       "      <td>22.475030</td>\n",
+       "      <td>4694.815045</td>\n",
+       "      <td>678.715487</td>\n",
+       "      <td>agent_attestations_1  agent_attestations_0 ...</td>\n",
+       "      <td>{'agent_attestations_1': 0.5108411557663454, '...</td>\n",
+       "      <td>0</td>\n",
+       "      <td>3</td>\n",
+       "      <td>5</td>\n",
+       "      <td>96</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1487</th>\n",
+       "      <td>306.664937</td>\n",
+       "      <td>1.119913</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.119918</td>\n",
+       "      <td>1.117907</td>\n",
+       "      <td>0.64</td>\n",
+       "      <td>2.213226</td>\n",
+       "      <td>606.043121</td>\n",
+       "      <td>0.531506</td>\n",
+       "      <td>200.095493</td>\n",
+       "      <td>...</td>\n",
+       "      <td>20.00158</td>\n",
+       "      <td>22.475305</td>\n",
+       "      <td>4695.109654</td>\n",
+       "      <td>678.718890</td>\n",
+       "      <td>agent_attestations_1  agent_attestations_0 ...</td>\n",
+       "      <td>{'agent_attestations_1': 0.4979930202683946, '...</td>\n",
+       "      <td>0</td>\n",
+       "      <td>3</td>\n",
+       "      <td>5</td>\n",
+       "      <td>97</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1492</th>\n",
+       "      <td>306.755191</td>\n",
+       "      <td>1.120009</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.119921</td>\n",
+       "      <td>1.123601</td>\n",
+       "      <td>0.58</td>\n",
+       "      <td>2.212874</td>\n",
+       "      <td>606.123705</td>\n",
+       "      <td>0.531508</td>\n",
+       "      <td>200.095493</td>\n",
+       "      <td>...</td>\n",
+       "      <td>20.00158</td>\n",
+       "      <td>22.475430</td>\n",
+       "      <td>4684.528890</td>\n",
+       "      <td>678.810626</td>\n",
+       "      <td>agent_attestations_1  agent_attestations_0 ...</td>\n",
+       "      <td>{'agent_attestations_1': 0.5109668943877075, '...</td>\n",
+       "      <td>0</td>\n",
+       "      <td>3</td>\n",
+       "      <td>5</td>\n",
+       "      <td>98</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1497</th>\n",
+       "      <td>306.855175</td>\n",
+       "      <td>1.120021</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.119931</td>\n",
+       "      <td>1.123941</td>\n",
+       "      <td>0.64</td>\n",
+       "      <td>2.212498</td>\n",
+       "      <td>606.212974</td>\n",
+       "      <td>0.531516</td>\n",
+       "      <td>200.095493</td>\n",
+       "      <td>...</td>\n",
+       "      <td>20.00158</td>\n",
+       "      <td>22.475941</td>\n",
+       "      <td>4673.267443</td>\n",
+       "      <td>678.916608</td>\n",
+       "      <td>agent_attestations_1  agent_attestations_0 ...</td>\n",
+       "      <td>{'agent_attestations_1': 0.4985032783348726, '...</td>\n",
+       "      <td>0</td>\n",
+       "      <td>3</td>\n",
+       "      <td>5</td>\n",
+       "      <td>99</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1502</th>\n",
+       "      <td>306.857628</td>\n",
+       "      <td>1.119928</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.119988</td>\n",
+       "      <td>1.109632</td>\n",
+       "      <td>0.88</td>\n",
+       "      <td>2.212603</td>\n",
+       "      <td>606.215165</td>\n",
+       "      <td>0.531566</td>\n",
+       "      <td>200.095493</td>\n",
+       "      <td>...</td>\n",
+       "      <td>20.00158</td>\n",
+       "      <td>22.478915</td>\n",
+       "      <td>4676.387145</td>\n",
+       "      <td>678.953996</td>\n",
+       "      <td>agent_attestations_1  agent_attestations_0 ...</td>\n",
+       "      <td>{'agent_attestations_1': 0.4938448398075422, '...</td>\n",
+       "      <td>0</td>\n",
+       "      <td>3</td>\n",
+       "      <td>5</td>\n",
+       "      <td>100</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>5 rows × 23 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "         reserve     price  realized_price  spot_price  private_price  \\\n",
+       "1482  306.664756  1.119910               0    1.119913       1.118950   \n",
+       "1487  306.664937  1.119913               0    1.119918       1.117907   \n",
+       "1492  306.755191  1.120009               0    1.119921       1.123601   \n",
+       "1497  306.855175  1.120021               0    1.119931       1.123941   \n",
+       "1502  306.857628  1.119928               0    1.119988       1.109632   \n",
+       "\n",
+       "      private_alpha     kappa      supply     alpha    supply_0  ...  \\\n",
+       "1482           0.68  2.213216  606.042960  0.531501  200.095493  ...   \n",
+       "1487           0.64  2.213226  606.043121  0.531506  200.095493  ...   \n",
+       "1492           0.58  2.212874  606.123705  0.531508  200.095493  ...   \n",
+       "1497           0.64  2.212498  606.212974  0.531516  200.095493  ...   \n",
+       "1502           0.88  2.212603  606.215165  0.531566  200.095493  ...   \n",
+       "\n",
+       "      attestations_0  attestations_1  invariant_V  invariant_I  \\\n",
+       "1482        20.00158       22.475030  4694.815045   678.715487   \n",
+       "1487        20.00158       22.475305  4695.109654   678.718890   \n",
+       "1492        20.00158       22.475430  4684.528890   678.810626   \n",
+       "1497        20.00158       22.475941  4673.267443   678.916608   \n",
+       "1502        20.00158       22.478915  4676.387145   678.953996   \n",
+       "\n",
+       "                                                 agents  \\\n",
+       "1482     agent_attestations_1  agent_attestations_0 ...   \n",
+       "1487     agent_attestations_1  agent_attestations_0 ...   \n",
+       "1492     agent_attestations_1  agent_attestations_0 ...   \n",
+       "1497     agent_attestations_1  agent_attestations_0 ...   \n",
+       "1502     agent_attestations_1  agent_attestations_0 ...   \n",
+       "\n",
+       "                                           chosen_agent  simulation run  \\\n",
+       "1482  {'agent_attestations_1': 0.5108411557663454, '...           0   3   \n",
+       "1487  {'agent_attestations_1': 0.4979930202683946, '...           0   3   \n",
+       "1492  {'agent_attestations_1': 0.5109668943877075, '...           0   3   \n",
+       "1497  {'agent_attestations_1': 0.4985032783348726, '...           0   3   \n",
+       "1502  {'agent_attestations_1': 0.4938448398075422, '...           0   3   \n",
+       "\n",
+       "     substep  timestep  \n",
+       "1482       5        96  \n",
+       "1487       5        97  \n",
+       "1492       5        98  \n",
+       "1497       5        99  \n",
+       "1502       5       100  \n",
+       "\n",
+       "[5 rows x 23 columns]"
+      ]
+     },
+     "execution_count": 6,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "experiments.dataset[0].tail()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "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>reserve</th>\n",
+       "      <th>price</th>\n",
+       "      <th>realized_price</th>\n",
+       "      <th>spot_price</th>\n",
+       "      <th>private_price</th>\n",
+       "      <th>private_alpha</th>\n",
+       "      <th>kappa</th>\n",
+       "      <th>supply</th>\n",
+       "      <th>alpha</th>\n",
+       "      <th>supply_0</th>\n",
+       "      <th>...</th>\n",
+       "      <th>attestations_0</th>\n",
+       "      <th>attestations_1</th>\n",
+       "      <th>invariant_V</th>\n",
+       "      <th>invariant_I</th>\n",
+       "      <th>agents</th>\n",
+       "      <th>chosen_agent</th>\n",
+       "      <th>simulation</th>\n",
+       "      <th>run</th>\n",
+       "      <th>substep</th>\n",
+       "      <th>timestep</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>1503</th>\n",
+       "      <td>300.000000</td>\n",
+       "      <td>1.000000</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.000000</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0.00</td>\n",
+       "      <td>2.000000</td>\n",
+       "      <td>600.000000</td>\n",
+       "      <td>0.500000</td>\n",
+       "      <td>200.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>20.0</td>\n",
+       "      <td>20.000000</td>\n",
+       "      <td>1200.000000</td>\n",
+       "      <td>650.000000</td>\n",
+       "      <td>agent_attestations_1  agent_attestations_0 ...</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1508</th>\n",
+       "      <td>300.000000</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.086274</td>\n",
+       "      <td>1.000000</td>\n",
+       "      <td>0.98</td>\n",
+       "      <td>2.172549</td>\n",
+       "      <td>600.000000</td>\n",
+       "      <td>0.502521</td>\n",
+       "      <td>200.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>20.0</td>\n",
+       "      <td>20.256580</td>\n",
+       "      <td>3618.655872</td>\n",
+       "      <td>651.764572</td>\n",
+       "      <td>agent_attestations_1  agent_attestations_0 ...</td>\n",
+       "      <td>{'agent_attestations_1': 0.2565801213118535, '...</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>5</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1513</th>\n",
+       "      <td>317.418514</td>\n",
+       "      <td>1.069006</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.087214</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0.96</td>\n",
+       "      <td>2.110915</td>\n",
+       "      <td>616.294114</td>\n",
+       "      <td>0.503750</td>\n",
+       "      <td>200.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>20.0</td>\n",
+       "      <td>20.350613</td>\n",
+       "      <td>2439.905200</td>\n",
+       "      <td>670.043483</td>\n",
+       "      <td>agent_attestations_1  agent_attestations_0 ...</td>\n",
+       "      <td>{'agent_attestations_1': 0.09403285712342058, ...</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>5</td>\n",
+       "      <td>2</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1518</th>\n",
+       "      <td>317.496671</td>\n",
+       "      <td>1.087143</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.088164</td>\n",
+       "      <td>1.069006</td>\n",
+       "      <td>0.92</td>\n",
+       "      <td>2.112486</td>\n",
+       "      <td>616.366007</td>\n",
+       "      <td>0.504587</td>\n",
+       "      <td>200.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>20.0</td>\n",
+       "      <td>20.412387</td>\n",
+       "      <td>2464.651535</td>\n",
+       "      <td>670.707222</td>\n",
+       "      <td>agent_attestations_1  agent_attestations_0 ...</td>\n",
+       "      <td>{'agent_attestations_1': 0.15580663229094682, ...</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>5</td>\n",
+       "      <td>3</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1523</th>\n",
+       "      <td>317.513328</td>\n",
+       "      <td>1.088149</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.090761</td>\n",
+       "      <td>1.087143</td>\n",
+       "      <td>0.85</td>\n",
+       "      <td>2.117470</td>\n",
+       "      <td>616.381314</td>\n",
+       "      <td>0.506874</td>\n",
+       "      <td>200.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>20.0</td>\n",
+       "      <td>20.575052</td>\n",
+       "      <td>2544.833231</td>\n",
+       "      <td>672.324803</td>\n",
+       "      <td>agent_attestations_1  agent_attestations_0 ...</td>\n",
+       "      <td>{'agent_attestations_1': 0.1626651559147385, '...</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>5</td>\n",
+       "      <td>4</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>5 rows × 23 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "         reserve     price  realized_price  spot_price  private_price  \\\n",
+       "1503  300.000000  1.000000               0    1.000000       0.000000   \n",
+       "1508  300.000000  0.000000               0    1.086274       1.000000   \n",
+       "1513  317.418514  1.069006               0    1.087214       0.000000   \n",
+       "1518  317.496671  1.087143               0    1.088164       1.069006   \n",
+       "1523  317.513328  1.088149               0    1.090761       1.087143   \n",
+       "\n",
+       "      private_alpha     kappa      supply     alpha  supply_0  ...  \\\n",
+       "1503           0.00  2.000000  600.000000  0.500000     200.0  ...   \n",
+       "1508           0.98  2.172549  600.000000  0.502521     200.0  ...   \n",
+       "1513           0.96  2.110915  616.294114  0.503750     200.0  ...   \n",
+       "1518           0.92  2.112486  616.366007  0.504587     200.0  ...   \n",
+       "1523           0.85  2.117470  616.381314  0.506874     200.0  ...   \n",
+       "\n",
+       "      attestations_0  attestations_1  invariant_V  invariant_I  \\\n",
+       "1503            20.0       20.000000  1200.000000   650.000000   \n",
+       "1508            20.0       20.256580  3618.655872   651.764572   \n",
+       "1513            20.0       20.350613  2439.905200   670.043483   \n",
+       "1518            20.0       20.412387  2464.651535   670.707222   \n",
+       "1523            20.0       20.575052  2544.833231   672.324803   \n",
+       "\n",
+       "                                                 agents  \\\n",
+       "1503     agent_attestations_1  agent_attestations_0 ...   \n",
+       "1508     agent_attestations_1  agent_attestations_0 ...   \n",
+       "1513     agent_attestations_1  agent_attestations_0 ...   \n",
+       "1518     agent_attestations_1  agent_attestations_0 ...   \n",
+       "1523     agent_attestations_1  agent_attestations_0 ...   \n",
+       "\n",
+       "                                           chosen_agent  simulation run  \\\n",
+       "1503                                                  0           1   1   \n",
+       "1508  {'agent_attestations_1': 0.2565801213118535, '...           1   1   \n",
+       "1513  {'agent_attestations_1': 0.09403285712342058, ...           1   1   \n",
+       "1518  {'agent_attestations_1': 0.15580663229094682, ...           1   1   \n",
+       "1523  {'agent_attestations_1': 0.1626651559147385, '...           1   1   \n",
+       "\n",
+       "     substep  timestep  \n",
+       "1503       0         0  \n",
+       "1508       5         1  \n",
+       "1513       5         2  \n",
+       "1518       5         3  \n",
+       "1523       5         4  \n",
+       "\n",
+       "[5 rows x 23 columns]"
+      ]
+     },
+     "execution_count": 7,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "experiments.dataset[1].head()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "303"
+      ]
+     },
+     "execution_count": 8,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "len(experiments.dataset[0])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1080x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1080x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "experiments = experiments.sort_values(by =['rules_price']).reset_index(drop=True)\n",
+    "# df = experiments\n",
+    "cols = 1\n",
+    "rows = 1\n",
+    "cc_idx = 0\n",
+    "\n",
+    "# len(experiments)\n",
+    "# df = experiments.copy()\n",
+    "# print(df.head())\n",
+    "\n",
+    "while cc_idx<len(experiments):\n",
+    "    cc = experiments.iloc[cc_idx]['rules_price']\n",
+    "    cc_label = experiments.iloc[cc_idx]['rules_price']\n",
+    "#     secondary_label = experiments.iloc[cc_idx]['string']\n",
+    "    sub_experiments = experiments[experiments['rules_price']==cc]\n",
+    "    cc_idx += len(sub_experiments)\n",
+    "    fig, axs = plt.subplots(ncols=cols, nrows=rows, figsize=(15*cols,7*rows))\n",
+    "    sub_experiments.reset_index(drop=False)\n",
+    "    for i, experiment in sub_experiments.iterrows():\n",
+    "        df = experiment['dataset'].copy()\n",
+    "        # filter out starting point\n",
+    "        df = df[df.timestep > 0]       \n",
+    "#         print(df['price'])\n",
+    "\n",
+    "        df = df.groupby('timestep').agg({'price': ['min', 'mean', 'max']}).reset_index()\n",
+    "\n",
+    "#         print(df.head())\n",
+    "        plot_label = experiment['rules_price']\n",
+    "        ax = axs\n",
+    "        title = 'Price' + '\\n' + 'Scenario: ' + str(cc_label)  + ' rules_price'\n",
+    "        ax.set_title(title)\n",
+    "        ax.set_ylabel('Price')\n",
+    "        colors = ['b','orange', 'g', 'magenta', 'r']\n",
+    "        \n",
+    "#         ax.plot(df.timestep, df['price'],color = colors[0], label='Price')\n",
+    "        df.plot(x='timestep', y=('price','mean'), label='Mean', ax=ax, legend=True)\n",
+    "        ax.fill_between(df.timestep, df[('price','min')], df[('price','max')], alpha=0.3) \n",
+    "\n",
+    "\n",
+    "        ax.legend()\n",
+    "\n",
+    "        ax.set_xlabel('Timesteps')\n",
+    "        ax.grid(color='0.9', linestyle='-', linewidth=1)\n",
+    "\n",
+    "        plt.tight_layout()\n",
+    "        \n",
+    "fig.tight_layout(rect=[0, 0, 1, .97])\n",
+    "fig.patch.set_alpha(1)\n",
+    "display(fig)\n",
+    "plt.close()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### FIRST RUN ONLY!"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAABDAAAAHhCAYAAACYzk47AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nOzdfZicdXko/nveN2EDEl7CKkJjbEC0NAI/LUrwAGJRiGJpFZWoKJZWW/UnQvVgUVNBFHsUuDwUbQkWehBEihxQLOALpxERKVjo4aWCRAJLSIFAXjbz+pw/dmd2J/uSzSYzs7vz+VxXrp155plnvs/Mg859z/29v6kkSZIAAAAAmMbSnR4AAAAAwLZIYAAAAADTngQGAAAAMO1JYAAAAADTngQGAAAAMO1JYAAAAADTXrbTAwAAusOaNWvi2GOPjcWLFze2JUkS733ve+OP//iPm/a97bbb4o477ojPfOYz7R4mADBNpZIkSTo9CABg9luzZk0sW7Ys7rnnnsa2tWvXxgknnBBXXHFFHHjggR0cHQAw3anAAAA6ZsGCBbH//vvHqlWrYsWKFTEwMBC9vb3x9re/PX74wx/GpZdeGuvWrYvPfvaz8eijj0Y6nY6TTz453vve98aGDRvi3HPPjYcffjjK5XIcfvjhcdZZZ0U26+sNAMxGemAAAB1zzz33xG9/+9vYsmVL/PrXv44rrrgirrjiiqZ9Pv/5z8fv/M7vxM033xxXX311XHPNNbF69eo477zz4pWvfGVcd911cf3118dzzz0XK1eu7NCZAACt5icKAKBttmzZEm9729siIqJarcbuu+8eF1xwQTzzzDNxwAEHRG9v76jn/OxnP4szzzwzIiLmzZsXN954Y0RE/OQnP4n77rsvrr322saxAYDZSwIDAGibnp6e+N73vjdq+3XXXRdz584d8znZbDZSqVTj/uOPPx6777571Gq1uPDCC2PRokUREfHCCy807QcAzC6mkAAA09rhhx8e3/3udyMiYsOGDfG+970vHnvssTjiiCPi8ssvjyRJolQqxZ//+Z/HlVde2eHRAgCtIoEBAExr55xzTjz66KOxbNmyeNe73hWnn356vOpVr4qzzz47Nm/eHMuWLYtly5bF4sWL47TTTuv0cAGAFrGMKgAAADDtqcAAAAAApj0JDAAAAGDak8AAAAAApj0JDACYJu69995Yvnx5LFu2LE444YQ47bTT4j//8z87Oqb77rsvPvrRj7bk2B/4wAfi2WefbcmxJ3LdddfF6aef3vbXHenss8+On/3sZx0dAwDMNNlODwAAiCiVSnH66afHZZddFq985SsjIuJ73/tefOhDH4rbbrstMplMR8b1e7/3e3HRRRe15NirVq1qyXFngnPPPbfTQwCAGUcCAwCmgYGBgdiwYUNs3ry5se2tb31r9Pb2RrVajUwmE9dee22sXLky0ul07L777vGlL30p+vr64kc/+lFccsklUS6Xo6enJ/7qr/4qXv3qV8fFF18cTzzxRKxbty6eeOKJWLBgQVxwwQWx9957x49//OO49NJLo1QqxbPPPhsnnnhifPzjH48777wzzj333Jg7d25s2rQpzjrrrPjSl74UN954Y2zYsCE+//nPx4MPPhipVCqWLl0an/jEJyKbzcaFF14YEREf+9jHms5r06ZN8elPfzpWr14d6XQ6XvnKV8aKFSvi7LPPjoiI973vffGNb3wj0ul0rFixIvr7+6NcLsfxxx8ff/ZnfxZr1qyJ5cuXx9KlS+NXv/pVJEkS55xzThx22GFNr7NmzZp4z3veE4sWLYonnngizj///PjABz4Q99xzT+PxZcuWNe7XbdiwIc4999x4+OGHo1wux+GHHx5nnXVWZLPZuOiii+KWW26JXC4Xu+++e3zxi1+Mvffee9zP8Lrrroubb745arVaPPnkk7FgwYI4//zzY8GCBbF8+fLYbbfd4tFHH413vetd8S//8i/xnve8J4477rj48Y9/HF/72teiVqvF3Llz4/Of/3wceOCB8W//9m/xla98JQYGBiKdTsdf/MVfxFFHHTX1iwwAZroEAJgWLrvssuTggw9Ojj766OSTn/xk8p3vfCfZvHlzkiRJ8sADDySvfe1rkyeffDJJkiRZuXJl8td//dfJb37zm+SEE05Inn322SRJkuThhx9OXv/61yebNm1KLrroouSYY45JNmzYkCRJkpx++unJhRdemNRqteSUU05JfvOb3yRJkiRPPfVU8opXvCJ55plnkp///OfJgQcemKxZsyZJkiT5+c9/nhx//PFJkiTJWWedlfzN3/xNUqvVkmKxmHzgAx9ILr300gnP6Z//+Z+TD3zgA0mSJEmlUknOPvvs5LHHHkuSJEkWL16cPPPMM0mSJMny5cuT2267LUmSJNmyZUuyfPny5Kabbkoef/zxZPHixckNN9yQJEmS/OQnP0le//rXJ6VSqel16vvdddddjftLlixperx+/7vf/W7yp3/6p0mSJMmnPvWp5B//8R8b4/vkJz+ZfOMb30iefPLJ5JBDDkmKxWKSJEnyD//wD8ktt9wy4bl+97vfTZYsWZI8+uijSZIkyQUXXJD85V/+ZZIkSXLKKackn/70pxv7nnLKKckPfvCDZN26dcmhhx6a/Md//EeSJEnywx/+MPngBz+YrF+/PnnTm96UPP74443P6Mgjj0yeeOKJCccAALOZCgwAmCZOPfXU+JM/+ZO466674q677opvfvOb8c1vfjOuvfbauOOOO+KII46Ivr6+iIh4//vfHxER//RP/xRPP/10435ERCqVit/+9rcREfGa17wment7IyLioIMOiueffz5SqVT83d/9XfzkJz+JG2+8MR555JFIkiQGBgYiIqKvry9e8pKXjBrf7bffHldddVWkUqnI5/Nx8sknx7e+9a340z/903HP6dBDD42vfvWrsXz58njd614X73vf+2L//fdv2mfz5s1x1113xfPPP9+o5Ni8eXM8+OCDcfDBB8duu+0Wy5Yti4iIN7zhDZHJZOKhhx6KV73qVU3HyWazsWTJksm+3RER8ZOf/CTuu+++uPbaayMiYsuWLRERsWDBgjjwwAPj7W9/exx55JFx5JFHxuGHH77N473+9a+PhQsXRkTEO97xjnjb297WeGzrqpGIiH/7t3+L3/3d342DDjooIiLe9KY3xZve9Kb46U9/GuvWrYuPfOQjjX1TqVQ89NBD8eIXv3i7zhEAZgsJDACYBu6+++6455574rTTToujjjoqjjrqqPjEJz4RJ5xwQqxatSoymUykUqnG/lu2bIknnngiarVaHH744fG1r32t8Vh/f3/svffeccstt0RPT09jeyqViiRJYvPmzfH2t7893vjGN8Zhhx0WJ510Utx6662RJElERMydO3fMMdZqtaYx1Gq1qFQqE57XS1/60rjlllvizjvvjJ///Odx6qmnxooVK+Loo49uOk6SJPHtb3875syZExERzz77bBQKhXjuuedG9f+o1Wpj9gTJ5/ORzWabzrWuXC6Pe04XXnhhLFq0KCIiXnjhhUilUpFOp+PKK6+M++67L+64444477zzYunSpXHWWWdNeL4jx7X1OMd6X7f+XJMkiYceeiiq1WosWrQovvOd7zQeW7t2bcyfP3/C1weA2cwqJAAwDcyfPz8uueSS+OUvf9nYtm7duti4cWMsXrw4Xvva18Ydd9wRTz/9dEREfPvb344LLrggDj/88Fi1alU88sgjERHx05/+NN761rc2KgnGsnr16ti4cWN8/OMfj6OPPjruvPPOKJVKUavVJhzjEUccEVdeeWUkSRKlUimuueaaeN3rXjfhc/7X//pf8elPfzqOOOKIOPPMM+OII46I//t//29EDAbvlUolent7Y8mSJbFy5cqIGEwivOtd74rbbrstIgaTGbfffntERPzoRz+KXC4XixcvnvB1d9111yiXy/HrX/86IiJuuummcc/p8ssvb5zTn//5n8eVV14ZDz74YJxwwgmxaNGiOP300+P9739/3HfffRO+ZkTEz3/+81i7dm1EDH5G2+pZ8fu///vxyCOPNFabue222+LMM8+MJUuWxOrVq+Ouu+6KiIgHHngg/vAP/7BxbADoRiowAGAaWLhwYXz961+Pr371q/HUU09FoVCIefPmxXnnnRcve9nLIiLizDPPjNNOOy0iIvbaa68477zzYsGCBbFixYr4xCc+EUmSRDabjUsuuSR22WWXcV/rgAMOiP/23/5bvPnNb458Ph+LFy+Ol7/85bF69erI5/PjPu8zn/lMfOELX4hly5ZFuVyOpUuXxp/92Z9FRIzbxPPEE0+MX/ziF/GWt7wl5syZE319fbF8+fKIiDjuuONi+fLlcfHFF8dXvvKV+Ju/+ZtYtmxZlEqlOOGEE+Ktb31rrFmzJgqFQnzve9+Lr3zlK9HT0xNf//rXt7kqy7x58+LMM8+MD33oQzF//vw47rjjxtzv7LPPjnPPPbdxTq973evitNNOi1wuF29+85vjpJNOirlz50ZPT0985jOfmfA1Iwannpx55pmxbt26ePnLXx4rVqyYcP8999wzvvKVr8Rf/dVfRbVajd7e3vjqV78a8+fPj4suuii+/OUvR7FYjCRJ4stf/nLsu+++2xwDAMxWqWRkfSUAwDQy3uoh09F1110XP/zhD+PSSy/t9FAAYFZSgQEAMEnvfve7Y9OmTWM+9sd//MdtHg0AdBcVGAAAAMC0p4knAAAAMO1JYAAAAADT3ozogXHvvfdGoVDo9DCmpFKpNNakh9nKdU43cJ3TDVzndAvXOt1gJl/nxWIxlixZMmr7jDibQqEQr3jFKzo9jCnp7++Pvr6+Tg8DWsp1TjdwndMNXOd0C9c63WAmX+cPPPDAmNtNIQEAAACmPQkMAAAAYNqTwAAAAACmvRnRA2Ms5XI51qxZE1u2bOn0UCZUrVZj/fr1O3ycnp6e2HfffSOXy+2EUQEAAMDMMmMTGGvWrIl58+bF7/zO70Qqler0cMZVKpUin8/v0DGSJIlnnnkm1qxZEwsXLtxJIwMAAICZY8ZOIdmyZUvsscce0zp5sbOkUqnYY489pn21CQAAALTKjE1gRERXJC/quulcAQAAYGszOoEBAAAAdAcJjB1w5513xgEHHBDf//73m7YvW7YsPvWpT3VoVAAAADD7SGDsoJe97GVx4403Nu4/9NBDMTAw0MERAQAAwOwzY1chGem7d6+Ja375+E495jsOe2mcdOi+29zvwAMPjMceeyxeeOGF2HXXXeOGG26IZcuWRX9/f/zgBz+Iyy67LLLZbBx66KHxyU9+Mp566qn43Oc+F8ViMdavXx8f+chH4o1vfGMsW7YsXvOa18RDDz0UqVQq/uf//J8xb968nXpOAAAAMFOpwNgJjj322LjlllsiSZL493//93j1q18d69evj4svvjj+/u//Pq666qpYu3ZtrFq1Kh599NE49dRTY+XKlfHXf/3X8U//9E8REbFp06Y4/vjj48orr4y99947br/99g6fFQAAAEwfs6IC46RD951UtUSrLFu2LD73uc/FS1/60jjssMMiIqJarcazzz4bH/7whyOdTsemTZvi8ccfj0MPPTQuueSSuPbaayOVSkWlUmkc56CDDoqIiL6+vigWix05FwAAAJiOVGDsBC996Utj8+bNccUVV8Rb3/rWiBhc9rSvry++8Y1vxBVXXBGnnHJK/P7v/35ceOGF8ba3vS0uuOCCeO1rXxtJkjSOY6lUAAAAGNusqMCYDt7ylrfE9773vVi4cGE8/vjjMX/+/Dj++OPj1FNPjSRJ4iUveUm8+c1vjuOOOy7OPffcuPTSS6Ovry+ee+65Tg8dAAAApr1UMrIEYJp64IEH4hWveMU2t01HpVIp8vn8TjnWTDlnuk9/f3/09fV1ehjQUq5zuoHrnG7hWqcbzOTrfLzYVwUGAEAblKu12FKuxpby4N9ipRaVWi0q1SQqtSSqtVqUq0lUa0mUq7Wo1ga3b/unpuEd6vsmW92frOfWPxe7r9vWqw2OKYmIJBm8XRv5d8T2yY98xLYp/rSWjHm07T92MrRjtZZEbatzq9+v1ernOfzaTe/9Tv59cOvPs36uY33eydBnUBsaf+PzGjH+8Y4fI46z80e/48dPkuZzT8bYPvLlkqbnJk3bBwYGoqfnqcb12th/69eYxHu+9auN3icZ8zlbPz4VSdP5JqO2jbXfdh1/xH/vQxtGX++TGmfz+z9yTEnzjpMY09i7T+Z/A6ZiUv+7MdZ7PuZ+7a0dOG7xbvHxGZrAGI8EBgBdKRkKUCpD/8qVWpSHAsjyUGBZqiRRqdWiXK3F0+s2xuNbnm06xliti4YDneagZ+tAb5vjm+Q51I8/GGwN/qvW6sFLEtVkjC+aW32ZHrGpcdyxxtH8RXHs/ev7jQz2RgZO9fei/p6XqoPvb7Ey+N6XKtXBz6Bai1KlttW4R77+BF/Ux3ifJqMekNYD0OH79QBw8p9LpZbEQKkaxcpwwqJSm/ZFr0N+0+kBzAo7u7VZ/XD1nmnD9+uPpxr306lUpFOD+6ZSg/um06lIpwb3GntszRt35vjHOtRUj5+K1Ihzbu4h19ieat5/rNesVWuRyRQH36MRg6wfc+T7O/K9bT6HsT+LprFs9dzUVjuM9dyJJMnoYzW9BxMcLzXmJzEJqRi+blIRqUgPvi9bneM2DzPWmLZ6H8bbb4whjTrG1tt3psl9PqN3GvtzaJ8Fu81p46u1x4xOYCRJ0jWNL2fATB9gB9RqSWypVGOgVI2IiGw6Hen0Vn9TYzf7rQfG1XowWxu8Xa0lUaxUo1iuNYKoYqXWtK1YGfyVd6xfTuu/3NWGorhGMNcU2DWPoR6MliqDf8vVZHDbiEA1YujLzhhfFutftOsBefOv04P360mHam3EL6RbBeuNYL428rlJVKq1pvt0RioVkU2nIp9JRy6bHvybSUe+fjs7+Fg2nW5cE/UvzhH1L8upMYOVxmvsQDC29fU48v7I151IPpOOQi4dPblMzMlloieXjp5sJubkM1HIZaInm45CLhO5dCoy6VTkMunIpFORTaciO3Q7l0kNBaLbfsHJBE6TsX79+njRi1607deL4eC4HhSn68Fyauv3bFvHmtyX/smYYowxSv19T6eGA/90avgcR34uTYFul3wvnQ2eeuqp2GeffTo9DGip/v7+Tg9hp5uxCYyenp545plnYo899pj1/2eRJEk888wz0dPT0+mhwJSUKrUYKFWjVB0MYuvBbalS/9V1eNvIct36r7b1ILoeJNe/JGZG/MKUSQ9/uUynU5EkyVCwXoviUKl2c/A+eLsRDFfrv8TXmu5Xa7VIIpq+nNcDmHRq+JegVGr4V+fh0uqIGFF6Wa0lsaVcjYFyNTaXBpMVA+Xhv5ORSQ+ed6SGg/bpGIdnhwKyfHYoMM2kIjd0O2Ksct3mxEhmKKjLjvg7HNSloyeXarwX6aHPPpNONV0X9aBj8LmpyKZHBoipyKTTjePnM+nIZgaPnWv8Hb6dzaRj88YNseuuuzbOcexf/pNRAU56REA3MtCbjMnslh55/dffk1Qq0ukY2t4ccI/7a2CM/yvWeL9mNt0ccXu8YG+8JBzTR39tY/QtmNfpYUDL+XEQZqYZm8DYd999Y82aNbFu3TYmanZYtVqNTCazw8fp6emJfffddyeMiNmgVktic7kaG7dUYmOxHBuL9duD/wZKlRgYmmddD5CLleFAeUt5MGGQTtcDjBHBRkTT/cGgcDhwHkwWNAeL6zdsjFqmPzYVK7GpWIkNWyqxqVSJTcVqbCxWGmXg00m+EVw3B7KDge5wsFs//3oZea02HGxHNCcsRiY0Irb6FXfo/ezJZWL+LvnYd/dM9OQyMTefibn5bON2TzYd6XQqKtXB5MTISoNKbbiiIIlkRKA6OmjPpIdfrzD0S28hmx6+n01HIZuJQi7dOE6qKdAePJd0o1x0uAy5+dfp4UqKdCoG39P04DnMNv39lejr27PTwwAA6FozNoGRy+Vi4cKFnR7GNs3kzq8Mq9WS2FCsxAsD5Xh+oDz8d8tg8mCgVInNpeFf1Tc3flUf3F6u1prmTzd+fR45tzpGVx40NUUbmiIwUBpMCkxWLjMYxNbLmOfkMtGTz0Q+k4qk2jwnfXSzsuHy/KbpCU1z7ZPIpiPmzclHbyEbu+Szse/uc6O3kInenmzsUshGbz47WDadHSoRb/wqP1wyXt+WGQp8678g14PokSXJEc3z60dOgahPKYiI6MkNBenZwVLu+u18ZnYG2AAAMJvN2AQG7AwbtpTjqee3xFMvbIn+57fE2ue3RP8Lg3/XbtgSzw+U4/nN5dhQrGyz6V42nYo5+eFf1OcM/aLeW8hGPpNumnrQ9Ot1xFa/fI8uva6XnKfTEXNy2ejtyQ4mCAq52KWQiXk9g4mDwe3ZoXFkoyc7WPreamvXro0FCxa0/HUAAIDuJYHBtFDvDbClXI0tldrw7fJg/4ItI7q4N/5WqrGlNHr/etO/6lDVQLWWDN4fseLAxqHExabS6L4D83fJx4Jde2LBroV4+V69sducXOw2Jxe7Dv3bbcS/XefkGtUF+WzrEwXTVa02/aaIAAAAs4sEBpOSJEkMlKvx7KZSrN9cHvw7UI5SpRaV6vBqA5X6EoTVwSRCuVaLLaVqbCxWB/sjlAZ7JGwemgZR/7sjPRJ6hrq892QHu7zXpyHUm/WlRzQCzOXSkUmn48W79cSRi/eKfXbtiX1264l9du2Jvt3mxN67FqInt+M9SwAAANi5JDC6WLlai2c2lmLdhmKs27gl1m0oxn817hdj/eZSPLupHM9tKsVzm0tR3M4kQzoVkc2kY05ucBrF3HwmdilkY5dCJvbsLTRu71LIxtxcdjgRMdSroCeXHlpqLjPisRHL0Q01I9TRHgAAYPaTwJgltpSrccO9T8Z/bSoOT7soD6840ZiaUarG+oHBJMVzm8tjHmteTzb27C3E/F3y8ZIX9cSrXrxrzN8lH7vvko/5cwf/7j43Fy+am4tCNtNYnnB4ucHUrF2FAAAAgM6QwJgF7l79bJx17b/HI+s2NbbVl0scWa1Qr25YuOcu8ZqF82Ov3p7Yc14+9uotxF7zBv/t2WsKBQAAANOPBMYMtrlUiS/f/FB8647H4sW7zYnLT/3/4g9etoclIgEAAJh1JDBmqFW//q/41HX/Ho8/OxDvPXz/OOu4A6O34OMEAABgdhLxzjAvbCnHeTc9EN++6/FYuOcucc3ph8drFs7v9LAAAACgpSQwZpDbHlgbZ//z/fH0hi1x+hteFv//GxfrVwEAAEBXkMCYATYWK/GZf74vrr/3yThwn3nxjfceGgfv+6JODwsAAADaRgJjBvj2L34b19/7ZHzsmN+Njxz18shn050eEgAAALSVBMYM8MJAOVKpiI+/8XcjlbK6CAAAAN3HT/kzQLFai1wmLXkBAABA15LAmAFKlVoUMj4qAAAAupeoeAYoV2v6XgAAANDVRMUzQKkyOIUEAAAAupWoeAYoVxMVGAAAAHQ1UfEMUKqYQgIAAEB3ExXPAEVTSAAAAOhyouIZQBNPAAAAup2oeAawjCoAAADdTlQ8A5SqtchlU50eBgAAAHSMBMYMUK7WIq8CAwAAgC4mKp4BrEICAABAtxMVzwAlq5AAAADQ5UTFM0DJKiQAAAB0OVHxDFCq6IEBAABAdxMVzwAqMAAAAOh2ouIZoKwCAwAAgC4nKp4BStVa5FRgAAAA0MVExdNcrZZEuZqowAAAAKCrtSwq/tWvfhXLly8ftf1HP/pRnHTSSfHOd74zrrnmmla9/KxRrtUiIvTAAAAAoKtlW3HQb37zm3HDDTfEnDlzmraXy+X44he/GNdee23MmTMn3vWud8VRRx0Ve+21VyuG0XHPbSpFkiQ7dIxSZSiBoQIDAACALtaSqHi//faLiy++eNT2Rx55JPbbb7/YbbfdIp/Px6GHHhq//OUvWzGEjtuwpRx/8MXb4l8fe2GHjtNIYKjAAAAAoIu1pALjD//wD2PNmjWjtm/cuDHmzZvXuL/LLrvExo0bt3m8SqUS/f39O3WMrfbc5nIUK7X4r83VHRr70xtLERFRLg7MuPeA7uL6pBu4zukGrnO6hWudbjDbrvOWJDDG09vbG5s2bWrc37RpU1NCYzzZbDb6+vpaObSdrndLOSLui3Il2aGxl5/ZHBERL9p13ox7D+ge/f39rk9mPdc53cB1TrdwrdMNZvJ1vn79+jG3t3VewqJFi2L16tWxfv36KJVK8ctf/jJe/epXt3MIbVOf8lGq1nboOPXnm0ICAABAN2tLBcb//t//OzZv3hzvfOc741Of+lR88IMfjCRJ4qSTTooFCxa0YwhtV2+6Wapq4gkAAAA7qmUJjH333bexTOqyZcsa248++ug4+uijW/Wy00YqlYp8Jh3lnVaBkdoZwwIAAIAZyc/6LZTPpqNU2bEKjHoCJJ/J7IwhAQAAwIwkgdFC+exOqMCwjCoAAABIYLRSPpOO4k5KYOQyppAAAADQvSQwWiifTUd5B6eQWIUEAAAAJDBaqpBN7/gyqlYhAQAAAAmMVspn0ztvGVUVGAAAAHQxUXEL7YwmnmVTSAAAAEACo5XymXSjgmKq6lNQcqaQAAAA0MVExS1kCgkAAADsHKLiFirshCkkjVVIVGAAAADQxUTFLbRTKzAkMAAAAOhiouIW2ik9MCq1yKZTkU6ndtKoAAAAYOaRwGihwQqMHV+FRP8LAAAAup3IuIUK2UyUd8IUEiuQAAAA0O1Exi2Uz+6cZVRVYAAAANDtRMYttHOaeCYaeAIAAND1RMYtlM+ko1JLolabehJDBQYAAABIYLRUPfGwI408y5WaCgwAAAC6nsi4hQo7IYGhAgMAAAAkMFqqUYGxA408B1chSe2sIQEAAMCMJIHRQvWpHzuUwFCBAQAAABIYrVRPPBR3sAIjn83srCEBAADAjCSB0UI7awpJ3hQSAAAAupwERgsVhiondiSBUTaFBAAAACQwWml4GdXqlI9RqtYiZxlVAAAAupzIuIXqTTx3uAeGBAYAAABdTmTcQjujB4YpJC2KqScAACAASURBVAAAACCB0VKFnZDAKFZMIQEAAACRcQsN98DYsSkkBRUYAAAAdDmRcQvVe1eYQgIAAAA7RmTcQvXEw1SbeFaqtaglYQoJAAAAXU9k3EI72sSzPvVEBQYAAADdTmTcQjvaxLNcSSIiLKMKAABA1xMZt9CONvEsVqsREZFTgQEAAECXExm3UL1yYqo9MMrVwQqMggoMAAAAupzIuIVSqVTkMqmp98Co6IEBAAAAERIYLZffCQkMq5AAAADQ7UTGLZbLpKM01Mtie5WtQgIAAAARIYHRcjsyhaRoCgkAAABEhARGy+UzqSk38RyeQpLamUMCAACAGUcCo8VymfSUKzDqU0gKKjAAAADociLjFtsZTTzzmczOHBIAAADMOBIYLZbPpKNUnWICY+h5uawpJAAAAHQ3CYwWy+1AD4zGKiSWUQUAAKDLiYxbbEemkBQbTTx9TAAAAHQ3kXGL7UgTz/rzNPEEAACg24mMWyyXSU25B0ZjCokEBgAAAF1OZNxiO2MVElNIAAAA6HYi4xbbGVNIVGAAAADQ7UTGLZbPpKJYqU7pueVqLVKpiGzaMqoAAAB0NwmMFtuRCoxitRa5TDpSKQkMAAAAupsERovls1Nv4lmq1KKg/wUAAABIYLRaPpOKcjWJWi3Z7ueWqzX9LwAAACAkMFquvoLIVKowSpWaFUgAAAAgJDBaLp8Z7F8xlQRGuZqowAAAAICQwGi5RgXGFBp5liqmkAAAAECEBEbLNSowppDAKJpCAgAAABEhgdFyuR1IYGjiCQAAAINExy1Wr6AoTnEKiWVUAQAAoEUJjFqtFuecc068853vjOXLl8fq1aubHv+Hf/iH+KM/+qM46aST4pZbbmnFEKaNHZlCUqrWIpdN7ewhAQAAwIyTbcVBb7311iiVSnH11VfHvffeG+eff35ccsklERHxwgsvxBVXXBH/8i//EgMDA3HiiSfGscce24phTAv1KSClanW7n1uu1mLXnpZ8RAAAADCjtKQC4+67746lS5dGRMSSJUvi/vvvbzw2Z86cePGLXxwDAwMxMDAQqdTsrjCoV2BMdQqJJp4AAADQogqMjRs3Rm9vb+N+JpOJSqUS2ezgy/X19cXxxx8f1Wo1Tj/99G0er1KpRH9/fyuG2nL1Jp7PPPd89PeXt+u5m4uliFp+xp473cV1SjdwndMNXOd0C9c63WC2XectSWD09vbGpk2bGvdrtVojeXH77bfH008/HbfddltERHzwgx+MQw45JA4++ODxB5nNRl9fXyuG2nIPPf1IREQU5uwSfX37bNdza/FA9M7tmbHnTvfo7+93nTLruc7pBq5zuoVrnW4wk6/z9evXj7m9JfMTDjnkkLj99tsjIuLee++NxYsXNx7bbbfdoqenJ/L5fBQKhZg3b1688MILrRjGtNBYRrU6tSkkeVNIAAAAoDUVGMcee2ysWrUqTj755EiSJM4777xYuXJl7LfffnHMMcfEz372s3jHO94R6XQ6DjnkkHj961/fimFMC7kdXIWk3gQUAAAAullLEhjpdDpWrFjRtG3RokWN2x/96Efjox/9aCteetqpV1BMJYFRVoEBAAAAEdGiKSQMy+3IKiTVWuRUYAAAAIAERqtNtQKjVkuiXE1UYAAAAEBIYLTcVJt4lmuD++uBAQAAABIYLZef4hSSesWGCgwAAACQwGi5VCoV+Ux6u6eQlKtJRKjAAAAAgAgJjLbIZ7c/gdGowJDAAAAAAAmMdshn01GqVrfrOfUERs4UEgAAAJDAaIepTCGpN/1UgQEAAAASGG2xQ1NIVGAAAACABEY7DE4hmWoFRqoVQwIAAIAZRQKjDfKZdBTL27sKSb0CI9OKIQEAAMCMIoHRBlOqwLAKCQAAADSIjtugkE1HcYo9MHIZU0gAAABAAqMNptTE0yokAAAA0CA6boOCVUgAAABgh4iO20APDAAAANgxouM2yGe2vwKjbAoJAAAANIiO22BHemDkTCEBAAAACYx2MIUEAAAAdozouA3ymUwUy9Xtek5jFRIVGAAAACCB0Q47VIEhgQEAAAASGO2Qz6ajXE2iVksm/ZxSpRbZdCrS6VQLRwYAAAAzgwRGGxSG+lhsTxVGuVrT/wIAAACGiJDbYCoJjFKlZgUSAAAAGCJCboN6JcX2LKVaqiYqMAAAAGCICLkN6o04tyuBUalp4AkAAABDRMhtMLUKDD0wAAAAoE6E3Ab5qTTxVIEBAAAADSLkNqgnIoplFRgAAAAwFSLkNhiuwKhO+jmDq5CkWjUkAAAAmFEkMNqgnsAo6oEBAAAAUyJCboNCNhMRU1iFZOh5AAAA0O0kMNqgMJVVSCq1yJtCAgAAABEhgdEWU1qFxBQSAAAAaBAht0F9FZLtqsCo1iJnGVUAAACICAmMtshPeQqJjwcAAAAiJDDawhQSAAAA2DEi5DaYSgVGsWIKCQAAANSJkNugPhWkuJ1TSAoqMAAAACAiJDDaYioJDFNIAAAAYJgIuQ3S6VTkM+lJTyGpVGtRS8IUEgAAABgiQm6TfHbyCYxyNWk8BwAAAJDAaJt8Nh2lanVS+9YTHZZRBQAAgEEi5DbZnikkxaFER04FBgAAAESEBEbbTGUKSUEFBgAAAESEBEbbDE4hmVwCozGFRAUGAAAARIQERttszxSS+n5WIQEAAIBBIuQ2yWfTUZz0FBIVGAAAADCSCLlNtieBUTSFBAAAAJqIkNuksB1NPIenkKRaOSQAAACYMSQw2mR7Ehj1KSQFFRgAAAAQERIYbTOVVUg08QQAAIBBIuQ22a5VSDTxBAAAgCYi5DbJT2EKSV4FBgAAAESEBEbbbM8UkqIpJAAAANBEhNwm+Uxmu1ch0cQTAAAABomQ2ySfTUexUp3UvmU9MAAAAKCJCLlN8tl0lKtJ1GrJNve1CgkAAAA0EyG3SX06yGT6YNQTGCowAAAAYFC2FQet1Wrxuc99Lh566KHI5/PxhS98Ifbff//G4z/96U/j61//ekREHHTQQfHZz342UqlUK4YybYxMYPTkMhPuW67WIpWKyKZn93sCAAAAk9WSn/hvvfXWKJVKcfXVV8cZZ5wR559/fuOxjRs3xgUXXBB/93d/F9dcc0285CUvieeee64Vw5hW6tUUk2nkWazWIpdJz/qkDgAAAExWSxIYd999dyxdujQiIpYsWRL3339/47F77rknFi9eHF/60pfi3e9+d+y5554xf/78VgxjWslnJp/AKFeSKOh/AQAAAA0tmUKycePG6O3tbdzPZDJRqVQim83Gc889F3feeWdcf/31MXfu3HjPe94TS5YsiYULF457vEqlEv39/a0Yalv09/fH5o0vRETE2nXPRGpg4ikk61/YENlMxNq1a6NWm9zSq9BpM/m/UZgs1zndwHVOt3Ct0w1m23XekgRGb29vbNq0qXG/VqtFNjv4Ui960Yvi937v92KvvfaKiIjDDjssHnjggQkTGNlsNvr6+lox1Jbr7++Pvr6+2HtdEhGrY5ddd4u+BfMmfE628HQUstlYsGBBewYJO6h+ncNs5jqnG7jO6RaudbrBTL7O169fP+b2lsxTOOSQQ+L222+PiIh77703Fi9e3HjsVa96VTz88MPx7LPPRqVSiV/96lfx8pe/vBXDmFa2awpJNbECCQAAAIzQkgqMY489NlatWhUnn3xyJEkS5513XqxcuTL222+/OOaYY+KMM86I0047LSIijjvuuKYEx2xVT0gUJ5HAKFVqEhgAAAAwQksSGOl0OlasWNG0bdGiRY3bxx9/fBx//PGteOlpaziBUd3mvsXK4CokAAAAwKBJR8mPPfZY/PSnP42nnnoqkiRp5ZhmpcJ2LKNarqrAAAAAgJEmVYFx5ZVXxi233BLPP/98nHjiifHb3/42zjnnnFaPbVbJZwZXHplMAqNUqVlGFQAAAEaYVJR80003xeWXXx7z5s2L97///fGrX/2q1eOadQq5oQqM6iQSGNVa5LKpVg8JAAAAZoxJJTDqU0ZSqcGgOp/Pt25Es9T2rUJSa+wPAAAATHIKyQknnBDvec974sknn4wPfehD8cY3vrHV45p18tvRA6OkiScAAAA0mVQC45RTTok/+IM/iP/8z/+Ml73sZXHAAQe0elyzTiOBMZkpJJZRBQAAgCaTipKvueaauOaaa+LNb35zfOlLX4rrr7++1eOadbarAsMqJAAAANBkUlHyVVddFWeccUZERFx66aVx1VVXtXRQs1G9p0VxklNI9MAAAACAYZOKktPpdBQKhYiIyOVyjWaeTN52JTBUYAAAAECTSfXAOOaYY+Ld7353HHzwwfEf//EfcfTRR7d6XLNOOp2KXCY1uVVIVGAAAABAk0klMD784Q/HUUcdFb/5zW/ixBNPjAMPPLDV45qV8pn0pHtg5FRgAAAAQMOEUfJ3vvOdiIj427/92/jBD34QDz74YHz/+9+P//E//kdbBjfbFHKZKFWrE+5TqyVRriYqMAAAAGCECSsw9tlnn4iI2H///SOTybRlQLPZZCowyrXBx/XAAAAAgGETJjCWLl0aERHf//7347LLLmvLgGazfHbbCYz64yowAAAAYNikemDMmzcvbr311li4cGGk04OB9cKFC1s6sNkon01HqbqNCoxq0tgXAAAAGLTNBMbGjRtjzZo18a1vfauxLZVKxT/+4z+2dGCz0WSmkDQqMCQwAAAAoGHCBMaVV14Zl112WWQymfjYxz4WRx55ZLvGNSvls+koTjKBkTOFBAAAABomjJJvvPHGuPnmm+Pqq69WcbETTCqBUVWBAQAAAFubMErO5/ORz+dj/vz5US6X2zWmWaugiScAAABMyaSj5CRJWjmOrjCpHhiNCoxUO4YEAAAAM8KEPTB+/etfxxlnnBFJkjRu1/3t3/5tywc320xuFZJ6BUamHUMCAACAGWHCBMbXvva1xu2TTz655YOZ7bZnCkkuowIDAAAA6iZMYLzmNa9p1zi6Qn57emBo4gkAAAANouQ2mswUEquQAAAAwGii5DbKZzJWIQEAAIApECW3kSkkAAAAMDWi5DaqTyGZaEnasikkAAAAMIoouY0KQ0mJ4gRVGPUeGDlTSAAAAKBBlNxG9b4WEzXyNIUEAAAARhMlt1E9KTFRH4zGKiQqMAAAAKBBlNxGhckkMKxCAgAAAKOIkttoMhUY5WotsulUpNOpdg0LAAAApj0JjDZqJDC20QND/wsAAABoJlJuo0YTz21MIbECCQAAADQTKbdRflLLqCYqMAAAAGArIuU2mtQqJJWaBp4AAACwFZFyGxUaFRjVcfcpVfXAAAAAgK2JlNson8lExDZWIVGBAQAAAKOIlNtoUquQqMAAAACAUUTKbTTZHhi5TKpdQwIAAIAZQQKjjQqTSWCowAAAAIBRRMptNKkpJJVa5PTAAAAAgCYi5Taa7BSSggoMAAAAaCJSbqP66iLFiVYhMYUEAAAARhEpt1E9gbGtHhimkAAAAEAzkXIbpdOpyGVSE1ZglCq1RqIDAAAAGCRSbrN8Jj1hBYYpJAAAADCaSLnN8tl0lKrVcR8vWoUEAAAARhEpt1k+O3EFhlVIAAAAYDSRcpsVshlTSAAAAGA7iZTbbHAKydgJjEq1FrUkTCEBAACArYiU22yiJp7lajK4jwoMAAAAaCJSbrN8Nj3uMqr1xIZlVAEAAKCZSLnNJmriWRxanSSnAgMAAACaiJTbrDBBD4z6FJKCCgwAAABoIlJus3wmHcXyNqaQqMAAAACAJiLlNptoFZJ6AsMqJAAAANBMpNxmE/XAKFdVYAAAAMBYRMptVpioiacpJAAAADCmlkTKtVotzjnnnHjnO98Zy5cvj9WrV4+5z2mnnRZXXXVVK4YwbU1uCkmqnUMCAACAaa8lCYxbb701SqVSXH311XHGGWfE+eefP2qfr33ta/H888+34uWntXwms80pJAUVGAAAANCkJZHy3XffHUuXLo2IiCVLlsT999/f9PjNN98cqVQqjjzyyFa8/LQ2UQ8MTTwBAABgbNlWHHTjxo3R29vbuJ/JZKJSqUQ2m42HH344brzxxrjooovi61//+qSOV6lUor+/vxVDbYuRYy8NbIpStRZr166NWq05kbH2v56LiIjNGzdEf/9AW8cIO2om/zcKk+U6pxu4zukWrnW6wWy7zluSwOjt7Y1NmzY17tdqtchmB1/q+uuvj7Vr18b73ve+eOKJJyKXy8VLXvKSCasxstls9PX1tWKoLdff39809j123xQRT8WL9tgzCtlM0769awcTGnvvsXv07dUbMFNsfZ3DbOQ6pxu4zukWrnW6wUy+ztevXz/m9pYkMA455JD48Y9/HG95y1vi3nvvjcWLFzceO+ussxq3L7744thzzz27aipJfmh6SLFSG5XAKJpCAgAAAGNqSQLj2GOPjVWrVsXJJ58cSZLEeeedFytXroz99tsvjjnmmFa85IxRXyJ1rD4Y9W2aeAIAAECzliQw0ul0rFixomnbokWLRu33l3/5l614+WltogRGfRWSvAQGAAAANBEpt1l9CslEFRimkAAAAEAzkXKbFXJDCYyqCgwAAACYLJFym22rAiOVisimU+0eFgAAAExrEhhtVq+uKI6RwChWa5HLpCOVksAAAACAkSQw2mzCJp6VJAr6XwAAAMAoouU2qy+ROlYPjFK1qv8FAAAAjEG03Gb5TCYixu+BYQUSAAAAGE203GbDPTCqox4rVxMVGAAAADAG0XKbTdQDo1SpSWAAAADAGETLbTZRAqNoCgkAAACMSbTcZhM18SxXVWAAAADAWETLbbbNKSSZVLuHBAAAANOeBEab5TP1Jp5jLaOqAgMAAADGIlpus3oCY6wKjHK11ngcAAAAGCZabrN0OhW5TGrMHhglTTwBAABgTKLlDshn0pZRBQAAgO0gWu6AfDYdxUp11HY9MAAAAGBsouUOyGcnqMAwhQQAAABGES13wLgJDBUYAAAAMCbRcgfkM+kxm3iWVWAAAADAmETLHVDIZsatwMipwAAAAIBRRMsdMNjEszmBkSRJlKuJCgwAAAAYg2i5A8bqgVGfUqIHBgAAAIwmWu6AQnZ0D4x6QkMFBgAAAIwmWu6AfGZ0BUa5mgw+pgIDAAAARhEtd8BYPTAaFRgSGAAAADCKaLkDxuyBMXQ/ZwoJAAAAjCJa7oCxppBo4gkAAADjEy13QF4TTwAAANguouUOKGQzE1RgpDoxJAAAAJjWJDA6YKweGOV6AiOT6cSQAAAAYFqTwOiA+hSSJEka24abeKrAAAAAgK1JYHRAYahR58g+GJZRBQAAgPGJljug3qhz5DQSq5AAAADA+ETLHVBPUjQlMKxCAgAAAOMSLXdAPYFRHCuBoQIDAAAARhEtd8BYU0jKppAAAADAuETLHZAfq4lntb4KiY8EAAAAtiZa7oAJe2CowAAAAIBRRMsdUBirB0ZVE08AAAAYj2i5A6xCAgAAANtHtNwBhTF6YJSrtcimU5FOpzo1LAAAAJi2JDA6IJ/JRMToCgz9LwAAAGBsIuYOGG8KiRVIAAAAYGwi5g7IN5p4VhvbStVEBQYAAACMQ8TcAeNVYGjgCQAAAGMTMXdAPVExsolnqaoHBgAAAIxHxNwBY1VglFVgAAAAwLhEzB1QaPTAUIEBAAAAkyFi7oDGFJJRq5CkOjUkAAAAmNYkMDognU5FLpPSAwMAAAAmScTcIflMeowKDB8HAAAAjEXE3CH57OgERkEFBgAAAIxJxNwh+Ww6ipVq437ZFBIAAAAYl4i5Q0ZVYFRNIQEAAIDxiJg7JJ9JNzfxrNQaq5MAAAAAzUTMHZLPZpoqMEwhAQAAgPGJmDukkE1HcUQCo2gVEgAAABiXiLlDtu6BUa5ahQQAAADGk23FQWu1Wnzuc5+Lhx56KPL5fHzhC1+I/fffv/H45ZdfHjfddFNERLzhDW+Iv/iLv2jFMKa1QjYdG4uVxv1SxRQSAAAAGE9LIuZbb701SqVSXH311XHGGWfE+eef33js8ccfjxtuuCG+/e1vx9VXXx3/+q//Gg8++GArhjGt5TPDFRiVai1qSZhCAgAAAONoSQXG3XffHUuXLo2IiCVLlsT999/feGyfffaJv//7v49MJhMREZVKJQqFQiuGMa2NnEJSriaNbQAAAMBoLUlgbNy4MXp7exv3M5lMVCqVyGazkcvlYv78+ZEkSXz5y1+Ogw46KBYuXDjh8SqVSvT397diqG0x1tir5WIMlMrx1FNPxfMD5YiIqBS3zOjzpLu5dukGrnO6geucbuFapxvMtuu8JQmM3t7e2LRpU+N+rVaLbHb4pYrFYvz3//7fY5dddonPfvaz2x5kNht9fX2tGGrL9ff3jzn23Xqfjkr/QOyzzz6R3rBlcNuuvTP2POlu413nMJu4zukGrnO6hWudbjCTr/P169ePub0lcxYOOeSQuP322yMi4t57743Fixc3HkuSJD784Q/HAQccECtWrGhMJek2+Ww6StXmKSQFPTAAAABgTC2pwDj22GNj1apVcfLJJ0eSJHHeeefFypUrY7/99otarRa/+MUvolQqxf/5P/8nIiI+8YlPxKtf/epWDGXaGtkDo/5XDwwAAAAYW0sSGOl0OlasWNG0bdGiRY3b9913XytedkYZK4FhFRIAAAAYm4i5QwrZTJSqtUiSJMpVFRgAAAAwERFzhxSGkhWlai2KjQqMVCeHBAAAANOWBEaH5Iemi5QqNT0wAAAAYBtEzB1ST1aUKrXGFJKCBAYAAACMScTcIfkRU0g08QQAAICJiZg7pD6FpFiuRUkTTwAAAJiQiLlDRlZgNFYhUYEBAAAAYxIxd8jIHhhFU0gAAABgQiLmDqknMIqaeAIAAMA2iZg7pJC1jCoAAABMloi5QwpWIQEAAIBJEzF3SD6TiYjBCoyyVUgAAABgQiLmDslvNYUklYrIplMdHhUAAABMTxIYHTK8jGo1itVa5DLpSKUkMAAAAGAsEhgd0liFpFyLciWJgv4XAAAAMC5Rc4fkMyOaeFar+l8AAADABETNHbJ1DwwrkAAAAMD4RM0dUl9GtVipRbmaqMAAAACACYiaO6QxhWSoAkMCAwAAAMYnau6QdDoVuUwqStVaFE0hAQAAgAmJmjson0lHqVKLclUFBgAAAExE1NxB+Wx6eApJJtXp4QAAAMC0JYHRQY0EhgoMAAAAmJCouYPy2XSUqkNTSPTAAAAAgHGJmjson0lHsVKNkiaeAAAAMCFRcwflsxnLqAIAAMAkiJo7KJ9NR1EPDAAAANgmUXMHFTIjVyHxUQAAAMB4RM0dVMiNaOKpAgMAAADGJWruoLwKDAAAAJgUUXMH5bNDCYxqLXIqMAAAAGBcouYOqjfxLFcTFRgAAAAwAVFzB+Uz6dhUrAzeVoEBAAAA4xI1d1A+m44N9QSGCgwAAAAYl6i5g+o9MOq3AQAAgLGJmjtoZNJCAgMAAADGJ2ruoMKIaSM5U0gAAABgXKLmDlKBAQAAAJMjau6gQjbTuK2JJwAAAIxP1NxBzRUYqQ6OBAAAAKY3CYwOakpgZDIT7AkAAADdTQKjg/JNTTxVYAAAAMB4JDA6SBNPAAAAmBxRcwdJYAAAAMDkiJo7qLkHho8CAAAAxiNq7qBCRgUGAAAATIaouYNMIQEAAIDJETV3UCE7vHRqzhQSAAAAGJeouYNUYAAAAMDkiJo7SBNPAAAAmBxRcwdJYAAAAMDkiJo7qJ60yKZTkU6nOjwaAAAAmL4kMDqoXoGh/wUAAABMTOTcQYWhxIUVSAAAAGBiIucOqk8hUYEBAAAAExM5d1A6nYpsOqWBJwAAAGyDyLnD8tm0CgwAAADYBpFzhxWyaRUYAAAAsA0i5w7L/7/27j+mqvqP4/jrwOXS9GKMLVtOKS8/NltjgI50Idmcs9z6xzVC2jXnXzaaomgQhl2DITdcP2atmc5oFIlO11qrtrTyZjlqFDmYyWaFC6+F9ot7k8u93vP947vu90vx1U2533vu8fn4i3M/2+e+YS8d98Xn3uNIU4aDW6gCAAAAAHAlCSkwYrGYtm3bpocfflgej0dDQ0MT1g8cOKCVK1eqsrJSH3/8cSJGSBlOTmAAAAAAAHBVjkRseuTIEY2Pj6u7u1t9fX1qa2vTK6+8IkkaGRlRZ2enDh06pHA4rOrqat1zzz1yOp2JGMXynOl8BgYAAAAAAFeTkFfOvb29Wrx4sSSpuLhY/f398bWTJ0+qpKRETqdTWVlZys3N1bfffpuIMVKC05GuDE5gAAAAAABwRQk5gREMBuVyueLX6enpikajcjgcCgaDysrKiq9Nnz5dwWDwivtFo1EFAoFEjPp/caXZlxfMUPb0zJT+/gDpyjkH7IKc40ZAznGjIOu4Edgt5wkpMFwul0KhUPw6FovJ4XBMuhYKhSYUGpMO6XDotttuS8SoCRcIBK44e22Kfl/Af7tazgE7IOe4EZBz3CjIOm4EqZzz3377bdLHE/LehdLSUvn9fklSX1+fCgsL42tFRUXq7e1VOBzW6Oiozpw5M2EdAAAAAADg7xJyAmPZsmX67LPPVFVVJdM01draqtdee025ublaunSpPB6PqqurZZqmNm7cqMzMzESMAQAAAAAAbCIhBUZaWpqeeeaZCY/l5eXFv66srFRlZWUinhoAAAAAANgQt78AAAAAAACWR4EBAAAAAAAsjwIDAAAAAABYHgUGAAAAAACwPAoMAAAAAABgeRQYAAAAAADA8igwAAAAAACA5VFgAAAAAAAAy6PAAAAAAAAAlkeBAQAAAAAALI8CAwAAAAAAWB4FBgAAAAAAsDwKDAAAAAAAYHkUGAAAAAAAwPIM0zTNZA9xNX19fcrMzEz2GAAAAAAAIMHC4bCKi4v/8XhKFBgAAAAAAODGxltIAAAAAACA5VFgAAAAAAAAy6PAAAAAAAAAlkeBAQAAAAAAPQbWfQAAB3FJREFULI8CAwAAAAAAWJ4j2QPYUSwWk9fr1enTp+V0OtXS0qLbb7892WMBUyISiaixsVHDw8MaHx/XY489pvz8fDU0NMgwDBUUFOjpp59WWhr9KFLfxYsXtXLlSu3bt08Oh4Ocw3Z2796tjz76SJFIRKtWrVJZWRk5h+1EIhE1NDRoeHhYaWlpam5u5v902Mo333yjnTt3qrOzU0NDQ5Nm+6WXXtInn3wih8OhxsZGFRUVJXvsa8K/0gQ4cuSIxsfH1d3drbq6OrW1tSV7JGDKvPPOO8rOzlZXV5f27Nmj5uZm7dixQ7W1terq6pJpmjp69GiyxwSuWyQS0bZt23TTTTdJEjmH7fT09Ojrr7/WW2+9pc7OTp0/f56cw5aOHTumaDSq/fv3q6amRi+88AJZh23s2bNHTz31lMLhsKTJf18ZGBjQF198oYMHD+q5557T9u3bkzz1taPASIDe3l4tXrxYklRcXKz+/v4kTwRMnfvvv18bNmyIX6enp2tgYEBlZWWSpIqKCn3++efJGg+YMj6fT1VVVZo5c6YkkXPYzvHjx1VYWKiamhqtW7dOS5YsIeewpblz5+ry5cuKxWIKBoNyOBxkHbaRm5urXbt2xa8ny3Zvb6/Ky8tlGIZmzZqly5cv65dffknWyNeFAiMBgsGgXC5X/Do9PV3RaDSJEwFTZ/r06XK5XAoGg1q/fr1qa2tlmqYMw4ivj46OJnlK4PocPnxYOTk58TJaEjmH7fz666/q7+/Xiy++qO3bt2vz5s3kHLY0bdo0DQ8P64EHHlBTU5M8Hg9Zh20sX75cDsd/Phlismz//fVpKmeez8BIAJfLpVAoFL+OxWITQgWkukAgoJqaGlVXV+vBBx9Ue3t7fC0UCmnGjBlJnA64focOHZJhGDpx4oROnTql+vr6CX+pIOewg+zsbLndbjmdTrndbmVmZur8+fPxdXIOu+jo6FB5ebnq6uoUCAT06KOPKhKJxNfJOuzkvz/L5a9s//31aSgUUlZWVjLGu26cwEiA0tJS+f1+SVJfX58KCwuTPBEwdS5cuKC1a9dqy5YteuihhyRJd955p3p6eiRJfr9fCxYsSOaIwHV788039cYbb6izs1Pz5s2Tz+dTRUUFOYetzJ8/X59++qlM09RPP/2kS5cuadGiReQctjNjxoz4i7Wbb75Z0WiU311gW5Nlu7S0VMePH1csFtO5c+cUi8WUk5OT5EmvjWGappnsIezmr7uQDA4OyjRNtba2Ki8vL9ljAVOipaVF77//vtxud/yxrVu3qqWlRZFIRG63Wy0tLUpPT0/ilMDU8Xg88nq9SktLU1NTEzmHrTz77LPq6emRaZrauHGjZs+eTc5hO6FQSI2NjRoZGVEkEtHq1at11113kXXYxo8//qhNmzbpwIED+v777yfN9q5du+T3+xWLxfTkk0+mbGlHgQEAAAAAACyPt5AAAAAAAADLo8AAAAAAAACWR4EBAAAAAAAsjwIDAAAAAABYHgUGAAAAAACwPEeyBwAAAKmrra1NAwMDGhkZ0djYmObMmSOHw6H58+fr8ccfn9Ln+vDDD1VUVKRbb711SvcFAACpgduoAgCA63b48GF999132rx5c8Kew+PxyOv1Ki8vL2HPAQAArIsTGAAAYEr19PRo//79ev7557Vs2TKVlJRoaGhICxcu1OjoqE6ePKm5c+eqvb1dgUBATU1NCofDyszMVHNzs3JycrRhwwYFg0GNjY1py5YtunTpkk6dOqX6+np1dXWpu7tb7777rgzD0IoVK7R69Wo1NDTINE0FAgH9+eef8vl8mj179j/2uvvuu5P9IwIAANeAAgMAACTM8PCwXn/9dd1yyy0qKyvTwYMH1dTUpKVLl+qPP/6Qz+eTx+PRvffeqxMnTmjnzp1at26dLly4oI6ODl28eFE//PCDlixZonnz5snr9ers2bN677331NXVJcMwtGbNGpWXl0uS5syZI5/Pp2PHjqm9vV11dXX/2AsAAKQmCgwAAJAw2dnZmjVrliRp2rRpys/PlyRlZWUpHA5rcHBQu3fv1t69e2WapjIyMlRQUKBHHnlEmzZtUjQalcfjmbDn4OCgzp07pzVr1kiSfv/9d509e1aStHDhQklSSUmJWltbr7oXAABIHRQYAAAgYQzDuOK62+3W2rVrVVpaqjNnzujLL7/U6dOnFQqF9Oqrr+rnn39WVVWV7rvvPhmGIdM05Xa7lZ+fr71798owDHV0dKiwsFAffPCBBgYGtGDBAn311VcqKCj4n3sBAIDUQ4EBAACSpr6+Xl6vV+FwWGNjY9q6davuuOMOvfzyy3r77beVkZGh9evXS/r3qYonnnhC+/bt06JFi7Rq1SqNj49PuDOJ3+/X0aNHFYvFtGPHDs2cOXPSvQAAQOrhLiQAAMAWGhoatGLFClVUVCR7FAAAkABpyR4AAAAAAADgajiBAQAAAAAALI8TGAAAAAAAwPIoMAAAAAAAgOVRYAAAAAAAAMujwAAAAAAAAJZHgQEAAAAAACyPAgMAAAAAAFjevwAiUCT3K06mDQAAAABJRU5ErkJggg==\n",
+      "text/plain": [
+       "<Figure size 1080x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1080x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "experiments = experiments.sort_values(by =['rules_price']).reset_index(drop=True)\n",
+    "# df = experiments\n",
+    "cols = 1\n",
+    "rows = 1\n",
+    "cc_idx = 0\n",
+    "\n",
+    "# len(experiments)\n",
+    "# df = experiments.copy()\n",
+    "# print(df.head())\n",
+    "\n",
+    "while cc_idx<len(experiments):\n",
+    "    cc = experiments.iloc[cc_idx]['rules_price']\n",
+    "    cc_label = experiments.iloc[cc_idx]['rules_price']\n",
+    "#     secondary_label = experiments.iloc[cc_idx]['string']\n",
+    "    sub_experiments = experiments[experiments['rules_price']==cc]\n",
+    "    cc_idx += len(sub_experiments)\n",
+    "    fig, axs = plt.subplots(ncols=cols, nrows=rows, figsize=(15*cols,7*rows))\n",
+    "    sub_experiments.reset_index(drop=False)\n",
+    "    for i, experiment in sub_experiments.iterrows():\n",
+    "        df = experiment['dataset'].copy()\n",
+    "        # filter out starting point\n",
+    "        df = df[df.timestep > 0] \n",
+    "        # FIRST RUN ONLY\n",
+    "        df = df[df.run == 1]       \n",
+    "#         print(df['price'])\n",
+    "\n",
+    "        df = df.groupby('timestep').agg({'price': ['min', 'mean', 'max']}).reset_index()\n",
+    "\n",
+    "#         print(df.head())\n",
+    "        plot_label = experiment['rules_price']\n",
+    "        ax = axs\n",
+    "        title = 'Price' + '\\n' + 'Scenario: ' + str(cc_label)  + ' rules_price'\n",
+    "        ax.set_title(title)\n",
+    "        ax.set_ylabel('Price')\n",
+    "        colors = ['b','orange', 'g', 'magenta', 'r']\n",
+    "        \n",
+    "#         ax.plot(df.timestep, df['price'],color = colors[0], label='Price')\n",
+    "        df.plot(x='timestep', y=('price','mean'), label='Mean', ax=ax, legend=True)\n",
+    "        ax.fill_between(df.timestep, df[('price','min')], df[('price','max')], alpha=0.3) \n",
+    "\n",
+    "\n",
+    "        ax.legend()\n",
+    "\n",
+    "        ax.set_xlabel('Timesteps')\n",
+    "        ax.grid(color='0.9', linestyle='-', linewidth=1)\n",
+    "\n",
+    "        plt.tight_layout()\n",
+    "        \n",
+    "fig.tight_layout(rect=[0, 0, 1, .97])\n",
+    "fig.patch.set_alpha(1)\n",
+    "display(fig)\n",
+    "plt.close()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1080x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1080x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "experiments = experiments.sort_values(by =['rules_price']).reset_index(drop=True)\n",
+    "# df = experiments\n",
+    "cols = 1\n",
+    "rows = 1\n",
+    "cc_idx = 0\n",
+    "\n",
+    "# len(experiments)\n",
+    "# df = experiments.copy()\n",
+    "# print(df.head())\n",
+    "\n",
+    "while cc_idx<len(experiments):\n",
+    "    cc = experiments.iloc[cc_idx]['rules_price']\n",
+    "    cc_label = experiments.iloc[cc_idx]['rules_price']\n",
+    "#     secondary_label = experiments.iloc[cc_idx]['string']\n",
+    "    sub_experiments = experiments[experiments['rules_price']==cc]\n",
+    "    cc_idx += len(sub_experiments)\n",
+    "    fig, axs = plt.subplots(ncols=cols, nrows=rows, figsize=(15*cols,7*rows))\n",
+    "    sub_experiments.reset_index(drop=False)\n",
+    "    for i, experiment in sub_experiments.iterrows():\n",
+    "        df = experiment['dataset'].copy()\n",
+    "        # filter out starting point\n",
+    "        df = df[df.timestep > 0] \n",
+    "        # FIRST RUN ONLY\n",
+    "        df = df[df.run == 1]  \n",
+    "\n",
+    "        df = df.groupby('timestep').agg({'spot_price': ['min', 'mean', 'max']}).reset_index()\n",
+    "\n",
+    "#         print(df.head())\n",
+    "        plot_label = experiment['rules_price']\n",
+    "        ax = axs\n",
+    "        title = 'spot_price' + '\\n' + 'Scenario: ' + str(cc_label)  + ' rules_price'\n",
+    "        ax.set_title(title)\n",
+    "        ax.set_ylabel('spot_price')\n",
+    "        colors = ['b','orange', 'g', 'magenta', 'r']\n",
+    "        \n",
+    "#         ax.plot(df.timestep, df['price'],color = colors[0], label='Price')\n",
+    "        df.plot(x='timestep', y=('spot_price','mean'), label='Mean', ax=ax, legend=True)\n",
+    "        ax.fill_between(df.timestep, df[('spot_price','min')], df[('spot_price','max')], alpha=0.3) \n",
+    "\n",
+    "\n",
+    "        ax.legend()\n",
+    "\n",
+    "        ax.set_xlabel('Timesteps')\n",
+    "        ax.grid(color='0.9', linestyle='-', linewidth=1)\n",
+    "\n",
+    "        plt.tight_layout()\n",
+    "        \n",
+    "fig.tight_layout(rect=[0, 0, 1, .97])\n",
+    "fig.patch.set_alpha(1)\n",
+    "display(fig)\n",
+    "plt.close()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1080x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1080x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "experiments = experiments.sort_values(by =['rules_price']).reset_index(drop=True)\n",
+    "# df = experiments\n",
+    "cols = 1\n",
+    "rows = 1\n",
+    "cc_idx = 0\n",
+    "\n",
+    "# len(experiments)\n",
+    "# df = experiments.copy()\n",
+    "# print(df.head())\n",
+    "\n",
+    "while cc_idx<len(experiments):\n",
+    "    cc = experiments.iloc[cc_idx]['rules_price']\n",
+    "    cc_label = experiments.iloc[cc_idx]['rules_price']\n",
+    "#     secondary_label = experiments.iloc[cc_idx]['string']\n",
+    "    sub_experiments = experiments[experiments['rules_price']==cc]\n",
+    "    cc_idx += len(sub_experiments)\n",
+    "    fig, axs = plt.subplots(ncols=cols, nrows=rows, figsize=(15*cols,7*rows))\n",
+    "    sub_experiments.reset_index(drop=False)\n",
+    "    for i, experiment in sub_experiments.iterrows():\n",
+    "        df = experiment['dataset'].copy()\n",
+    "        # filter out starting point\n",
+    "        df = df[df.timestep > 0] \n",
+    "        # FIRST RUN ONLY\n",
+    "        df = df[df.run == 1]  \n",
+    "        df = df.groupby('timestep').agg({'alpha': ['min', 'mean', 'max']}).reset_index()\n",
+    "\n",
+    "#         print(df.head())\n",
+    "        plot_label = experiment['rules_price']\n",
+    "        ax = axs\n",
+    "        title = 'Alpha' + '\\n' + 'Scenario: ' + str(cc_label)  + ' rules_price'\n",
+    "        ax.set_title(title)\n",
+    "        ax.set_ylabel('Alpha')\n",
+    "        colors = ['b','orange', 'g', 'magenta', 'r']\n",
+    "        \n",
+    "#         ax.plot(df.timestep, df['price'],color = colors[0], label='Price')\n",
+    "        df.plot(x='timestep', y=('alpha','mean'), label='Mean', ax=ax, legend=True)\n",
+    "        ax.fill_between(df.timestep, df[('alpha','min')], df[('alpha','max')], alpha=0.3) \n",
+    "\n",
+    "\n",
+    "        ax.legend()\n",
+    "\n",
+    "        ax.set_xlabel('Timesteps')\n",
+    "        ax.grid(color='0.9', linestyle='-', linewidth=1)\n",
+    "\n",
+    "        plt.tight_layout()\n",
+    "        \n",
+    "fig.tight_layout(rect=[0, 0, 1, .97])\n",
+    "fig.patch.set_alpha(1)\n",
+    "display(fig)\n",
+    "plt.close()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1080x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1080x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "experiments = experiments.sort_values(by =['rules_price']).reset_index(drop=True)\n",
+    "# df = experiments\n",
+    "cols = 1\n",
+    "rows = 1\n",
+    "cc_idx = 0\n",
+    "\n",
+    "# len(experiments)\n",
+    "# df = experiments.copy()\n",
+    "# print(df.head())\n",
+    "\n",
+    "while cc_idx<len(experiments):\n",
+    "    cc = experiments.iloc[cc_idx]['rules_price']\n",
+    "    cc_label = experiments.iloc[cc_idx]['rules_price']\n",
+    "#     secondary_label = experiments.iloc[cc_idx]['string']\n",
+    "    sub_experiments = experiments[experiments['rules_price']==cc]\n",
+    "    cc_idx += len(sub_experiments)\n",
+    "    fig, axs = plt.subplots(ncols=cols, nrows=rows, figsize=(15*cols,7*rows))\n",
+    "    sub_experiments.reset_index(drop=False)\n",
+    "    for i, experiment in sub_experiments.iterrows():\n",
+    "        df = experiment['dataset'].copy()\n",
+    "        # filter out starting point\n",
+    "        df = df[df.timestep > 0] \n",
+    "        # FIRST RUN ONLY\n",
+    "        df = df[df.run == 1]  \n",
+    "\n",
+    "        df = df.groupby('timestep').agg({'reserve': ['min', 'mean', 'max']}).reset_index()\n",
+    "\n",
+    "#         print(df.head())\n",
+    "        plot_label = experiment['rules_price']\n",
+    "        ax = axs\n",
+    "        title = 'Reserve' + '\\n' + 'Scenario: ' + str(cc_label)  + ' rules_price'\n",
+    "        ax.set_title(title)\n",
+    "        ax.set_ylabel('Amount')\n",
+    "        colors = ['b','orange', 'g', 'magenta', 'r']\n",
+    "        \n",
+    "#         ax.plot(df.timestep, df['price'],color = colors[0], label='Price')\n",
+    "        df.plot(x='timestep', y=('reserve','mean'), label='Mean', ax=ax, legend=True)\n",
+    "#         ax.fill_between(df.timestep, df[('reserve','min')], df[('reserve','max')], reserve=0.3) \n",
+    "\n",
+    "\n",
+    "        ax.legend()\n",
+    "\n",
+    "        ax.set_xlabel('Timesteps')\n",
+    "        ax.grid(color='0.9', linestyle='-', linewidth=1)\n",
+    "\n",
+    "        plt.tight_layout()\n",
+    "        \n",
+    "fig.tight_layout(rect=[0, 0, 1, .97])\n",
+    "fig.patch.set_alpha(1)\n",
+    "display(fig)\n",
+    "plt.close()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1080x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1080x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "experiments = experiments.sort_values(by =['rules_price']).reset_index(drop=True)\n",
+    "# df = experiments\n",
+    "cols = 1\n",
+    "rows = 1\n",
+    "cc_idx = 0\n",
+    "\n",
+    "# len(experiments)\n",
+    "# df = experiments.copy()\n",
+    "# print(df.head())\n",
+    "\n",
+    "while cc_idx<len(experiments):\n",
+    "    cc = experiments.iloc[cc_idx]['rules_price']\n",
+    "    cc_label = experiments.iloc[cc_idx]['rules_price']\n",
+    "#     secondary_label = experiments.iloc[cc_idx]['string']\n",
+    "    sub_experiments = experiments[experiments['rules_price']==cc]\n",
+    "    cc_idx += len(sub_experiments)\n",
+    "    fig, axs = plt.subplots(ncols=cols, nrows=rows, figsize=(15*cols,7*rows))\n",
+    "    sub_experiments.reset_index(drop=False)\n",
+    "    for i, experiment in sub_experiments.iterrows():\n",
+    "        df = experiment['dataset'].copy()\n",
+    "        # filter out starting point\n",
+    "        df = df[df.timestep > 0] \n",
+    "        # FIRST RUN ONLY\n",
+    "        df = df[df.run == 1]  \n",
+    "\n",
+    "        df = df.groupby('timestep').agg({'supply': ['min', 'mean', 'max']}).reset_index()\n",
+    "\n",
+    "#         print(df.head())\n",
+    "        plot_label = experiment['rules_price']\n",
+    "        ax = axs\n",
+    "        title = 'supply' + '\\n' + 'Scenario: ' + str(cc_label)  + ' rules_price'\n",
+    "        ax.set_title(title)\n",
+    "        ax.set_ylabel('Amount')\n",
+    "        colors = ['b','orange', 'g', 'magenta', 'r']\n",
+    "        \n",
+    "#         ax.plot(df.timestep, df['price'],color = colors[0], label='Price')\n",
+    "        df.plot(x='timestep', y=('supply','mean'), label='Mean', ax=ax, legend=True)\n",
+    "#         ax.fill_between(df.timestep, df[('supply','min')], df[('supply','max')], supply=0.3) \n",
+    "\n",
+    "\n",
+    "        ax.legend()\n",
+    "\n",
+    "        ax.set_xlabel('Timesteps')\n",
+    "        ax.grid(color='0.9', linestyle='-', linewidth=1)\n",
+    "\n",
+    "        plt.tight_layout()\n",
+    "        \n",
+    "fig.tight_layout(rect=[0, 0, 1, .97])\n",
+    "fig.patch.set_alpha(1)\n",
+    "display(fig)\n",
+    "plt.close()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### INTO AGENT LEVEL across AGENTS"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### This plot is across agents of the same variable (this case is agent_attestations_1)\n",
+    "### Can be repeated for all other columns in agents dataframe"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 32,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1080x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAABDAAAAHwCAYAAABQRJ8FAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nOzdd3xUVf7/8ddkMpPJZNKBFDokIYGAFAVRmqgYQOyo4IIB+bG7iAVYpIg0AwZ0jbpgRATZIFFYy6qYLwKrogJGEGkSSmiBdCC9Tbu/P4IRhIQ2mZvyeT4e40zu3Jz7vpcrzHzuuedoFEVREEIIIYQQQgghhKjDXNQOIIQQQgghhBBCCHElUsAQQgghhBBCCCFEnScFDCGEEEIIIYQQQtR5UsAQQgghhBBCCCFEnScFDCGEEEIIIYQQQtR5UsAQQgghhBBCCCFEnScFDCGEEELUur179zJ79uwbamPJkiVs3rzZIe2dOnWKZ5555qrXLywsZNiwYezbt++6tymEEEKIGyMFDCGEEELUutTUVLKzs2+ojeTkZKxWq0Pay8jI4Pjx41e17pYtWxg+fPhVry+EEEKI2uGqdgAhhBBCqMdut7Nw4UL27NlDSUkJiqIQExND27ZtmTFjBmlpafj4+NC0aVNCQ0N55plnOHr0KAsWLCA/Px+bzcaoUaN45JFHSE5OJi4ujpYtW3LkyBGsVivz5s0jODiYt956i6KiImbMmMErr7xSbZ7jx48zf/58SkpKyM3NJTw8nDfeeIOPP/6Y/fv3s3jxYsrLyy9p75tvviE+Ph6LxYLBYGDatGl069aNo0eP8uKLL2I2m1EUhUceeYTHH3+cWbNmkZ2dzVNPPcWKFStqPEYJCQm8+uqrPP/8844+/EIIIYS4BlLAEEIIIRqxPXv2kJOTw9q1a3FxceHdd99l+fLlGI1GQkJCWLZsGTk5OTz00EOEhoZitVp59tlnWbx4MZ06daKoqIjHHnuMkJAQoPLWjjlz5hAREcHKlSuJi4vjgw8+4Nlnn+Xrr7+usXgBsG7dOh544AHuv/9+LBYLDz30EN999x1PPPEEGzZs4IknniAqKgq73V7V3okTJ4iLiyMhIQFfX1+OHDnCmDFj2LhxIytWrGDgwIGMHz+e3NxcFi5cyIgRI4iJieHll1++YvECuKp1hBBCCFH7pIAhhBBCNGLdunXD29ubjz76iFOnTpGcnIyHhwc7duzgs88+A6BZs2ZERUUBcOLECdLS0pg5c2ZVG+Xl5Rw4cID27dsTHBxMREQEAB07dqxq42pNnTqVrVu3snz5ck6cOEFOTg6lpaU1/s7WrVvJyckhOjq6aplGoyEtLY27776badOmsXfvXnr37s2sWbNwcZE7aIUQQoj6SAoYQgghRCP23XffsWDBAsaMGcOdd95Ju3bt+OKLL3B1dUVRlKr1fv/Sb7PZ8PT05PPPP69678yZM3h6erJ7924MBkPVco1Gc1EbV2Py5MnYbDYGDx7MgAEDyMzMvGIbdrud3r1788Ybb1Qty8zMpFmzZoSHh/P111+zbds2tm/fztKlS/n000+vKZMQQggh6ga5BCGEEEI0Ylu3buWOO+5g5MiRREZGsnnzZmw2G/379+fjjz8GIC8vj82bN6PRaGjbti0Gg6GqgJGZmcm9997L/v37a9yOVqutGoCzJj/++CNPP/00Q4YMASpvcbHZbJe0ceHr3r17s3XrVo4ePQpUDrp53333UV5ezpQpU0hKSmLo0KHMmTMHk8lEWloaWq0Wi8VyHUdMCCGEEGqRHhhCCCFEI/b4448zZcoUhg0bhtVq5fbbb2fjxo0sXbqUWbNmMWzYMHx8fAgODsZgMKDX63n77bdZsGAB7733Hlarleeee44ePXqQnJxc7Xa6du3K0qVLmThxIkuWLKl2vUmTJvH0009jNBoxmUzccsstpKWlATBw4EBef/11LBYL3bt3v6i9+fPnM3nyZBRFwdXVlfj4eDw8PJgwYQIvvvgia9euRavVctddd3HLLbdQUFCAm5sbjzzyCP/5z3/QaDQOP7ZCCCGEcCyNcq19O4UQQgjR4K1Zs4aOHTvSrVs3zGYzI0eO5JlnnqF///5qRxNCCCFEIyU9MIQQQghxiZCQEF5++WXsdjsWi4WoqCiHFC+OHTvGpEmTLvte27ZtLxrHwhmKi4t54oknLvueh4cHiYmJTs0jhBBCiOpJDwwhhBBCCCGEEELUeTKIpxBCCCGEEEIIIeo8KWAIIYQQKti9ezejRo1i2LBh3HvvvYwbN44jR46ommnfvn08++yzqmb4s7179zJ79myg9vMNHDiQffv21Vr7V1IXj78QQghRl8gtJEIIIYSTmc1m+vbty8qVK+nUqRMAn3/+OXFxcfzvf/9Dq9WqnLDu+PTTT/n6669ZtmxZrW9r4MCBvPnmm3Tu3LnWtyWEEEKIayeDeAohhBBOVlZWRlFREaWlpVXL7rvvPkwmEzabDa1Wy8cff8z777+Pi4sLvr6+LFq0iKCgIL755hvi4+OxWCwYDAamTZtGt27d+Ne//kV6ejq5ubmkp6cTEBDAq6++SrNmzfj2229ZtmwZZrOZc+fO8cADD/D888+TnJzMggULMBqNlJSU8MILL7Bo0SLWr19PUVER8+bN4+DBg2g0Gvr27cvkyZNxdXXlzTffBOC55567aL+Sk5N5/fXXCQoK4vjx47i7uzN+/HhWr17N8ePHGTRoEDNnzsRut7Nw4UL27NlDSUkJiqIQExNDjx49mD59Ovn5+Zw6dYqbbrqJbdu2UVRUxIwZM3jggQd4+eWXWb9+PdOnT8dkMnHo0CGysrLo0KEDixYtwsPD46JMF7Y3YMAAzp49S2hoKE899VTV+xf+/LvqjvPRo0d58cUXMZvNKIrCI488Uu0goL8bOHAgQ4cOZevWrRQVFTFmzBhGjhxZ4/EvKSkhJiaGXbt2VU3/OmnSJCwWC6+99ho7duzAZrPRsWNHZs2ahclkuu7zUQghhKgvpIAhhBBCOJm3tzdTp05l3LhxNGnShO7du9OrVy+GDh2KXq/n4MGDvPbaa3z22WcEBQWxatUq4uPjGTt2LHFxcSQkJODr68uRI0cYM2YMGzduBGDnzp3897//xWQy8be//Y2PPvqIZ555hpUrVxIbG0ubNm3Izs7mjjvuYPTo0QAcOXKEzZs307x5c5KTk6syxsTE4OPjw5dffonFYuHvf/87K1euZPz48ZcULi60b98+5syZQ8eOHRk3bhzvvvsuCQkJFBcX069fP5566ikyMjLIyclh7dq1uLi48O6777J8+XJ69OgBQHl5OV999RXwRw+MV1555aJ8APv37ychIQGNRsOjjz7Khg0bePjhhy/JdGF706dPv+Kfz4kTJ6o9zitWrGDgwIGMHz+e3NxcFi5cyIgRI3Bxqfmu3IKCAj755BOys7N54IEHqva1uuP/1ltvUVFRQVJSEjabjbFjx/Lzzz+zY8cOtFotn376KRqNhtdff53XXnuNuXPnXnG/hBBCiPpOChhCCCGECsaMGcPw4cPZsWMHO3bsYPny5SxfvpyPP/6Y7du306dPH4KCggCIjo4GYM2aNeTk5FT9DKDRaEhLSwOgZ8+eVVfiO3bsSEFBARqNhnfeeYfvvvuO9evXc/ToURRFoaysDICgoCCaN29+Sb7vv/+eDz/8EI1Gg16v5/HHH+ff//4348ePr3G/WrRoQceOHQFo1aoVnp6e6PV6/Pz88PDwoKCggG7duuHt7c1HH33EqVOnSE5OvqjnxO9f7q+kb9++6PV6AMLCwigoKLjselfb3u+2bt1a7XG+++67mTZtGnv37qV3797MmjXrisULgJEjR6LRaAgMDKRv375s3bqVTp06VXv8t23bxowZM9BqtWi1Wj744AMAXn31VYqKiti2bRsAFosFf3//a9o/IYQQor6SAoYQQgjhZL/88gu//vor48aN44477uCOO+5g8uTJ3HvvvWzduhWtVotGo6lav7y8nPT0dOx2O7179+aNN96oei8zM5NmzZqxadMmDAZD1XKNRoOiKJSWlvLggw9y1113cfPNN/Pwww+zefNmfh8Cy2g0Xjaj3W6/KIPdbsdqtV5x334vKPzO1fXSjxrfffcdCxYsYMyYMdx55520a9eOL774our96jL92eX293IubO/P61kslkvWr+k4h4eH8/XXX7Nt2za2b9/O0qVL+fTTTwkMDKwx64XHwW63VxU9qttXV1fXi45/ZmYmBoMBu93OzJkz6d+/PwAlJSVUVFTUuG0hhBCioZBZSIQQQggn8/PzIz4+np07d1Yty83Npbi4mLCwMHr16sX27dvJyckB4KOPPuLVV1+ld+/ebN26laNHjwKwZcsW7rvvPsrLy6vd1smTJykuLub5559n4MCBJCcnYzabsdvtNWbs06cPH3zwAYqiYDabWbduHbfddpsD9r6yh8Mdd9zByJEjiYyMZPPmzdhstsuuq9Vqr6pwcrV8fX3Zv38/ANnZ2fz888+XrFPTcZ4yZQpJSUkMHTqUOXPmYDKZqnrA1OS///0vABkZGWzdupV+/frVuH7v3r357LPPsNvtmM1mnn32WXbs2EGfPn1Ys2ZN1Z/hSy+9xOuvv36th0EIIYSol6QHhhBCCOFkbdu2ZenSpcTFxZGVlYWbmxuenp4sXLiQdu3aAVSNkQHQtGlTFi5cSEBAAPPnz2fy5MkoioKrqyvx8fGXDFx5oQ4dOjBgwAAGDx6MXq8nLCyMkJAQTp48eUlviQvNmjWLmJgYhg0bhsVioW/fvvztb38DqHYQz6v1+OOPM2XKFIYNG4bVauX2229n48aNly2qdO3alaVLlzJx4kRGjRp1Xdu70KhRo/jHP/7BPffcQ4sWLbj11lsvWSckJKTa4zxhwgRefPFF1q5dWzW45i233HLF7Z4+fZqHHnqI8vJyZs2aRbt27cjNza12/YkTJ7JgwQLuv/9+bDYbQ4YMYdCgQfTr149Fixbx4IMPYrPZiIiIuKpxPYQQQoiGQKZRFUIIIYSoRTI9qxBCCOEY0gNDCCGEEOIGfPHFF6xYseKy7w0bNszJaYQQQoiGS3pgCCGEEEIIIYQQos6TQTyFEEIIIYQQQghR50kBQwghhBBCCCGEEHVevRgDY/fu3bi5uakd4yJWq/Wyc9sL0RDJ+S4aGznnRWMi57toTOR8F41JfT7fKyoq6Nq16yXL68XeuLm5ERERoXaMi2RmZhIUFKR2DCGcQs530djIOS8aEznfRWMi57toTOrz+Z6SknLZ5XILiRBCCCGEEEIIIeo8KWAIIYQQQgghhBCizpMChhBCCCGEEEIIIeq8ejEGxuVYLBZOnz5NeXm5Ktu32Wzk5+ersu2rYTAYaNGiBTqdTu0oQgghhBBCCCHEDau3BYzTp0/j6elJmzZt0Gg0Tt++2WxGr9c7fbtXQ1EUzp49y+nTp2nbtq3acYQQQgghhBBCiBtWb28hKS8vx9/fX5XiRV2n0Wjw9/dXrXeKEEIIIYQQQgjhaPW2gAFI8aIGcmyEEEIIIYQQQjQk9bqAIYQQQgghhBBCiMZBChh1zKZNm8jOzr6q9aZMmeKEREIIIYQQQgghhPqkgFHHJCQkUFxcXOM6MTEx/POf/8RutzsplRBCCCGEEEIIoa56OwvJhT755TTrdp5yaJuP3tySh3u0qPb94uJi5s2bR1FREXl5eQwfPpzIyEjmzZuHh4cH/v7+uLm5ERsby+rVq1m/fj0ajYYhQ4YwevRopk+fjl6vJz09nZycHGJjY8nNzSUlJYVp06aRmJhY7Swn3bt356677mLt2rUO3WchhBBCCCGEEKKuahAFDDWkpaUxdOhQBg0aRHZ2NqNGjcLDw4PFixcTGhpKXFwc2dnZpKamkpSURGJiIhqNhujoaPr06QNAcHAw8+fPZ926daxdu5b58+cTERHB3Llza5yidciQISQnJztrV4UQQgghhBBCCNU1iALGwz1a1NhbojY0adKExMRENm7ciMlkwmq1kpOTQ2hoKAA9evQgKSmJw4cPk5GRQXR0NAAFBQWkpaUBEBERAUBgYCC7du1yan4hhBBCCCGEEKI+kTEwrtOqVavo2rUrr732GlFRUSiKQmBgIKmpqQDs2bMHgHbt2hESEkJCQgKrV6/moYceIiwsDLj8VKcajQZFUZy3I0IIIYQQQgghRD3QIHpgqGHAgAHExMTw5Zdf4uPjg1arZfbs2cycOROj0YhOpyMgIIDw8HB69+7NiBEjMJvNdOnShYCAgGrb7datGy+88AIrV67Ex8fHiXskhBBCCCGEEELUXRqlHlzuT0lJqbrdoqZlzmQ2my8Zp2LNmjUMHjwYPz8/4uLi0Ol0TJw4UaWE6h8j0XBkZmYSFBSkdgwhnEbOedGYyPkuGhM530VjUp/P9+q+y0oPDAfy9/dn7NixGI1GPD09iY2Nve62Jk6cSEFBwUXLTCYT8fHxNxpTCCGEEEIIIRonRQFLGZhLgDp/Lf/yFAUUG9htFzzbwW69aJnG5q12UoeTAoYDRUVFERUV5ZC2lixZ4pB2hBBCCCFEI2O3Q3EW5J2E/DTIPwkVhWqnqhM8i0vA5OG8DSrK+Yf9jwd/+rnud4ivf2wWMBdDRdH55+KLnxWb2gmdwrv9EBj1odoxHEoKGEIIIYQQQlyJolR+KbrienawloG1ovIqr7X8/HNF5XJLeeUyxe6YXHYbFKb/UajIOwkFp8Bmvng9V3e4zADyjY1RUZx/HDRa0LhUblejOf/6ggeN58+lquOA5TIPqwM35OICOkPlee9qAF0zcG1d+drkXvme1gAu9fnYu1TuZ9W5pL3k/CqJuA13tWM6mBQwhBBCCCGEUBQozvmjEJCfdvGj4FRl4aGucvcD39YQGAnhQytf+/z+aAm6hvY15vpk1+MxAX6nWK3Y8vOxnjuH7Vwetrxzf7wuLEQpL0cxV2Avr0CpqLjotb2iHKXC7NxeH4qCYjZjLy3FXlrqpG3bgdLzj8ZLNzANbu6rdgyHkgKGEEIIIYSov8wlUJgJRRmXfy4vuHIbdisUZV5aoDD6g08rCOgEHQaD4SruJ9dozl/1dassGrgazj+7VS7XGSqXabTXt7+X255nILh5Oqa9RsZuNmM+fgLz0VQqUlOpSD2KJSur7t3WoSjYS0qwnTuHrbCw2nwunp64GAxo3NzQuLnhcv5Z4+aG1tOz8rVej0br4tT4Gr0eF6MRjdGI1sMDjdGIS9XDAxcPIy4Gg/QScrBzBoPaERxOChhCCCGEEOIPZfmQkwI5Byof2QegOFvtVJdS7FB6DiouU6Bw8wLPIPAKqux9cKUu+hqXyiKAb5vKgoVPK/BuCW6m2kjeqNnLy7EVFmIvKsJeVIRid9CtNFdDUTCnpJBz9izm88UKc1oa2M6Ph+Digr5VK3TNm4OTv+BfDV2LFrj6+aL19UPr54urn9/Fr3180LjK1zvxB01mptoRHE7O8Dpm06ZNdOnShYCAgMu+X1RUxNSpUykuLsZisTB9+nS6devm5JRCCCGEqLN+H8CxKIsrjrBvt+N+bBfsT68sVOSkQOHpP95384JmERDclTp5n77R73yhIvji50ZSeLBkZHB25fvYi4rUjnIJxWLBVlSEvbAQW1FRZdGisBDFbL7yL9eyUq0WfevWuIWE4Bl1D24hIbiFhKBv0wYXNze14zUI5RYbeaVm8kos5JeaqbA5sVAlqgS4WqjfN0xdSgoYdUxCQgJz586ttoDx/vvvc+uttxIdHc2xY8eYMmUKn332mZNTCiGEEEJVNmvlmAx5x+HcMTh3/PzjGOSdqBws8ir5ALjooGkHaH0bBHSEZucf3i2kS3cdpNhs5K1JJOeNN8Bmw7VJE7UjXcpVi9bLG62nJ65BQWi9vNB6eeLi6YXW2wsXT0+0np6gddCtNFcpX6Mh+OabcdHrnbrd66UoCimZRWxOySanqO6NwWKzQ2GZpbJYUVpZrMgrNVNukYJFXXB3mC/L27dSO4ZD1UoBw263M3fuXA4dOoRerycmJobWrVtXvR8TE8OuXbvw8Kicwujtt9/G0/MG7tvb/SH8+sGNxr5Yt79A1xHVvl1cXMy8efMoKioiLy+P4cOHExkZybx58/Dw8MDf3x83NzdiY2NZvXo169evR6PRMGTIEEaPHs306dPR6/Wkp6eTk5NDbGwsubm5pKSkMG3aNBITE9Ff5i/W6OjoquU2mw03qRILIYRoDErPwe5EyE1RO8mlFCpnfLCWV840Yas4P+PE+Z9/fzhsiH0FSs9e3J6rAXzbgl9bCLmz8tkzGFyu9OVQQ67FQNPw3qDVOSifqE3lhw6TOfslyvfsxaNvXwLnzEHfornaseqNoszMOl+8sNrs7DyZx8bfstl4IIvTeWVoNOBr1Ne5flAaDXi56/A16mnuY6BTsBe+Rh0+Rj2+Rn3Va4Ou7t2S0xh4KSVqR3C4WilgbN68GbPZzNq1a9m9ezexsbHEx8dXvf/bb7/x3nvv4efnVxubd4q0tDSGDh3KoEGDyM7OZtSoUXh4eLB48WJCQ0OJi4sjOzub1NRUkpKSSExMRKPREB0dTZ8+fQAIDg5m/vz5rFu3jrVr1zJ//nwiIiKYO3fuZYsXAF5eXgDk5uYydepUZs6c6bR9FkIIIZwuYzf8vBz2f1xZEPAMOj/tYB2j1Z8fpNGtspjg6gZ60wU/G66imHANjH7g167y4du28ri4XN9xsWZmSvGiHrBXVHAmPp6z761A6+lJ8Kuv4nXvUDTSQ6ZBKDPb+OFILhsPZPO/lGzySi3oXV3oG9KEZwaGMDA8gKaecuFSXJvMzLrXa+dG1UoB45dffqFv38rpWrp27cr+/fur3rPb7Zw8eZLZs2dz5swZHnnkER555JEb22DXETX2lqgNTZo0ITExkY0bN2IymbBareTk5BAaGgpAjx49SEpK4vDhw2RkZBAdHQ1AQUEBaWlpAERERAAQGBjIrl27rnrbhw4dYvLkybzwwgv07NnTsTsmhBBCqM1aAb/9F3Ysh9M7QGeEm0ZAz/9XORuEEI1Myc8/kzV7DuYTJ/C+/36aTZ+Gq6+v2rFENRRF4beMQjYdyOb4mStfAS8qt7D92FnKLXa8DK7cGRHAoI4B9Atrioeb3PEvxIVq5f+I4uJiTKY/Bk/SarVYrVZcXV0pLS3lL3/5C2PGjMFmszF69GgiIyMJDw+vtj2r1Urmn0ZQtdlsmFUcBGjVqlV07tyZxx57jJ9//pnvvvuOgIAAUlJSaN++Pbt27cJms9GyZUvat29PfHw8Go2GhIQE2rZti81mq9oHq9V60f6YzeZq9+3o0aM8//zzvPbaa3To0KHGY2Cz2S45bkJcLzmXRGMj57zzuRRl4JGyFveU/6AtP4fVuw0lt82kLOxBFDdPsAPy51Ir5Hyvm+zFxZS/swzzV1/hEhSEx+JFaG65hdzycvl/4QbUxvlusdn59XQx3x8r4Mdj+WQXW3DRQJCXGy5X6CSj02q4t6M//dp50625J67ayl8oPJdLocOTisamof39XisFDJPJREnJH9VGu92O6/kpfdzd3Rk9ejTu7u4A3HrrrRw8eLDGAoarqytBQRePn5qfn1/tbRbOMGDAAGJiYkhKSsLHxwdXV1fmzJnDnDlzMBqN6HQ6AgIC6Ny5M7fddhtPPvkkZrOZLl260KJFC7RaLa6uruj1elxdXdFqtej1erp3786LL77IypUr8fHxuWS7b731FhaLhcWLFwOVx/rC23MupNVqLzluQlyPzMxMOZdEoyLnvAPZrHDmEJQXQkURVJx/Nhef//n8ssIMOPZd5e+EDYae43BtOwBvFxe8Vd2Bhk/O92unKAr2ggKsublYz5ypfM6tfLbl54Nyhdlfrm4rlGzbhu3sOfzGjqXpxKdxMRod0G7j5sjzvaDMwneHcth0IJsth3IpqrBi0LnQL7Qpd3UM4M7wZvib5LYPoZ76/Pd7fn7+ZZfXSgGje/fufPvttwwZMoTdu3cTFhZW9d6JEyeYNGkSn332GXa7nV27dvHggw/WRoxa1bNnTzZs2HDRsjVr1vDOO+/g5+dHXFwcOl3l/aTjxo1j3LhxF60bGxtb9bpfv37069cPgEmTJjFp0qRqt1tdsUIIIYSoc86kwidPQebualbQgJtn5cPgDbc/BzePBZ+GNWK6uHqKolD4VRIFX3wOVpvacf5EwVZSUlmkyD2DYrFcsobGYEDr64vmOscj+TO31m1oFv8O7pGOu3WquMJKdmE52YXl5BRWnH9dQXZROWeLKxxTe6nDzGYzev2JG26nwmpnf3oBVrtCE5OeIZ2DuLtjAH1Cm2DQOXdmFSEak1opYNx9991s3bqVxx9/HEVRWLhwIe+//z6tWrXizjvvZNiwYTz66KPodDruv//+qnEj6jt/f3/Gjh2L0WjE09PzoiLFtZo4cSIFBQUXLaupt4UQQghRZygK7EqADdMrB7G8N65yoEk3L3Az/VG00Hlc98CTouEpP3yY7JdjKN2xA13rVrj6/jHYu11RsNkVrHYFq03BardjtSuVM8A4kc1gwNKmI5abfLH4+GHx9sPi/ftrX+wG41VNO6sAZqudMrONMouNUrONckvl69+XlZltWGx2+C4PvvvxhnIrVBYucgorKK64dDYco15LoJcBf5MeFxkU9KoYdC481bctgzoG0LWlL9or3ScihHAIjaLU/TprSkpK1YCXNS1zpsrqbd2egkntYyQajvrc/UyI6yHn/A0oPQdfPgcpX0DbfvDgMvAKVjuV+JOCMgvfHMzmm4O5nMkvxs2gXjd3fUUpfX78jB6/bKbCzZ0t/Yezq1M/zpVbyS+1kFdqptRc13pjOI5Oq8Gg0+Ku02LUaytf6yt/1mldrqYeclWMei0BXgYCvAwEehlo5uVW9bOpEQ0UKX+/i8akPp/v1X2XbTx/WwkhhBCidh3/AT77KxRnw93zofcz0sOiDjlbXMGmA9n83/4sth09g8Wm0MzTDX+jFr1NhavHikK3Qz8xZNvHeJQW8XOnvmy89UFK3U1orXYCvAx0CPTE16jH16jDx6iveu3rocfHqEOnrb/nl5PgNU4AACAASURBVN7VpapIIYQQ4upIAUMIIYQQN8ZmgW8Xwo9x4N8exm2G4G5qpxJAVkE5X/+Wxf/tz+Tn4+ewK9DSz50xt7clKjKQri18yM7OcvoVuvKDB8l6OYayX37B0KULgS+9RKfOkYxxagohhBD1jRQwhBBCiFqgKAq5RRWV9+k3YNr84/j+39/RZ++mJPIJCvvPR9F5QH6Z2tHqnAqrnaJyC0XlVorKLRSWWyks+/3nymUlZqvDBlHMLChn96nKUdxDmpl4+o4QoiID6RjkhcbB4xwoNhu2c+dQ7Paa17NYOLfq3+QlJqL19iYo5mW8H3rIYYNeCiGEaNikgCGEEEI4kN2u8L+DOSz9NrXqy2ND1Jxc7tb+wj9c11GKlmctz7NhZ0/Ymax2tHrJ5OaKp8EVDzdXHDUWoMnNlX8MCiMqMpCQZp431JZiNmPJzsaSno4lPQNLxgWP9HQs2dlwmVk5LsvFBd/HH6Pps8+ivcyU8UIIIUR1pIBRx2zatIkuXboQEBBw2fdLS0uZMmUKBQUFuLu78+qrr+Ln53fZdYUQQjiP1Wbnq32ZvP3tUQ5lF9HSz50Zg8PxMerUjuYQ7mVZBJz9maZnd9Ds7A5MZekAZPvdQnLXhdzhHsgdKmdUg+u5M7RbNAP9mewrr3y+14Pm/H80Va9rd/wJK3CwhvcVRaHgChmU8nIu6hqi0eDarBm64GDcb7oJr+bNcQ0MQON65Y+W7jfdhKFDh6sLL4QQQlxAChh1TEJCAnPnzq22gLFu3To6derExIkT+fTTT3n77beZNWuWk1MKIYT4XYXVxie/pPPOlqOknSslLMDEG4915d4uQbjW18H57DYozICT2+DE93DiR8g7Ufmeuy+0vh3aPAtt+hAQ0In7Gum0i4qicGp8DKV5ufg+MbLWCxG1paSkGA8PU43ruBgM6Jo3R9c8GF1wMLrAQDR1fDY2IYQQDU+DKGB8cfQLPjvymUPbfDD0Qe5rf1+17xcXFzNv3jyKiorIy8tj+PDhREZGMm/ePDw8PPD398fNzY3Y2FhWr17N+vXr0Wg0DBkyhNGjRzN9+nT0ej3p6enk5OQQGxtLbm4uKSkpTJs2jcTExMtO0xodHY3NVjmVWEZGBk2aNHHofgshhLg6JRVWPvw5jeU/HCO7sIKbWngza2gP7ooIwMVR9wDUhuM/QNpPUJ5f+SjLh/KCP57L86Gi8I/1Dd7Qug/0+hu06QPNOsnMIuflr/sPJT/8QMBLs/B74gm141y3zMxMAurpNHtCCCEalwZRwFBDWloaQ4cOZdCgQWRnZzNq1Cg8PDxYvHgxoaGhxMXFkZ2dTWpqKklJSSQmJqLRaIiOjqZPnz4ABAcHM3/+fNatW8fatWuZP38+ERERzJ0797LFi99ptVpGjx7N4cOHef/99521y0KIBqq4wsqPR86wNfUMJWar2nHqhLKyMtzdq78lwG5X+O5wLvmlFm5r78/rj3bltvb+Dh8Y0aHyT8HXMyDly8qfdUYw+IC7T2WRwrsFBEZWvjb4gEcTaNkLAjqBi1bd7HWQ+dQpshctwuO23viOGKF2HCGEEKJRaBAFjPva31djb4na0KRJExITE9m4cSMmkwmr1UpOTg6hoaEA9OjRg6SkJA4fPkxGRgbR0dEAFBQUkJaWBkBERAQAgYGB7Nq165q2n5CQwNGjR/nrX//K5s2bHbdjQohG4cSZEr45mMO3h3JIPnYOs82Oh16Lr4d0CQew2WxotTXPotGzjR9/G9Ce7q18nZTqOlkrYNu/4PvXKn8e+BLcOgH0RnVz1WOK3U7GjBloXFwIWrBAZtAQQgghnKRBFDDUsGrVKrp27crIkSP56aef2LJlC4GBgaSmphISEsKePXsAaNeuHSEhIbz33ntoNBpWrVpFWFgYGzZsuOyVOo1Gg1LD/GnLli0jICCABx54AKPRiFYrV8WEEFdmttrZeeIc3xzM4ZuDORw7UwJA+6YePHlba+4Ib8YtbfzQ1dcxGxwsMzOToIbQpT51MyS9AOeOQsR9cM9C8Gmpdqp679y/Eyjb+QtBr7yCriGcJ0IIIUQ9IQWM6zRgwABiYmL48ssv8fHxQavVMnv2bGbOnInRaESn0xEQEEB4eDi9e/dmxIgRmM3mGmcYAejWrRsvvPACK1euxOcyU4s9/PDDTJs2jU8++QSbzcbChQtrczeFEHVMhdXGwcwi9qUXsD+9gH3pBRSWX3nqwnPFZkrMNvRaF25t78/o3q0ZGB5AK3+5Ct8g5afBhhlwcD34h8BfPoWQO9VO1SBUHD1KblwcpoED8X7gfrXjCCGEUFl5iYVju3Ox26q/CK0Wva+FhlZn1yg1Xe6vI1JSUqput6hpmTOZzeZLxqlYs2YNgwcPxs/Pj7i4OHQ6HRMnTlQpofrHSDQcDeZqdD1TYbVxKOuPYsXe0wUczi7Ccv4fSB+jjs7NvWlqcrtiWyaDK31CmnB7SBM83KR2fSX19py3VsC2t+D7f1b+3H8q9J4Irlc+R8SVKRYLJ0aMxHL6NO2+/ALXpk3VjuQQ9fZ8F+I6yPkuHOnY7ly2JB6itNCsdpTLat3Vm3v/1kPtGNeluu+y8inWgfz9/Rk7dixGoxFPT09iY2Ovu62JEydSUFBw0TKTyUR8fPyNxhRCqGh/egEvrz/ArrS8K65rtSv8XmL2dq8sVozr247Ozb3p3NybFr7udXvQSGcpSIevpkDhaYc12cRiBV09/CeyOBeKs+R2kVpyZvlyyvfvp/kbbzSY4oUQQohrV1Zk5oe1hzmyMwf/FiYG/60znv4GtWNdoqD4rNoRHK4efjqru6KiooiKinJIW0uWLHFIO0KIuuFscQWvbTzERztO4WfUE31bmyuON+GqdaFDgCedm3vT0k+KFZd1eid8NBLMpZVTfDroGNnKy9EZ6t4HkSvybQs9noSQu9RO0uCU/fYbZ96Ox2voULyi7lE7jhBCCBUoikLqLzl8/9FhzGVWet3Xlm73tEZbR8cQKyxteJ8dpYAhhBC1yGKzk7D9JG9sPkyZ2cZTt7fl2btC8TLo1I5W/+39D3z+NHgGwujPoZnjbpnLky7G4gJ2s5nM6dNx9fUl8KVZascRQgihgpKCCr7/8DDHdufSrLUnA0dH4N/cpHasRkcKGEIIUUu+P5zL/PUHSM0ppn9YU166tyMhzeQfuhtmt8O3MfDDP6F1H3g0ATz81U4lGrAzb71FxZFUWi57B+1lBtgWQgjRcCmKwqHkLH5cdwSrxc5tD4Vw050tcFGh14XZZuajgx+RcCCBEkvJFdcf1mIYM4NmOiGZ80gBQwghHOzk2RJeXp/C5pRs2vgbWfHkzQwMbya3gDhCRTF8Oh4OfQXdn4Qhr4Gr/sq/J8R1Kt31K2dXrMRn+HBM/furHUcIIcQVlBdbyMsudUhbdpudXV+nkfbbWYJCvBk4KgKfAOfP4Gaz20g6nsSSX5eQUZJBr8BehPqGXvH3Ohs7OyGdc0kBQ4gGqMJqY81PaRSUXXl6zatRXFyEyVTkkLYaurMlFazbcRqdVsP0weGMub0Nbq5atWM1DPlp8OEIyDkAUYug118dNuaFEJdjLy0lY/p0dMHBNJs2Te04QgghqpGfU8rxPWc4sfcMman5OHKeTVc3LX0fC6Nz/+ZoXJz7uUNRFH5I/4E3d73J4bzDRPhFMPe2ufQO7n1Vv5+ZmVnLCZ1PChh1zKZNm+jSpQsBAQE1rnf06FEeffRRtm3bhpubTI8n/lBhtfG31b/w7aFcB7ec5eD2Giati4b7uwYzPSqcZl71cBDIuirtJ/joCbBZ4ImPIeROtROJBs6Wn8/pSZOwpKXRKuHfaE0eakcSQghxnt2ukH28kBN7czm+5wx5WZU9Lvybm+gxuA0Bbb1w0Tqm2OAfbMLDx/nft/bm7iXulzh2Zu+khakFr/Z7lUFtBuGiqZsDhjqLFDDqmISEBObOnVtjAaO4uJhFixah10u3aXGxC4sXCx/szMherRzSrsyZLi7iyMsaV2N3Inz5XOWUoCPWQtMw525fNDoVR49yasIErBmZBL3yCh49e6odSQghBJBxJI+D27M4se8MZUUWXFw0BIf5ENm/OW06N8GribvaEW/Y8YLj/OvXf7Hp5Cb8DH7M7DWTR0IfQaeVAeChgRQw8v/7Xwo++dShbXo//BA+DzxQ7fvFxcXMmzePoqIi8vLyGD58OJGRkcybNw8PDw/8/f1xc3MjNjaW1atXs379ejQaDUOGDGH06NFMnz4dvV5Peno6OTk5xMbGkpubS0pKCtOmTSMxMfGyBQpFUXjppZeYPHkyEyZMcOg+i/qttooXQgBgKYf9n0ByPGTtc/722/aH4avA6Of8bYtGpfj770mfPAWNmxut/v1vjN27qR1JCCEEcDa9mP++/is6gyutI/1p26UJrTr54Wasf1/sFUXhTNkZThWd4lTRKdKK0jhVdIrTRac5cPYAeq2eCTdNYHSn0XjopAfghRpEAUMNaWlpDB06lEGDBpGdnc2oUaPw8PBg8eLFhIaGEhcXR3Z2NqmpqSQlJZGYmIhGoyE6Opo+ffoAEBwczPz581m3bh1r165l/vz5REREMHfu3Gp7VyxZsoT+/fsTHh7uzN0VdZwUL0StKcqCHStg50ooPQNNI6DfVHBx4j8fHk0qB+yUKw+iFimKwrn3V5Hz2mu4dehAy6VL0AUHqx1LCCEE58eCWHcEvdGVv8zrjcFUvz4TKIrCBykfsDNrJ6eKKwsVZdayqvddNC4EeQTRwrMFozuNZnTH0TRxb6Ji4rqrQRQwfB54oMbeErWhSZMmJCYmsnHjRkwmE1arlZycHEJDK0eD7dGjB0lJSRw+fJiMjAyio6MBKCgoIC0tDYCIiAgAAgMD2bVr11Vt94svviAwMJBPPvmE3Nxcxo4dy5o1axy/g6LekOKFqBXpuyD5Hdj/KditEBYFt/6tsieEDJwpGhi72UzWnLkUfPYZnvfcQ/ArC3ExOn+UeSGEEJd3fM8Z0g/l0fexsHpZvHjl51f48OCHtPFqQxuvNvQK7EUrr1a09GxJS8+WBHsEyy0iV6lBFDDUsGrVKrp27crIkSP56aef2LJlC4GBgaSmphISEsKePXsAaNeuHSEhIbz33ntoNBpWrVpFWFgYGzZsuOyUihqNBqWG+8s3bdpU9XrgwIGsXLnS8Tsn6g0pXgiHslkh5YvKwsWpZNB7wi1PQc/x4N9e7XRC1Aprbi6nn3mWst27aTJxIk0m/B2NS+MeIE0IIeoSm8XO1k9S8Q3yILJf/eoZpygKi3cs5sODH/JkxyeZcvOUy34HFFdPChjXacCAAcTExPDll1/i4+ODVqtl9uzZzJw5E6PRiE6nIyAggPDwcHr37s2IESMwm81XnGGkW7duvPDCC6xcuRIfHx8n7pGoby4sXix4MFKKFw1N1j74TzSU5Ttvm9YKMBeBb1uIioWuT4DBy3nbF8LJyg8c4NSEp7Hl59P8jTfwirpH7UhCCCH+ZM83pyjMLWPYszfhoq0/BWZFUfjnzn/yQcoH/CXiL1K8cBCNUtPl/joiJSWl6naLmpY5k9lsvmScijVr1jB48GD8/PyIi4tDp9MxceJElRKqf4xE7flz8eKJXq1rdXsyC4mTKQr8exhk/wadHnTedjUu0H4ghN0DLlrnbbcOqm/nvL2sjIojR6g4egzFYlY7Tr1gyy/gzNtvo/X1peXSJRg6dlQ7kmrq2/kuxI2Q871+KSmoYM2cn2ge5svQCV3UjnPVFEUhblcc7+9/nxHhI5jRc4YqxYv6fL5X911WemA4kL+/P2PHjsVoNOLp6UlsbOx1tzVx4kQKCgouWmYymYiPj7/RmEIl+aVmrPYbrxfa7QrTPtnrtOKFUMGRTXDiBxj8KvQar3YaUYcoioI1K4vygwepOHSI8oOHqDh0CPPJk2C3qx2v3nHv1o0W/3oL1yYyUJoQQtRFyZ8fw2axc/vDIWpHuWqKovDWr2/x/v73eazDY6oVLxoqKWA4UFRUFFFRUQ5pa8mSJQ5pR9QNn/16mklr9zi0TSleNFB2G2yeU3kbR49otdOIOqJ461bOLnuXikOHsF1Q3Na1aIFbeAe8hgzBLbwDhtBQNO7uKiatTzS4NmsqHyqFEKKOyk0rImV7Jl3vbIlPQP0ZWHnp7qW8t+89Hgl7hJm9Zsq/Mw4mBQwhapmiKCz//jjtmnow5rY2DmmzfTMTt7WXK4YN0p4PIecADF8FrpefTlk0LuUHD3J64jO4+vvjec89lYWK8HDcwsLQmkxqxxNCCCEcrnLa1MO4m3TcPLSt2nGuWvyeeJbtXcZDoQ/x0q0v4aKpP2N21BdSwBCilu1Ky+NAZqH0mBBXZimDbxdC8x7Q0blTQ4u6yXr2LKcmTEDr5UXrxDXomjVTO5IQQghR61J/ySEztYABT3TAzb1+fGV9d++7vL37be5vfz9zes+R4kUtqR9ngxD1WML2k3i6ufJA1+ZqRxF1XfI7UJgOD70L0t2w0VPMZk4/9xy2s+do/cEHUrwQQgjRKFjNNrZ9mop/CxMRt6s7baqiKJRZy664XuLBRP71678Y1m4Y826bJ8WLWiQFDCFqUW5RBUn7MnmiV2s83OR/N1GD0nPwQxyERUGbPmqnESpTFIWsl1+mbOcvBP/zNdw7R6odSQghhHCKXzelUXyugruiO+Liou4FnQn/m8CP6T9e1bpD2g7h5dtfRtvIZ3KrbfKNqo7ZtGkTXbp0ISAg4LLvK4pCv379aNOmDQBdu3ZlypQpTkworsXaHWlYbAqjesutI+IKvn8NzEVw11y1k4g6IO+DNeT/52P8//pXvIcOVTuOEEII4RTFeRXs+vok7bs3pXmYr6pZdmbt5Mf0HxnWbhihvqE1ruvj5sOw9sOkeOEEUsCoYxISEpg7d261BYy0tDQ6derEO++84+Rk4lpZbXbWJKfRJ6QJ7ZvKQHuiBnkn4Od3oesT0OzS+a5F41KybRvZsbGY7ryTps89q3YcIYQQwmm2/zcVxQ63PaT+tKnL9i7D3+DPS71fwt1VZviqKxpEAePgT5mkbM10aJsRtwcRfmtQte8XFxczb948ioqKyMvLY/jw4URGRjJv3jw8PDzw9/fHzc2N2NhYVq9ezfr169FoNAwZMoTRo0czffp09Ho96enp5OTkEBsbS25uLikpKUybNo3ExET0+ktnIPjtt9/Izs5m1KhRGAwGZsyYQbt27Ry678IxNqfkkFlQztz7OqkdRdR138SAiyvcMVPtJEJl5pMnOT1pMm7t2hK8aBEaF7mHVgghROOQdayAw8nZ9IhqjVcTdQsGu3N281PmT/zj5n9I8aKOaRAFDDWkpaUxdOhQBg0aVFVQ8PDwYPHixYSGhhIXF0d2djapqakkJSWRmJiIRqMhOjqaPn0q728PDg5m/vz5rFu3jrVr1zJ//nwiIiKYO3fuZYsXAE2bNmX8+PEMHjyYnTt3MnXqVD755BNn7rq4Sqt/OkGwt4E7w2XgPVGDjN2w7z/Qdwp4qTtQlVCXraiIUxOeRqPR0OLtt9GaPNSOJIQQQjiFYlf4Yd0RjN56ukepf+v1O3vfwdfNl+Fhw9WOIv6kQRQwwm+tubdEbWjSpAmJiYls3LgRk8mE1WolJyeH0NDK+6N69OhBUlIShw8fJiMjg+joaAAKCgpIS0sDICKisqt4YGAgu3btuqrtRkZGotVW3lt18803k52djaIoaGTGgjolNaeYralnmXpPB1y1cgVVVENRYNNscPeD259TO41QkWKzkfGPqZhPnqTVihXoW7ZUO5IQQghRo/zsUta9sgNLuc1hbd4ZHYHeoO5X1H25+9iavpXnuz+PUWdUNYu4VIMoYKhh1apVdO3alZEjR/LTTz+xZcsWAgMDSU1NJSQkhD179gDQrl07QkJCeO+999BoNKxatYqwsDA2bNhw2aKDRqNBUZRqt7tkyRJ8fHz4f//v/3Hw4EGCg4OleFEHffDTSfRaFx67Rb6EiBoc/R8c3wJRsWDwVjuNUFFuXBzFW7YQOGc2Hr16qh1HCCGEuKIDP2ZgM9u5eUgbcMDXEU8/Ax16Bt54Qzdo2d5leLt583j442pHEZchBYzrNGDAAGJiYvjyyy/x8fFBq9Uye/ZsZs6cidFoRKfTERAQQHh4OL1792bEiBGYzeYaZxgB6NatGy+88AIrV67Ex8fnkvfHjx/P1KlT2bJlC1qtlldeeaU2d1Nch5IKK5/8cpohnQNpYnJTO46oq+w22DQHfNvAzU+pnUYAdrMZa04utrNnsObkUJqZ5ZTtlu/fx9n3VuAz4nF8R4xwyjaFEEKIG2G32TmUnEXrzv70uq/hjMd34OwBtpzewjPdnsFDJ7dy1kVSwLhOPXv2ZMOGDRctW7NmDe+88w5+fn7ExcWh0+kAGDduHOPGjbto3djY2KrX/fr1o1+/fgBMmjSJSZMmVbtdb29v3n33XUfthqgFn/2aTlGFlVG926gdRdRle9dB9n54eAW4Xn7Mm/pKURTK9+zBVlSsdpRL2MtKsebkYs3NxZqTc9HDVlBw0brOTG/s2ZPAmTKIqxBCiPoh7cA5SgvNhPd27m38tW3ZnmV46j0ZES4XFOoqKWA4kL+/P2PHjsVoNOLp6XlRkeJaTZw4kYI/fZg2mUzEx8ffaExRixRFYfX2k3QK9qJ7q0t70AgBgKW8cuaR4G7Q6SG10zjc2WXLyH3jTbVj1EyrxbVpU1ybNUPXuhXGW26u+lnr709eYSF+fv5OiaLRuuDevTua80VvIYQQoq47uD0Lg0lH60jn/FvpDIfOHeKbU98w4aYJeOo91Y4jqiEFDAeKiooiKirKIW0tWbLEIe0I5/r5+DkOZRex6OHOMjbJ9TKXwncLoThH7SS1pzADCk/Dg/HQwKbJLPrf/8h94028hgzBd9Rf1I5zCRc3t8oihZ9fjVOUFmdmYgpqWFeVhBBCCEcoL7FwfG8ukf2ao3VtOJ9jlu1dhklnYmTESLWjiBrU6wKGzL5RvZoGAhW1J+Gnk3i767jvpuZqR6m/kqbC7jXgq/4UWrXq1gnQtp/aKRyq/PBhMqa+gCEykqCFC3AxGNSOJIQQQggHO7IjG7tVaVC3jxzJO8Kmk5sY32U83m4ysHpdVm8LGAaDgbNnz+Lv7y9FjD9RFIWzZ89ikC8PTpVTWM7X+7OIvq0N7nqt2nHqp18/gN0fQL8XYOCLaqcR18Cal8fppyei8TDSYukSKV4IIYQQDdTB7Zn4tzDRtGXDuc3i3b3vYnQ1MipilNpRxBXU2wJGixYtOH36NLm5uaps32azodXW3S+pBoOBFi1aqB2jUUn8OQ2rXeEvtzbwngO1Jfs3+Ooflb0SBkxXO424BorFQvqkyVizsmi9OgFdDTMtCSGEEKL+OptRTM7JIvoMD1U7isMcyz/G1ye+ZmzkWHwMMoZdXVdvCxg6nY62bduqtv3MzEyC5P5ocZ7FZicxOY3+YU1p00SmXLpmFUWw7kkweFXOyuFSd4uD4lLZixZT+tNPBL3yCu5du6odRwghhBC15ND2LFxcNIT1bDgXK97d9y4GVwOjO41WO4q4Cg1n1BUhVLTxt2xyiioY3Vt6X1wzRYEvn4NzR+GRlWBqpnYicQ3y/vMf8j74AL8nn8TnwQfUjiOEEEKIWmK32TmUnEXrzv64ezaMKeBPFJzg/47/H491eAw/g5/accRVkAKGEA6QsP0EzX3cGdBBvnxfs50rYP8nMHAWtOmjdhpxDUp/+YWs+S/jcfvtNJv6D7XjCCGEEKIWpR04R2mhuUEN3rl833J0Ljqe7PSk2lHEVZIChhA36FBWEcnHz/GXW1ujdZEBZa9Jxq+wYQaEDoLbJ6mdRlwDS0YGp599Dn1wMM1f/yca13p7R6IQQgghrsLB7Vm4e+po3dlf7SgOcarwFF8d+4rhYcNp4t5E7TjiKsknTlFvnM4r5f2tJ7DZ69YUsfvSC9C7uvDYLS3VjlK/lOVXjnvh0QweXAYuUk+tL+xlZZyaOBGlooIWCf9G6y3TjQkhhBANWXmJheN7c+ncrwVabcP4zPbe/vfQarSMiRyjdhRxDaSAIeqND39OY8WPx/F216kd5RJjbm+Dn0fDuBfQKRQFPn8aCtNhzAYwyj2H9YWiKGS++CIVKQdp+U48bu3bqx1JCCGEcKqUsylM+2EaZ0rPqB3FaUIzetLTeh8L86eRn5ildhyHKLYU83j44zQzyi3g9YkUMES9sedUAZ2Cvfjq2b5qRxE3avtSOLge7nkFWt6idpoGr3TXr+T8859gs91wW/bycioOHqTplMmY+vd3QDohhBCi/tiWsY1J307Cy82L+0PuVzuO07gfiMDmV0L/rr3UjuIwOhcd0ZHRascQ10gKGKJesNsV9pzOZ9hNwWpHETcqLRk2z4GIYXDr39VO0ygUfP455fv3Y+zR44bbcjEa8Xz6afzHjXNAMiGEEKL++PLol8zeOpt2Pu2Ivyu+0Vy5P5tRzEdnfqbP8FBu6jlM7TiikZMChqgXjp8toajcStcWPmpHETei5Cx8PAa8W8D9S0Ejg546Q9mePRhvvplWK95TO4oQQghR7yiKwor9K3hz15v0CupF3IA4PPWeasdymoPbs3Bx0RDWM0DtKELILCSifthzKh+Am1pKAaNe+/F1KM6B4f8Ggwz86Az2khIqDh/G/aYuakcRQggh6h2b3caC5AW8uetNhrYbSvyd8Y2qeGG32TmUnEXrzv64e8p4b0J90gND1At7TuVj1GsJafb/2bvz8Djv8lz896ya0WjfF2uz5H13nDibnY2QYHICp01JCiSUFNJTCNCFE672RyHk5IS0XPQUOJy0h15NSgIlUCAJIQHiOCc2JiHYsUaxJcuSLGubTSPNvs+87++PsUQMdqSR5p3vO+/cn38mtma5LU/kmWeeGl3h7QAAIABJREFU7/NUiI5CqzH2CtB9DdC2U3SSkhE7dQqQJFh37BAdhYiIqKjE03F87vDncGjqEO7dei8+s/sz0OtK6/PfycF5xIJJbLyqVXQUIgAsYFCR6J8OYFt7NQx6HjkoWhEv4DkFbPuC6CQlJWa3AwAs29mBQUREtFy+uA+fOvQpDMwO4G+u+Bt8cNMHRUcS4vRrTlgrTejaVi86ChEAHiGhIpBMSxhyBLGTx0eK27kj2csebq4opJjdDnNXF4y1taKjEBERFYXp0DTuefEenJ4/jX+8/h9LtngRD6cwPuDF+stbYDDwbSOpAzswSPVOu4JIZiRs5wDP4jZ+GDBXAq08PlIosiwjZrej4uqrRUchIiJakdccr+HlyZfzdn/RaBTlE+XveJ2DEweRklL41ru/hV1Nu/L22MVm5JgbUlrGxqtbREchWsQCBqnebwd4cuhjURs/DHRdDRj4Y6dQ0g4HMrNeWDj/goiIipAsy3j49YfhirhgM9nycp+SJEGvf+dugmZbM/5+399jbc3avDxmsTr9mhMNHRVoWFM6Q0tJ/fhOglSvfyqAhgoz2musoqPQSgUdwNwocNlHRScpKQvzLzjAk4iIitGpuVOYDE3iS1d/CX+w7g/ycp9OpxOtrRxIuZS5mTA8EyFc+0frREchugALGKR69mk/dqypgU7HAZ5Fa3xh/sV+sTlKTMxuh66sDJb160VHISIiytkL4y/AqDfips6bVn1fgdkYDn17CN6ZEPT60Tyk07ZMSoJer8P6K5pFRyG6AAsYpGrBeApjs2HcvqNNdBRajfHDgLUWaN4qOklJifXbYdm6FTqTSXQUIiKinGSkDH4+/nNc234tqstWd4z4bP8sXv73Ieh0QPumStgq8nMcReuauqpgrTSLjkF0ARYwSNVOTgcgy8AObiApXrIMjL8KdF8LLHHmlPJHSiYRHxpC7Yc/LDoKERFRzt70vAlPzIPP9nx2xfeRyUh47cdjsB+cQlNXJW75+FZEUn4eISEqYixgkKr1T58f4LmGAzyLlu8cEJgCrvmM6CQlJXH6NORkkvMviIioKL0w/gKsRiuuW7Oy9ethXxw//9YpuM4GsO26dlxzxzoYTHpEnP48JyWiQmIBg1RtYCqA7vpy1JSzfa1oneP8CxFi/ecHeO5kAYOIiIpLKpPCSxMv4fqO61FueueVpxczOTiHl/5tEJmUhHd/bAvW7eEcByKtYAGDVM0+7ccVPXWiY9BqjB8GKpqBBg6SLKSY3Q5jSwtMzXzRRkRExeU152sIJAI40HMgp9tJkozf/HQcx144h7pWG269bytqWzjvgkhLWMAg1XIH43AG4tixhvMvipYsZwsY3fsAbpEpqJjdzuMjRERUlF4YfwFV5ipc03bNsm8TDSbx0r+dwvRpHzZe2YL9H9wAk9mgYEoiEkGRAoYkSXjwwQcxPDwMs9mMhx9+GF1dXb93nfvuuw833XQT/viP/1iJGFTk7FPn5190cP5F0fKeAcJuHh8psLTXi9T0NGr5s5VIc+KRFF5/Zgzzzkje7jOZTMJsduTt/ohWQ5IlmGfX4Q7LXvzkfw0s+3Z+dxTJeAY33L0Rm6/h9joirVKkgHHw4EEkk0k8/fTT6O/vx6OPPorHHnvsguv80z/9EwKBgBIPTxphn/bDoNdhSxsLGEVr/HD2kgWMgooNZF/wcf4FkbZMDs7h0L8PIRZKoaW3Gro8LXbS63XQG9glR+rgiwWQRgr15XU5PS+be6qx9/YeNKypVDAdEYmmSAHj+PHj2LdvHwBg586dOHny5AVf/9nPfgadTof9+/mmhi7NPhXAxpZKWExs/yta44eB6g6gtlt0kpIS67cDRiMsmzeLjkJEeZBKZvDaD0fx1qszqG0px4FPbEdTV1Xe7t/pdHKtJKnGZw59Bm9538L/uOMlGPR8DUhEF1KkgBEOh1FRUbH4a4PBgHQ6DaPRiDNnzuD555/H17/+dXzzm99c1v2l02k4nU4loq6KGjNphSTL6J/04V3ra/l9Vomc/x5kCc1nX0W860YEXC5lQtFFhX/zGxjWroXb7wf8XBe3UvzZQ2owPxPDsR87EJ5LondvLbbe1ISMKQJnHo+QAHy+kzqEU2EcmT6C2zpug8ftUexx+HynUqK157siBYyKigpEIr/9h1WSJBiN2Yd65pln4Ha78ZGPfAQzMzMwmUxob29/x24Mo9Gouk8G+GmFssZmwwgnM7h6Qxu/zyqwoue7cwBIBFC+5VaU8++wYORMBmeGh1H9/vejhd/3FePPeBItk5Fw/IVzOPbiBGzVZtz+FzvRsVGZrVx8vpNa/Hjkx0jJKfzR1j9Ca6Myz0k+36mUFPPz3X+JD+EUKWDs3r0br7zyCg4cOID+/n6sX//b9YkPPPDA4n9/4xvfQENDA4+S0O8ZmF4Y4MkNJEXr3JHsZfc+sTlKTGJ0DFI0yvkXREXM54rg4OOD8EyEsH5vM/bfuR5l5SbRsYgU98L4C+io7MDWhq2ioxCRSilSwLj55ptx9OhR3HXXXZBlGY888ggef/xxdHZ24qabblLiIUlj7FMBlJsN6GuqWPrKpE7jh4G6XqC6XXSSkhKz9wOAZlaoyrIM93gQmZRU0Medm49ACvoK+phEADA7FcLrz56F0azHLR/fir7LmkRHIioIb8yLN1xv4E+3/il0XL1ORJegSAFDr9fjoYceuuD3ent7f+96n/rUp5R4eNKA/ik/trVXw6DnP2BFKZMGzh0Ftt0hOknJidntMNTUwNTZKTpKXrz24zGc+MWkoEcX9bhU6jq31OPGezbCVl0mOgpRwfz83M8hyRIO9BwQHYWIVEyRAgbRaiTTEgYdQXz0mm7RUWilnP1AMsT1qQLE7HZYdmzXxKdX844I7AensG5PE7bsK2wnz9z8HOrr6gv6mEQAYDQb0NRdqYn/h4ly8eL4i1hXuw59tX2ioxCRirGAQapz2hVEMiNx/kUxGz+cveT8i4LKBINIjo6h6kDxf3olyzIOP30GJosB++5cD2uluaCPr3fG0dpaW9DHJCIqVTPhGdhn7fjM7s+IjkJEKqcXHYDod9mnsgM8t6+pFpyEVmz8MNC0GahoFJ2kpMTeeguANuZfjB73YGbYh723ry148YKIiArrxfEXAQC3dt8qOAkRqR0LGKQ6/VMBNFSY0V5jFR2FViKdACZf5/ERAeIDA4BOB+v27aKjrEoynsbR/xxFQ0cFtuznEFgiIq17cfxFbG/cjjWVa0RHISKVYwGDVMc+7ceONTU8/1uspo8B6RgLGALE+u0w966FobJSdJRVOf7iBCL+BPbftQF6DvIlItK0Mf8YzvjOcHgnES0LCxikKqF4CmOzYc6/KGbnjgA6PdB1jegkJUWWZcTsdli3F/fxEZ8rgv6Dk9h4ZQtae3mMjIhI614YfwF6nR63dN8iOgoRFQEWMEhV3poJQJbBAkYxGz8MtGwHrPw7LKTU5CQyfn9Rz7+QZRlHvj8Co9mAq/6AU+iJiLROlmW8cPYFXNFyBRqsDaLjEFERYAGDVMU+FQAA7OAAz+KUjAJTb/D4iAAxux0AYN1ZvAWMs/2zmBqcxxX/pQflVRzcSUSkdSe9JzEdnubxESJaNhYwSFXsU35015ejppxvXorS1OuAlAJ6rhOdpOTE+u3QlZejrK84OxdSyQx++YMR1LfbsO06Du4kIioFL4y/AJPehJu6bhIdhYiKhFF0AKK3s0/7cUVPnegYtFLjhwG9Eei8UnSSkhOz22Hdtg06g0F0lBV582cTCM8ncPNfb4bewNo6EVExmwxO4j9H/hOyLL/j9V4YfwHXtl+LKnNVgZIRUbFjAYNUwx2MwxmIY8cazk4oWuNHgPbLgLIK0UlKihSPIz48jPp77xUdZUX8nije/MUE1l/RjLZ1taLjEBHRKj019BT+4/R/wGq0vuP1jDojPrDhAwVKRURawAIGqYZ9yg+AAzyLVjwAON4E9v216CQlJz44CKTTRTn/QpZl/PL7IzAY9Lj6D4vz+AsREV1oeH4Yu5p24dvv+bboKESkMezTJdWwT/th1OuwpY1thEVp4jVAljjAU4BY//kBntu3C06Su3MDXkycnMPlt/XAVl0mOg4REa2SLMsY8Y1gfe160VGISINYwCDVsE8FsKGlEhZTcZ7hL3njhwFDGbDmCtFJSk7MboepvR3GhuJaQZdOZnDk+yOobbVh+41rRMchIqI8cEQcCKVCLGAQkSJ4hIRUQZJkDEz7cduONtFR6HcFnag6/EWgzPTO1xs9CHRcAZgshclFi2J2O8p37xYdI2dv/mISobk43veXu2Dg4E4iIk04M38GALChboPgJESkRSxgkCqcm4sgGE9jJwd4qk//d2AbehqoWuITcoMJ2HV3YTLRopTLhbTLVZTzL4aOOtC1rR5rNnBwJxGRVgz7hqGDDutq1omOQkQaxAIGqYJ9mgM8VevcEaTqNsD06TdEJ6GLiNkHAADWHcVVwIiHUwj7Eth2A4+OEBFpyRnfGXRUdqDcVC46ChFpEHt2SRXsUwGUmw3oa+L6TVVJJ4DJ15Fs2ys6CV1CzG6HzmRC2aZNoqPkxDsdAgA0rOH/80REWjI8P8zjI0SkGHZgkCr0T/mxrb0aBr1OdBR6u+ljQDqORPuVsInOoqC5f3scsbcGRMdYkdjxN2HZvBl6s1l0lJx4p8MAgIY1lYKTEBFRvkRTUUyFpnBb722ioxCRRrGAQcIl0hkMOoL46DXdoqPQ7xo/DOj0SLbuEZ1EMVIyCc8//iMMVVUw1BTfESZ9RQWq7/hD0TFy5p0Oo7zKjPKq4iq8EBHRpY34RyBDxoZadmAQkTJYwCDhBh1BJDMSdnVykJ/qnDsCtO6AXFYlOoliEiMjQDqNli/8HapuvVV0nJLhnQ7z+AgRkcYMzw8D4AYSIlIOZ2CQcCcmswM8d3UW36ffmpaMAlNvAN37RCdRVGJoCABgKbIZEsUsk5bgc0bQ0MECBhGRlpzxnUGFqQJttjbRUYhIo1jAIOFOTPnRXmNFc5VFdBR6u6lfA1IK6LlOdBJFxQeHoLfZYOroEB2lZPhcEUgZGfXswCAi0pQzvjNYX7seOh1nmhGRMljAIOFOTPqwk90X6jN+GNAbgc4rRSdRVHxwEGWbNkKn54/DQuEATyIi7ZFkabGAQUSkFL5iJ6E8oTimfTHs6mABQ3XOHQHadgNl2v2UXM5kEB8ehmXTZtFRSop3KgyDSY+aJqvoKERElCeOsAORVATr61jAICLlsIBBQvUvzr/gAE9VSYSAmTeBnv2ikygqOTEBORbj/IsC806HUd9mg97Af4KIiLRi2Hd+gCc3kBCRgvjqkYQ6MeWHyaDDljbtbrkoShOvAXIG6NH2AM/44PkBnptZwCgUWZYxxw0kRESac2b+DHTQoa+mT3QUItIwFjBIqBOTPmxuq4bFZBAdhd5u/FXAYAY69opOoqj40CB0JhPKentFRykZEX8C8UgK9Zx/QUSkKWd8Z9BV1YVyU7noKESkYSxgkDDpjISB6QDnX6jRuSPAmisAk7ZnFCSGhlC2fj10JpPoKCXjtwM82YFBRKQlw75hrKtdJzoGEWkcCxgkzLA7hGgyg13cQKIu0XnAOaD5+ReyLCM+OMTjIwW2UMDgClUiIu2IpCKYCk1x/gURKY4FDBLmxMIAzw4O8FSViV8BkDU//yLtdCLj96OMAzwLyjsVRlWDBWVWo+goRESUJyO+EQDAhjoWMIhIWSxgkDAnJv2ot5nRUaftYwpFZ/wwYLQC7XtEJ1FUfOj8AE8WMApqbiaMBs6/ICLSlDO+MwCA9bVcoUpEymIBg4Q5MeXDrs4a6HQ60VHo7c4dATqvBIxm0UkUFR8cAnQ6WDbw06JCSSUy8HuiPD5CRKQxw/PDqDRXotXWKjoKEWkcCxgkhD+axNnZCHZ18viIqoRnAc+g5udfANkODHNPD/TlnJZeKHMzYUDmAE8iIq0Z9g1jfe16fihFRIpjAYOE6J9amH/BAZ6qcu5I9rJEChg8PlJY3EBCRKQ9kixhxDfC4yNEVBAsYJAQJyb90OmA7SxgqMv4YcBcCbTuFJ1EUWmfD2mnE5bNm0VHKSlz02GYrUZU1ltERyEiojyZCc0gmo5yAwkRFQQLGCTEiSk/NjRXoqKMmwhU5dwRoOtqwKDtv5fEwgBPrlAtKO90CPXtNrYYExFpyLBvGAA3kBBRYbCAQQUnSTL6J7MDPElFgg5gbrQ0jo8MDgLgBpJCkiUZ3pkIGjq4gYSISEvO+M5Ar9Ojt6ZXdBQiKgEsYFDBnfVGEIynsauDAzxVZXxh/sU+sTkKID44BGNbKww1LKIVSmA2hnQiw/kXREQaMzw/jM7KTliNVtFRiKgEsIBBBXdi0gcA7MBQm3OHAUsN0LxNdBLFZQd4cv5FIXGAJxGRNg37hnl8hIgKhgUMKrgTU35UlhnR28g3MqoyfhjovhbQa/vHghSJIHnuHI+PFNjcTBg6vQ51bTbRUYiIKE/CyTBmwjMc4ElEBaPtdyqkSicm/djZWQO9noP8VMN3DvBPAj3XiU6iuPjwGUCWOcCzwLxTIdQ0l8NoMoiOQkREeTLiHwEArlAlooJhAYMKKpJIY9gVxC6uT1WXUpp/MXR+gCdXqBaUdzrM4yNERBozPM8NJERUWCxgUEENTAcgycCuTg7wVJVzRwBbI9C4UXQSxcWHhmCorYWxuVl0lJIRj6QQ9iVYwCAi0phh3zCqzFVoLue/qURUGCxgUEGdmMoO8NzJDgz1kOXz8y/2ATrtH+uJDw7CsmkTdCXwZ1WLxQGeHSxgEBFpyRnfGayvXc9/U4moYFjAoII6MelHT4MNtTaz6Ci0YG4MCDlL4viInEwiMTLK+RcFNre4gaRScBIiIsoXSZYw4hvh8REiKigWMKhgZFnGiUk/51+ozfir2csSGOCZGBsDUimUcQNJQXmnQ7BWmVFexcIlEZFWTIWmEEvHuIGEiAqKBQwqmGlfDN5wArs6WcBQlXNHgMo2oG6t6CSKiw8OAQAsmzjAs5C802E0cv4FEZGmnPGdAcANJERUWCxgUMGcmPID4ABPVZHl7AaSnv2lMf9iaAi68nKYu7tERykZmYyEeWcE9SxgEBFpyvD8MPQ6PXprekVHIaISwgIGFcyJSR8sJj02tPAcvGp4hoCotyTmXwDZAoZl40bo9PzRVyh+VxRSWuYGEiIijRn2DaO7qhsWo0V0FCIqIXwVTwVzYtKP7e01MBn4tFON8cPZy579YnMUgCxJSAwNwcL5FwXlnQoB4ABPIiKtGfGN8PgIERWcUXQAKg2JdAaDjiA+ek236ChAKg4c+h/AmZ+LTiJexAPUdAE1naKTKC45MQEpGuUGkgLzTodhMOpR02wVHYWIiPIklAxhJjyDO9bfIToKEZUYFjCoIE45gkhmJPEDPN2ngB9+DPAMAuveDZjZ1o5Nt4lOUBCJoYUBnixgFJJ3Ooz6dhv07LwiItIMDvAkIlFYwKCCODEpeICnJAG/fgw4+CBgrQU+/EOg711ispAQ8aEhwGRCWV+f6CglQ5ZleKfD6NnRIDoKERHl0UIBgytUiajQWMCggjgx6UNbtQXNVQIGPQWdwDN/Dpx9BdhwALj9G4CNb6hKTXxwCGV9fdCZzaKjlIxoIIl4OMUBnkREGjM8P4zqsmo0lTeJjkJEJYYFDCqIE5N+Md0XQz8Bnvs0kIoBt/0v4LKPlsS6ULqQLMuIDw2h4vrrRUcpKbMc4ElEpElnfGewoXYDdHxNRUQFxkPJpDhPMI4Zf6yw8y8SYeC5TwFPfxio6QD+7DCw514WL0pU2uNBZn4els2bRUcpKXMzYQBAPTswiIg0IyNluIGEiIRhBwYp7sTUwvyLAhUwZt7MDuqcPwtc+5fA9X8LGHlsoJTFBwcBgBtICsw7HUZlvQVlVv5TQ0SkFVOhKcQzcRYwiEgIvqokxZ2Y9MNk0GFLW7XyDxYPAk/cBlhrgI/8BOjZp/xjkurFBwcBnQ6WDRw2VkjeqTDnXxARacywbxgAsKGO/6YSUeHxCAkp7sSkD5tbq2AxGZR/sPFXgVQE+K//wuIFLYoPDcHc1QW9zSY6SslIJTMIeKIsYBARaczw/DAMOgN6a3pFRyGiEsQCBikqnZEwMB0o3ADP0ZcBcwXQsbcwj0dFITE4xOMjBTY/E4EsAw0dHOBJRKQlI74RdFd1o8xQJjoKEZUgFjBIUWfcYcRSmcLMv5BlYOxloGc/Z17Qoozfj5TDgbJNLGAUknd6YQMJOzCIiLRk2DeM9XWcf0FEYrCAQYoa90YAAOuaCvAp7NwY4J8Eem9U/rGoaMRPnwYAWDZxA0kheafDMFsMqKy3iI5CRER5Ek6G4Yw4OcCTiIRhAYMUNeOPAgDaa63KP9jYy9nLvncp/1hUNOKDQwC4gaTQ5qbDqF9TAR1XFxMRacZYYAwA0FvN+RdEJAYLGKQohz+OijIjqiwFWHgzehCoWwvU9Sj/WFQ04kNDMLa0wFhXJzpKyZAlGd7pMBrWcP4FEZGWjPmzBYy+mj7BSYioVHGNKilqxh9De41V+U9h0wng3C+BnR9S9nGo6MQHB2HJw/yLuZkw+l+egpyR85BK29IpCalEhvMviIg0ZtQ/CovBgvbKdtFRiKhEsYBBiprxxdBWU4Az8JOvAako0HeT8o9FRUOKxZAcH0fVLbes6n4yaQk//9ZJhHwJlFea8pRO2+rabFizsUDbh4iIqCDG/GPoqe6BXscmbiISgwUMUpQjEMPurgJsIBl9GdCbgO59yj+WhgSeew6hV15Z8nrxWBzT1uIbxiiFI4AkrXr+xcChafhcUbz3E9vRvb0hT+mIiIiKy6hvFHtbuaqeiMRRpIAhSRIefPBBDA8Pw2w24+GHH0ZXV9fi17/zne/gRz/6EXQ6HT75yU/ihhtuUCIGCRZJpOGPptBWU4gBnoeAziuBMras52L2a19HJhSCseGd35Rn0mkkjMVZ77Ts2I7yPXtWfPuIP4Hf/HQcXdvqWbwgIqKSFUwG4Yl50FfL+RdEJI4i70gOHjyIZDKJp59+Gv39/Xj00Ufx2GOPAQDm5+fx3e9+F8888wwSiQTe+9734vrrr+ekeg1y+GMAgHalCxhBJ+A+CbzrQWUfR2PSXi9SMzNoeuAB1N/70Xe8rtPpRGtra4GSqcvRH45CysjY94F1oqMQEREJwwGeRKQGihxgO378OPbty7by79y5EydPnlz8Wl1dHZ599lmYTCZ4vV5UVVWxeKFR04UqYIwdyl72cv5FLmIDAwAA647tgpOo18wZH0Z+48aud3eiurFcdBwiIiJhRv2jAIDeGq5QJSJxFOnACIfDqKj4bSu/wWBAOp2G8XwLutFoxFNPPYVvfOMbuPvuu5e8v3Q6DafTqUTUVVFjJjUZmpgFAJiSITidCcUep+bk8zBbG+CR6wH+nSxb7JdHAYMB/ro6BJbxfSu157skyTj01DjKq01o21lWcn9+Kr3nPJU2Pt9pKQMzA7AYLEAQcIaK+/nC5zuVEq093xUpYFRUVCASiSz+WpKkxeLFgg9/+MP4wAc+gI9//ON4/fXXceWVV146pNGouvb1Um6pX66wPQCjXoetfZ0w6BXqspEygON1YN270drGlV65mDh7FpYNG9DW07PkdUvx+W5/eQpBTwLv+bNt6OhsFB2HCqwUn/NUuvh8p+VwvuVEX00f2ov89Raf71RKivn57vf7L/r7ihwh2b17Nw4fPgwA6O/vx/r16xe/dvbsWdx///2QZRkmkwlmsxl6PVcxaZHDH0NLtUW54gUAOPuB2DzQ9y7lHkOD5EwG8YEBWHfuEB1FlSKBBN74yVl0bq5Dz04O7iQiIhrzj/H4CBEJp0gHxs0334yjR4/irrvugizLeOSRR/D444+js7MTN910EzZu3Ig777wTOp0O+/btwxVXXKFEDBJsxh9TfgPJ6CEAOqCXm2xykRgbgxSNwrKd8y8u5vUfjyGdkrDvzvWc0UNERCXPH/fDG/NygCcRCadIAUOv1+Ohhx664Pd6e39bsb3//vtx//33K/HQpCIOfxx7e+qUfZCxl4HWHYCNn5LnIma3AwCsO9iB8bucYwGcft2F3bd0oaaZgzuJiIgWBnhyhSoRicazG6SIdEaCKxhXtgMjHgCm3gD6uH0kV/GBAeirq2Hu7hYdRVUkScbh7w2jorYMew50i45DRESkClyhSkRqwQIGKcIdSiAjyWivVbCAcfZVQM5wfeoKxPrtsG7fzuMRv+PU4Rl4p8K4+g/7YCoziI5DRESkCqP+UVSYKtBc3iw6ChGVOBYwSBEzvhgAKNuBMfYyYK4EOjhDJReZcASJ0VFYOf/iArFQEr9+7izaN9Si77Im0XGIiIhUYywwhrU1a/nBBxEJxwIGKcLhzxYw2pUqYMhydoDn2usAg0mZx9Co+MmTgCxzA8nveP2ZMaTiGezn4E4iIqILjPnHeHyEiFSBBQxSxIx/oQPDoswDzI0CgUmg90Zl7l/DFgd4btsmOIl6uMeDGPyVE9tvXIO6NpvoOERERKoxF5vDfHwevdVcoUpE4imyhYRoxh9Dnc2McrNCT7HRg9lLDvDMWWxgAOauLhhqakRHUdz0sA+//P4ZpJLSO14vHk6hvMqMy2/rKVAyIiKi4sABnkSkJixgkCJmfDHlui8AYPRloK4XqO1W7jE0SJZlxOx2VFxztegoinOPB/HT/zMAW5UZLT1V73xlHbB1/xqYLfyRSERE9HZcoUpEasJX66QIhz+GtY0KteKn4sC5XwK771bm/jUs7XAg4/XCovEBnnOOMH7yv/tRXmnCf/3sbtiqy0RHIiIiKkpj/jFUmivRaG0UHYWIiDMwKP9kWcaMP6bcBpLJ14B0DOh7lzL3r2GL8y927BScRDlBbww/+Vo/DEY9bv9pd2ztAAAgAElEQVTMLhYviIiIVmHUP4q+mj4OuCYiVWABg/IuEEshmswot4Fk7GXAYAa6r1Xm/jUsZh+ArqwMlg3rRUdRRCSQwLNf60c6JeH2T+9EdaOCa3yJiIg0TpZljAXG0FvDAZ5EpA4sYFDeTfsUXqE6egjovBIwc1tErmJ2OyxbtkBn0t7q2XgkhZ983Y5oMInb7t+B+vYK0ZGIiIiK2lx8DoFEgAM8iUg1WMCgvHOcX6HaXqtAASPoADyngF5uH8mVnEwiPjgIqwbnX6QSGfz0m3b43BEc+G/b0LK2WnQkIiKiojfiGwEAdmAQkWosWcD4wQ9+cMGvv/3tbysWhrRh5nwBQ5EZGGOHspdcn5qz+PAw5GQS1p07REfJq0xKwov/8hbc40G8+0+3oGNTnehIREREmsAVqkSkNpfcQvL888/j0KFD+PWvf43XX38dAJDJZDAyMoJ77rmnYAGp+Dj8MZQZ9ai3mfN/56MvAxXNQPPW/N+3xsXsAwCgqQ4MSZLx0uODmBqcxw13b0TvribRkYiIiDRj1D+K6rJq1FvqRUchIgLwDgWMffv2obGxEX6/H3feeScAQK/Xo6Ojo2DhqDjN+GNor7Hmf1q1lAHOvgKsfw/ASdg5i9ntMDY2wtjaKjpKXsiyjFe/O4yxNz24+g/7sPmaNtGRiIiINGXMP8YNJESkKpcsYFRXV2Pv3r3Yu3cv5ubmkEgkAGS7MKgETLwGhJwrumm3axibygzAyfn8Zgo6gJiPx0dWKDZgh2XHds28CDn+swkM/tKBy27twq6bO0XHISIi0hRZljHmH8OBtQdERyEiWnTJAsaCL33pS3j11VfR1NQEWZah0+nwve99rxDZSJToPPD4ewDIK7r5Awv/8Z/5CvQ2Rguw9gYF7ljb0j4fUhOTqLnjDtFR8mboqAMdm2qx931rRUchIiLSHE/Ug1AqxAGeRKQqSxYw7HY7Dh48CL2eC0tKRmAagAzc+vfA2utzumkincF7v/FL/MlV3fjwlV35z2atBWw8h5mr+MDC/AttDPAMzccR9Max/YYOzXSUEBERqQkHeBKRGi1ZwOjq6kIikYDVqsBGCVKnhaMj7ZcBTRtzuqnTG8GofA6W9i1A0xoFwtFKxOwDgF4P69YtoqPkhXPUDwBoW1cjOAkREZE2jfi5QpWI1GfJAobT6cQNN9yArq7sp+k8QlICgo7sZVXuwx4XVqi2K7FClVYsZrejbN066G020VHyYmbED7PFgPo1FaKjEBERadKYfwx1ljrUWbienIjUY8kCxle/+tVC5CA1CTkB6LLrSnPEAob6yJKE2FtvoeqWW0RHyRvniB+tfTXQ63l8hIiISAlj/jF2XxCR6ixZwPjxj3/8e793//33KxKGVCLkBGyNgMGU801nfDHodEBLtUWBYLQSyXPnIAWDsO7UxvyLaDAJnyuKjVdpYx0sERGR2siyjLHAGG7vvV10FCKiCyxZwGhoaACQ/UE2ODgISZIUD0WCBZ0rOj4CAA5/DE2VZTAbOfRVLWL2hQGe2wUnyQ/OvyAiIlKWK+JCJBXhAE8iUp0lCxh33XXXBb/+2Mc+plgYUomQE6juWNFNZ/wxHh9RmZi9H/qKCph7tdEG6hjxw2jSo7GzUnQUIiIiTRr1jwLgAE8iUp8lCxjj4+OL/z07Owun06loIFKBoAPouGJFN3X4Y9jaXp3nQLQasYEBWLZthU4jq5Ado340r62GgV0+REREilgoYLADg4jUZskCxhe+8IXF/y4rK8MDDzygaCASLBUHYvNAZVvON5UkGQ5/HLdsbVEgGK2EFIshMXwG9R/XRudUIpqCdzqMy9/bIzoKERGRZo36R9FgbUB1GT+UIiJ1WbKA8eSTT8Ln82Fqagpr1qxBXR1XKWla6HyHzQpmYHgjCSQzEo+QqEj81Ckgk4F1uzYGeDrHAoAMtHP+BRERkWK4gYSI1GrJHuwXX3wRd911F/75n/8Zd955J5599tlC5CJRQq7sZWXuXRQzPq5QVZuY3Q4AsO7QxgBPx4gfeoMOzT1VoqMQERFpkiRLOBs4y+MjRKRKS3ZgPPHEE/jRj34Em82GcDiMj3zkI3jf+95XiGwkQsiRvVzBERKHPw4AaGMBQzVi9gGY1qyBsb5edJS8cIz40dxdBaPZIDoKERGRJjnCDsTSMRYwiEiVluzA0Ol0sNlsAICKigqUlZUpHooECq78CMmMPwoAaK9lAUMtYnY7rDu0cXwkGU9jdiKEVh4fISIiUsyYfwwAB3gSkTot2YHR2dmJRx99FHv27MGxY8fQ2dlZiFwkSsgJGK2AJfc3iQ5/HJVlRlRZTAoEo1yl3G6k3W7NHB9xjwchSTLaWMAgIiJSzMIGkrU1awUnISL6fUt2YDzyyCPo6OjAr371K3R0dODhhx8uRC4SJejIdl/odDnfdNoXY/eFivx2/oU2OjAcI37odEDrWk5EJyIiUsqofxRN5U2oMnPeFBGpz5IFjFgshpaWFuzZswe1tbX4xS9+UYhcJErICVTmfnwEABz+GOdfqEjMbofOZELZpk2io+SFY8SPho5KmK1LNo4RERHRCo35x3h8hIhUa8l3Avfeey/6+vpQWVkJIDsT48CBA4oHI0FCTqB9z4puOuOP4bKu2jwHopWK2wdQtnkT9Gaz6CirlklJcI8HsfW6dtFRiIiINCsjZXA2cBYfaPmA6ChERBe1ZAGjsrISX/7ylwuRhUST5ewQz025d2CEE2kEYil2YKxScnoGaY979Xcky4idOoWaO+5Y/X2pgHsiiExa4vwLIiIiBc2EZ5DIJNiBQUSqtWQB49prr8V//Md/oK/vtz/ILr/8ckVDkSAxH5BJrHCFagwAN5CshpzJYPx974MUieTtPssv2523+xLJMeIHALT1sYBBRESklIUBnixgEJFaLVnAOHbsGJLJJH7zm98AyB4hYQFDo4KO7OWKVqieL2DUWPKZqKSkvXOQIhHUfeQe2PbvX/X96cvKYN21Kw/JxHOO+FHXZoOlghtuiIiIlLKwQrW3pldwEiKii1uygBGNRvHEE08UIAoJF3JmL1fQgTHjWyhglOczUUlJu10AgPIrr0TFNdcITqMeUkaCcyyADVe2iI5CRESkaaP+UbTaWmEz2URHISK6qCULGOvWrcPzzz+PzZs3Q3d+tWZPT4/iwUiAxQJG7m8UHf4YjHodGivL8hyqdKRc2QKGqYVv1N9udiqMVCLD+RdEREQKG/WPsvuCiFRtyQLG6dOncfr0aeh0Ovh8Ppw7dw5vvfVWIbJRoQUXChgrO0LSWmOBQa/Lc6jSkXZlh3caWcC4AOdfEBERKS8tpTEeGMfVbVeLjkJEdElLFjCefPJJDAwM4KmnnsLY2Bju0MhWA7qIkAMobwCMua/ddPhjaKvmAM/VSLld0JWVwVDDN+pv5xjxo7rRClsNu3uIiEgZ/rgf4VRYdAyhnBEnUlKKHRhEpGqXLGAkk0n89Kc/xXe/+12YTCaEw2G8/PLLsFg4pFGzgs4VDfAEAIc/jr1r6/IcqLSknS4YW5oXj2oRIEsynKN+rN3ZKDoKERFp1GuO1/CJlz+BtJQWHUUV1teuFx2BiOiSLlnAuPHGG3HbbbfhK1/5Crq7u/Gxj32MxQutCzlWNMAznZHgCsbRXsMOjNVIud0wNfP4yNvNOyNIRNOcf0FERIoIJAL4/NHPY03FGnxs28dExxGuylyFTXWbRMcgIrqkSxYw7rnnHjz//POYmZnBHXfcAVmWC5mLRAg6gfbLcr6ZO5RARpJZwFiltMsF62W7RcdQlcX5FyxgEBGRAh5941HMxebwtQNfw9aGraLjEBHREvSX+sJ9992H5557DnfffTeef/55nDx5El/5yldw5syZQuajQkkngah3ZQM8z69QbWMBY8VkSULK42EHxu9wjPhRUVuGynp2fxERUX69NPESnj/7PO7bfh+LF0REReKSBYwFV1xxBb7yla/gpZdeQktLCx544IFC5KJCC2dXeK6kgOHwZwsY7bUsYKxUZn4eSKVgbGUBY4Esy3CM+NG2roZzQYiIKK+8MS8eeu0hbK7fjI9v/7joOEREtExLFjAWVFVV4e6778YzzzyjZB4SZWGFalXuMzBmzhcwuIVk5VLObAHJxBWqiwKeGKLBJI+PEBFRXsmyjC/96kuIpqL48rVfhklvEh2JiIiWadkFDNK4kCN7uZIjJP4Y6m1mWM2GPIcqHWl3toBhbG4WnEQ9HKOcf0FERPn3zOgz+H/T/w9/cdlfYG3NWtFxiIgoByxgUNZqOjB8Mc6/WKWUyw2AHRhv5zjjh7XShJrmctFRiIhII6ZD03j0jUdxecvl+NCmD4mOQ0REOWIBg7JCDsBQBlhrc76pwx/jBpJVSrtd0JlMMNTm/v3XKseIH219nH9BRET5IckSPn/089DpdHj4moeh1/FlMBFRseFPbsoKuYDKFiDHN4uyLGPGzw6M1Uq53DC2tECn5/+SABCciyE0H0crj48QEVGePDn4JI67j+Nzl38ObRW5d5wSEZF4fLdEWUHnio6PBGIpRJMZbiBZpbTTCRPnXyxyjgYAcP4FERHlx5h/DF9/8+u4vuN6vL/v/aLjEBHRCrGAQVkhx4oGeE77zq9QrbHkO1FJSbmzHRiU5Rjxw2w1or69QnQUIiIqcikphb858jewmWz44lVf5NFEIqIixgIGAbK84g4Mh3+hgMFBiyslyzLSLhdMLezAWOAY8aO1rxp6PV9kEhHR6vzfgf+LofkhfPGqL6LB2iA6DhERrYJRdABSgbgfSMdWvEIVANrYgbFiGZ8PcioFY3PxdmDIsowffeVNeKdCebm/dErCpqtzfz4SEZG6nfKewgnPiYI9Xiwdw7cGvoXbe2/HTV03FexxiYhIGSxgUHaAJ5Ad4pkjhz8Gi0mPOps5z6FKR9qV/f6bWou3gBHxJ+A6G0DnlnrUt9lWfX96ow6brmEBg4hIS8b8Y/iTn/0J4pl4QR+3u6obn7vicwV9TCIiUgYLGAQEHdnLFRwhWdhAwvOkK5c6X8Ao5g4Mz0S28+Ly27rR0lMtOA0REalNIpPA5w5/DuWmcvzw9h+iuqxw/1bYTDYY9XzJS0SkBfxpTkDImb1c0RGSONq5QnVVFgoYxTwDw3MuCL1ehwYO3SQioov4p+P/hGHfML550zfRWdUpOg4RERUpDvGk7ABPYGUFDF+MBYxVSrvcgNEIQ3296Cgr5pkMoa7dBqPZIDoKERGpzJHpI3hq6Cl8cOMHsX/NftFxiIioiLGAQdkVqtY6wJTbIM54KgNvOMECxiql3S6Ympqg0xfn/46yLMMzEURTV5XoKEREpDLemBefP/p59NX04a/2/JXoOEREVOR4hIRWvELVGcgO4WpjAWNVUi43jC3FO/8i6I0jEUmjqatSdBQiIlIRWZbxd0f/DpFUBP/67n9FmaFMdCQiIipyxfmRL+VXyLniDSQA0F7LAsZqpFxOmIq4gOGZCAIAOzCIiOgC3z39Xfxy5pf47J7PYl3tOtFxiIhIA1jAoPMFjJXNvwDAIySrIMsy0kXegeGZCMFg1KMuD+tTiYhIG4bnh/HVY1/F9Wuux50b7hQdh4iINIIFjFKXSQFhz4pXqOp0QEt1brMz6Lcyfj/kRKKoN5DMTgRRv6YCBiN/nBARERBLx/C5w59DTVkNHrrmIa5aJyKivOE7jlIXdgOQV7hCNYbmSgtMBj6NVirtdgMAjM3F2YEhSzI8kyE0c/4FERGd99VjX8VYYAz/89r/iVpLreg4RESkIXznWeoWVqiuoANjxB1CbxOPDaxGyuUCgKLtwPB7okjFM2jk/AsiIgJwaPIQnh5+Gh/d8lFc1XaV6DhERKQxLGCUupAje5njEM90RsJpVwibW/nGdTXS5wsYxpbcO2DUwDMRAgA0dbMDg4io1LkjbnzhV1/AprpN+NSuT4mOQ0REGsQ1qqUulH0DjcrcOjDGvREk0hI2t7GAsRoplwswGGBsqBcdZUU854IwmvWobWEnDhGpQ1pK48j0ESQyCdFRiobP70NtfPVHPZ4efhrJTBL/sP8fYDKY8pCMiIjoQooUMCRJwoMPPojh4WGYzWY8/PDD6OrqWvz6E088gZ/+9KcAgOuuuw7333+/EjFoOYIOQG8CynN7A33KkV2dubm1WolUJSPtcsPY1ASdwSA6yop4JkJo7KyEXs8BbUSkDkdnjuLTr3xadIyS9dDVD6G7ult0DCIi0ihFChgHDx5EMpnE008/jf7+fjz66KN47LHHAABTU1N47rnn8IMf/AA6nQ4f/OAH8a53vQsbN25UIgotZWGFqj6300SDziDMRj16G/nJ+2qk3C6Ymotz/oWUkeCdCmHL/nbRUYiIFk2HpwEAT77nSVSZ2SW4HLOzs2hsbFz1/ZSbytFiK86h1EREVBwUKWAcP34c+/btAwDs3LkTJ0+eXPxaS0sL/vVf/xWG8584p9NplJWVKRGDliPoAKpyn78w6AhiY0sljNxAsipplxtlGzaIjrEi884o0ikJTdxAQkQq4o66YdKbsKNxB9d3LpM1ZkVrTXHOYiIiotKiSAEjHA6joqJi8dcGgwHpdBpGoxEmkwl1dXWQZRn/8A//gM2bN6Onp+cd7y+dTsPpdCoRdVXUmClXjb4ppOo3wp/Dn0WWZZyc8WP/2mpNfA9EkWUZSacTuj17iuL7+LsZzw34AQC68kRR5CfKFZ/XxWlibgL1ZfVwnR+STMvD5zuVEj7fqZRo7fmuSAGjoqICkUhk8deSJMFo/O1DJRIJ/O3f/i1sNhu++MUvLh3SaERrq7o+GXA6narLtCIxL4yNPbDm8GdxBeLwx9LY09eije+BIJlAAIF4HNW9a1Gv8u/jxZ7vw68EYbYa0be5EzrOwCCN0czP+BIUtAfRVtnGv78c8PlOpYTPdyolxfx89/v9F/19Rfr/d+/ejcOHDwMA+vv7sX79+sWvybKMT3ziE9iwYQMeeuihxaMkJEA8CCTDOR8hOeUIAABXqK5SyuUGAJhaivO88OxEEE1dlSxeEJGqeKIeNJcX52whIiIiemeKdGDcfPPNOHr0KO666y7IsoxHHnkEjz/+ODo7OyFJEt544w0kk0kcOXIEAPBXf/VX2LVrlxJR6J2EzrcT5bhCddARhE4HbGQBY1XS7mx7s7EIh3hmUhK802HsfFeH6ChERItkWYY76sYNHTeIjkJEREQKUKSAodfr8dBDD13we729vYv//dZbbynxsJSroCN7mWMHxqAziO56GyrKFHn6lIzU+fPZxdiBMecIQ8rIaOxkEYuI1COYDCKRSaDZVnyFYSIiIloaV0iUssUOjNwLGDw+snpplwvQ62HMw+q6QvNMhAAATd3cQEJE6uGKZAvDPEJCRESkTSxglLKFDowcChjBeAoTc1FsbmMBY7VSLjeMjY3QGYuvk8VzLghLhQmVdRbRUYiIFrmj2dlCTeVNgpMQERGREljAKGUhF2CpBszly77JaWf2k3d2YKxe2uWCsaU4PyX0TITQ1FUFnY4DPIlIPTxRDwCgxVZ8R/OIiIhoaSxglLKQcwUDPLMbSLawA2PVUm43TM3F9yI7lcxg3hlBUxePjxCRurijbuigQ721XnQUIiIiUgALGKUs6FjRAM+GCjMaK8sUClU6irUDwzsVhizJLGAQkep4oh40WBtg0ptERyEiIiIFsIBRylbQgXHKEcSmVh4dWK1MKAQpEinKDgzPRBAA0NTFLhwiUhd3xM35F0RERBrGAkapyqSBsDunDoxkWsKIO8wBnnmQXlih2lqcBQxbtRm2GnbhEJG6uKNubiAhIiLSMBYwSlXEA8hSThtIxmbDSGYkDvDMg5QrOynf2FJ8BYzZiRCauvkcICL1cUfZgUFERKRlLGCUqpAze5lDAWPQkT06sKWtWolEJSXtPt+B0VxcnxQmY2n43FHOvyAi1YmmogglQ2i2FdfPVSIiIlo+FjBKVfB8ASOHIySDziAsJj16GmwKhSodKZcb0OlgbGwUHSUns5MhQAYaOf+CiFRmYYUqj5AQERFpFwsYpWqxA2P5QzxPOQLY2FIFg54DPFcr5XLC0FAPndksOkpOPBMhAGAHBhGpjjuaPZrHAgYREZF2sYBRqoIOQG8EbMvrAJBlGYOOIAd45kna5YapJbcVtmrgmQiist4Ca0VxFV6ISPsWOzB4hISIiEizWMAoVSEnUNEC6Jf3FJjxxxCMp7GFBYy8SLtdMLUU34tsz0SQ61OJSJUWOjA4xJOIiEi7WMAoVUEHULn8DRgLAzy5gSQ/Ui43jM3FtYEkHk4h6I3z+AgRqZI74kaVuQpWo1V0FCIiIlIICxilKuTKaYDnKUcQeh2wsYUFjNXKhCOQQqGi68DwTGaLWCxgEJEacYUqERGR9rGAUapCzpwGeA46g+hpsMFqNigYqjQsrFAttg6MhQGe3EBCRGrkjro5/4KIiEjjWMAoRYkwkAjmtkLVEcTmtmoFQ5WOlCtbwDC1FlkB41wQNc3lKLMaRUchIvo9nqiHG0iIiIg0jgWMUpTjCtVANIUZf4wDPPMk7coOmjO2FFcBY3YyxOMjRKRKKSmFudgcCxhEREQaxwJGKQo6spfL7MA45QwA4ADPfEktHCFpKp6z2vFwGmFfghtIiEiVvFEvZMgsYBAREWkcCxilKJR9A43K5RUwFjaQbGIBIy/SLjcM9fXQm82ioyybzxEDADSyA4OIVIgrVImIiEoDCxilKHS+A2O5BQxnEE2VZWisLFMwVOlIuZwwNRfXp4Q+Rxw6HdDYwQIGEanPQgGDQzyJiIi0jQWMUhR0AmVVQFnFsq6eHeDJ7ot8SbvcRTf/wueIobbVBlMZt9AQkfq4I+cLGDxCQkREpGksYJSikGPZ3ReJdAajnjAHeOZRyu2GqYgKGLIsw++Io6mbzwEiUid31A2LwYIqM39OERERaRkLGKUo6Fz2AM8RdxhpScbmVq5QzQcpGoUUCBRVB0bYl0AikkFTJ4+PEJE6eaIeNJU3QafTiY5CRERECjKKDkAChJxAw3XLuurCAE8eIbm0Fx4bwLjdu/wbXP9N4E0A/+2QYpmUwA0kRKRW7qib8y+IiIhKAAsYpUaSsltIKpfXATDoDKLcbEBXXbnCwYpTaD6OcbsX3dvq0biMDoXk9DQCzz6L6ve/H+b29gIkzI9kJoambnZgEJE6eaIe7GraJToGERERKYwFjFITmQXkDFDVtqyrDzqC2NRaBb2ebbkXM/amBwBwzR3rUNO8dJHH/+MBOM+9gN73fBrmri6l4+WN0+lkazYRqZIkS3BH3VyhSkREVAI4A6PU5LBCVZJkDDqDHOD5DkaOedDYWbms4gUApN0uACiqGRhERGrmi/uQltLcQEJERFQCWMAoNUFn9nIZQzynfFGEE2lsbmUB42ICszF4zgXRd9nyP/VLuVww1NZCX1amYDIiotLhjnKFKhERUangERKlyDLw2v8GAtOik1xo9nT2snLpIyQc4PnORo9nXzTnUsBIu9zsviAiyiN35HwBg0M8iYiINI8FDCXZvwcEpkSn+H2tO4GKpd90DzqDMOh1WN/M4Y0XM3LMg+aeKlQ1WJd9m5TLBRMLGEREeeOJZmcRcQYGERGR9rGAoRSdDvjzo6JTrMopRxB9jRWwmAyio6iOzxXB3HQY1/7Rupxul3a5YN25Q6FURESlxx11w6AzoN5SLzoKERERKYwzMOiSBh1BHh+5hNHjHkCX2/ERKR5Hxu+HqWXp+SNERLQ87qgbDdYGGPQsthMREWkdCxh0UXPhBFzBOAd4XoQsyxj5jRttfTWw1Sx/GGfanT2nbWzhOW0ionxxR92cf0FERFQiWMCgixpyhgBwgOfFzDsi8LmiWLcnt/PWKWd2hSpnYBAR5Y8n6uEGEiIiohLBAgZd1KAzAADswLiIkd+4odMBa3flVsBIu7MFDGMzX2gTEeWDLMtwRVwsYBAREZUIFjDook45gmirtqDWZhYdRVVkWcbIcQ/WbKxFeVVu35uUK3uEhB0YRET5EU6FEUvHWMAgIiIqESxg0EVxgOfFzU6GEJyNoW9P7i+W024XDNXV0FuXv3aViIgujStUiYiISgvXqGrIJ7/7Jo6OevNyX/5oCrduZafA7xo55oHeoMPanY053zblcsPI7gsiorxxR7KdbRziSUREVBpYwNCQV4dn0VVfjj1dtau+L4Nejzsv78hDKu2QJRmjx93o2FwHi82U8+1TLic3kBAR5ZE7mi1gsAODiIioNLCAoRHxVAbhRBoHtrXikzf0iY6jSa7xIMLzCVx5+9oV3T7tcsO6dVueUxERlS4WMIiIiEoLZ2BohDecAAA0VHDoplJGj7lhMOrRsyP34yNSIoHM/Dw7MIiI8sgT9aC2rBZlhjLRUYiIiKgAWMDQCG84CQBoqOCLOCVIkozR4x50ba2H2Zp741Lakx00Z2rmDAwionxxR92cf0FERFRCWMDQiLnzHRj1LGAowjniRzSYRN+elbUpp10uAICplQUMIqJ8cUfcXKFKRERUQjgDQyPmFjsweIQkXxJnx+H6wheQCYVwsup6GKybID/8SZyV0znflxQKAQC3kBAR5ZEn6sH2xu2iYxAREVGBsIChEbMLHRg2dmDkQ9rrxdR990GKRFC2+zK4MxvQonPBumblBQjb/n0wd3XlMSURUelKZBLwJXzswCAiIiohLGBoxFw4CZvZAKvZIDpK0ZOiUUz9+SeQ9nrR9eS3MWtoQ/Lrdmz/swPo2PkR0fGIiAjZ7guAG0iIiIhKCWdgaMRcJIGGSnZfrJacyWDms/8d8VOn0P6PX4V12zaMHPPAbDGgc0ud6HhERHSeO5JdocohnkRERKWDBQyN8IYTqLdx/sVqyLIM9yNfRvjQITT/f3+LyhtvRCYlYbx/Fj07G2E0sbuFiMnobAsAAB5vSURBVEgtFjoweISEiIiodLCAoRFz4SRXqK7S/BP/Dt93voO6e+9F3Yc+BACYHJpHIppG32VsUSYiUhN39HwHBgsYREREJYMFDI3whhNcoboKwZ/9DJ6//3tU3normj7714u/P3rMjTKbER2beHyEiEhN3FE3yo3lqDBXiI5CREREBcIhnhqQkWTMR5JLrlCVZRnu8SBcZwOQ5QKFKwKpmWn4vvciTFfeg9gtd2L25enFr43bvVi3pwkGI2t9RERq4ol6OP+CiIioxLCAoQH+aBKSjEseIQnMxnDmDReGf+1CwBMrcLoi0f2+7OWz5y74bZ0O2HhVa+HzEBHRO3JH3Dw+QkREVGJYwNAAbzgJAKh/WwdGPJLC6HEPzvzaBedYANAB7etrcNmtXeje1gCDiR0F6fl5TP7JvZCiEXQ+/jjMHWt+7zp6vQ5GrqYlIlIdd9SNva17RccgIiKiAmIBQzBZlpFJS0teT0okgXT6ol+bdc7BlkqgNpPC2OtTGD7uxcSgH1JGRm2zFXsPdGD97npU1C50aKSATB7/EEVISibh/stPAe5p9Pz7E7Cu6xYdiYiIlikjZeCNedmBQUREVGJYwBDsyPdH8NYr00tfcQmfQA36/2UMAGBKBtHmOYYW1xuoDE9B9zQws+pH0CCdDu1f/xqsO3eKTkJERDmYi88hI2dYwCAiIioxLGAI5j4bQG1LOTZc2XLJ68RPnkLwFz+Hbe9e6Mosv/f1aV8UI+4wru1rQGNlGi3VMeh1awGsVTB58bNs3gTbVVeJjkFERDlyR86vUOUQTyIiopLCAoZgkUASHZv///buP7at+t7/+Mu/kjZ2mpBfJYyWNgmd2FBVCupAFMbUy48hTVdiExSmMNRx72Vi4ldhdEC7QqvSrmi/gIsYvZR9uxUKAk37TvtFYaPAUDd1FNQOqOp05VKCHTu1m3OSHNvxuX84CRRiO9DYx/F5Pv5a6iP7reSzrXn1/Xm/m3T2ZfPyPhM79Hv1vfu8Pv//N8tb+8lBnZv/+Lb+30s9umfNV+X1ekpYLQAAzosM5gKMtro2hysBAADlxCRHB2WztgaPpRRsKLz+NB2NytfYOGF4IUmxgZSagjWEFwAAVxgLMLhCAgCAuxBgOGhoICU7ayvUOHEwMSYTicrflv9fmeKmlXeFKgAA1SYyGFHAG9BJM05yuhQAAFBGBBgOMhOWJKmuoUiAEY3KPzv/vzL1GanjVqgCAFDNooNRtdW1yevhrzEAALgJ/8/voLEA48P1phPLRCLyt7XmfT1u0IEBAHCPiBnh+ggAAC5EgOGgsQAjWKADw85klInHFSjQgRE3UmqhAwMA4BJjHRgAAMBdCDAcZCZT8ng9mjkrf/iQicelbDbvDAzTymgoPaJmOjAAAC5g27Yig3RgAADgRgQYDjISlupmFd4ekonkJq372yb+i1rcSEkSV0gAAK6QtJKyRiw6MAAAcCECDAeZCavoCtVMNCpJ8s+e+C9qfUbuGgpDPAEAbjC+QjVIBwYAAG5DgOEgM2EpWGSFanq0AyOQ5wpJfDTAaAnSgQEAqH7jAQZXSAAAcB0CDAdNJsDIRPskn0++5uYJX4+bo1dI6unAAABUv+hgrjORAAMAAPcpSYCRzWa1Zs0aXXXVVeru7tbhw4c/8Ux/f78uueQSWZZVihIqXiY1ImswUzzAiETkb22Vxzvxjyo2kPv+NQUJMAAA1S8yGJFHHrXUtThdCgAAKLOSBBg7d+5UKpXSjh07tHLlSm3cuPG4119++WWtWLFCsVisFB8/LZjJ4itUpdwMjHzzL6RcB8asGX7V+n1TWh8AAJUoOhhV88xmBbwBp0sBAABl5i/Fm+7Zs0cXXHCBJGnRokXat2/fca97vV5t3bpVX//61yf1fplMRr29vVNe54k6kZpihwclSVbWKPg+Q0eOyDd3Tt5n/rcvocYZvor8/qC6cMbgNpz5yvRu/7tqCjTx85lifD/hJpx3uEm1nfeSBBiGYSgUCo1/7fP5lMlk5PfnPu7888//VO/n9/vV3t4+pTWeqN7e3hOqyXgvN4RsbscpamoP5n3uWH+/QkuX6uQ8n2WO/EuzG+sq7vuD6nKi5x2YbjjzlSsxktCchjn8fKYQ5x1uwnmHm0zn855IJCb885JcIQmFQjJNc/zrbDY7Hl4gx0iMXiFpzD+7Ijs4qOzAgPyz8w8qixsptYTYQAIAcIfIYIQBngAAuFRJAozFixdr165dkqS9e/dqwYIFpfiYac1MWvLXeFUzM3+wk4nmJq3721rzPhMzLDWHGOAJAKh+g+lBDaQGNDtIgAEAgBuVpC3i4osv1quvvqrly5fLtm1t2LBBW7du1dy5c7Vs2bJSfOS0YyYsBRtq5fF48j6THg0wAnk6MDIjWR0dTKs5SAcGAKD6sUIVAAB3K0mA4fV6dd999x33Z52dnZ947sUXXyzFx08LZsKaxArVsQ6MibeQ9JspSVJLPQEGAKD6RQZz86Pa6vJv5wIAANWrJFdIUNykAoyxKyR5OjBixmiAEeQKCQCg+tGBAQCAuxFgOMC2bZnJ1CQCjIg8dXXyBifeUhI3c4NA6cAAALgBHRgAALgbAYYDrMGMRtJZBRsKd06kI1EF2tryzsmIGbkAo5kODACAC0TMiOpr6lUXqHO6FAAA4AB2mzrAHF+hWvwKSbEVqpLUzBpVoCLsj+/XTS/epPhQ3OlScIJs2fIo/5BlOGPEHlFXY5fTZQAAAIcQYDhgLMAIFR3iGdHMxYvzvt5nWKrxeTVrBj9GwGnv9L+j/3r+vxT0B7XizBVOl4MTZBiGQqGQ02VgAkvalzhdAgAAcAi/+TrAmEQHhm3bykSjCszOf883bqTUHKopuIoVQOn1JHr0n8//p2p9tdpy6RbNqZ/jdEk4Qb29vWpvb3e6DAAAAHwEMzAcMJgcDTAa8gcYI4mE7HQ67wpVSYobllq4PgI46vCxw7r+T9fL6/Hqfy75H8ILAAAAoETowHCAkUhpRjAgXyB/fjS+QrUt/wyM2GgHBgBnvDfwnr79x28rk81o62VbNa9hntMlAQAAAFWLDgwHmAmr+ADPSG5VXLEOjOYgHRiAEz4wP9D1f7peQ5khPXbJY+ps7HS6JAAAAKCq0YHhgFyAUWyFai7AyDcDw7ZtxYyUWurpwADKrW+wT9/+47eVtJLacskWfb7p806XBAAAAFQ9OjAcMKkOjLErJK2tE74+YGWUGsmqhQ4MoKziQ3Fd/6frFRuK6ZF/e0RfbPmi0yUBAAAArkAHRpllR7IaHEhN4gpJVL6mJnlqJu6wiBspSaIDAyijxHBC//H8f6jX7NV/L/tvLWpb5HRJAAAAgGsQYJTZ4LGUZBfeQCLlOjD8swsN8MxtMmEGhvsMpAb0Rt8bsm27bJ/Z39+vppGmsn1eJbJl66HXH9Lh5GE9/G8P65yTz3G6JAAAAMBVCDDKzEzkOidCRTow0tGI/G0TXx+RcgM8JbGFxIUefP1BPfn2k06X4Up+r18/+8rPdG77uU6XAgAAALgOAUaZmYlc8FB8BkafZn7xzLyvx0avkLSG6MBwm7f739YZTWfonnPvKdtnxmIxtbS0lO3zKlVbXZtODp7sdBkAAACAKxFglJmZLB5g2Om0RuLxgitUx66QnBSkA8NNbNtWOBHWpfMu1cLWhWX73N5Mr9pb28v2eQAAAADwcWwhKTMjYcnr9WhmKJD3mUxfn2Tb8udZoSrlhnieVBdQwMeP0E3iw3EdSx1TZ2On06UAAAAAQFnx22+ZmQlLdQ018ng9eZ8ZW6EaKDLEs5nrI65zMHFQktTR0OFwJQAAAABQXgQYZWYmrKLzL9KRXIBR6ApJ3EipmesjrhNOhCVJXY1dDlcCAAAAAOVFgFFmkwkwxjowCq5RNS211NOB4TY9iR7V19SrZSYDNQEAAAC4CwFGmZnJ1CQCjIgUCMjX2Jj3mdiApRY6MFznYOKgOhs65fHkv4IEAAAAANWIAKOM0taIUkMZBRsKBw+ZaFSB1lZ5vBP/eFKZrI4NZ9TCDAxXsW1b4WSYAZ4AAAAAXIkAo4zMRG71aWgSMzAKzr8wc+/DEE936R/uV9JKEmAAAAAAcCUCjDIaCzDqil0hiUQKzr+IGylJUnOIKyRuMjbAs7OBAAMAAACA+xBglJGZnFwHRiYalX92/g6MmJF7H66QuEs4ORpg0IEBAAAAwIUIMMrIGO3ACDbkDx5GDFNZ01SgwBWS2GgHRgsdGK4SToQVCoTUVpf/bAAAAABAtSLAKCMzYSlQ61PNTH/eZyazQjVOB4YrhRNhdTR2sIEEAAAAgCsRYJSRmZjkClVJ/tbCV0hmBLyqq/FNaX2obD3JHnU1djldBgAAAAA4ggCjjMyEpWBj8RWqkgrOwIgbKTUHa/mXeBfpH+5X/3C/Oho6nC4FAAAAABxBgFFGZtIq2oGRjuQ6MArOwDBTaqnn+oib9CR6JDHAEwAAAIB7EWCUiW3buQCjwABPScpEovKGQvIGg3mfiQ1YagkywNNNxlaocoUEAAAAgFsRYJTJsJlWNmNPYgZGtOAAT0mKmxYDPF0mnAyrzl+n2XWFzwYAAAAAVCsCjDIxJ7FCVZIykYj8ba15X89m7dwMDFaoukpPokedjZ3MPQEAAADgWgQYZWIczQUYoZOKzMDoiyrQlv9f2Y8Np5XJ2mqmA8NVDiYOMv8CAAAAgKsRYJTJYDIlSQWvkNjZrDLRPvkLDfA0cu/TQgeGaySGE4oPx9XZQIABAAAAwL0IMMrEGL1CUjcrf/AwcvSolMkUnIERM3LvwwwM9+hJ5jaQdDSyQhUAAACAexFglImZtDSzPiCfP/+3PDO6QrXQDIz4eAcGAYZbHEwclMQGEgAAAADuRoBRJmbCKrqBJD0aYAQm0YHBEE/36En2aKZ/pk4Onux0KQAAAADgGAKMMplMgJGJ9klSwSskccOS1yOdVEeA4RbhRFgdDR3yevivKwAAAAD34jeiMjET1qRWqMrjkb+5Oe8zMTOlpmCNfF7WabpFOBFmAwkAAAAA1yPAKIORTFZDA+niHRh9UflamuUJBPI+Exuw1Bxk/oVbJK2k+ob6CDAAAAAAuB4BRhkMHhtdodpQ+NpHOhJRoDX/ClVJipsp5l+4yKHkIUlihSoAAAAA1yPAKANzdIXqZGZgFJp/IeWGeLKBxD3GNpDQgQEAAADA7QgwymAswAidVHwGhr+tSAeGQQeGm4QTYc3wzdApoVOcLgUAAAAAHEWAUQbGWAdGgSGe2VRKI0ePyj87f4AxnB6RYWXowHCRnmSP5jfMZwMJAAAAANfjt6IyGExa8vo8mhHKP5xzbIVqoMAVkpiRC0Ja6MBwjXAirK7GLqfLAAAAAADHEWCUgTG6QtXjyb/6NBONSFLBKyRxIzcMlC0k7jCQGlBkMKKOxg6nSwEAAAAAxxFglIGZsCYxwDMqSfK3TaIDo54Aww16kj2S2EACAAAAABIBRlmYiZSCjYWvfWQiYx0YrXmf+bADgyskbtCTyAUYXCEBAAAAAAKMsphMB0Y6GpWnpka+xsa8z8TMsRkYdGC4wcHEQdX6atlAAgAAAAAiwCi51HBGaWuk+BWSSFT+traCczJiAykFa3yaWeOb6jJRgcLJsOY3zJfPy88bAAAAAAgwSsycxApVKXeFxF9gA4kkxU1LzXRfuEZPokedjcy/AAAAAACJAKPkxgKM0CSGeAZm599AIuWGeLJC1R3MtKles5cBngAAAAAwigCjxMY7MAoEGLZtKx2Nyt9aOMCIGyk6MFxibIAnK1QBAAAAIIcAo8SM0QCjriF/50TWMGQPDRW9QhIzUgzwdIlwMiyJDSQAAAAAMIYAo8TMZEo1M3yqmeHP+8yHK1Tzd2CMZG31m1whcYtwIqwab40+F/qc06UAAAAAQEUgwCixyaxQzUSjklRwBkZiMKWsLTUHCTDcIJwIa17DPPm9+YMvAAAAAHATAowSm0yAkY7kAoxCHRgxIyVJaqnnCokb9CTZQAIAAAAAH0WAUWKT6sAYu0JSYAZG3MjN0mgOEmBUu8H0oI4YR9hAAgAAAAAfQYBRQnbW1mAyNakrJN6GBnlnzMj7TMzMdWC01nOFpNodSh6SJDowAAAAAOAjCDBKaMhIK5u1FWwocoUkGlGgrbXgM7EBOjDcYmwDCQEGAAAAAHyIAKOEzNEVqqGiHRh98rcVXqEaNy35vB41zAxMWX2oTAcTB+X3+jWnfo7TpQAAAABAxSDAKKGxAKOusfC1j0wkUnCApyTFjZSagzXyej1TVh8qU0+iR/NmsYEEAAAAAD6KAKOEjEl0YNgjI8rEYvIXWKEqSTHDUnOI6yNuEE6E1dXY5XQZAAAAAFBRCDBKyExakkeaOSt/B0YmHpdGRop2YMSMlFpCDPCsdkOZIR0xjqijscPpUgAAAACgohBglJCZsFRXXyOfL/+3OROJSpICBVaoSrkOjBY6MKreoeQh2bJZoQoAAAAAH0OAUUJmYhIrVPtyAUbRIZ6jMzBQ3cKJ3AYSrpAAAAAAwPEIMErITFjFA4xIRJIKXiEZTGU0lB5RSz0dGNWuJ9kjv8evObPYQAIAAAAAH1WSACObzWrNmjW66qqr1N3drcOHDx/3+tNPP60rrrhCV155pf785z+XooSKYCaLBxjpaFTyeuVvac77TGwgJUl0YLjAwcRBnTbrNAW8rMsFAAAAgI8qyZ7GnTt3KpVKaceOHdq7d682btyoRx55RJLU19enbdu26dlnn5VlWbrmmmt0/vnnq6amun45Tw2nNWyk9fY/durggYfzPtf593+qvn6mVu18PO8zRwdTCjRGdWAoqmcPzCpFuagQ/4z9U4vaFjldBgAAAABUnJIEGHv27NEFF1wgSVq0aJH27ds3/tqbb76ps846SzU1NaqpqdHcuXP19ttva+HChXnfL5PJqLe3txSlnpBCNVnDafkyQzrzry+quf+fBd9n73yPfvfBzwo+M6NdeqpHUs9nqRTTyfwZ86fdeQeqEWcebsJ5h5tw3uEm1XbeSxJgGIahUCg0/rXP51Mmk5Hf75dhGKqvrx9/LRgMyjCMwkX6/Wpvby9FqZ9Zb29v0Zr+fZ1PfX2rlC7yXl88qVG/ChS+MjDT71f9zJL8uFBBvB6vWme2yuPxOF3KcSZz3oFqwpmHm3De4Sacd7jJdD7viURiwj8vyW/EoVBIpmmOf53NZuX3+yd8zTTN4wKNatLe/jm1t3/O6TIAAAAAAJj2SjLEc/Hixdq1a5ckae/evVqwYMH4awsXLtSePXtkWZYGBgYUDoePex0AAAAAAODjStKBcfHFF+vVV1/V8uXLZdu2NmzYoK1bt2ru3LlatmyZuru7dc0118i2bd16662qrWU9KAAAAAAAyK8kAYbX69V999133J91dnaO/+crr7xSV155ZSk+GgAAAAAAVKGSXCEBAAAAAACYSgQYAAAAAACg4hFgAAAAAACAikeAAQAAAAAAKh4BBgAAAAAAqHgEGAAAAAAAoOIRYAAAAAAAgIpHgAEAAAAAACoeAQYAAAAAAKh4BBgAAAAAAKDiEWAAAAAAAICKR4ABAAAAAAAqHgEGAAAAAACoeAQYAAAAAACg4nls27adLqKYvXv3qra21ukyAAAAAABAiVmWpUWLFn3iz6dFgAEAAAAAANyNKyQAAAAAAKDiEWAAAAAAAICKR4ABAAAAAAAqHgEGAAAAAACoeAQYAAAAAACg4hFgAAAAAACAiud3uoDpJpvNau3atXrnnXdUU1Oj9evX67TTTnO6LGDKpNNp3XXXXTpy5IhSqZS+853vqKurS6tWrZLH49Hpp5+uH/zgB/J6yT9RPeLxuK644go9/vjj8vv9nHdUtUcffVQvvvii0um0rr76ai1ZsoQzj6qUTqe1atUqHTlyRF6vV+vWreN/41GV3njjDT3wwAPatm2bDh8+POEZf+ihh/SXv/xFfr9fd911lxYuXOh02Z8J/239lHbu3KlUKqUdO3Zo5cqV2rhxo9MlAVPqN7/5jRobG7V9+3Y99thjWrdune6//37dcsst2r59u2zb1gsvvOB0mcCUSafTWrNmjWbMmCFJnHdUtd27d+v111/Xk08+qW3btumDDz7gzKNqvfTSS8pkMnrqqad044036ic/+QnnHVXnscce0z333CPLsiRN/PeY/fv3629/+5ueeeYZ/ehHP9K9997rcNWfHQHGp7Rnzx5dcMEFkqRFixZp3759DlcETK3LLrtMN9988/jXPp9P+/fv15IlSyRJF154of761786VR4w5TZt2qTly5erra1NkjjvqGqvvPKKFixYoBtvvFE33HCDLrroIs48qtb8+fM1MjKibDYrwzDk9/s576g6c+fO1YMPPjj+9URnfM+ePVq6dKk8Ho9OOeUUjYyMqL+/36mSTwgBxqdkGIZCodD41z6fT5lMxsGKgKkVDAYVCoVkGIZuuukm3XLLLbJtWx6PZ/z1gYEBh6sEpsZzzz2npqam8WBaEucdVe3o0aPat2+ffvrTn+ree+/V7bffzplH1aqrq9ORI0f01a9+VatXr1Z3dzfnHVXn0ksvld//4WSIic74x3+Hnc5nnxkYn1IoFJJpmuNfZ7PZ4w4MUA16e3t144036pprrtHXvvY1bd68efw10zQ1a9YsB6sDps6zzz4rj8ej1157TW+99ZbuvPPO4/5FgvOOatPY2KiOjg7V1NSoo6NDtbW1+uCDD8Zf58yjmjzxxBNaunSpVq5cqd7eXn3rW99SOp0ef53zjmr00ZkuY2f847/Dmqap+vp6J8o7YXRgfEqLFy/Wrl27JEl79+7VggULHK4ImFqxWEwrVqzQHXfcoW984xuSpC984QvavXu3JGnXrl0655xznCwRmDK/+tWv9Mtf/lLbtm3TGWecoU2bNunCCy/kvKNqnX322Xr55Zdl27YikYiGhoZ03nnnceZRlWbNmjX+S1pDQ4MymQx/p0HVm+iML168WK+88oqy2azef/99ZbNZNTU1OVzpZ+Oxbdt2uojpZGwLyYEDB2TbtjZs2KDOzk6nywKmzPr16/X73/9eHR0d43929913a/369Uqn0+ro6ND69evl8/kcrBKYet3d3Vq7dq28Xq9Wr17NeUfV+uEPf6jdu3fLtm3deuutOvXUUznzqEqmaequu+5SX1+f0um0rr32Wp155pmcd1Sd9957T7fddpuefvppHTp0aMIz/uCDD2rXrl3KZrP6/ve/P23DOwIMAAAAAABQ8bhCAgAAAAAAKh4BBgAAAAAAqHgEGAAAAAAAoOIRYAAAAAAAgIpHgAEAAAAAACqe3+kCAADA9LVx40bt379ffX19Gh4e1pw5c+T3+3X22Wfru9/97pR+1vPPP6+FCxdq9uzZU/q+AABgemCNKgAAOGHPPfecenp6dPvtt5fsM7q7u7V27Vp1dnaW7DMAAEDlogMDAABMqd27d+upp57Sj3/8Y1188cU666yzdPjwYZ177rkaGBjQm2++qfnz52vz5s3q7e3V6tWrZVmWamtrtW7dOjU1Nenmm2+WYRgaHh7WHXfcoaGhIb311lu68847tX37du3YsUO//e1v5fF4dPnll+vaa6/VqlWrZNu2ent7NTg4qE2bNunUU0/9xHt96UtfcvpbBAAAPgMCDAAAUDJHjhzRL37xC7W2tmrJkiV65plntHr1ai1btkzHjh3Tpk2b1N3drS9/+ct67bXX9MADD+iGG25QLBbTE088oXg8rn/961+66KKLdMYZZ2jt2rV699139bvf/U7bt2+Xx+PRddddp6VLl0qS5syZo02bNumll17S5s2btXLlyk+8FwAAmJ4IMAAAQMk0NjbqlFNOkSTV1dWpq6tLklRfXy/LsnTgwAE9+uij2rJli2zbViAQ0Omnn65vfvObuu2225TJZNTd3X3cex44cEDvv/++rrvuOklSMpnUu+++K0k699xzJUlnnXWWNmzYUPS9AADA9EGAAQAASsbj8RR8vaOjQytWrNDixYsVDof197//Xe+8845M09TPf/5zRaNRLV++XF/5ylfk8Xhk27Y6OjrU1dWlLVu2yOPx6IknntCCBQv0hz/8Qfv379c555yjf/zjHzr99NPzvhcAAJh+CDAAAIBj7rzzTq1du1aWZWl4eFh333235s2bp4cffli//vWvFQgEdNNNN0nKdVV873vf0+OPP67zzjtPV199tVKp1HGbSXbt2qUXXnhB2WxW999/v9ra2iZ8LwAAMP2whQQAAFSFVatW6fLLL9eFF17odCkAAKAEvE4XAAAAAAAAUAwdGAAAAAAAoOLRgQEAAAAAACoeAQYAAAAAAKh4BBgAAAAAAKDiEWAAAAAAAICKR4ABAAAAAAAq3v8B86+EQSfCuGwAAAAASUVORK5CYII=\n",
+      "text/plain": [
+       "<Figure size 1080x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "experiments = experiments.sort_values(by =['rules_price']).reset_index(drop=True)\n",
+    "\n",
+    "cols = 1\n",
+    "rows = 1\n",
+    "cc_idx = 0\n",
+    "# config_labels = ['RULE 1,'RULE 2']\n",
+    "\n",
+    "\n",
+    "while cc_idx<len(experiments):\n",
+    "    cc = experiments.iloc[cc_idx]['rules_price']\n",
+    "    cc_label = experiments.iloc[cc_idx]['rules_price']\n",
+    "#     secondary_label = experiments.iloc[cc_idx]['string']\n",
+    "    sub_experiments = experiments[experiments['rules_price']==cc]\n",
+    "    cc_idx += len(sub_experiments)\n",
+    "    fig, axs = plt.subplots(ncols=cols, nrows=rows, figsize=(15*cols,7*rows))\n",
+    "    sub_experiments.reset_index(drop=False)\n",
+    "    for i, experiment in sub_experiments.iterrows():\n",
+    "        df = experiment['dataset'].copy()\n",
+    "        # filter out starting point\n",
+    "        df = df[df.timestep > 0] \n",
+    "        # FIRST RUN ONLY\n",
+    "        df = df[df.run == 1]  \n",
+    "        \n",
+    "\n",
+    "# MAKE A FOR LOOP or FUNCTION FOR ALL AGENTS\n",
+    "        df['agent_1_attest_1'] = df.agents.apply(lambda x: np.array(x['agent_attestations_1'][0]))\n",
+    "        df['agent_2_attest_1'] = df.agents.apply(lambda x: np.array(x['agent_attestations_1'][1]))\n",
+    "        df['agent_3_attest_1'] = df.agents.apply(lambda x: np.array(x['agent_attestations_1'][2]))\n",
+    "        df['agent_4_attest_1'] = df.agents.apply(lambda x: np.array(x['agent_attestations_1'][3]))\n",
+    "        df['agent_5_attest_1'] = df.agents.apply(lambda x: np.array(x['agent_attestations_1'][4]))\n",
+    "    \n",
+    "#         df = df.groupby('timestep').agg({'supply': ['min', 'mean', 'max']}).reset_index()\n",
+    "    \n",
+    "#         df['agent_attest_1'] = np.array(df.agent_attest_1,dtype = float)\n",
+    "#         print(df['agent_attest_1'])\n",
+    "#         print(df['agent_attest_1'][10])\n",
+    "        plot_label = experiment['rules_price']\n",
+    "        ax = axs\n",
+    "        title = 'agent_attest_1' + '\\n' + 'Scenario: ' + str(cc_label)  + ' rules_price'\n",
+    "        ax.set_title(title)\n",
+    "        ax.set_ylabel('Amount')\n",
+    "        colors = ['b','orange', 'g', 'magenta', 'r']\n",
+    "        \n",
+    "#         ax.plot(df.timestep, df['price'],color = colors[0], label='Price')\n",
+    "        df.plot(x='timestep', y='agent_1_attest_1', label='agent_1', ax=ax, legend=True)\n",
+    "        df.plot(x='timestep', y='agent_2_attest_1', label='agent_2', ax=ax, legend=True)\n",
+    "        df.plot(x='timestep', y='agent_3_attest_1', label='agent_3', ax=ax, legend=True)\n",
+    "        df.plot(x='timestep', y='agent_4_attest_1', label='agent_4', ax=ax, legend=True)\n",
+    "        df.plot(x='timestep', y='agent_5_attest_1', label='agent_5', ax=ax, legend=True)\n",
+    "\n",
+    "#         ax.fill_between(df.timestep, df[('supply','min')], df[('supply','max')], supply=0.3) \n",
+    "\n",
+    "        ax.legend()\n",
+    "\n",
+    "        ax.set_xlabel('Timesteps')\n",
+    "        ax.grid(color='0.9', linestyle='-', linewidth=1)\n",
+    "\n",
+    "        plt.tight_layout()\n",
+    "        \n",
+    "fig.tight_layout(rect=[0, 0, 1, .97])\n",
+    "fig.patch.set_alpha(1)\n",
+    "display(fig)\n",
+    "plt.close()\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### INTO AGENT LEVEL across columns in AGENTS dataframe\n",
+    "\n",
+    "### THIS CASE FIRST AGENT\n",
+    "## CAN ADD MORE VARS and REPEAT FOR OTHER AGENTS"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 37,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1080x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1080x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "experiments = experiments.sort_values(by =['rules_price']).reset_index(drop=True)\n",
+    "\n",
+    "cols = 1\n",
+    "rows = 1\n",
+    "cc_idx = 0\n",
+    "# config_labels = ['RULE 1,'RULE 2']\n",
+    "\n",
+    "\n",
+    "while cc_idx<len(experiments):\n",
+    "    cc = experiments.iloc[cc_idx]['rules_price']\n",
+    "    cc_label = experiments.iloc[cc_idx]['rules_price']\n",
+    "#     secondary_label = experiments.iloc[cc_idx]['string']\n",
+    "    sub_experiments = experiments[experiments['rules_price']==cc]\n",
+    "    cc_idx += len(sub_experiments)\n",
+    "    fig, axs = plt.subplots(ncols=cols, nrows=rows, figsize=(15*cols,7*rows))\n",
+    "    sub_experiments.reset_index(drop=False)\n",
+    "    for i, experiment in sub_experiments.iterrows():\n",
+    "        df = experiment['dataset'].copy()\n",
+    "        # filter out starting point\n",
+    "        df = df[df.timestep > 0] \n",
+    "        # FIRST RUN ONLY\n",
+    "        df = df[df.run == 1]  \n",
+    "\n",
+    "# MAKE A FOR LOOP or FUNCTION FOR ALL AGENTS\n",
+    "        df['agent_1_reserve'] = df.agents.apply(lambda x: np.array(x['agent_reserve'][0]))\n",
+    "        df['agent_1_supply_1'] = df.agents.apply(lambda x: np.array(x['agent_supply_1'][0]))\n",
+    "        df['agent_1_supply_0'] = df.agents.apply(lambda x: np.array(x['agent_supply_0'][0]))\n",
+    "        df['agent_1_supply_free'] = df.agents.apply(lambda x: np.array(x['agent_supply_free'][0]))\n",
+    "        df['agent_1_attest_1'] = df.agents.apply(lambda x: np.array(x['agent_attestations_1'][0]))\n",
+    "        df['agent_1_attest_0'] = df.agents.apply(lambda x: np.array(x['agent_attestations_0'][0]))\n",
+    "\n",
+    "    \n",
+    "#         df = df.groupby('timestep').agg({'supply': ['min', 'mean', 'max']}).reset_index()\n",
+    "    \n",
+    "#         df['agent_attest_1'] = np.array(df.agent_attest_1,dtype = float)\n",
+    "#         print(df['agent_attest_1'])\n",
+    "#         print(df['agent_attest_1'][10])\n",
+    "        plot_label = experiment['rules_price']\n",
+    "        ax = axs\n",
+    "        title = 'agent_attest_1' + '\\n' + 'Scenario: ' + str(cc_label)  + ' rules_price'\n",
+    "        ax.set_title(title)\n",
+    "        ax.set_ylabel('Amount')\n",
+    "        colors = ['b','orange', 'g', 'magenta', 'r']\n",
+    "        \n",
+    "#         ax.plot(df.timestep, df['price'],color = colors[0], label='Price')\n",
+    "        df.plot(x='timestep', y='agent_1_reserve', label='agent_reserve', ax=ax, legend=True)\n",
+    "        df.plot(x='timestep', y='agent_1_supply_1', label='agent_supply_1', ax=ax, legend=True)\n",
+    "        df.plot(x='timestep', y='agent_1_supply_0', label='agent_supply_0', ax=ax, legend=True)\n",
+    "        df.plot(x='timestep', y='agent_1_supply_free', label='agent_supply_free', ax=ax, legend=True)\n",
+    "        df.plot(x='timestep', y='agent_1_attest_1', label='agent_attestations_1', ax=ax, legend=True)\n",
+    "        df.plot(x='timestep', y='agent_1_attest_0', label='agent_attestations_0', ax=ax, legend=True)\n",
+    "\n",
+    "#         ax.fill_between(df.timestep, df[('supply','min')], df[('supply','max')], supply=0.3) \n",
+    "\n",
+    "        ax.legend()\n",
+    "\n",
+    "        ax.set_xlabel('Timesteps')\n",
+    "        ax.grid(color='0.9', linestyle='-', linewidth=1)\n",
+    "\n",
+    "        plt.tight_layout()\n",
+    "        \n",
+    "fig.tight_layout(rect=[0, 0, 1, .97])\n",
+    "fig.patch.set_alpha(1)\n",
+    "display(fig)\n",
+    "plt.close()\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.6.5"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}
diff --git a/__pycache__/run2.cpython-36.pyc b/__pycache__/run2.cpython-36.pyc
index e043eb714fe5299cb8cdb6bcf5816f3adf359507..43a5cb2da6f9cdbbc1595686225a88c9e7e1912c 100644
GIT binary patch
delta 86
zcmaFC_Ki*1n3tDJfk7;uk%NKZF#{4{2C^N1xVUbjat0?u3R4bCE^8F)#tsE0M!v}_
WnO2Jl0cAxPd4P}uOafU<JWK#cZ42B0

delta 74
zcmeyy_JU2>n3tC;O_e|X2|EMBV+JI^3}ib1adFW^<&2HfWtkXxCvRn1Eg}FE6Jg{5
LLJlwqWHAE(1#1gz

diff --git a/mutliscale_ABM_main.ipynb b/mutliscale_ABM_main.ipynb
new file mode 100644
index 0000000..437289c
--- /dev/null
+++ b/mutliscale_ABM_main.ipynb
@@ -0,0 +1,1663 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# NOTES\n",
+    "## USING run.py and NOT run2.py\n",
+    "## sys path (or run2.py)  may interfere with overwriting configs, NEED TO LOOK INTO MORE\n",
+    "## WILL ONLY WORK FOR SIMULATIONS WHERE N>1. CAN STILL VISUALIZE SINGLE RUN DATA AS SEEN BELOW, BUT cadCAD RUN MUST BE N>1."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {
+    "scrolled": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\n",
+      "<class 'list'>\n",
+      "[1000]\n",
+      "\n",
+      "CHECKPOINT 1\n",
+      "CHECKPOINT 2\n",
+      "Initial Conditions (config.py) :  {'reserve': 300, 'price': 1, 'realized_price': 0, 'spot_price': 1, 'private_price': 0, 'private_alpha': 0, 'kappa': 0, 'supply': 600.0, 'alpha': [0.5], 'supply_0': 200, 'supply_1': 200, 'supply_free': 600.0, 'attestations': 40, 'attestations_0': 20, 'attestations_1': 20, 'invariant_V': 1200.0, 'invariant_I': 650.0, 'agents':    agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply_1  \\\n",
+      "0                     0                     0             50              10   \n",
+      "1                     0                     0             50              10   \n",
+      "2                     0                     0             50              10   \n",
+      "3                     0                     0             50              10   \n",
+      "4                     0                     0             50              10   \n",
+      "\n",
+      "   agent_supply_0  agent_supply_free  \n",
+      "0              10                 30  \n",
+      "1              10                 30  \n",
+      "2              10                 30  \n",
+      "3              10                 30  \n",
+      "4              10                 30  , 'chosen_agent': 0}\n",
+      "Params (config.py) :  {'starting_kappa': 2, 'starting_alpha': 0.5, 'money_raised': 1000, 'C': 700, 'f': 0.03, 'm': 0.15, 'period': 2000, 'rules_price': 'martin'}\n",
+      "Params (config.py) :  {'starting_kappa': 2, 'starting_alpha': 0.5, 'money_raised': 1000, 'C': 700, 'f': 0.03, 'm': 0.15, 'period': 2000, 'rules_price': 'martin'}\n",
+      "Params (config.py) :  {'starting_kappa': 2, 'starting_alpha': 0.5, 'money_raised': 1000, 'C': 700, 'f': 0.03, 'm': 0.15, 'period': 2000, 'rules_price': 'martin'}\n",
+      "Params (config.py) :  {'starting_kappa': 2, 'starting_alpha': 0.5, 'money_raised': 1000, 'C': 700, 'f': 0.03, 'm': 0.15, 'period': 2000, 'rules_price': 'step'}\n",
+      "Params (config.py) :  {'starting_kappa': 2, 'starting_alpha': 0.5, 'money_raised': 1000, 'C': 700, 'f': 0.03, 'm': 0.15, 'period': 2000, 'rules_price': 'step'}\n",
+      "Params (config.py) :  {'starting_kappa': 2, 'starting_alpha': 0.5, 'money_raised': 1000, 'C': 700, 'f': 0.03, 'm': 0.15, 'period': 2000, 'rules_price': 'step'}\n",
+      "\n",
+      "                  ___________    ____\n",
+      "  ________ __ ___/ / ____/   |  / __ \\\n",
+      " / ___/ __` / __  / /   / /| | / / / /\n",
+      "/ /__/ /_/ / /_/ / /___/ ___ |/ /_/ /\n",
+      "\\___/\\__,_/\\__,_/\\____/_/  |_/_____/\n",
+      "by cadCAD\n",
+      "\n",
+      "Execution Mode: multi_proc\n",
+      "Configuration Count: 6\n",
+      "Dimensions of the first simulation: (Timesteps, Params, Runs, Vars) = (100, 8, 1, 19)\n",
+      "Execution Method: parallelize_simulations\n",
+      "Execution Mode: parallelized\n",
+      "Total execution time: 9.21s\n"
+     ]
+    }
+   ],
+   "source": [
+    "# import sys\n",
+    "# sys.path.append('../')\n",
+    "import pandas as pd\n",
+    "\n",
+    "import matplotlib.pyplot as plt\n",
+    "# import run2\n",
+    "\n",
+    "from src.sim import run\n",
+    "\n",
+    "import seaborn as sns\n",
+    "\n",
+    "# For analysis\n",
+    "import numpy as np\n",
+    "import pandas as pd\n",
+    "\n",
+    "sns.set_style(\"whitegrid\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "cadCAD                        0.4.17             \n"
+     ]
+    }
+   ],
+   "source": [
+    "!pip list | grep cadCAD"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {
+    "scrolled": false
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "config_ids =  [{'M': {'starting_kappa': 2, 'starting_alpha': 0.5, 'money_raised': 1000, 'C': 700, 'f': 0.03, 'm': 0.15, 'period': 2000, 'rules_price': 'martin'}, 'simulation_id': 0, 'run_id': 0}, {'M': {'starting_kappa': 2, 'starting_alpha': 0.5, 'money_raised': 1000, 'C': 700, 'f': 0.03, 'm': 0.15, 'period': 2000, 'rules_price': 'martin'}, 'simulation_id': 0, 'run_id': 1}, {'M': {'starting_kappa': 2, 'starting_alpha': 0.5, 'money_raised': 1000, 'C': 700, 'f': 0.03, 'm': 0.15, 'period': 2000, 'rules_price': 'martin'}, 'simulation_id': 0, 'run_id': 2}, {'M': {'starting_kappa': 2, 'starting_alpha': 0.5, 'money_raised': 1000, 'C': 700, 'f': 0.03, 'm': 0.15, 'period': 2000, 'rules_price': 'step'}, 'simulation_id': 1, 'run_id': 0}, {'M': {'starting_kappa': 2, 'starting_alpha': 0.5, 'money_raised': 1000, 'C': 700, 'f': 0.03, 'm': 0.15, 'period': 2000, 'rules_price': 'step'}, 'simulation_id': 1, 'run_id': 1}, {'M': {'starting_kappa': 2, 'starting_alpha': 0.5, 'money_raised': 1000, 'C': 700, 'f': 0.03, 'm': 0.15, 'period': 2000, 'rules_price': 'step'}, 'simulation_id': 1, 'run_id': 2}]\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "C:\\Users\\mbarl\\Projects\\ICF_Internal\\src\\sim\\run.py:78: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
+      "  sub_df = df[df.simulation == config_id['simulation_id']][df.run == config_id['run_id'] + 1]\n",
+      "C:\\Users\\mbarl\\Projects\\ICF_Internal\\src\\sim\\run.py:78: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
+      "  sub_df = df[df.simulation == config_id['simulation_id']][df.run == config_id['run_id'] + 1]\n",
+      "C:\\Users\\mbarl\\Projects\\ICF_Internal\\src\\sim\\run.py:78: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
+      "  sub_df = df[df.simulation == config_id['simulation_id']][df.run == config_id['run_id'] + 1]\n",
+      "C:\\Users\\mbarl\\Projects\\ICF_Internal\\src\\sim\\run.py:78: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
+      "  sub_df = df[df.simulation == config_id['simulation_id']][df.run == config_id['run_id'] + 1]\n",
+      "C:\\Users\\mbarl\\Projects\\ICF_Internal\\src\\sim\\run.py:78: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
+      "  sub_df = df[df.simulation == config_id['simulation_id']][df.run == config_id['run_id'] + 1]\n",
+      "C:\\Users\\mbarl\\Projects\\ICF_Internal\\src\\sim\\run.py:78: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
+      "  sub_df = df[df.simulation == config_id['simulation_id']][df.run == config_id['run_id'] + 1]\n"
+     ]
+    }
+   ],
+   "source": [
+    "experiments = run.run()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "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>index</th>\n",
+       "      <th>starting_kappa</th>\n",
+       "      <th>starting_alpha</th>\n",
+       "      <th>money_raised</th>\n",
+       "      <th>C</th>\n",
+       "      <th>f</th>\n",
+       "      <th>m</th>\n",
+       "      <th>period</th>\n",
+       "      <th>rules_price</th>\n",
+       "      <th>dataset</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>0</td>\n",
+       "      <td>2</td>\n",
+       "      <td>0.5</td>\n",
+       "      <td>1000</td>\n",
+       "      <td>700</td>\n",
+       "      <td>0.03</td>\n",
+       "      <td>0.15</td>\n",
+       "      <td>2000</td>\n",
+       "      <td>martin</td>\n",
+       "      <td>reserve     price  realized_price  sp...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>0</td>\n",
+       "      <td>2</td>\n",
+       "      <td>0.5</td>\n",
+       "      <td>1000</td>\n",
+       "      <td>700</td>\n",
+       "      <td>0.03</td>\n",
+       "      <td>0.15</td>\n",
+       "      <td>2000</td>\n",
+       "      <td>step</td>\n",
+       "      <td>reserve     price  realized_price  sp...</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   index  starting_kappa  starting_alpha  money_raised    C     f     m  \\\n",
+       "0      0               2             0.5          1000  700  0.03  0.15   \n",
+       "1      0               2             0.5          1000  700  0.03  0.15   \n",
+       "\n",
+       "   period rules_price                                            dataset  \n",
+       "0    2000      martin           reserve     price  realized_price  sp...  \n",
+       "1    2000        step           reserve     price  realized_price  sp...  "
+      ]
+     },
+     "execution_count": 4,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "experiments"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "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>reserve</th>\n",
+       "      <th>price</th>\n",
+       "      <th>realized_price</th>\n",
+       "      <th>spot_price</th>\n",
+       "      <th>private_price</th>\n",
+       "      <th>private_alpha</th>\n",
+       "      <th>kappa</th>\n",
+       "      <th>supply</th>\n",
+       "      <th>alpha</th>\n",
+       "      <th>supply_0</th>\n",
+       "      <th>...</th>\n",
+       "      <th>attestations_0</th>\n",
+       "      <th>attestations_1</th>\n",
+       "      <th>invariant_V</th>\n",
+       "      <th>invariant_I</th>\n",
+       "      <th>agents</th>\n",
+       "      <th>chosen_agent</th>\n",
+       "      <th>simulation</th>\n",
+       "      <th>run</th>\n",
+       "      <th>substep</th>\n",
+       "      <th>timestep</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>300.000000</td>\n",
+       "      <td>1.000000</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.000000</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0.00</td>\n",
+       "      <td>2.000000</td>\n",
+       "      <td>600.000000</td>\n",
+       "      <td>0.500000</td>\n",
+       "      <td>200.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>20.0</td>\n",
+       "      <td>20.000000</td>\n",
+       "      <td>1200.000000</td>\n",
+       "      <td>650.000000</td>\n",
+       "      <td>agent_attestations_1  agent_attestations_0 ...</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>5</th>\n",
+       "      <td>300.026889</td>\n",
+       "      <td>1.000022</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.084917</td>\n",
+       "      <td>1.001076</td>\n",
+       "      <td>0.74</td>\n",
+       "      <td>2.169738</td>\n",
+       "      <td>600.026888</td>\n",
+       "      <td>0.501361</td>\n",
+       "      <td>200.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>20.0</td>\n",
+       "      <td>20.153041</td>\n",
+       "      <td>3554.198921</td>\n",
+       "      <td>650.979671</td>\n",
+       "      <td>agent_attestations_1  agent_attestations_0 ...</td>\n",
+       "      <td>{'agent_attestations_1': 0.15304119675024097, ...</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>5</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>10</th>\n",
+       "      <td>301.512069</td>\n",
+       "      <td>1.083468</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.086216</td>\n",
+       "      <td>0.993533</td>\n",
+       "      <td>0.71</td>\n",
+       "      <td>2.166573</td>\n",
+       "      <td>601.397654</td>\n",
+       "      <td>0.502480</td>\n",
+       "      <td>200.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>20.0</td>\n",
+       "      <td>20.272443</td>\n",
+       "      <td>3482.997889</td>\n",
+       "      <td>653.248033</td>\n",
+       "      <td>agent_attestations_1  agent_attestations_0 ...</td>\n",
+       "      <td>{'agent_attestations_1': 0.11940162751985284, ...</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>5</td>\n",
+       "      <td>2</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>15</th>\n",
+       "      <td>301.564639</td>\n",
+       "      <td>1.086165</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.087566</td>\n",
+       "      <td>1.082990</td>\n",
+       "      <td>0.73</td>\n",
+       "      <td>2.169062</td>\n",
+       "      <td>601.446053</td>\n",
+       "      <td>0.503640</td>\n",
+       "      <td>200.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>20.0</td>\n",
+       "      <td>20.391865</td>\n",
+       "      <td>3538.916623</td>\n",
+       "      <td>654.112491</td>\n",
+       "      <td>agent_attestations_1  agent_attestations_0 ...</td>\n",
+       "      <td>{'agent_attestations_1': 0.1194224597725313, '...</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>5</td>\n",
+       "      <td>3</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>20</th>\n",
+       "      <td>301.717442</td>\n",
+       "      <td>1.087418</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.089257</td>\n",
+       "      <td>1.078198</td>\n",
+       "      <td>0.71</td>\n",
+       "      <td>2.171841</td>\n",
+       "      <td>601.586572</td>\n",
+       "      <td>0.505093</td>\n",
+       "      <td>200.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>20.0</td>\n",
+       "      <td>20.533524</td>\n",
+       "      <td>3602.415418</td>\n",
+       "      <td>655.282373</td>\n",
+       "      <td>agent_attestations_1  agent_attestations_0 ...</td>\n",
+       "      <td>{'agent_attestations_1': 0.14165863612184193, ...</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>5</td>\n",
+       "      <td>4</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>5 rows × 23 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "       reserve     price  realized_price  spot_price  private_price  \\\n",
+       "0   300.000000  1.000000               0    1.000000       0.000000   \n",
+       "5   300.026889  1.000022               0    1.084917       1.001076   \n",
+       "10  301.512069  1.083468               0    1.086216       0.993533   \n",
+       "15  301.564639  1.086165               0    1.087566       1.082990   \n",
+       "20  301.717442  1.087418               0    1.089257       1.078198   \n",
+       "\n",
+       "    private_alpha     kappa      supply     alpha  supply_0  ...  \\\n",
+       "0            0.00  2.000000  600.000000  0.500000     200.0  ...   \n",
+       "5            0.74  2.169738  600.026888  0.501361     200.0  ...   \n",
+       "10           0.71  2.166573  601.397654  0.502480     200.0  ...   \n",
+       "15           0.73  2.169062  601.446053  0.503640     200.0  ...   \n",
+       "20           0.71  2.171841  601.586572  0.505093     200.0  ...   \n",
+       "\n",
+       "    attestations_0  attestations_1  invariant_V  invariant_I  \\\n",
+       "0             20.0       20.000000  1200.000000   650.000000   \n",
+       "5             20.0       20.153041  3554.198921   650.979671   \n",
+       "10            20.0       20.272443  3482.997889   653.248033   \n",
+       "15            20.0       20.391865  3538.916623   654.112491   \n",
+       "20            20.0       20.533524  3602.415418   655.282373   \n",
+       "\n",
+       "                                               agents  \\\n",
+       "0      agent_attestations_1  agent_attestations_0 ...   \n",
+       "5      agent_attestations_1  agent_attestations_0 ...   \n",
+       "10     agent_attestations_1  agent_attestations_0 ...   \n",
+       "15     agent_attestations_1  agent_attestations_0 ...   \n",
+       "20     agent_attestations_1  agent_attestations_0 ...   \n",
+       "\n",
+       "                                         chosen_agent  simulation run substep  \\\n",
+       "0                                                   0           0   1       0   \n",
+       "5   {'agent_attestations_1': 0.15304119675024097, ...           0   1       5   \n",
+       "10  {'agent_attestations_1': 0.11940162751985284, ...           0   1       5   \n",
+       "15  {'agent_attestations_1': 0.1194224597725313, '...           0   1       5   \n",
+       "20  {'agent_attestations_1': 0.14165863612184193, ...           0   1       5   \n",
+       "\n",
+       "    timestep  \n",
+       "0          0  \n",
+       "5          1  \n",
+       "10         2  \n",
+       "15         3  \n",
+       "20         4  \n",
+       "\n",
+       "[5 rows x 23 columns]"
+      ]
+     },
+     "execution_count": 5,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "experiments.dataset[0].head()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "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>reserve</th>\n",
+       "      <th>price</th>\n",
+       "      <th>realized_price</th>\n",
+       "      <th>spot_price</th>\n",
+       "      <th>private_price</th>\n",
+       "      <th>private_alpha</th>\n",
+       "      <th>kappa</th>\n",
+       "      <th>supply</th>\n",
+       "      <th>alpha</th>\n",
+       "      <th>supply_0</th>\n",
+       "      <th>...</th>\n",
+       "      <th>attestations_0</th>\n",
+       "      <th>attestations_1</th>\n",
+       "      <th>invariant_V</th>\n",
+       "      <th>invariant_I</th>\n",
+       "      <th>agents</th>\n",
+       "      <th>chosen_agent</th>\n",
+       "      <th>simulation</th>\n",
+       "      <th>run</th>\n",
+       "      <th>substep</th>\n",
+       "      <th>timestep</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>1482</th>\n",
+       "      <td>306.664756</td>\n",
+       "      <td>1.119910</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.119913</td>\n",
+       "      <td>1.118950</td>\n",
+       "      <td>0.68</td>\n",
+       "      <td>2.213216</td>\n",
+       "      <td>606.042960</td>\n",
+       "      <td>0.531501</td>\n",
+       "      <td>200.095493</td>\n",
+       "      <td>...</td>\n",
+       "      <td>20.00158</td>\n",
+       "      <td>22.475030</td>\n",
+       "      <td>4694.815045</td>\n",
+       "      <td>678.715487</td>\n",
+       "      <td>agent_attestations_1  agent_attestations_0 ...</td>\n",
+       "      <td>{'agent_attestations_1': 0.5108411557663454, '...</td>\n",
+       "      <td>0</td>\n",
+       "      <td>3</td>\n",
+       "      <td>5</td>\n",
+       "      <td>96</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1487</th>\n",
+       "      <td>306.664937</td>\n",
+       "      <td>1.119913</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.119918</td>\n",
+       "      <td>1.117907</td>\n",
+       "      <td>0.64</td>\n",
+       "      <td>2.213226</td>\n",
+       "      <td>606.043121</td>\n",
+       "      <td>0.531506</td>\n",
+       "      <td>200.095493</td>\n",
+       "      <td>...</td>\n",
+       "      <td>20.00158</td>\n",
+       "      <td>22.475305</td>\n",
+       "      <td>4695.109654</td>\n",
+       "      <td>678.718890</td>\n",
+       "      <td>agent_attestations_1  agent_attestations_0 ...</td>\n",
+       "      <td>{'agent_attestations_1': 0.4979930202683946, '...</td>\n",
+       "      <td>0</td>\n",
+       "      <td>3</td>\n",
+       "      <td>5</td>\n",
+       "      <td>97</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1492</th>\n",
+       "      <td>306.755191</td>\n",
+       "      <td>1.120009</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.119921</td>\n",
+       "      <td>1.123601</td>\n",
+       "      <td>0.58</td>\n",
+       "      <td>2.212874</td>\n",
+       "      <td>606.123705</td>\n",
+       "      <td>0.531508</td>\n",
+       "      <td>200.095493</td>\n",
+       "      <td>...</td>\n",
+       "      <td>20.00158</td>\n",
+       "      <td>22.475430</td>\n",
+       "      <td>4684.528890</td>\n",
+       "      <td>678.810626</td>\n",
+       "      <td>agent_attestations_1  agent_attestations_0 ...</td>\n",
+       "      <td>{'agent_attestations_1': 0.5109668943877075, '...</td>\n",
+       "      <td>0</td>\n",
+       "      <td>3</td>\n",
+       "      <td>5</td>\n",
+       "      <td>98</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1497</th>\n",
+       "      <td>306.855175</td>\n",
+       "      <td>1.120021</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.119931</td>\n",
+       "      <td>1.123941</td>\n",
+       "      <td>0.64</td>\n",
+       "      <td>2.212498</td>\n",
+       "      <td>606.212974</td>\n",
+       "      <td>0.531516</td>\n",
+       "      <td>200.095493</td>\n",
+       "      <td>...</td>\n",
+       "      <td>20.00158</td>\n",
+       "      <td>22.475941</td>\n",
+       "      <td>4673.267443</td>\n",
+       "      <td>678.916608</td>\n",
+       "      <td>agent_attestations_1  agent_attestations_0 ...</td>\n",
+       "      <td>{'agent_attestations_1': 0.4985032783348726, '...</td>\n",
+       "      <td>0</td>\n",
+       "      <td>3</td>\n",
+       "      <td>5</td>\n",
+       "      <td>99</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1502</th>\n",
+       "      <td>306.857628</td>\n",
+       "      <td>1.119928</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.119988</td>\n",
+       "      <td>1.109632</td>\n",
+       "      <td>0.88</td>\n",
+       "      <td>2.212603</td>\n",
+       "      <td>606.215165</td>\n",
+       "      <td>0.531566</td>\n",
+       "      <td>200.095493</td>\n",
+       "      <td>...</td>\n",
+       "      <td>20.00158</td>\n",
+       "      <td>22.478915</td>\n",
+       "      <td>4676.387145</td>\n",
+       "      <td>678.953996</td>\n",
+       "      <td>agent_attestations_1  agent_attestations_0 ...</td>\n",
+       "      <td>{'agent_attestations_1': 0.4938448398075422, '...</td>\n",
+       "      <td>0</td>\n",
+       "      <td>3</td>\n",
+       "      <td>5</td>\n",
+       "      <td>100</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>5 rows × 23 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "         reserve     price  realized_price  spot_price  private_price  \\\n",
+       "1482  306.664756  1.119910               0    1.119913       1.118950   \n",
+       "1487  306.664937  1.119913               0    1.119918       1.117907   \n",
+       "1492  306.755191  1.120009               0    1.119921       1.123601   \n",
+       "1497  306.855175  1.120021               0    1.119931       1.123941   \n",
+       "1502  306.857628  1.119928               0    1.119988       1.109632   \n",
+       "\n",
+       "      private_alpha     kappa      supply     alpha    supply_0  ...  \\\n",
+       "1482           0.68  2.213216  606.042960  0.531501  200.095493  ...   \n",
+       "1487           0.64  2.213226  606.043121  0.531506  200.095493  ...   \n",
+       "1492           0.58  2.212874  606.123705  0.531508  200.095493  ...   \n",
+       "1497           0.64  2.212498  606.212974  0.531516  200.095493  ...   \n",
+       "1502           0.88  2.212603  606.215165  0.531566  200.095493  ...   \n",
+       "\n",
+       "      attestations_0  attestations_1  invariant_V  invariant_I  \\\n",
+       "1482        20.00158       22.475030  4694.815045   678.715487   \n",
+       "1487        20.00158       22.475305  4695.109654   678.718890   \n",
+       "1492        20.00158       22.475430  4684.528890   678.810626   \n",
+       "1497        20.00158       22.475941  4673.267443   678.916608   \n",
+       "1502        20.00158       22.478915  4676.387145   678.953996   \n",
+       "\n",
+       "                                                 agents  \\\n",
+       "1482     agent_attestations_1  agent_attestations_0 ...   \n",
+       "1487     agent_attestations_1  agent_attestations_0 ...   \n",
+       "1492     agent_attestations_1  agent_attestations_0 ...   \n",
+       "1497     agent_attestations_1  agent_attestations_0 ...   \n",
+       "1502     agent_attestations_1  agent_attestations_0 ...   \n",
+       "\n",
+       "                                           chosen_agent  simulation run  \\\n",
+       "1482  {'agent_attestations_1': 0.5108411557663454, '...           0   3   \n",
+       "1487  {'agent_attestations_1': 0.4979930202683946, '...           0   3   \n",
+       "1492  {'agent_attestations_1': 0.5109668943877075, '...           0   3   \n",
+       "1497  {'agent_attestations_1': 0.4985032783348726, '...           0   3   \n",
+       "1502  {'agent_attestations_1': 0.4938448398075422, '...           0   3   \n",
+       "\n",
+       "     substep  timestep  \n",
+       "1482       5        96  \n",
+       "1487       5        97  \n",
+       "1492       5        98  \n",
+       "1497       5        99  \n",
+       "1502       5       100  \n",
+       "\n",
+       "[5 rows x 23 columns]"
+      ]
+     },
+     "execution_count": 6,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "experiments.dataset[0].tail()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "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>reserve</th>\n",
+       "      <th>price</th>\n",
+       "      <th>realized_price</th>\n",
+       "      <th>spot_price</th>\n",
+       "      <th>private_price</th>\n",
+       "      <th>private_alpha</th>\n",
+       "      <th>kappa</th>\n",
+       "      <th>supply</th>\n",
+       "      <th>alpha</th>\n",
+       "      <th>supply_0</th>\n",
+       "      <th>...</th>\n",
+       "      <th>attestations_0</th>\n",
+       "      <th>attestations_1</th>\n",
+       "      <th>invariant_V</th>\n",
+       "      <th>invariant_I</th>\n",
+       "      <th>agents</th>\n",
+       "      <th>chosen_agent</th>\n",
+       "      <th>simulation</th>\n",
+       "      <th>run</th>\n",
+       "      <th>substep</th>\n",
+       "      <th>timestep</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>1503</th>\n",
+       "      <td>300.000000</td>\n",
+       "      <td>1.000000</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.000000</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0.00</td>\n",
+       "      <td>2.000000</td>\n",
+       "      <td>600.000000</td>\n",
+       "      <td>0.500000</td>\n",
+       "      <td>200.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>20.0</td>\n",
+       "      <td>20.000000</td>\n",
+       "      <td>1200.000000</td>\n",
+       "      <td>650.000000</td>\n",
+       "      <td>agent_attestations_1  agent_attestations_0 ...</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1508</th>\n",
+       "      <td>300.000000</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.086274</td>\n",
+       "      <td>1.000000</td>\n",
+       "      <td>0.98</td>\n",
+       "      <td>2.172549</td>\n",
+       "      <td>600.000000</td>\n",
+       "      <td>0.502521</td>\n",
+       "      <td>200.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>20.0</td>\n",
+       "      <td>20.256580</td>\n",
+       "      <td>3618.655872</td>\n",
+       "      <td>651.764572</td>\n",
+       "      <td>agent_attestations_1  agent_attestations_0 ...</td>\n",
+       "      <td>{'agent_attestations_1': 0.2565801213118535, '...</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>5</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1513</th>\n",
+       "      <td>317.418514</td>\n",
+       "      <td>1.069006</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.087214</td>\n",
+       "      <td>0.000000</td>\n",
+       "      <td>0.96</td>\n",
+       "      <td>2.110915</td>\n",
+       "      <td>616.294114</td>\n",
+       "      <td>0.503750</td>\n",
+       "      <td>200.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>20.0</td>\n",
+       "      <td>20.350613</td>\n",
+       "      <td>2439.905200</td>\n",
+       "      <td>670.043483</td>\n",
+       "      <td>agent_attestations_1  agent_attestations_0 ...</td>\n",
+       "      <td>{'agent_attestations_1': 0.09403285712342058, ...</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>5</td>\n",
+       "      <td>2</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1518</th>\n",
+       "      <td>317.496671</td>\n",
+       "      <td>1.087143</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.088164</td>\n",
+       "      <td>1.069006</td>\n",
+       "      <td>0.92</td>\n",
+       "      <td>2.112486</td>\n",
+       "      <td>616.366007</td>\n",
+       "      <td>0.504587</td>\n",
+       "      <td>200.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>20.0</td>\n",
+       "      <td>20.412387</td>\n",
+       "      <td>2464.651535</td>\n",
+       "      <td>670.707222</td>\n",
+       "      <td>agent_attestations_1  agent_attestations_0 ...</td>\n",
+       "      <td>{'agent_attestations_1': 0.15580663229094682, ...</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>5</td>\n",
+       "      <td>3</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1523</th>\n",
+       "      <td>317.513328</td>\n",
+       "      <td>1.088149</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.090761</td>\n",
+       "      <td>1.087143</td>\n",
+       "      <td>0.85</td>\n",
+       "      <td>2.117470</td>\n",
+       "      <td>616.381314</td>\n",
+       "      <td>0.506874</td>\n",
+       "      <td>200.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>20.0</td>\n",
+       "      <td>20.575052</td>\n",
+       "      <td>2544.833231</td>\n",
+       "      <td>672.324803</td>\n",
+       "      <td>agent_attestations_1  agent_attestations_0 ...</td>\n",
+       "      <td>{'agent_attestations_1': 0.1626651559147385, '...</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>5</td>\n",
+       "      <td>4</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>5 rows × 23 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "         reserve     price  realized_price  spot_price  private_price  \\\n",
+       "1503  300.000000  1.000000               0    1.000000       0.000000   \n",
+       "1508  300.000000  0.000000               0    1.086274       1.000000   \n",
+       "1513  317.418514  1.069006               0    1.087214       0.000000   \n",
+       "1518  317.496671  1.087143               0    1.088164       1.069006   \n",
+       "1523  317.513328  1.088149               0    1.090761       1.087143   \n",
+       "\n",
+       "      private_alpha     kappa      supply     alpha  supply_0  ...  \\\n",
+       "1503           0.00  2.000000  600.000000  0.500000     200.0  ...   \n",
+       "1508           0.98  2.172549  600.000000  0.502521     200.0  ...   \n",
+       "1513           0.96  2.110915  616.294114  0.503750     200.0  ...   \n",
+       "1518           0.92  2.112486  616.366007  0.504587     200.0  ...   \n",
+       "1523           0.85  2.117470  616.381314  0.506874     200.0  ...   \n",
+       "\n",
+       "      attestations_0  attestations_1  invariant_V  invariant_I  \\\n",
+       "1503            20.0       20.000000  1200.000000   650.000000   \n",
+       "1508            20.0       20.256580  3618.655872   651.764572   \n",
+       "1513            20.0       20.350613  2439.905200   670.043483   \n",
+       "1518            20.0       20.412387  2464.651535   670.707222   \n",
+       "1523            20.0       20.575052  2544.833231   672.324803   \n",
+       "\n",
+       "                                                 agents  \\\n",
+       "1503     agent_attestations_1  agent_attestations_0 ...   \n",
+       "1508     agent_attestations_1  agent_attestations_0 ...   \n",
+       "1513     agent_attestations_1  agent_attestations_0 ...   \n",
+       "1518     agent_attestations_1  agent_attestations_0 ...   \n",
+       "1523     agent_attestations_1  agent_attestations_0 ...   \n",
+       "\n",
+       "                                           chosen_agent  simulation run  \\\n",
+       "1503                                                  0           1   1   \n",
+       "1508  {'agent_attestations_1': 0.2565801213118535, '...           1   1   \n",
+       "1513  {'agent_attestations_1': 0.09403285712342058, ...           1   1   \n",
+       "1518  {'agent_attestations_1': 0.15580663229094682, ...           1   1   \n",
+       "1523  {'agent_attestations_1': 0.1626651559147385, '...           1   1   \n",
+       "\n",
+       "     substep  timestep  \n",
+       "1503       0         0  \n",
+       "1508       5         1  \n",
+       "1513       5         2  \n",
+       "1518       5         3  \n",
+       "1523       5         4  \n",
+       "\n",
+       "[5 rows x 23 columns]"
+      ]
+     },
+     "execution_count": 7,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "experiments.dataset[1].head()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "303"
+      ]
+     },
+     "execution_count": 8,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "len(experiments.dataset[0])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1080x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1080x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "experiments = experiments.sort_values(by =['rules_price']).reset_index(drop=True)\n",
+    "# df = experiments\n",
+    "cols = 1\n",
+    "rows = 1\n",
+    "cc_idx = 0\n",
+    "\n",
+    "# len(experiments)\n",
+    "# df = experiments.copy()\n",
+    "# print(df.head())\n",
+    "\n",
+    "while cc_idx<len(experiments):\n",
+    "    cc = experiments.iloc[cc_idx]['rules_price']\n",
+    "    cc_label = experiments.iloc[cc_idx]['rules_price']\n",
+    "#     secondary_label = experiments.iloc[cc_idx]['string']\n",
+    "    sub_experiments = experiments[experiments['rules_price']==cc]\n",
+    "    cc_idx += len(sub_experiments)\n",
+    "    fig, axs = plt.subplots(ncols=cols, nrows=rows, figsize=(15*cols,7*rows))\n",
+    "    sub_experiments.reset_index(drop=False)\n",
+    "    for i, experiment in sub_experiments.iterrows():\n",
+    "        df = experiment['dataset'].copy()\n",
+    "        # filter out starting point\n",
+    "        df = df[df.timestep > 0]       \n",
+    "#         print(df['price'])\n",
+    "\n",
+    "        df = df.groupby('timestep').agg({'price': ['min', 'mean', 'max']}).reset_index()\n",
+    "\n",
+    "#         print(df.head())\n",
+    "        plot_label = experiment['rules_price']\n",
+    "        ax = axs\n",
+    "        title = 'Price' + '\\n' + 'Scenario: ' + str(cc_label)  + ' rules_price'\n",
+    "        ax.set_title(title)\n",
+    "        ax.set_ylabel('Price')\n",
+    "        colors = ['b','orange', 'g', 'magenta', 'r']\n",
+    "        \n",
+    "#         ax.plot(df.timestep, df['price'],color = colors[0], label='Price')\n",
+    "        df.plot(x='timestep', y=('price','mean'), label='Mean', ax=ax, legend=True)\n",
+    "        ax.fill_between(df.timestep, df[('price','min')], df[('price','max')], alpha=0.3) \n",
+    "\n",
+    "\n",
+    "        ax.legend()\n",
+    "\n",
+    "        ax.set_xlabel('Timesteps')\n",
+    "        ax.grid(color='0.9', linestyle='-', linewidth=1)\n",
+    "\n",
+    "        plt.tight_layout()\n",
+    "        \n",
+    "fig.tight_layout(rect=[0, 0, 1, .97])\n",
+    "fig.patch.set_alpha(1)\n",
+    "display(fig)\n",
+    "plt.close()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### FIRST RUN ONLY!"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1080x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1080x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "experiments = experiments.sort_values(by =['rules_price']).reset_index(drop=True)\n",
+    "# df = experiments\n",
+    "cols = 1\n",
+    "rows = 1\n",
+    "cc_idx = 0\n",
+    "\n",
+    "# len(experiments)\n",
+    "# df = experiments.copy()\n",
+    "# print(df.head())\n",
+    "\n",
+    "while cc_idx<len(experiments):\n",
+    "    cc = experiments.iloc[cc_idx]['rules_price']\n",
+    "    cc_label = experiments.iloc[cc_idx]['rules_price']\n",
+    "#     secondary_label = experiments.iloc[cc_idx]['string']\n",
+    "    sub_experiments = experiments[experiments['rules_price']==cc]\n",
+    "    cc_idx += len(sub_experiments)\n",
+    "    fig, axs = plt.subplots(ncols=cols, nrows=rows, figsize=(15*cols,7*rows))\n",
+    "    sub_experiments.reset_index(drop=False)\n",
+    "    for i, experiment in sub_experiments.iterrows():\n",
+    "        df = experiment['dataset'].copy()\n",
+    "        # filter out starting point\n",
+    "        df = df[df.timestep > 0] \n",
+    "        # FIRST RUN ONLY\n",
+    "        df = df[df.run == 1]       \n",
+    "#         print(df['price'])\n",
+    "\n",
+    "        df = df.groupby('timestep').agg({'price': ['min', 'mean', 'max']}).reset_index()\n",
+    "\n",
+    "#         print(df.head())\n",
+    "        plot_label = experiment['rules_price']\n",
+    "        ax = axs\n",
+    "        title = 'Price' + '\\n' + 'Scenario: ' + str(cc_label)  + ' rules_price'\n",
+    "        ax.set_title(title)\n",
+    "        ax.set_ylabel('Price')\n",
+    "        colors = ['b','orange', 'g', 'magenta', 'r']\n",
+    "        \n",
+    "#         ax.plot(df.timestep, df['price'],color = colors[0], label='Price')\n",
+    "        df.plot(x='timestep', y=('price','mean'), label='Mean', ax=ax, legend=True)\n",
+    "        ax.fill_between(df.timestep, df[('price','min')], df[('price','max')], alpha=0.3) \n",
+    "\n",
+    "\n",
+    "        ax.legend()\n",
+    "\n",
+    "        ax.set_xlabel('Timesteps')\n",
+    "        ax.grid(color='0.9', linestyle='-', linewidth=1)\n",
+    "\n",
+    "        plt.tight_layout()\n",
+    "        \n",
+    "fig.tight_layout(rect=[0, 0, 1, .97])\n",
+    "fig.patch.set_alpha(1)\n",
+    "display(fig)\n",
+    "plt.close()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1080x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAABDAAAAHwCAYAAABQRJ8FAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nOzdeXwV9b3/8fecPcvJCglh3wQEW0FUREEFF1pt3LBKtbSlcq3de21VVCrYAmr1ar3WlVv1irbAT6mCemvFDUWLIG4ou2yBEBLIdpKcbWZ+f5zkkJgAERPOSfJ6Ph7nMcuZzPnMyTyUeecz3zFs27YFAAAAAACQxByJLgAAAAAAAOBICDAAAAAAAEDSI8AAAAAAAABJjwADAAAAAAAkPQIMAAAAAACQ9AgwAAAAAABA0iPAAAAACfeXv/xFy5cvb7P9vfbaa5ozZ06b7Q8AACSeK9EFAAAArFq1SoMHD26z/Z1zzjk655xz2mx/AAAg8QgwAADAUampqdHNN9+sHTt2yOFwaMSIEbrwwgt17733qmfPnvriiy/k8/l05513atCgQaqurtbtt9+uDRs2yDAMjR8/Xtdff70WLVqkdevW6U9/+pOcTqfOO++8Q37m8OHD9R//8R96++23VVtbq+uvv17nn3++lixZomeffVZ1dXVKT0/XpZdeqldeeUWPPvqoSktLNWvWLH3xxRdyOByaMmWKfvCDH6i6ulpz587Vpk2bFIlENHbsWN14441yufjnEQAAyYhbSAAAwFF59dVXVVNToxdeeEHPPvusJKmoqEjr1q3T1KlTtWzZMl122WW64YYbJElz5sxRVlaWli1bpueee04bN27U448/rquvvlonnHCCbrzxxsOGF5JkmqZSUlK0ZMkS/fnPf9Ytt9yiAwcOSJK2bNmiBQsWaMGCBU1+5vbbb1f//v31z3/+U4sWLdLixYu1Y8cOzZs3TyNGjNCSJUv0/PPPq7y8XE888UQ7fFMAAKAtEGAAAICjMnr0aG3ZskVTp07VY489ph/+8Ifq27evhg0bppNPPlmSNHnyZK1fv17l5eVasWKFvv/978swDHk8Hk2ZMkUrVqz4yp/7/e9/X5I0bNgwDRkyRKtXr5YkDR06VOnp6c22f/fdd3XllVdKkvx+v1588UX169dPb775phYtWqSLL75Yl112mT755BNt2rTpaL8OAADQzuiRBAAAR6VPnz569dVXtWrVKv373//WtGnT9Ic//EFOp7PZtk6nU5ZlyTCM+DrLshSNRr/y5zbev2VZ8eXU1NQWt3e5XE0+d9euXcrOzpZlWbr//vs1aNAgSVJVVVWT7QAAQHKhAwMAAByVv/3tb7r55ps1btw43XDDDRo3bpw+//xzbdiwQRs2bJAkLVq0SKNGjVJGRobGjRunp59+WrZtKxwOa/HixTr99NMlxUKJ1oYZzz//vCTps88+07Zt23TKKaccdvuxY8fqueeekyRVV1frhz/8obZv365x48bpySefjNfz05/+VE8//fTRfh0AAKCd0YEBAACOyiWXXKL3339fF1xwgVJSUlRQUKChQ4eqW7du+vOf/6zdu3crJydHf/rTnyRJM2fO1Jw5c1RYWKhIJKLx48fruuuukyRNnDhR9957ryKRiC699NLDfu7atWu1ePFiWZal++67T5mZmYfd/rbbbtPs2bNVWFgo27b1k5/8RCeccIJuvfVWzZ07N17P6aefrunTp7fNlwMAANqcYdu2negiAABA57Bq1Sr98Y9/1Isvvtgu+x86dKjee+895eTktMv+AQBA8qIDAwAAJI3/+Z//0bJly1p875prrjnG1QAAgGRCBwYAAAAAAEh6DOIJAAAAAACSHgEGAAAJ9tFHH2nq1KkqLCzUd77zHU2fPl2bN29OaE2ffvqpfvWrXyW0hi/75JNPdNttt0lq//omTpyoTz/9tN32fyTJ+P0DAJBo3EICAEAChcNhjR8/Xo8//rhGjBghSXrhhRd033336bXXXpPT6UxwhcljyZIleuWVV/Too4+2+2dNnDhR999/v77xjW+0+2cBAIDWYRBPAAASqK6uTtXV1aqtrY2vu+iii5Seni7TNOV0OvXss8/qiSeekMPhUHZ2tu666y4VFBTo9ddf18MPP6xIJCKfz6ebbrpJo0aN0gMPPKDdu3ertLRUu3fvVn5+vu6++27l5eXpjTfe0KOPPqpwOKwDBw7okksu0W9+8xutWrVKc+fOVWpqqmpqanTjjTfqrrvu0osvvqjq6mrdfvvt2rBhgwzD0Pjx43X99dfL5XLp/vvvlyT9+te/bnJcq1at0r333quCggJt27ZNKSkpuvbaa7VgwQJt27ZN559/vm655RZZlqV58+bp448/Vk1NjWzb1pw5czR69GjNmDFDFRUV2rVrl0488US9++67qq6u1s0336xLLrkk/rSTGTNmKD09XRs3btTevXs1dOhQ3XXXXUpLS2tSU+P9nX322dq/f7+OO+64+OCgM2bMaLLc4FDf89atW3XrrbcqHA7Ltm1dfvnluvrqqw/7+544caIuvPBCrVy5UtXV1Zo2bZquuuqqw37/NTU1mjNnjtauXSun06lzzz1X//mf/6lIJKJ77rlHq1evlmmaGj58uGbOnKn09PSjPh8BAEhmBBgAACRQZmambrjhBk2fPl3dunXTSSedpDFjxujCCy+Ux+PRhg0bdM899+gf//iHCgoK9OSTT+rhhx/Wj3/8Y91333166qmnlJ2drc2bN2vatGn617/+JUlas2aNnn/+eaWnp+u6667TwoUL9ctf/lKPP/647rzzTvXv318lJSWaMGGCfvCDH0iSNm/erOXLl6tXr15atWpVvMY5c+YoKytLy5YtUyQS0U9/+lM9/vjjuvbaa5sFF419+umnmjVrloYPH67p06frscce01NPPaVAIKAzzzxT11xzjfbs2aN9+/Zp0aJFcjgceuyxxzR//nyNHj1akhQMBvXSSy9JOtiBcccddzSpT5LWrVunp556SoZh6IorrtA///lPTZ48uVlNjfc3Y8aMI/5+tm/ffsjv+a9//asmTpyoa6+9VqWlpZo3b56+973vyeE4/B26lZWVeu6551RSUqJLLrkkfqyH+v7/+7//W6FQSC+//LJM09SPf/xjvf/++1q9erWcTqeWLFkiwzB077336p577tHs2bOPeFwAAHREBBgAACTYtGnT9N3vflerV6/W6tWrNX/+fM2fP1/PPvus3nvvPY0bN04FBQWSpB/96EeSpGeeeUb79u2LL0uSYRjauXOnJOnUU0+N/yV++PDhqqyslGEYeuSRR/Tmm2/qxRdf1NatW2Xbturq6iRJBQUF6tWrV7P6VqxYob///e8yDEMej0dTpkzR//7v/+raa6897HH17t1bw4cPlyT17dtXfr9fHo9HOTk5SktLU2VlpUaNGqXMzEwtXLhQu3bt0qpVq5p0TjRc3B/J+PHj5fF4JElDhgxRZWVli9u1dn8NVq5cecjv+bzzztNNN92kTz75RGPHjtXMmTOPGF5I0lVXXSXDMNSjRw+NHz9eK1eu1IgRIw75/b/77ru6+eab5XQ65XQ69fTTT0uS7r77blVXV+vdd9+VJEUiEeXm5n6l4wMAoCMhwAAAIIE++OADffjhh5o+fbomTJigCRMm6Prrr9d3vvMdrVy5Uk6nU4ZhxLcPBoPavXu3LMvS2LFj9ec//zn+XnFxsfLy8vTqq6/K5/PF1xuGIdu2VVtbq0svvVTnnnuuTj75ZE2ePFnLly9Xw3BYqampLdZoWVaTGizLUjQaPeKxNQQKDVyu5v/sePPNNzV37lxNmzZN55xzjgYOHKilS5fG3z9UTV/W0vG2pPH+vrxdJBJptv3hvudhw4bplVde0bvvvqv33ntPDz74oJYsWaIePXocttbG34NlWfHQ41DH6nK5mnz/xcXF8vl8sixLt9xyi8466yxJUk1NjUKh0GE/GwCAjoynkAAAkEA5OTl6+OGHtWbNmvi60tJSBQIBDRkyRGPGjNF7772nffv2SZIWLlyou+++W2PHjtXKlSu1detWSdJbb72liy66SMFg8JCftWPHDgUCAf3mN7/RxIkTtWrVKoXDYVmWddgax40bp6efflq2bSscDmvx4sU6/fTT2+DoYx0OEyZM0FVXXaUTTjhBy5cvl2maLW7rdDpbFZy0VnZ2ttatWydJKikp0fvvv99sm8N9z7/97W/18ssv68ILL9SsWbOUnp4e74A5nOeff16StGfPHq1cuVJnnnnmYbcfO3as/vGPf8iyLIXDYf3qV7/S6tWrNW7cOD3zzDPx3+Hvf/973XvvvV/1awAAoMOgAwMAgAQaMGCAHnzwQd13333au3evvF6v/H6/5s2bp4EDB0pSfIwMSerevbvmzZun/Px8/eEPf9D1118v27blcrn08MMPNxu4srGhQ4fq7LPP1re//W15PB4NGTJEgwcP1o4dO5p1SzQ2c+ZMzZkzR4WFhYpEIho/fryuu+46STrkIJ6tNWXKFP32t79VYWGhotGozjjjDP3rX/9qMVQZOXKkHnzwQf3iF7/Q1KlTj+rzGps6dap+97vfadKkSerdu7dOO+20ZtsMHjz4kN/zz372M916661atGhRfHDNU0455YifW1RUpMsuu0zBYFAzZ87UwIEDVVpaesjtf/GLX2ju3Lm6+OKLZZqmLrjgAp1//vk688wzddddd+nSSy+VaZo6/vjjWzWuBwAAHRWPUQUAADhGeDwrAABHjw4MAACANrJ06VL99a9/bfG9wsLCY1wNAACdCx0YAAAAAAAg6TGIJwAAAAAASHoEGAAAAAAAIOl16DEwPvroI3m93kSX8ZVFo9Emz4AHOhvOcXQFnOfo7DjH0dlxjqOz68jneCgU0siRI5ut75hHU8/r9er4449PdBlfWXFxsQoKChJdBtBuOMfRFXCeo7PjHEdnxzmOzq4jn+Pr169vcT23kAAAAAAAgKRHgAEAAAAAAJIeAQYAAAAAAEh6HXoMjJZEIhEVFRUpGAwmupRDMk1TFRUVbbIvn8+n3r17y+12t8n+AAAAAABIRp0uwCgqKpLf71f//v1lGEaiy2lROByWx+P52vuxbVv79+9XUVGRBgwY0AaVAQAAAACQnDrdLSTBYFC5ublJG160JcMwlJubm9TdJgAAAAAAtIVOF2BI6hLhRYOudKwAAAAAgK6rUwYYAAAAAACgcyHAaAerVq3S0KFD9fLLLzdZX1hYqBkzZiSoKgAAAAAAOi4CjHYycOBAvfjii/HljRs3qq6uLoEVAQAAAADQcXW6p5A09twHRVq8Zleb7vOKk/to8ujeR9xu2LBh2r59u6qqqpSRkaGlS5eqsLBQxcXFeuWVV/T000/L4XBo9OjR+t3vfqe9e/dq9uzZCoVCqqio0M9//nOde+65Kiws1KmnnqqNGzfKMAw99NBD8vv9bXpMAAAAAAAkOzow2tF5552nV199VbZt65NPPtGoUaNUUVGhhx56SE8++aT+/ve/q6SkRCtXrtQXX3yhadOm6YknntDvf/97PfPMM5KkmpoaXXjhhXr66aeVl5enFStWJPioAAAAAAA49jp1B8bk0b1b1S3RXgoLCzV79mz16dNHJ598siTJNE2Vl5fr2muvlRQLKHbt2qXRo0fr4Ycf1rPPPivDMBSNRuP7GT58uCSpoKBAoVDo2B8IAAAAAAAJ1m4dGB9//LGmTp3a4nt1dXWaMmWKtm7dKkmKRCK64YYbdNVVV+nyyy/Xa6+91l5lHVN9+vRRbW2tFixYoIsuukhS7LGnPXr00OOPP64FCxbo+9//vk488UTdf//9uvjii3X33XdrzJgxsm07vh8elQoAAAAA6OrapQNj/vz5Wrp0qVJSUpq99+mnn2rWrFkqKSmJr1u6dKmysrJ09913q7y8XJdeeqnOOeec9ijtmLvgggv0wgsvaMCAAdq1a5dycnI0adIkTZ06VaZpqlevXvr2t7+tb33rW5o7d64effRRFRQUqLy8PNGlAwAAAACQNAy78Z/628grr7yioUOH6sYbb9TixYubvPfBBx+oZ8+euvHGGzV79mwNGjRINTU1sm1b6enpKi8vb3UXxvr163X88ccfcV2yCYfD8ng8bba/jnDM6FqKi4tVUFCQ6DKAdsV5js6OcxydHec4OruOfI4f6hq3XTowJk2apKKiohbfGz16dLN1aWlpkqRAIKBf/epX+s1vftOqz4lGoyouLm6yzjRNhcPhr1jxsdeWNZqm2ex7ABKNcxJdAec5OjvOcXR2nOPoCGzbVsi0VRc2VRexVBuxFDUP9iHYajTfqD0hz++ROtk5njSDeBYXF+vnP/+5rrrqKhUWFrbqZ1wuV7NEqaKiok27G9pDW3dgOJ3ODpusoXPqyGkv0Fqc5+jsOMfR2XGOo63Ztq3qUFT7A2HtD4RUFgipLBCOLdeEdKAmrKhpy7RtWVZsalq2rIapJZm2rWDEVG3YVE0oqtqwqdpwVNZR3DeRnerSh7dNavsDPQYqKipaXJ8UAUZZWZl+/OMf67bbbtPYsWO/9v5s2+4yA1+2wx1AAAAAANCl2bat0uqQNpZUa29lUIFQVIFgVIFQVFX100AwokAoqupgVJV1Ee2vCSsctVrcX2aKWzlpHrmdhhyGIacj9orPG4YcDsntcCg71a1Uj0upHqdSPS6leZ2Nlp1K87rkdjrU+Iq38eVvw3yGM9J+X1CCHJMAY9myZaqtrdWVV17Z4vuPPPKIqqqq9NBDD+mhhx6SFBsI1OfzfeXP8vl82r9/v3Jzczt9iGHbtvbv339U3xMAAAAAQKqsjWhjSbU2llRr0976aUm1KmqbBwBel0N+n1t+n0vpXpf8Ppf65qTK73Orm9+jbmle5aZ71C394DQ71SOPq90eAHpInfEWqXYZxPNYaWlgj0gkoqKiIgWDwQRVdWSmacrpdLbJvnw+n3r37i23290m+wPaAi2Z6Ao4z9HZcY6js+Mc7zhs21Z5bUTbygLaVlarkqqgwlFLYdNSOGopUj9tWA5HLUXrb82wbcmqv+S17dh4EQ3rLEvacaBGJVWh+Gf5vS4N6eHXkHy/huana0gPv3pnpcrvcynN60pIEHG0OvI5fkwH8Uwkt9utAQMGJLqMw+rIJxIAAAAANGbbtqyGUKA+NGgICWyp0br6af3PxKb1g1DWry+tDmlbWY22ldVoe1mNvqifr6xr3g3hdhryOB1yuxzyOB3yNJq66m/VMAxDhmK3VcSmseXYe9IZg7tpaL5fQ3r4NTTfr4JMX6fv5O/IOl2AAQAAAABoH7Zt64uyGq3dUa61Oyv04c5ybSqpPqpBJo+kZ6ZPA7qnqfDEAvXPTdPA7mka0C1dBZk+eZwOORwEDV0NAQYAAAAAfA2WZStsWgo1vp2hfr5hndWKO/ejpq2IaStsmgpHYz8b24+tcNRU2IzNW9bBzgarUWdDfLn+NgmpaddDbPngYzd9bqeyUtzKbHilHpzPSvUozeNUbdjUx7sqtHZnuT7YUa4Pd1XEx4bw+1wa1TdbE4flyed2ymHUdzgYkiFDDuNgp4N0cD7eCVE/r0ZdEjmpHg3onqZ+OWlK8bTNbffoPAgwAAAAAHQJEdNSdTCqqrqIdpXWar9VqagVe4Rlk5dty7Ri4UF5bUQVtRFV1IZVXhuuXw6rojai8tqIKuvCipiJHVbQqA8KHPXBgQwdnFfT2yckxZ9eURcxFT1M64TLYcRDEUkanJeu84fna3S/bJ3UN1uDuqfTBYFjigADAAAAQNKybVtbSwN6f1u5Pt1dqYhpxS/AG/7S3/AX/tjUUF049qjLqrqIqoIRVdVFVRWMqDZsHnUdHlfs8ZbZqR5lpbo1qHu6stPcykzxyOduOv5Cw9TdaNnZigt9p8M4+HP1P+t1Nd1Xw2M444HF1xivwbZt1YZNVdbFQprKuoZXOD7vcjg0qm+WRvXJVmYqDw5AYhFgAAAAAEga4aildXsqtXrbAa3eXq4PdhxQef0tC1mpbqW6Y7cVNBkAMj4fm6Z6nMpIcSnD59bAbunx+YwUtzJ8LmWkuGWG6pSZ4Y8P9uhyxEKGxi+P06GsVLeyUt1KcTs73eCOhmEozRt7ukbPrJRElwMcEQEGAAAAgHZh27bW7qzQ5vpBHm3V347QeOwGSZYtHagJac32cn20q0KhqCVJGtAtTecNz9fJ/XN0Sv8c9c9NbbMQIfZkwB5tsi8AxwYBBgAAAIA2FTEtvfxpsR5/Z5s+Lqps1c84HYZG9MzQ1WP66dQB2RrdL0fd/d52rhRAR0KAAQAAAKBNVNZFtPD9nXry3e0qrgxqQLc0/fGSEzRxWJ6c9WM2qNETKoxGA0963Q753Dx1AsChEWAAAAAA+Fp27K/REyu3a/GaXaoNmxo7MFdzLjlBE4bm8ZQKAG2GAAMAAADAV2bbttbsKNf/vP2F/vV5iVwOQ4Un9tQ14wZoRM/MRJcHoBMiwAAAAADQapZla/n6Ej3y1lat3VmhrFS3fn72YE0d20/5Gb5ElwegEyPAAAAAAHBE4ailFz7arUdXfKEt+wLqnZ2iP1w8Qt8d3UcpHsauAND+CDAAAAAAHFIgFNXC93fqr+9sU3FlUMN6+HX/lJG68BsFcjkdiS4PQBdCgAEAAACgmbJASE+u3K6n3tuuqmBUpw3M0R2XfUNnDekuw2BgTgDHHgEGAAAA0Iht2wqEogpFLVmWLdO2ZdmxsR9My5Zlx16mpfi8bUu2fXDZsmP7serX2bZky67/AMlWbPvYYsP7sZ+x44U0eq/+/dpwVDUhU4FQRIGQqZpQVIFgVIFwND4fNi2Z9bXatmL1x4/DltWobtNqqPNgrZYVmw9FTdmSzh+er+vOGqRRfbOP9a8CAJogwAAAAECnYtcHBtaXAoSGC/hQxFJJVVB7K4PaWz8trgxqb1VdbFoZVG3YTPRhtEqax6k0r0vpPpfSvS6leVzy+1xyGIYcDkNOw5DDITkMQ06HEVtvGHIYktNhyGg07zAMGYbi76e4nbpoZC8NzktP9GECgCQCDAAAABxDtm2rOhRVSWVQJVUh7a0Kal91UNXBqGpDUdWEY10F8WkoqppwVLUhU2HTkhoCCalJ54Otpl0PX4XTYSjf71WPTJ+O75GhCUPzlJ/hlc/tbHThfzAEaLjwdxqGnA7JMAwZqr/wr19uCAEc9e/JkAzFAoL6xdjPNVmOzRlGbFk6uG+j/udTPM5YUOF1Ks3jksPBrRwAug4CDAAAgE7qy7cNmHbstoKIaSkcjb0ipqVQ1FLYtBSpn4ajlkr3VyijTPFbERpe0fp9mqYlsz4wMK2mtx80vuUiFDW1rzqkvZXB+LQu0ry7weUwYp0EXpdSPU6lel1K9zqVk5YaX+d2Ohp1CRwMAAwZ9cv18w2dB4bkcDTtKnAYhjwuh/L8PhVkxl656V45CQIAIOkRYAAAALShmlBUxZV12l0RVElVMD4OQcPIBg1jGTQsNIyFYDeat+zG28Z+Phy1FAhFVV0/zkFNo/lAKPaqDUebBBWJ5nQY8jgdysvwKt/v04ieGTpnWJ7yM3zKz/Qp3++NzWf4eAwnAOCICDAAAACOwLJitz1U1UVUWRdRVV1E+2vC2lNRpz0VsbBiT0Wd9lTWqaI20m51eF0O+evHOmgY86Bnli++nOpx1d/moPqxD4z4tGF9QweCx+mQx+WQu3765XW11VXKycmSq/5nXQ6HHA7J5XDEb6NoeXyFg2MrAADQlggwAABAl7DrQK3u+ucG7asOxW43UP14BTp4i0HDNBgxVRWMhRWVtRFVh6LxJ0Z8WYbPpZ5ZKeqVlaLR/bLVMytFPbN86pWVovwMn9xOhyQ1GdegYTyExuubjHVgGM3WNwQNx0pxcZ0KemQcs88DAOBICDAAAECnZtu2/v7+Ls196XMZhqERPTPqB3u0ZJuNB4SUVD92Q8MYCcfl+ZXhcykzxa2MFHeTaU6aRwWZPvl97kQfIgAAXQIBBgAA6LSKK+t003OfasWmUo0b3E13Xf5N9cpKSXRZAADgKBBgAACATse2bS1Zu1uzl32mqGnrj5ecoO+P6cu4DAAAdGAEGAAAoFPZVx3ULUvWafn6Ep3aP0d3f/eb6pebluiyAADA10SAAQAAOo1lH+/RbS+sU03Y1MwLj9e0MwbI6aDrAgCAzoAAAwAAHHOWZSsQjioQjCoUtWTXD6QZe9KHLdtWfNmWLdOyFYxYCkVNheqnwUbTYMTU2p3leuWzEp3YJ0v/9d0TNTgvPbEHCQAA2hQBBgCgU4mYlqqDUVXVxR6BWRWMqKouqsq6iEJRU4Ykh6P+AZaGcfBxmkbD4ywNWfVPorBsW3aT+dhUkrxWUKOsVPXvlqZ0b8f536lt2wpFLdWGTdWEorFpOKraUP00HFUoYsWe0tEkSKifr/8+bElR01LUshUxLUVNWxErNo2aliJWbFoTMlUdiqo6GFEgGFV1MKpAKPZqa16XQzdMGqqfnDlQLuexe9woAAA4NjrOv7gAAF1eVTCiogN1KiqvVVF5Xf0rNl9eG1ZVXUQ1YfMYVrRNktTd79WA3DT1y40FGgO6xeZT3E7VRQ52CAQjZny5LmIqFDFl21K6z6V0r0t+X+yV7nXH16V7XXIYUlVdVKWBoPZVh1QWCKu0OnTwFQiprDqksGnJtOz4K2pZMi3JtGJBg2XZCkZj27Q1p8OQy2HI7XTI5TTkcjjkdhpK8x48toJMX/0xuRsdq0tet0OGDDWMr2kYsYCpIVAyDMlhGPK5HfK5nfK6Dj0luAAAoPMiwAAAJKWIaWn55yV68ZNibSurUVF5raqCTf9qn+pxqnd2inplpWhEzwxlpriVkeJWhs+lzFS3MnwNy25lprjldTni3QQNHQayFZ+36jsMnA5DDuPghbOjUaeG4ZBsS1q3fY8qTY+2ldVoe1mNduyv1ZubSlX6QVGbfxcuh6FoC6GD22moe7pX3f1e9cj0yed2yOlwyFVfv8thyOk05DQMOR2xl8/tUKrHpTSPU6lel9I8LqV6nbGpx6k0r0telyN+/LEg4eD8wfVGLKhwGnI7HHIwzgQAAGhnBESO4gAAACAASURBVBgAgKSyc3+t/r56p/7fmiKVBULKz/BqRM9Mndw/W72zU9Q7OzU+zU51J+yxmAOzPSooKGi2PhCKasf+WKARMS15XU6leJzyuRyxqdupFLdT3vpuAkmqCcXGgqhquL0iGFUgFFF1/S0XYdNSbppH3f3eeGDR3e9VZkrijh8AAOBYI8AAACRcOGpp+foS/f39nXp7c5kchjRxWL6uGtNHZw3J61BPkUj3ujSiZ6ZG9Mxs9c9k+NxS6zcHAADokggwAAAJs72sRgtX79KzH+xSWSCsXlkpuv68Ibri5D7qkelLdHkAAABIIgQYAICvreHJFo0fb1kVjKg0EBtksiwQUll1OD7YZFkgNvBkRW1EToehc4bl6Xtj+urM47p3qG4LAAAAHDsEGAAA1YajWr5+n97eVKraiKlI1FLEtBQx7fpp0/lQNPYKRkyFopbCUeuIn5Hmcaqb36tu6V4N6p6uMQNz1C8nTReN7Kn8DLotAAAAcHgEGADQRQUjpt7YsE8vflKs1zaUKBixlJPmUXaqW26nQx6XI/5YzDSvKz7vdjrkdTnkrX9spdftkNfllK9+2vBIy3SvKz7oZDe/R6ke/pcDAACAo8e/JgGgCwlFTb21sVQvfVqs5Z+XqCZsKjfNo8tH99Z3vtlTp/TP4RYOAAAAJCUCDADooIIRU7sO1GrngVpV1EZk2rZMy1bUsmWaVmxq2bH1pq1t+2v06mclqg5FlZXq1kUje+rCb/TUaQNz5HI6En04AAAAwGERYABAErIsW3URUzXhqPZUBLVjf4127q/VjgO12rk/FlrsrQp+pX36fS6dP6KHCk8s0BmDu8lNaAEAAIAOhAADANpYxLS0PxBWZV2kxVdVo2lNOKq6sKna+lddxFRtOKpgpOVBMfP8XvXLTdUZg7upX26q+uWmqk9OqnJSPXI5DbkcDjkdRvzl+tLUMLg9BAAAAB0TAQYAfAW2bauyLqLdFXXaUxHUnoo67amoq1+OrSupDsq2D70Pv8+lzBS3MnxupXmdykz1qCDTqVSPUymehqlLqR6n0jxO9chMiQUV2alK8TiP3cECAAAASYQAAwBa4YvSgJ5bW6TnP9yj3RV1Td7zOB3qmeVTz6wUjTuum3pmpSjP71VWqluZKU1ffp+bQTIBAACAo0CAAQCHUFkb0bJP9ui5tUX6cGeFHIZ05pDumnZGf/XKSlHP+ldumkcOQgkAAACgXRFgAEAjEdPSik2lem5tkZZ/vk9h09LQfL9uuWCYLhnZS3kZvkSXCAAAAHRJBBgAugzbtlUTNlVeExtgs7w2rPLaiCrrp8WVQb36+V6VBcLKTfPo6tP6avJJvTWiZwaDXwIAAAAJRoABoNMqrqzTik2lemtTqT7YUa4DNWFFzEOPrun3ujTuuG6afFJvnTW0O48ZBQAAAJIIAQaATiMYMbV6+wG9tbFUKzaXalNJQJLUI8On0wd1U36GT9mpbmWnepRZP81OdSsr1aPMFLc8LgILAAAAIFkRYADocGzbVnUoqtLqkEqrQ/p8T5VWbC7Vv7/Yr2DEksfp0JiBOfru6D46a2h3HZeXzi0gAAAAQAdHgAEgKZUFQlq5pUzbymriQUVpIBSfD0WtJtsP7J6mKaf01VlDu+u0AblK8TgTVDkAAACA9kCAASAphKOW1uw4oBWbyvT25lJ9tqcq/l5Omkfd073q7veqf/80dfd748vd/V71y01V7+zUBFYPAAAAoL0RYABICNu2ta2sRis2lertzWV674v9qg2bcjkMndQvWzdMGqrxx3XTsB4ZjE0BAAAAgAADwLFjWrY+2FGuVz7bq399vle7DtRJkvrlpmrySb115pDuOm1gjvw+d4IrBQAAAJBsCDAAtKtw1NK7W8v0ymclevXzvSoLhOVxOjTuuG669sxBOvO4buqXm5boMgEAAAAkOQIMAG2uLmLq/z4t1j8/26vXN+xTdTCqNI9TE4bl6Vsn9NDZQ/OU7uU/PwAAAABajysIAF+bZdnasLda72wp1Ttb9mvV1jKFTFvZqW59+4Qe+tYJPXT6oG7yuXkyCAAAAICjQ4AB4KjsqajTO5vL9M6WMr27tUxlgbAk6bi8dF02Ml8XndRfp/TPlsvJAJwAAAAAvr52CzA+/vhj3XPPPVqwYEGz9+rq6jRt2jTNnTtXgwYNkmVZmj17tjZu3CiPx6M5c+aoX79+7VUagKMQjJh6d2uZ3txYqnc2l+mLshpJUne/V+OP665xg7vpjMHd1CPTp+LiYhUU5Ca4YgAAAACdSbsEGPPnz9fSpUuVkpLS7L1PP/1Us2bNUklJSXzd8uXLFQ6HtWjRIn300Ue688479fDDD7dHaQC+grJASK9v2Kfln5fo7c1lqouYSvU4NWZAjq4a01fjj+uuIfnpMgwj0aUCAAAA6OTaJcDo27evHnjgAd14443N3guHw3rwwQebvPfBBx9o/PjxkqSRI0dq3bp17VEWgCOwbVtbSwN69fN9Wr6+RGt3lsu2pZ6ZPn335N469/h8jRmYI6+LsSwAAAAAHFvtEmBMmjRJRUVFLb43evToZusCgYDS09Pjy06nU9FoVC7X4cuLRqMqLi7+esUmSEetGx2TbdsKRW1VBaOxV8hUVdBsNB9VRV1UHxYFVFQZkiQNy0/TT07vrbMG52hQrlemaUqK6kDpvlZ9Juc4ugLOc3R2nOPo7DjH0dl1tnM8KQbxTE9PV01NTXzZsqwjhheS5HK5VFBQ0J6ltYvY+AAdr250LKZl692tZfrH2t169fMSVYeih9zW5TCUlerW8J6Z+smEfJ17fJ4KMpvfAtZanOPoCjjP0dlxjqOz4xxHZ9eRz/GKiooW1ydFgHHSSSfpjTfe0AUXXKCPPvpIQ4YMSXRJQIdk27Y+21Ol5z/craUf79G+6pD8Ppe+/Y0eGtAtXVmpbmWluJWZ4lZmqltZqR5lpbiV6nEyjgUAAACApHZMAoxly5aptrZWV155ZYvvn3feeVq5cqWmTJki27Y1b968Y1EW0GnsrqjTCx/t1j/W7tbmfQG5nYYmDM3TpaN6acKwPPncjFkBAAAAoGNrtwCjd+/eWrx4sSSpsLCw2fuNH6/qcDj0hz/8ob1KATqtNzfu08NvbtWqbQckSSf3y9bcS0/Qhd8oUFaqJ8HVAQAAAEDbSYpbSAB8NVv2BTTnpc/15sZS9c5O0fXnDdElI3upb25qoksDAAAAgHZBgAF0IJW1Ef35tU1a8N4OpbiduvWC4/XD0/vL43IkujQAAAAAaFcEGEAHEDUt/e39nbr31U2qqotoyql9df15Q9Qt3Zvo0gAAAADgmCDAAJLc25tL9ccXP9emkoDGDszVbYXDdXxBRqLLAgAAAIBjigADSDK2bWt/TVibSqr1+DvbtHz9PvXLTdWjU0fr/OH5PO4UAAAAQJdEgAEkiGXZ2l1Rpy2lAW3dF9CWhldpQBW1EUlSutelGd8epmln9JfXxaNQAQAAAHRdBBhAO7NtW8WVQW0sqdamvdWxaUm1tu6rUV3EjG+Xm+bRoLx0XfCNAg3unq7Beek6sXeWMlPdCaweAAAAAJIDAQbwNdm2rZqwqcq6iCprIzpQE9aWfdXaWBLQpvrQojoUjW+fn+HVkHy/vndqrgbnpcdfOWmeBB4FAAAAACQ3Agx0ebZta+eBWu2uqFNd2FRt2FRtOFo/bTQfMlUTjqqyLqKqukgssKiLqCoYlWnZzfablerWkHy/LhnVS0N6+DU0368h+enKSiWoAAAAAICvigADXU5tOKqPd1Vq7c5yfbizXB/urND+mvAht3c6DKW6nUrxOJXmdSkjxa3MVI/65qYpM8WlzBT3l14eDeqepu5+LwNuAgAAAEAbIcBAp7frQK3W7DigtTsqtHZnuTbsrY53TAzsnqazh+bppH5ZGtgtXWlep1I9TqV4XPHQwutyEEQAAAAAQIIRYKDTsSxbHxVVaPnnJVq+vkSbSgKSpDSPUyP7ZulnZw/SSX2zNbJPlrIZdwIAAAAAOgQCDHQKdWFTK7eUafn6Ei1fv09lgZCcDkOn9s/RzAv76IzB3TQk3y+ng04KAAAAAOiICDDQYZVUBfXmxn169fN9emdLqYIRS36vS2cN7a7zhufr7CF5PIIUAAAAADoJAgx0GBHT0gc7yvXmxlK9talU64urJEm9slJ05cl9dO7wfI0ZkCuPy5HgSgEAAAAAbY0AA0ltT0Wd3tpUqjc37tPKLfsVCEXlchga3S9bN31rmM4e2l3DevgZZBMAAAAAOjkCDCSdurCpZ1bt0OI1u+IDcBZk+lR4YoHOGpKnMwbnyu/j1hAAAAAA6EoIMJA0ghFTf1u1Uw+/tVWl1SGd3C9bt1wwTGcPzdNxeel0WQAAAABAF0aAgYQLRkwtfH+nHnpzq/ZVh3TawBz95XujNGZgbqJLAwAAAAAkCQIMJEwoamrx6l168I2t2lsV1KkDcnT/lFEaO4jgAgAAAADQFAEGjrlw1NLiNbv00BtbtKcyqJP7Zeu/rjhRpw/K5TYRAAAAAECLCDBwTFTWRvTW5lK9vr5Eb24qVUVtRCf1zdJdl39T4wZ3I7gAAAAAABwWAQbahW3b2loa0Gvr9+n1Dfu0Zke5TMtWTppHE4fl6ZKRvTT+OIILAAAAAEDrEGCgzdi2rVXbDuif6/bq9Q37tPNArSRpWA+/rjtroCYOy9fIPllyOggtAAAAAABfDQEG2sTHuyp0x/+t17+/OCCPy6EzBuXqP84cqInD8tQrKyXR5QEAAAAAOjgCDHwtO/bX6O5XNurFT4qVm+bR7ReN0HdP7q1UD6cWAAAAAKDtcJWJo7I/ENIDr2/RM6t2yOVw6JcTB+vaMwfK73MnujQAAAAAQCdEgIGvpDYc1ePvbNMjb32huoipK07uo/889zjlZfgSXRoAAAAAoBMjwIBs29b+mrCipq2IaSlq2YqaliKmrajVsGxrY0m1Hnhts/ZVh3Te8Hzd9K2hGpznT3T5AAAAAIAugACji/twZ7lmLf1MnxRVtmr7k/pm6cGrT9Ip/XPauTIAAAAAAA4iwOii9gdCuuufG7R4TZHy/F7N+PYwZfjccjkNuZ2GXA6HXA5DLqcjts7hkN/n0jd7Z8oweAwqAAAAAODYIsDoYqKmpWdW7dR//WujasOmfnLmQP3ynOOU7uVUAAAAAAAkL65au5D3tx3QbS+s04a91Ro3uJtmXzRCg/PSE10WAAAAAABHRIDRBZRUBXXHy+v1/Ed71CsrRY98/yRNGtGDW0EAAAAAAB0GAUYnVh2M6Kn3duihN7YoYtr65cTB+tnZg5XicSa6NAAAAAAAvhICjE7oQE1YT6zcpv99d7uqglGdMyxPv//OcPXvlpbo0gAAAAAAOCoEGJ3Inoo6zX/7Cy18f5fqIqYmjcjXz84erBP7ZCW6NAAAAAAAvhYCjE7gi9KAHnlrq/7x4W5ZtnTxyJ766VmDdFy+P9GlAQAAAADQJggwOrB1uyv18Jtb9fK6YnmcDn3v1L76j/ED1ScnNdGlAQAAAADQpggwOhjbtvX25jI9tuILvbOlTH6vS9edNUg/PmOAuvu9iS4PAAAAAIB2QYDRQURMSy9+skePrdim9cVV6u736sZvDdXVY/opM8Wd6PIAAAAAAGhXBBhJLhCKauH7O/X4O9u0pzKowXnp+tPkb+riUT3ldfE4VAAAAABA10CAkaRKqoJ6YuV2PbNqh6qDUY0ZkKM/XnKCJgzNk8NhJLo8AAAAAACOKQKMJLR2Z7mmPPpvRS1L3zqhh649c5BG8ihUAAAAAEAXRoCRhLaUBBQ2Lf3jZ6drVN/sRJcDAAAAAEDCORJdAJoLRk1J4nGoAAAAAADUI8BIQqGIJUnyuvj1AAAAAAAgEWAkpWAk1oHhc/OUEQAAAAAAJAKMpBSKWnIYkounjQAAAAAAIIkAIykFI6Z8bqcMgwADAAAAAACJACMphaIW418AAAAAANAIV8lJqKEDAwAAAAAAxBBgJKFg1CLAAAAAAACgEQKMJBSKmNxCAgAAAABAI1wlJ6Fg1JKXDgwAAAAAAOIIMJIQHRgAAAAAADTFVXISYgwMAAAAAACaarcA4+OPP9bUqVObrX/99dc1efJkXXnllVq8eLEkqbq6WtOnT9fVV1+tH/3oRyotLW2vsjoEOjAAAAAAAGiqXa6S58+fr5kzZyoUCjVZH4lEdMcdd+jxxx/XggULtGjRIpWWlmrJkiUaMmSInnnmGV1wwQX661//2h5ldRghOjAAAAAAAGiiXQKMvn376oEHHmi2fuvWrerbt68yMzPl8Xg0evRorVmzRkOGDFFNTY0kKRAIyOVytUdZHUYwYspHBwYAAAAAAHHtkhRMmjRJRUVFzdYHAgH5/f74clpamgKBgAYMGKCVK1fqggsuUGVlpZ555plWfU40GlVxcXGb1X0sHa7uunBUMiMd9tgA6fDnONBZcJ6js+McR2fHOY7OrrOd48e01SE9PT3eaSFJNTU18vv9+stf/qLp06drypQp2rBhg375y19q2bJlR9yfy+VSQUFBe5bcLoqLiw9bd9j8WJnpqR3y2ADpyOc40BlwnqOz4xxHZ8c5js6uI5/jFRUVLa4/pvcpDBo0SDt27FBFRYXC4bDWrFmjUaNGKSMjI96ZkZub2yTk6IpCUUteN7eQAAAAAADQ4Jh0YCxbtky1tbW68sorNWPGDF1zzTWybVuTJ09Wfn6+fv3rX2vmzJn629/+pmg0qj/+8Y/HoqykFDEtmZYtn4tBPAEAAAAAaNBuAUbv3r3jj0ktLCyMr584caImTpzYZNv8/HzNnz+/vUrpUEJRS5LowAAAAAAAoBGukpNMMGJKEo9RBQAAAACgEQKMJBMPMLiFBAAAAACAOAKMJMMtJAAAAAAANMdVcpJp6MDw0oEBAAAAAEAcAUaSoQMDAAAAAIDmuEpOMoyBAQAAAABAcwQYSYYODAAAAAAAmuMqOcmE6MAAAAAAAKAZAowkE4zEOjB8dGAAAAAAABDHVXKSCUXrn0LipgMDAAAAAIAGBBhJJt6B4eJXAwAAAABAA66SkwwdGAAAAAAANEeAkWTowAAAAAAAoDmukpNMKGrK6TDkcvKrAQAAAACgAVfJSSYYsei+AAAAAADgS7hSTjLBiCkf418AAAAAANAEAUaSCUUteenAAAAAAACgCa6UkwwdGAAAAAAANEeAkWRCUUseOjAAAAAAAGiCK+UkQwcGAAAAAADNEWAkGcbAAAAAAACgOa6Uk0yIDgwAAAAAAJohwEgywYgln5tfCwAAAAAAjXGlnGRCUVNeFx0YAAAAAAA0RoCRZOjAAAAAAACgOa6UkwwdGAAAAAAANEeAkWTowAAAAAAAoDmulJOIbdsKRXkKCQAAAAAAX0aAkUQipi3Llrwufi0AAAAAADTGlXISCUZNSaIDAwAAAACALyHASCKhiCWJDgwAAAAAAL6MK+UkEozEOjC8dGAAAAAAANAEAUYSCUXpwAAAAAAAoCVcKSeRhg4MxsAAAAAAAKApAowkEmIQTwAAAAAAWkSAkUQYxBMAAAAAgJZxpZxEeIwqAAAAAAAtI8BIInRgAAAAAADQMq6UkwgdGAAAAAAAtIwAI4nQgQEAAAAAQMu4Uk4iPEYVAAAAAICWuVqz0aZNmzR79mxVV1ersLBQxx13nCZMmNDetXU5wWisA8PnJlcCAAAAAKCxVl0pz507V3fccYeysrJ0+eWX64EHHmjvurqkg7eQ0IEBAAAAAEBjrf5Tf79+/WQYhnJycpSWltaeNXVZwagpt9OQ02EkuhQAAAAAAJJKqwKMzMxMLVy4UHV1dXrppZeUkZHR3nV1SaGIRfcFAAAAAAAtaFWAMW/ePBUVFSk7O1vr1q3T3Llz27uuLikYNRn/AgAAAACAFrTqavnAgQMaNmyYHnvsMTmdTgUCgfauq0uiAwMAAAAAgJa1KsC48cYb1b17d0nSWWedpVtvvbVdi+qqglFTXjowAAAAAABoptVXy2PGjJEknXLKKbIsq90K6spCEVM+OjAAAAAAAGjG1ZqNMjIytGjRIo0cOVKffPIJTyFpJ6GoRQcGAAAAAAAtaNXV8p133qktW7bo7rvv1tatWzVv3rz2rqtLCtKBAQAAAABAiw7bgbF371716NFDlZWVuuqqq+LrKysrlZOT0+7FdTWhqKW0tFY1xQAAAAAA0KUc9mr5iSee0M0336zbbrtNhmFIkmzblmEYeuqpp45JgV0JHRgAAAAAALTssAHGzTffLCn25JHp06cfk4K6MsbAAAAAAACgZa26Wl6xYoVM02zvWro8OjAAAAAAAGhZqwZcKC8v1/jx49W7d28ZhiHDMLRw4cL2rq3LCUYs+ejAAAAAAACgmVYFGI888kh71wFJoagpr5sODAAAAAAAvqxVf+4Ph8OaM2eOpk+frrvuuqtVO/744481derUZutff/11TZ48WVdeeaUWL14sSTJNU3PmzNGUKVN02WWX6Y033vgKh9A52LYd68Bw0YEBAAAAAMCXtaoD46abbtLPf/5znXTSSfrggw80Y8YMLViw4JDbz58/X0uXLlVKSkqT9ZFIRHfccYeeffZZpaSk6Hvf+54mTJigt99+W9FoVAsXLlRJSYn+7//+7+sdVQcUNi1JogMDAAAAAIAWtOrP/SkpKTrrrLPk9/t19tlny+E4/I/17dtXDzzwQLP1W7duVd++fZWZmSmPx6PRo0drzZo1euedd9SjRw9de+21mjlzpiZOnHh0R9OBBSP1AQYdGAAAAAAANNOqDoyCggI99NBDOu200/TZZ5/J4/HonXfekSSNGzeu2faTJk1SUVFRs/WBQEB+vz++nJaWpkAgoPLycu3YsUOPPvqoVq9erZtvvlnPPPPMEeuKRqMqLi5uzSEknS/Xvb8mIkkyw8EOe0xAY5zH6Ao4z9HZcY6js+McR2fX2c7xVgUYhmFo165d2rVrlySpW7dueumllyS1HGAcSnp6umpqauLLNTU18vv9ysrK0tlnny3DMHTqqadq+/btrSve5VJBQUGrPz9ZFBcXN6s7eqBWkpST6e+QxwQ01tI5DnQ2nOfo7DjH0dlxjqOz68jneEVFRYvrWxVg3HHHHS2unzVr1lcqYtCgQdqxY4cqKiqUmpqqNWvW6JprrlFZWZneeustTZo0SRs2bOiwX/LXEYyYkiQfY2AAAAAAANBMqwKMQ9m2bVurtlu2bJlqa2t15ZVXasaMGbrmmmtk27YmT56s/Px8XXHFFZo1a5auuOIK2bat22+//euU1SGFooyBAQAAAADAoXytAMO27UO+17t37/hjUgsLC+PrJ06c2GyQTo/Hc8guj66CDgwAAAD8//buPcjO+i4D+HOSs9kNuSEOpHRChAQYyzidhmCkI0JJdaJ0mFLTaZraxQyog1KhJQViIJoO2BAp3lIvbToZFAnlqgOOOlPJSOxQ0WCLU4YBTTVMYKGhNIVdOGfPzT9itqbJ7p5tcvY92Xw+/zD7btj9zp7fMnyfPO97ABjdUf11f6lUOlZznPA0MAAAAGB0tuUuoYEBAAAAozuqAGOsW0iYmJEGRo9MCQAAAH7QmNtyo9HI8PBwPvGJT6RWq2V4eDjVajVXXnllkmTbtm2TMuSJYKSBUdbAAAAAgB805kM8H3744fz5n/95XnvttaxYsSJJMm3atFxwwQVJkp6ens5PeIKo1A40MNxCAgAAAIcbM8D4yEc+ko985CN56KGH8uEPf3iyZjohVesHGhge4gkAAACHa2tb/umf/ulcd911+cAHPpBrr702e/fu7fRcJxwNDAAAABhdWwHGhg0b8sEPfjD33XdfPvShD+WWW27p9FwnHA0MAAAAGF1b23K1Ws373//+zJ07Nz/7sz+bRqPR6blOOJVaMzOmT8u0aaWiRwEAAICu01aA0Wg08vzzzyfJyD85tqr1hvYFAAAAjGLMh3gedOutt2b9+vXZt29fTjvttNx2222dnuuEU6k10+v5FwAAAHBEbQUY5513XrZu3ZoXX3wxZ555Zk4++eROz3XCqdYa6evRwAAAAIAjaSvAuPfee/OXf/mXOeecc/Kf//mf+Y3f+I188IMf7PRsJ5RqvekWEgAAABhFWwHGgw8+mEcffTS9vb15++238/GPf1yAcYxVag1voQoAAACjaOuv/H/0R38006cfWK77+vrcQtIBGhgAAAAwurYaGK1WK1dccUWWLFmS5557LrVaLWvXrk2S3HXXXR0d8EShgQEAAACjayvA+NCHPpQ33ngj06dPz5NPPpn+/v6cd955nZ7thFKtNzOnr62XAwAAAE44bd2z8Mgjj2Tx4sV58sknc8MNN+Txxx/PsmXLsmzZsk7Pd8LQwAAAAIDRtRVg1Ov1/ORP/mTeeOONfOADH0iz2ez0XCecSl2AAQAAAKNpK8Co1WrZtGlTLrjggvzLv/xLGo1Gp+c64VRrHuIJAAAAo2lrY77jjjty1lln5dd+7dfy+uuv58477+z0XCcct5AAAADA6Np6auSZZ56ZM888M0ly2WWXdXKeE5a3UQUAAIDR2Zi7QKvVOhBgaGAAAADAEQkwukC1fuChqBoYAAAAcGQ25i5QrR0IMDwDAwAAAI5MgNEFKvUD7+rS1+PlAAAAgCOxMXeBgw2M3rIGBgAAAByJAKMLaGAAAADA2GzMXUADAwAAAMYmwOgCGhgAAAAwNhtzF9DAAAAAgLEJMLpApaaBAQAAAGOxMXeB799CooEBAAAARyLA6ALfv4XEywEAAABHYmPuAhoYAAAAMDYBRhfQwAAAAICx2Zi7gAYGAAAAjE2A0QU0MAAAAGBsNuYuUKk3MqM8LaVSqehRAAAAoCsJMLpAtdZMn/YFAAAAjMrW3AWq9UZ6Pf8CAAAARiXA6AKVWjN9PV4KAAAAGI2tuQtU6430z2qawQAAFcRJREFUljUwAAAAYDQCjC6ggQEAAABjszV3gUqtkT4NDAAAABiVAKMLVOvN9GpgAAAAwKhszV1AAwMAAADGJsDoAhoYAAAAMDZbcxfQwAAAAICxCTC6gAYGAAAAjM3W3AUqtUZ6NTAAAABgVAKMLlCtNdPXI8AAAACA0QgwCtZstjLcaKa37KUAAACA0diaC1atN5NEAwMAAADGIMAoWLXeSBINDAAAABiDrblglZoGBgAAAIxHgFEwDQwAAAAYX8e25meeeSb9/f2HXd+xY0dWrlyZVatW5YEHHjjkc7t3787SpUtTrVY7NVbX0cAAAACA8ZU78UW3bt2aRx99NDNnzjzkeq1Wy6ZNm/LQQw9l5syZWb16dS699NKceuqpGRwczObNmzNjxoxOjNS1KrUDDYy+Hg0MAAAAGE1HtuaFCxdmy5Yth13fvXt3Fi5cmHnz5mXGjBlZunRpdu3alVarlQ0bNuSGG244LPSY6g6+C0lvWQMDAAAARtORBsaKFSuyd+/ew64PDg5mzpw5Ix/PmjUrg4OD+fznP59LLrkkP/7jPz6h71Ov1zMwMHDU8xbh4Nwvv/pGkmT47cEMDNSKHAmOqeP1dxMmwjlnqnPGmeqccaa6qXbGOxJgjGb27NkZGhoa+XhoaChz5szJF7/4xbzjHe/Iww8/nH379uWqq67KvffeO+7XK5fLOf300zs5ckcMDAyMzD3ruwdKMKee8iM5/fR5RY4Fx8z/P+MwVTnnTHXOOFOdM85Udzyf8f379x/x+qQGGIsXL86ePXuyf//+nHTSSdm1a1euvvrqfOUrXxn5M8uXL8+2bdsmc6xCeQYGAAAAjG9SAozHHnssb731VlatWpV169bl6quvTqvVysqVKzN//vzJGKFreQYGAAAAjK9jAcaCBQtG3ib18ssvH7m+fPnyLF++fNR/b8eOHZ0aqStpYAAAAMD4bM0FOxhg9PZoYAAAAMBoBBgF+/4tJF4KAAAAGI2tuWDVgw0MAQYAAACMytZcsGq9md7ytJRKpaJHAQAAgK4lwChYpdZIn+dfAAAAwJgEGAU72MAAAAAARmdzLpgGBgAAAIxPgFGwSq2Zvh4vAwAAAIzF5lywar2R3rIGBgAAAIxFgFEwDQwAAAAYn825YBoYAAAAMD4BRsE0MAAAAGB8NueCaWAAAADA+AQYBavUmunVwAAAAIAx2ZwLVq030tejgQEAAABjEWAUrFprprfsZQAAAICx2JwLVtHAAAAAgHEJMArUaLZSa7Q0MAAAAGAcNucCVeuNJNHAAAAAgHEIMApUrTWTRAMDAAAAxmFzLlBFAwMAAADaIsAoUOX/Ghh9PV4GAAAAGIvNuUAHn4HRW9bAAAAAgLEIMAqkgQEAAADtsTkXqFrTwAAAAIB2CDAKVKlrYAAAAEA7bM4F0sAAAACA9ggwCqSBAQAAAO2xOReoooEBAAAAbRFgFKj6fw2MXg0MAAAAGJPNuUAHn4HR16OBAQAAAGMRYBRopIFR9jIAAADAWGzOBarUGimVkhnTvQwAAAAwFptzgar1ZnrL01IqlYoeBQAAALqaAKNAlVrD8y8AAACgDQKMAlVqjfR5C1UAAAAYlwCjQNV601uoAgAAQBtszwXSwAAAAID2CDAKpIEBAAAA7bE9F0gDAwAAANojwCiQBgYAAAC0x/ZcoEqtmV4NDAAAABiXAKNA1VojfRoYAAAAMC7bc4GqdQ0MAAAAaIcAo0AVDQwAAABoi+25QBoYAAAA0B4BRoE0MAAAAKA9tueC1BvN1Jut9PVoYAAAAMB4BBgFqdabSZLespcAAAAAxmN7Lkil1kgSDQwAAABogwCjIBoYAAAA0D7bc0E0MAAAAKB9AoyCaGAAAABA+2zPBdHAAAAAgPYJMAoy0sDo8RIAAADAeGzPBTnYwOgta2AAAADAeAQYBanUDjQw+jQwAAAAYFy254JU6xoYAAAA0K6OBRjPPPNM+vv7D7u+Y8eOrFy5MqtWrcoDDzyQJHnzzTdzzTXX5OMf/3hWrVqVr3/9650aq2tUNTAAAACgbeVOfNGtW7fm0UcfzcyZMw+5XqvVsmnTpjz00EOZOXNmVq9enUsvvTT33XdfLrzwwqxZsybf+ta3snbt2vz1X/91J0brGhoYAAAA0L6OBBgLFy7Mli1bctNNNx1yfffu3Vm4cGHmzZuXJFm6dGl27dqVNWvWZMaMGUmSRqOR3t7etr5PvV7PwMDAsR1+knz79f1JksHvfTf1wVbB08Cxd7z+bsJEOOdMdc44U50zzlQ31c54RwKMFStWZO/evYddHxwczJw5c0Y+njVrVgYHBzN37twkyb59+3LjjTdm/fr1bX2fcrmc008//dgMPYkGBgYyo29WkmTB6aelZ7rbSJhaBgYGjsvfTZgI55ypzhlnqnPGmeqO5zO+f//+I16f1M159uzZGRoaGvl4aGhoJNB4/vnns2bNmnzqU5/KsmXLJnOsQlTrzUwrJeVppaJHAQAAgK43qQHG4sWLs2fPnuzfvz/Dw8PZtWtXlixZkv/6r//K9ddfn7vuuiuXXHLJZI5UmEqtkb6e6SmVBBgAAAAwno7cQvKDHnvssbz11ltZtWpV1q1bl6uvvjqtVisrV67M/Pnzs3HjxgwPD+d3f/d3kxxoavzZn/3ZZIxWmGq9md6yW0cAAACgHR0LMBYsWDDyNqmXX375yPXly5dn+fLlh/zZqR5WHMnBBgYAAAAwPhWAgmhgAAAAQPts0AXRwAAAAID2CTAKUqk30yvAAAAAgLYIMApSrTXcQgIAAABtskEXpFJvuoUEAAAA2iTAKIgGBgAAALTPBl2QqgYGAAAAtE2AURANDAAAAGifDbogB56B4ccPAAAA7bBBF6RSa6Sv7BYSAAAAaIcAoyDVejO9GhgAAADQFht0AeqNVhrNlgYGAAAAtEmAUYBqo5kkGhgAAADQJht0Aar1AwGGt1EFAACA9ggwCjBcbyWJt1EFAACANtmgC3DwFhINDAAAAGiPAKMAB28h6fUQTwAAAGiLAKMAI7eQeIgnAAAAtMUGXYCRh3hqYAAAAEBbBBgFGPY2qgAAADAhNugCaGAAAADAxAgwClD1DAwAAACYEBt0AUYaGN5GFQAAANoiwChAtXHwFhI/fgAAAGiHDboA338bVQ0MAAAAaIcAowDff4inHz8AAAC0wwZdgOFGM9OnlVKe7scPAAAA7bBBF6Bab2lfAAAAwATYogswXG96/gUAAABMgACjANV6UwMDAAAAJsAWXYBqo5U+DQwAAABomwCjAMP1ZmZoYAAAAEDbbNEFqNabGhgAAAAwAQKMAlQbrfRqYAAAAEDbbNEF0MAAAACAiRFgFGC43tTAAAAAgAmwRRfAu5AAAADAxAgwCnDgFhI/egAAAGiXLboAw/VWessaGAAAANAuAUYBNDAAAABgYmzRk6zVamW40dTAAAAAgAkQYEyyWqOVZisaGAAAADABtuhJVq03kkQDAwAAACZAgDHJKrVmEg0MAAAAmAhb9CSr1P6vgdGjgQEAAADtEmBMsmr9QAOjt+xHDwAAAO2yRU+ygw2MPg0MAAAAaJsAY5JpYAAAAMDE2aInWVUDAwAAACZMgDHJNDAAAABg4mzRk8wzMAAAAGDiBBiTrFIXYAAAAMBECTAmWbXmFhIAAACYKFv0JHMLCQAAAEycAGOSeYgnAAAATJwtepJV3EICAAAAE2aLnmTVeiPlaaWUp/vRAwAAQLs6ukU/88wz6e/vP+z6jh07snLlyqxatSoPPPBAkqRSqeQ3f/M387GPfSy/+qu/mtdff72ToxWmUmtqXwAAAMAEdWyT3rp1a2699dZUq9VDrtdqtWzatCnbtm3LPffck/vvvz/79u3Lfffdl3PPPTfbt2/PFVdckT/90z/t1GiFqtQbAgwAAACYoHKnvvDChQuzZcuW3HTTTYdc3717dxYuXJh58+YlSZYuXZpdu3bl6aefzq/8yq8kSS6++OK2Aox6vZ6BgYFjP3wHzcxwzviRvrz66qtpNptFjwMdc7z9bsIPwzlnqnPGmeqccaa6qXbGOxZgrFixInv37j3s+uDgYObMmTPy8axZszI4OHjI9VmzZuXNN98c93uUy+Wcfvrpx27oSXDrFe/IK6++mvnz5xc9CnTMwMDAcfe7CRPlnDPVOeNMdc44U93xfMb3799/xOuTfi/D7NmzMzQ0NPLx0NBQ5syZc8j1oaGhzJ07d7JHmxSlUilptYoeAwAAAI4rkx5gLF68OHv27Mn+/fszPDycXbt2ZcmSJTn//PPzxBNPJEl27tyZpUuXTvZoAAAAQJfq2C0kP+ixxx7LW2+9lVWrVmXdunW5+uqr02q1snLlysyfPz+rV6/OzTffnNWrV6enpyd33XXXZI0GAAAAdLmOBhgLFiwYeZvUyy+/fOT68uXLs3z58kP+7MyZM/PHf/zHnRwHAAAAOE55P08AAACg6wkwAAAAgK4nwAAAAAC6ngADAAAA6HoCDAAAAKDrCTAAAACArifAAAAAALqeAAMAAADoegIMAAAAoOsJMAAAAICuJ8AAAAAAup4AAwAAAOh6AgwAAACg6wkwAAAAgK5XarVaraKH+GF94xvfSG9vb9FjAAAAAMdItVrNe97znsOuH9cBBgAAAHBicAsJAAAA0PUEGAAAAEDXE2AAAAAAXU+AAQAAAHQ9AQYAAADQ9QQYAAAAQNcrFz3AiaTZbGbjxo15/vnnM2PGjNx+++35sR/7saLHgqNSq9Wyfv36vPTSSxkeHs6v//qv5+yzz866detSKpVyzjnn5Hd+53cybZq8lOPbd77znfziL/5itm3blnK57Iwz5XzhC1/Ijh07UqvVsnr16ixbtsw5Z8qo1WpZt25dXnrppUybNi233Xab/5YzZTzzzDP53Oc+l3vuuSd79uw54rn+/Oc/n3/6p39KuVzO+vXr8+53v7vosX8ofkMn0T/+4z9meHg4999/f9auXZs77rij6JHgqD366KM5+eSTs3379mzdujW33XZbNm3alE9+8pPZvn17Wq1WHn/88aLHhKNSq9Xy27/92+nr60sSZ5wp56mnnsrXv/713HfffbnnnnvyyiuvOOdMKU888UTq9Xq+/OUv59prr80f/uEfOuNMCVu3bs2tt96aarWa5Mj/j/Lss8/mX//1X/Pggw/m93//9/OZz3ym4Kl/eAKMSfT000/nZ37mZ5Ik73nPe/LNb36z4Ing6P38z/98rr/++pGPp0+fnmeffTbLli1Lklx88cV58sknixoPjonNmzfnox/9aE477bQkccaZcr761a/m3HPPzbXXXptrrrkm73vf+5xzppSzzjorjUYjzWYzg4ODKZfLzjhTwsKFC7Nly5aRj490rp9++ulcdNFFKZVKeec735lGo5HXX3+9qJGPigBjEg0ODmb27NkjH0+fPj31er3AieDozZo1K7Nnz87g4GCuu+66fPKTn0yr1UqpVBr5/JtvvlnwlPDDe+SRR3LKKaeMBNBJnHGmnO9+97v55je/mT/6oz/KZz7zmXz60592zplSTjrppLz00kv5hV/4hWzYsCH9/f3OOFPCihUrUi5//8kQRzrXP7iHHs/n3TMwJtHs2bMzNDQ08nGz2TzksMHxamBgINdee20+9rGP5fLLL8+dd9458rmhoaHMnTu3wOng6Dz88MMplUr52te+lueeey4333zzIX9r4YwzFZx88slZtGhRZsyYkUWLFqW3tzevvPLKyOedc453d999dy666KKsXbs2AwMD+eVf/uXUarWRzzvjTBX//zkuB8/1D+6hQ0NDmTNnThHjHTUNjEl0/vnnZ+fOnUmSb3zjGzn33HMLngiO3muvvZarrroqN954Yz784Q8nSc4777w89dRTSZKdO3fmggsuKHJEOCr33ntv/uqv/ir33HNP3vWud2Xz5s25+OKLnXGmlKVLl+af//mf02q18uqrr+btt9/Oe9/7XuecKWPu3LkjC9u8efNSr9f9/wpT0pHO9fnnn5+vfvWraTabefnll9NsNnPKKacUPOkPp9RqtVpFD3GiOPguJC+88EJarVY++9nPZvHixUWPBUfl9ttvz9///d9n0aJFI9duueWW3H777anValm0aFFuv/32TJ8+vcAp4djo7+/Pxo0bM23atGzYsMEZZ0r5vd/7vTz11FNptVr51Kc+lQULFjjnTBlDQ0NZv3599u3bl1qtliuvvDI/8RM/4YwzJezduzc33HBDHnjggfz3f//3Ec/1li1bsnPnzjSbzfzWb/3WcRvYCTAAAACArucWEgAAAKDrCTAAAACArifAAAAAALqeAAMAAADoegIMAAAAoOuVix4AAJga7rjjjjz77LPZt29fKpVKzjjjjJTL5SxdujSf+MQnjun3+spXvpJ3v/vdmT9//jH9ugBA9/I2qgDAMfXII4/kW9/6Vj796U937Hv09/dn48aNWbx4cce+BwDQXTQwAICOeeqpp/LlL385f/AHf5Cf+7mfy5IlS7Jnz55ceOGFefPNN/Mf//EfOeuss3LnnXdmYGAgGzZsSLVaTW9vb2677baccsopuf766zM4OJhKpZIbb7wxb7/9dp577rncfPPN2b59e+6///787d/+bUqlUi677LJceeWVWbduXVqtVgYGBvLWW29l8+bNWbBgwWFf66d+6qeK/hEBAG0SYAAAk+Kll17KX/zFX+TUU0/NsmXL8uCDD2bDhg15//vfnzfeeCObN29Of39/Lrnkknzta1/L5z73uVxzzTV57bXXcvfdd+c73/lO/ud//ifve9/78q53vSsbN27Miy++mL/7u7/L9u3bUyqVsmbNmlx00UVJkjPOOCObN2/OE088kTvvvDNr16497GsBAMcPAQYAMClOPvnkvPOd70ySnHTSSTn77LOTJHPmzEm1Ws0LL7yQL3zhC/nSl76UVquVnp6enHPOOfmlX/ql3HDDDanX6+nv7z/ka77wwgt5+eWXs2bNmiTJ9773vbz44otJkgsvvDBJsmTJknz2s58d92sBAN1NgAEATIpSqTTm5xctWpSrrroq559/fnbv3p1/+7d/y/PPP5+hoaF88YtfzLe//e189KMfzaWXXppSqZRWq5VFixbl7LPPzpe+9KWUSqXcfffdOffcc/MP//APefbZZ3PBBRfk3//933POOeeM+rUAgOODAAMA6Ao333xzNm7cmGq1mkqlkltuuSVnnnlm/uRP/iR/8zd/k56enlx33XVJDrQqbrrppmzbti3vfe97s3r16gwPDx/yziQ7d+7M448/nmazmU2bNuW000474tcCAI4P3oUEAJhy1q1bl8suuywXX3xx0aMAAMfItKIHAAAAABiPBgYAAADQ9TQwAAAAgK4nwAAAAAC6ngADAAAA6HoCDAAAAKDrCTAAAACArve/ATQmhIjx1TMAAAAASUVORK5CYII=\n",
+      "text/plain": [
+       "<Figure size 1080x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "experiments = experiments.sort_values(by =['rules_price']).reset_index(drop=True)\n",
+    "# df = experiments\n",
+    "cols = 1\n",
+    "rows = 1\n",
+    "cc_idx = 0\n",
+    "\n",
+    "# len(experiments)\n",
+    "# df = experiments.copy()\n",
+    "# print(df.head())\n",
+    "\n",
+    "while cc_idx<len(experiments):\n",
+    "    cc = experiments.iloc[cc_idx]['rules_price']\n",
+    "    cc_label = experiments.iloc[cc_idx]['rules_price']\n",
+    "#     secondary_label = experiments.iloc[cc_idx]['string']\n",
+    "    sub_experiments = experiments[experiments['rules_price']==cc]\n",
+    "    cc_idx += len(sub_experiments)\n",
+    "    fig, axs = plt.subplots(ncols=cols, nrows=rows, figsize=(15*cols,7*rows))\n",
+    "    sub_experiments.reset_index(drop=False)\n",
+    "    for i, experiment in sub_experiments.iterrows():\n",
+    "        df = experiment['dataset'].copy()\n",
+    "        # filter out starting point\n",
+    "        df = df[df.timestep > 0] \n",
+    "        # FIRST RUN ONLY\n",
+    "        df = df[df.run == 1]  \n",
+    "\n",
+    "        df = df.groupby('timestep').agg({'spot_price': ['min', 'mean', 'max']}).reset_index()\n",
+    "\n",
+    "#         print(df.head())\n",
+    "        plot_label = experiment['rules_price']\n",
+    "        ax = axs\n",
+    "        title = 'spot_price' + '\\n' + 'Scenario: ' + str(cc_label)  + ' rules_price'\n",
+    "        ax.set_title(title)\n",
+    "        ax.set_ylabel('spot_price')\n",
+    "        colors = ['b','orange', 'g', 'magenta', 'r']\n",
+    "        \n",
+    "#         ax.plot(df.timestep, df['price'],color = colors[0], label='Price')\n",
+    "        df.plot(x='timestep', y=('spot_price','mean'), label='Mean', ax=ax, legend=True)\n",
+    "        ax.fill_between(df.timestep, df[('spot_price','min')], df[('spot_price','max')], alpha=0.3) \n",
+    "\n",
+    "\n",
+    "        ax.legend()\n",
+    "\n",
+    "        ax.set_xlabel('Timesteps')\n",
+    "        ax.grid(color='0.9', linestyle='-', linewidth=1)\n",
+    "\n",
+    "        plt.tight_layout()\n",
+    "        \n",
+    "fig.tight_layout(rect=[0, 0, 1, .97])\n",
+    "fig.patch.set_alpha(1)\n",
+    "display(fig)\n",
+    "plt.close()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1080x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1080x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "experiments = experiments.sort_values(by =['rules_price']).reset_index(drop=True)\n",
+    "# df = experiments\n",
+    "cols = 1\n",
+    "rows = 1\n",
+    "cc_idx = 0\n",
+    "\n",
+    "# len(experiments)\n",
+    "# df = experiments.copy()\n",
+    "# print(df.head())\n",
+    "\n",
+    "while cc_idx<len(experiments):\n",
+    "    cc = experiments.iloc[cc_idx]['rules_price']\n",
+    "    cc_label = experiments.iloc[cc_idx]['rules_price']\n",
+    "#     secondary_label = experiments.iloc[cc_idx]['string']\n",
+    "    sub_experiments = experiments[experiments['rules_price']==cc]\n",
+    "    cc_idx += len(sub_experiments)\n",
+    "    fig, axs = plt.subplots(ncols=cols, nrows=rows, figsize=(15*cols,7*rows))\n",
+    "    sub_experiments.reset_index(drop=False)\n",
+    "    for i, experiment in sub_experiments.iterrows():\n",
+    "        df = experiment['dataset'].copy()\n",
+    "        # filter out starting point\n",
+    "        df = df[df.timestep > 0] \n",
+    "        # FIRST RUN ONLY\n",
+    "        df = df[df.run == 1]  \n",
+    "        df = df.groupby('timestep').agg({'alpha': ['min', 'mean', 'max']}).reset_index()\n",
+    "\n",
+    "#         print(df.head())\n",
+    "        plot_label = experiment['rules_price']\n",
+    "        ax = axs\n",
+    "        title = 'Alpha' + '\\n' + 'Scenario: ' + str(cc_label)  + ' rules_price'\n",
+    "        ax.set_title(title)\n",
+    "        ax.set_ylabel('Alpha')\n",
+    "        colors = ['b','orange', 'g', 'magenta', 'r']\n",
+    "        \n",
+    "#         ax.plot(df.timestep, df['price'],color = colors[0], label='Price')\n",
+    "        df.plot(x='timestep', y=('alpha','mean'), label='Mean', ax=ax, legend=True)\n",
+    "        ax.fill_between(df.timestep, df[('alpha','min')], df[('alpha','max')], alpha=0.3) \n",
+    "\n",
+    "\n",
+    "        ax.legend()\n",
+    "\n",
+    "        ax.set_xlabel('Timesteps')\n",
+    "        ax.grid(color='0.9', linestyle='-', linewidth=1)\n",
+    "\n",
+    "        plt.tight_layout()\n",
+    "        \n",
+    "fig.tight_layout(rect=[0, 0, 1, .97])\n",
+    "fig.patch.set_alpha(1)\n",
+    "display(fig)\n",
+    "plt.close()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1080x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1080x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "experiments = experiments.sort_values(by =['rules_price']).reset_index(drop=True)\n",
+    "# df = experiments\n",
+    "cols = 1\n",
+    "rows = 1\n",
+    "cc_idx = 0\n",
+    "\n",
+    "# len(experiments)\n",
+    "# df = experiments.copy()\n",
+    "# print(df.head())\n",
+    "\n",
+    "while cc_idx<len(experiments):\n",
+    "    cc = experiments.iloc[cc_idx]['rules_price']\n",
+    "    cc_label = experiments.iloc[cc_idx]['rules_price']\n",
+    "#     secondary_label = experiments.iloc[cc_idx]['string']\n",
+    "    sub_experiments = experiments[experiments['rules_price']==cc]\n",
+    "    cc_idx += len(sub_experiments)\n",
+    "    fig, axs = plt.subplots(ncols=cols, nrows=rows, figsize=(15*cols,7*rows))\n",
+    "    sub_experiments.reset_index(drop=False)\n",
+    "    for i, experiment in sub_experiments.iterrows():\n",
+    "        df = experiment['dataset'].copy()\n",
+    "        # filter out starting point\n",
+    "        df = df[df.timestep > 0] \n",
+    "        # FIRST RUN ONLY\n",
+    "        df = df[df.run == 1]  \n",
+    "\n",
+    "        df = df.groupby('timestep').agg({'reserve': ['min', 'mean', 'max']}).reset_index()\n",
+    "\n",
+    "#         print(df.head())\n",
+    "        plot_label = experiment['rules_price']\n",
+    "        ax = axs\n",
+    "        title = 'Reserve' + '\\n' + 'Scenario: ' + str(cc_label)  + ' rules_price'\n",
+    "        ax.set_title(title)\n",
+    "        ax.set_ylabel('Amount')\n",
+    "        colors = ['b','orange', 'g', 'magenta', 'r']\n",
+    "        \n",
+    "#         ax.plot(df.timestep, df['price'],color = colors[0], label='Price')\n",
+    "        df.plot(x='timestep', y=('reserve','mean'), label='Mean', ax=ax, legend=True)\n",
+    "#         ax.fill_between(df.timestep, df[('reserve','min')], df[('reserve','max')], reserve=0.3) \n",
+    "\n",
+    "\n",
+    "        ax.legend()\n",
+    "\n",
+    "        ax.set_xlabel('Timesteps')\n",
+    "        ax.grid(color='0.9', linestyle='-', linewidth=1)\n",
+    "\n",
+    "        plt.tight_layout()\n",
+    "        \n",
+    "fig.tight_layout(rect=[0, 0, 1, .97])\n",
+    "fig.patch.set_alpha(1)\n",
+    "display(fig)\n",
+    "plt.close()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAABDAAAAHhCAYAAACYzk47AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nOzde3xU1bn/8e9cMhMgQQQEc0qwBEG8VLFQONSIhGrxAqKCQChRREUo0FpfEPGnBq9416pHhdaip1Cxgra0eKpF0HIRY1RUikKtBSUSKIVEcyMzs2f9/kgyJHIJtkySvfbn/aqvmdmzZ+9nOwvL88xaz/YZY4wAAAAAAABaMX9LBwAAAAAAANAUChgAAAAAAKDVo4ABAAAAAABaPQoYAAAAAACg1aOAAQAAAAAAWj0KGAAAAAAAoNWjgAEAAKxSWFio4cOHt3QYAADgKKOAAQAAAAAAWj0KGAAAoNlUVlbqJz/5iUaOHKlLL71Ut9xyi9avX99oxkTDGRSPP/64Zs6cqQkTJmjYsGH66U9/qoqKCknS0KFD9dBDD+myyy7Teeedp+eee67RuaqqqvS9731PW7duTWybOHGiXnvttWa4UgAAcLRRwAAAAM1mxYoVqqys1LJly7R06VJJUnFx8WE/U1RUpJ///Of605/+pGAwqCeeeCLx3pdffqkXX3xRCxcu1GOPPaYtW7Yk3mvbtq0uueQSLVmyRJL0+eefa9u2bcrJyUnClQEAgGSjgAEAAJpNv3799Pe//115eXn6xS9+oSuvvFLdu3c/7GfOP/98de7cWX6/X6NHj9batWsT740fP14+n0/HH3+8zj77bK1bt67RZ8ePH69ly5YpGo3qt7/9rUaPHq1AIJCUawMAAMlFAQMAADSbzMxMrVixQpMnT1ZFRYWuuuoqFRUVyRiT2CcajTb6TMOCQzwel9+//68vwWDwkO9JUo8ePXTSSSdp5cqVWr58uS6//PKjfUkAAKCZUMAAAADN5rnnntNNN92k7OxszZo1S9nZ2ZKkHTt2aM+ePTLG6OWXX270mZUrV6q8vFzxeFwvvPBCoyUgv//97xOfX7dunQYPHnzAOcePH6/7779fp59+urp27ZrEqwMAAMkUbHoXAACAo+OSSy7R22+/rQsvvFBt2rRRRkaG8vLyVFlZqVGjRum4447TkCFDtHHjxsRnOnfurGuvvValpaX63ve+pylTpiTeKy4u1mWXXaZ9+/bplltuUVZWlnbv3t3onDk5Obrllls0bty4ZrtOAABw9PlMwzmbAAAArcjjjz+u0tJSFRQUHPDe0KFD9eijj+o73/nOYY+xYcMG3XLLLVq+fLl8Pl+yQgUAAEnGDAwAAGCtG2+8UW+//bYeeeQRihcAALgcMzAAAAAAAECrRxNPAAAAAADQ6lHAAACglXv//feVl5enESNGaPjw4brmmmv0ySeftGhMGzdu1E9+8pOkHHvSpEnau3dvUo59OC+99JKuu+66Zj9vQzfffLPefPPNFo0BAIDWih4YAAC0YpFIRNddd50WLFigU089VZK0bNkyXXvttVq5cqUCgUCLxPWd73xHjz32WFKOvW7duqQc1w3uvvvulg4BAIBWiwIGAACtWHV1tcrLy1VVVZXYdvHFFystLU2O4ygQCGjp0qV65pln5Pf7deyxx+q+++5TRkaGVq1apaeeekrRaFSpqam68cYbdeaZZ+rxxx/XF198od27d+uLL75Q165d9cADD6hLly56/fXXNX/+fEUiEe3du1eXXHKJrr/+ehUWFuruu+9W27ZtVVlZqfz8fN13331avny5ysvLdfvtt2vz5s3y+Xw6++yzdcMNNygYDOrRRx+VJP30pz9tdF2VlZW66aab9Nlnn8nv9+vUU0/VHXfcoZtvvlmSdOWVV+oXv/iF/H6/7rjjDpWUlCgajeqiiy7SlClTVFxcrLy8PJ199tn64IMPZIxRQUGB+vfv3+g8xcXF+tGPfqSePXvqiy++0L333qtJkyZpw4YNifdHjBiReF2vvLxcd999t/72t78pGo1q0KBBys/PVzAY1GOPPaYVK1YoJSVFxx57rO655x516dLlkN/hSy+9pFdeeUXxeFw7duxQ165dde+996pr167Ky8vTMccco3/84x/Kzc3Vn//8Z/3oRz/S+eefr9dff10///nPFY/H1bZtW91+++3q06eP3nvvPT344IOqrq6W3+/X9OnTlZOT8+8PMgAA3MIAAIBWbcGCBeb00083Q4cONTNnzjRLliwxVVVVxhhjPv74YzNw4ECzY8cOY4wxzzzzjLn11lvN1q1bzfDhw83evXuNMcb87W9/M2eddZaprKw0jz32mPnBD35gysvLjTHGXHfddebRRx818XjcTJgwwWzdutUYY8zOnTvNySefbPbs2WPeeust06dPH1NcXGyMMeatt94yF110kTHGmPz8fHPnnXeaeDxuampqzKRJk8z8+fMPe02/+93vzKRJk4wxxsRiMXPzzTebbdu2GWOM6d27t9mzZ48xxpi8vDyzcuVKY4wx+/btM3l5eebll18227dvN7179zZ/+MMfjDHGvPHGG+ass84ykUik0Xnq9ysqKkq87tu3b6P361+/+OKLZvLkycYYY2bPnm1+/etfJ+KbOXOm+cUvfmF27Nhhvvvd75qamhpjjDG/+tWvzIoVKw57rS+++KLp27ev+cc//mGMMeaBBx4wM2bMMMYYM2HCBHPTTTcl9p0wYYL505/+ZHbv3m369etnNm3aZIwx5tVXXzVXX321KSsrMz/84Q/N9u3bE9/R4MGDzRdffHHYGAAAsAEzMAAAaOWuuuoqXX755SoqKlJRUZF++ctf6pe//KWWLl2q9evXKzs7WxkZGZKkiRMnSpJ+85vf6J///GfitST5fD59/vnnkqQBAwYoLS1NknTKKafoyy+/lM/n07x58/TGG29o+fLl+vTTT2WMUXV1tSQpIyND3/rWtw6Ib/Xq1Vq8eLF8Pp9CoZDGjRun//3f/9XkyZMPeU39+vXTI488ory8PH3/+9/XlVdeqRNOOKHRPlVVVSoqKtKXX36ZmMlRVVWlzZs36/TTT9cxxxyjESNGSJLOOeccBQIBbdmyRaeddlqj4wSDQfXt2/dI/3VLkt544w1t3LhRS5culSTt27dPktS1a1f16dNHl156qQYPHqzBgwdr0KBBTR7vrLPOUo8ePSRJY8aM0ciRIxPvfX3WiCS999576tWrl0455RRJ0g9/+EP98Ic/1F/+8hft3r1b06ZNS+zr8/m0ZcsW/dd//dc3ukYAANyGAgYAAK3Yu+++qw0bNuiaa65RTk6OcnJydMMNN2j48OFat26dAoGAfD5fYv99+/bpiy++UDwe16BBg/Tzn/888V5JSYm6dOmiFStWKDU1NbHd5/PJGKOqqipdeumlOvfcc9W/f3+NGjVKr732mkzdHdfbtm170Bjj8XijGOLxuGKx2GGvKzMzUytWrFBhYaHeeustXXXVVbrjjjs0dOjQRscxxuj5559XmzZtJEl79+5VOBxWaWnpAf0/4vH4QXuChEIhBYPBRtdaLxqNHvKaHn30UfXs2VOS9NVXX8nn88nv92vRokXauHGj1q9fr7lz5+rss89Wfn7+Ya+3YVxfj/Ng/16//r0aY7RlyxY5jqOePXtqyZIlifd27dqljh07Hvb8AADYgLuQAADQinXs2FFPPfWU3nnnncS23bt3q6KiQr1799bAgQO1fv16/fOf/5QkPf/883rggQc0aNAgrVu3Tp9++qkk6S9/+YsuvvjixEyCg/nss89UUVGh66+/XkOHDlVhYaEikYji8fhhY8zOztaiRYtkjFEkEtELL7yg73//+4f9zHPPPaebbrpJ2dnZmjVrlrKzs/XRRx9Jqk3eY7GY0tLS1LdvXz3zzDOSaosIubm5WrlypaTaYsbq1aslSatWrVJKSop69+592PO2b99e0WhUf//73yVJL7/88iGv6dlnn01c09SpU7Vo0SJt3rxZw4cPV8+ePXXddddp4sSJ2rhx42HPKUlvvfWWdu3aJan2O2qqZ8UZZ5yhTz/9NHG3mZUrV2rWrFnq27evPvvsMxUVFUmSPv74Yw0bNixxbAAAbMYMDAAAWrEePXroiSee0COPPKKdO3cqHA4rPT1dc+fOVVZWliRp1qxZuuaaayRJxx13nObOnauuXbvqjjvu0A033CBjjILBoJ566im1a9fukOc66aSTNGTIEF1wwQUKhULq3bu3TjzxRH322WcKhUKH/Nwtt9yiu+66SyNGjFA0GtXZZ5+tKVOmSNIhm3hecsklevvtt3XhhReqTZs2ysjIUF5eniTp/PPPV15enh5//HE9+OCDuvPOOzVixAhFIhENHz5cF198sYqLixUOh7Vs2TI9+OCDSk1N1RNPPNHkXVnS09M1a9YsXXvtterYsaPOP//8g+5388036+67705c0/e//31dc801SklJ0QUXXKBRo0apbdu2Sk1N1S233HLYc0q1S09mzZql3bt368QTT9Qdd9xx2P07d+6sBx98UDfeeKMcx1FaWpoeeeQRdezYUY899pjuv/9+1dTUyBij+++/X926dWsyBgAA3M5nGs6jBAAAcIFD3T2kNXrppZf06quvav78+S0dCgAArsYMDAAAgP/Q+PHjVVlZedD3Ro8e3czRAABgJ2ZgAAAAAACAVo8mngAAAAAAoNWjgAEAAAAAAFo963pgvP/++wqHwy0dxjcWi8US96gHbMQYh80Y37AZ4xs2Y3zDdm4d4zU1Nerbt+8B2913JU0Ih8M6+eSTWzqMb6ykpEQZGRktHQaQNIxx2IzxDZsxvmEzxjds59Yx/vHHHx90O0tIAAAAAABAq0cBAwAAAAAAtHoUMAAAAAAAQKtnXQ+Mg4lGoyouLta+fftaOpRDchxHZWVlR+VYqamp6tatm1JSUo7K8QAAAAAAaGmeKGAUFxcrPT1d3/72t+Xz+Vo6nIOKRCIKhUL/8XGMMdqzZ4+Ki4vVo0ePoxAZAAAAAAAtzxNLSPbt26dOnTq12uLF0eTz+dSpU6dWPdsEAAAAAIBvyhMFDEmeKF7U89K1AgAAAAC8wTMFDAAAAAAA4F4UMJpJYWGhTjrpJP3f//1fo+0jRozQ7NmzWygqAAAAAADcgQJGM8rKytLy5csTr7ds2aLq6uoWjAgAAAAAAHfwxF1IGnrx3WK98M72o3rMMf0zNapftyb369Onj7Zt26avvvpK7du31x/+8AeNGDFCJSUlevXVV7Vo0SL5/X7169dPM2fO1M6dO3XbbbeppqZGZWVlmjZtms4991yNGDFCAwYM0JYtW+Tz+fTkk08qPT39qF4TAAAAAACtCTMwmtl5552nFStWyBijDz/8UGeeeabKysr05JNP6tlnn9XixYu1a9curVu3Tv/4xz901VVX6ZlnntGtt96q3/zmN5KkyspKXXTRRVq0aJG6dOmi1atXt/BVAQAAAACQXJ6bgTGqX7cjmi2RLCNGjNBtt92mzMxM9e/fX5LkOI5KS0s1efJkSbUFiu3bt6tfv3566qmntHTpUvl8PsViscRxTjnlFElSRkaGampqmv9CAAAAAABoRszAaGaZmZmqqqrSwoULdfHFF0uqve3p8ccfrwULFmjhwoWaMGGCzjjjDD366KMaOXKkHnjgAQ0cOFDGmMRxuFUqAAAAAMBLPDcDozW48MILtWzZMvXo0UPbt29Xx44dNWzYMOXl5clxHH3rW9/SBRdcoPPPP19333235s+fr4yMDJWWlrZ06AAAAAAAtAifafizvgU+/vhjnXzyyU1ua20ikYhCodBRO54brhneUlJSooyMjJYOA0gKxjdsxviGzRjfsJ1bx/ih8llmYACAJGOMjJHixshIiedq8Nyo7tFIarDNiRsZY+QYU/e8dptjjOJxo3iDY9UfL/Fcjbc7caNY3ChujGJO3WPcyInH5cQlJx5X1DGKOnHFHKOIE1fMqdsWjysaq33PyCTOY752XiOjuv8llqaZxGs1/uwR1Li/vsfBPlJ/nfXn2b9fg+0Nzl3/71oNnpsGMZv9B258bNP4fF+Pp2FoxjTe2jg2c9BrOxhjpJqaGoXD++9wdbDPWfZ7QSNHcmmH+p4ajdPDjB0bfP2/MaZuY/xrf+5am2g0qpSUv7d0GC2mNX4vrTAk14pFowq2svFt8/9foHm1T03Rred+S+4rXxwaBQzAQvF4bTIbie1PdmufxxVxapPcWDyeSAbjRolEuz45jNcl47WPavB8/2P981i8NtmufYzXJdz7n8fiRmVflSsY/pcisQaxxGrjqanbtj9JNYkkp2ER4eBJ+YGJZqKAEK+9zvqiwP5tJlF0iFv4dwSfTwrU9cmpb5fjU+JJ4sHnq93ua7Stbk9fg9dH0HLn67scrE+PL/He/lc+X+Pt9fH4G8Tvr4vhYPGp4bYGx2h47QfE0eCNxjEd/PNH0nEoGo0pJd64rdTBPtec7Yuau1PSkfRmOtj3lBhrknz+BtsbfMqmtk/+Bn+mfD6f/F8b263xWvft8yk1NdzSYbSw1vfFtMax4kY1+3wKp6a2dBgH4OvF0ZCemqKUgF2jyTMFDGOMZxpfUrVtrD5JjdYn1nW/VMfrfiGv/5W7YTJen+hW1jgq3xdVRU1M5ftiKt8XVfm+mL7aF6vbFtW+qKOoU5usR+oeo06DX8nrEvnGvxjvj+2bfFsHTeK/9qt1a0vKg36fAn6fUvw+pYYCCgX8CgX9Sql7DAX9CgX8Sk8NKuD3fe0v8g0T1/3J7eGSctV9rv689Y+BRq/9CtQlSv66g/vUOFH2+7+WQKtx4tFw34DfJ7/fJ7/Pp4Cv/nnddp+vwXU1/DfTIIluEH8wUPuZoN8vv18K1sUa8PsVrDteKFj7fkrQrxS/TykBv4KB2seUgF8Bvzf+W9eauHV6JnAkGN+wGeMbtispKWnpEI4qTxQwUlNTtWfPHnXq1Mn6IoYxRnv27FFqM1SSnbjRl9VRxZx47VT5ul/x6wsBX//lvn4afMxp/Ct4/fb6X+Ej9b/M1/1T02BbTczRvmhc+6KOqqOO9kX3v65/XhNz9s8GqJtufzSFgn61Tw0qPTVFaeGg2qQElJriVzAcrEvK65LLgF8pdUlloEEyLB3sV+8jP3/DX6AbHqthkl+fzIbqYggFA3WP/kSSG/Tv/6XbX/8roK/httrXAX99Uq4Gz2sf6xP3lPriQN1xA3VJdf0xJP6CAAAAAOA/44kCRrdu3VRcXKzdu3e3dCiH5DiOAoHAUTlWamqqunXrdsT7G2MUdYz2xRzti9QXBuIq3xfV7vIa/bO8Rv8s35d4Xv+4p6KmWX7p9/ukcDCgUNCvcNCv1LqCQe1jQO3bpKhLelipKQG1SQkonOJX0F/7i3Sw7hf3YF0RISXgq/tFe/+v8Q2T8fp/6n81bxcKKD01RempQaWnBpWWGlQ4eHS+JwAAAADAkfNEASMlJUU9evRo6TAOK5m/TpdVRbR5Z7m27Cyve/xKu76qUU3MUXXE0b5YbY+Awwn4feqcFlKX9FQd3z5V3/nWMeqSHlbHdiGlBP2HnDpfWwjYP/294ZT+YMD3tWnxtQWK+iUF4ZTax2DAf9jYAAAAAAD280QBw0uKS6v09ta9iWLF5rpiRb0ObVN0Utd0DezRUW1CgcSshfoZDW1CAaUGax/bhYM6Li2sLu3D6tg2lOgJAAAAAABAc6OAYZlLn3xTu8trFAr4dWKXNJ3Vs7NOOj5dfTLaq8/x6eqSHra+DwgAAAAAwD4UMCzzZVVUE/67u+aMOFUpLL0AAAAAAFiCDNcixhhFnLg6tgtTvAAAAAAAWIUs1yL1twsNBVgiAgAAAACwCwUMi0SduCQpFORrBQAAAADYhUzXIpFYbQGD5SMAAAAAANuQ6VqkfgYGBQwAAAAAgG3IdC0SYQkJAAAAAMBSZLoWqV9CEmIGBgAAAADAMmS6FknchYQZGAAAAAAAy5DpWoQmngAAAAAAW5HpWiSSaOLpa+FIAAAAAAA4uihgWCRKE08AAAAAgKXIdC1CE08AAAAAgK3IdC3CDAwAAAAAgK3IdC1CE08AAAAAgK3IdC2yv4knXysAAAAAwC5kuhaJOkaSFGYJCQAAAADAMmS6FmEJCQAAAADAVmS6FqGJJwAAAADAVmS6Ftk/A8PXwpEAAAAAAHB0UcCwCE08AQAAAAC2ItO1SGIJCQUMAAAAAIBlyHQtEonFFfT75PezhAQAAAAAYBcKGBaJOnEaeAIAAAAArES2a5FILE7/CwAAAACAlch2LRJxDAUMAAAAAICVgsk68Pz587Vq1SpFo1Hl5ubq8ssvlyTNnTtXPXr0UG5uriTprrvu0nvvvad27dpJkp588kmlp6cnjvP+++/r7rvvViAQUHZ2tqZPn56skF0v6sQVZgkJAAAAAMBCSSlgFBYWasOGDVq8eLGqq6u1YMEC7d27V/n5+dq2bZuuvvrqxL6bNm3S008/rY4dOx70WHPmzNHjjz+uzMxMTZ48WZs2bdKpp56ajLBdr3YJCQ08AQAAAAD2ScrP9WvXrlXv3r01bdo0TZkyRUOGDFFlZaVmzJihkSNHJvaLx+P67LPPVFBQoHHjxmnp0qWNjlNRUaFIJKLu3bvL5/MpOztb69evT0bIVqCJJwAAAADAVkmZgVFaWqodO3Zo3rx5Ki4u1tSpU/XKK68oMzNTq1evTuxXVVWlCRMm6KqrrpLjOLriiit02mmnqU+fPpJqCxhpaWmJ/du1a6ft27cf9tyxWEwlJSXJuKyk+0/j/qqySoo7rr1+2I+xCZsxvmEzxjdsxviG7Wwa40kpYHTo0EFZWVkKhULKyspSOBzW3r171alTp0b7tWnTRldccYXatGkjSfrv//5vbd68OVHASEtLU2VlZWL/yspKtW/f/rDnDgaDysjIOMpXlHwlJSX/cdz+4Odqm+p35fXDfkdjjAOtFeMbNmN8w2aMb9jOrWO8rKzsoNuTst6gX79+WrNmjYwx2rVrl6qrq9WhQ4cD9tu2bZvGjx8vx3EUjUb13nvvNepvkZaWppSUFH3++ecyxmjt2rXq379/MkK2AktIAAAAAAC2SsoMjJycHBUVFWn06NEyxqigoECBQOCA/Xr27KkRI0ZozJgxSklJ0ciRI9WrVy+tX79e7777rqZPn67bb79dM2fOlOM4ys7O1hlnnJGMkK0QicXVNpS0G8sAAAAAANBifMYY09JBHE0ff/yxTj755JYO4xs7GlN7Rjy+Vselh7Vg4veOUlTA0ePW6WvAkWB8w2aMb9iM8Q3buXWMHyqvZ72BRbiNKgAAAADAVhQwLBJ14koJ8JUCAAAAAOxDtmuRCE08AQAAAACWItu1SCQWV4gZGAAAAAAAC5HtWoTbqAIAAAAAbEW2a5HaJp58pQAAAAAA+5DtWiTqGAoYAAAAAAArke1awhhDE08AAAAAgLXIdi0RdYwkKRTwtXAkAAAAAAAcfRQwLBF14pLEDAwAAAAAgJXIdi0RidUWMOiBAQAAAACwEdmuJepnYFDAAAAAAADYiGzXEhGWkAAAAAAALEa2a4n6JSQhZmAAAAAAACxEtmuJxF1ImIEBAAAAALAQ2a4laOIJAAAAALAZ2a4lIokmnr4WjgQAAAAAgKOPAoYlojTxBAAAAABYjGzXEjTxBAAAAADYjGzXEszAAAAAAADYjGzXEjTxBAAAAADYjGzXEvubePKVAgAAAADsQ7ZriahjJElhlpAAAAAAACxEtmsJlpAAAAAAAGxGtmsJmngCAAAAAGxGtmuJ/TMwfC0cCQAAAAAARx8FDEvQxBMAAAAAYDOyXUsklpBQwAAAAAAAWIhs1xKRWFxBv09+P0tIAAAAAAD2oYBhiagTp4EnAAAAAMBaZLyWiMTi9L8AAAAAAFiLjNcSEcdQwAAAAAAAWIuM1xJRJ64wS0gAAAAAAJYi47VE7RISGngCAAAAAOxEAcMSNPEEAAAAANiMjNcSNPEEAAAAANiMjNcSEYcCBgAAAADAXmS8lmAJCQAAAADAZmS8lojE4goxAwMAAAAAYCkyXktEHcMMDAAAAACAtch4LcFtVAEAAAAANqOAYYkoTTwBAAAAABYj47VEhCaeAAAAAACLkfFagiaeAAAAAACbkfFagtuoAgAAAABsRsZriahj6IEBAAAAALAWGa8lau9CwtcJAAAAALATGa8FjDE08QQAAAAAWI2M1wJRx0iSQgFfC0cCAAAAAEByBJN14Pnz52vVqlWKRqPKzc3V5ZdfLkmaO3euevToodzcXEnSs88+q5dfflmSdM4552j69OmNjrNp0yZNmTJF3/72tyVJubm5uvDCC5MVtitFnbgkMQMDAAAAAGCtpBQwCgsLtWHDBi1evFjV1dVasGCB9u7dq/z8fG3btk1XX321JGn79u36wx/+oCVLlsjn82n8+PE699xz1adPn8SxPvroI1111VWaNGlSMkK1Qn0Bgx4YAAAAAABbJaWAsXbtWvXu3VvTpk1TRUWF8vPzVVlZqRkzZmj16tWJ/Y4//ng9/fTTCgQCkqRYLKZwONzoWH/961+1detWrVy5UieccIL+3//7f0pLSzvkuWOxmEpKSpJxWUn378b9r8qoJKmqoty11w5vYHzCZoxv2IzxDZsxvmE7m8Z4UgoYpaWl2rFjh+bNm6fi4mJNnTpVr7zyijIzMxsVMFJSUtSxY0cZY3T//ffrlFNOUY8ePRod6/TTT9fll1+u0047TU899ZSeeOIJ3XjjjYe+oGBQGRkZybispCopKfm343ZKqyRJx3U61pXXDm/4T8Y40NoxvmEzxjdsxviG7dw6xsvKyg66PSlrDjp06KDs7GyFQiFlZWUpHA5r7969B923pqZGM2fOVGVlpebMmXPA++edd55OO+20xPOPPvooGSG7WiRW1wODJSQAAAAAAEslJePt16+f1qxZI2OMdu3aperqanXo0OGA/Ywx+vGPf6yTTjpJd9xxR2IpSUNXX321PvzwQ0nS+vXrdeqppyYjZFdL3IWEJp4AAAAAAEslZQlJTk6OioqKNHr0aBljVFBQcNDixGuvvVaKzDMAACAASURBVKa3335bkUhEa9askSTdcMMNSk9P16JFi3Tbbbfptttu05133qmUlBR17txZd955ZzJCdjWaeAIAAAAAbJe026jm5+cfdPuMGTMSz8877zxt3LjxoPvddtttkqRTTz1Vzz///FGPzyY1sfoChq+FIwEAAAAAIDn4yd4C9TMwWEICAAAAALAVGa8FaOIJAAAAALAdGa8FmIEBAAAAALAdGa8FaOIJAAAAALAdGa8F9jfx5OsEAAAAANiJjNcCUcdIksIsIQEAAAAAWIqM1wIRZmAAAAAAACxHxmsBmngCAAAAAGxHxmuB/U08fS0cCQAAAAAAyUEBwwI08QQAAAAA2I6M1wKJJSQUMAAAAAAAliLjtUAkFlfQ75PfzxISAAAAAICdKGBYIOrEaeAJAAAAALAaWa8Foo6h/wUAAAAAwGpkvRaoicUpYAAAAAAArEbWa4GoE1eYJSQAAAAAAIuR9VogEosrJUADTwAAAACAvShgWIAmngAAAAAA25H1WiDq0AMDAAAAAGA3sl4L0MQTAAAAAGA7sl4LsIQEAAAAAGA7sl4LRGJxhZiBAQAAAACwGFmvBaKOYQYGAAAAAMBqZL0WqG3iyW1UAQAAAAD2ooBhgQhNPAEAAAAAliPrtUCEJp4AAAAAAMuR9VqAJp4AAAAAANuR9VqA26gCAAAAAGxH1muBqGPogQEAAAAAsBpZrwVo4gkAAAAAsB1Zr8sZY2jiCQAAAACwHlmvy0UdI0kKBXwtHAkAAAAAAMlDAcPlok5cklhCAgAAAACwGlmvy9UXMFhCAgAAAACwGVmvy0VizMAAAAAAANiPrNflIszAAAAAAAB4AFmvy9XPwAgxAwMAAAAAYDGyXpervwsJS0gAAAAAADYj63U5mngCAAAAALyArNflahJNPH0tHAkAAAAAAMlDAcPlmIEBAAAAAPACsl6Xo4knAAAAAMALyHpdrn4GBk08AQAAAAA2I+t1OZaQAAAAAAC8gKzX5fY38eSrBAAAAADYi6zX5aKOkSSFmYEBAAAAALAYWa/LRZiBAQAAAADwALJel9vfxNPXwpEAAAAAAJA8SStgzJ8/X2PHjtVll12mJUuWJLbPnTtXixcvTrx+4YUXdNlll2nMmDF6/fXXDzjO+++/r8svv1zjxo3T//zP/yQrXNeiiScAAAAAwAuSkvUWFhZqw4YNWrx4sRYuXKidO3dq7969uuaaa7Rq1arEfrt379bChQv1/PPP61e/+pUefvhhRSKRRseaM2eOHnroIS1evFgffPCBNm3alIyQXYsmngAAAAAAL0hK1rt27Vr17t1b06ZN05QpUzRkyBBVVlZqxowZGjlyZGK/Dz/8UGeeeaZCoZDS09PVvXt3bd68OfF+RUWFIpGIunfvLp/Pp+zsbK1fvz4ZIbtWYgYGBQwAAAAAgMWCyThoaWmpduzYoXnz5qm4uFhTp07VK6+8oszMTK1evTqxX0VFhdLT0xOv27Vrp4qKikbvp6WlNXp/+/bthz13LBZTSUnJUbya5vPvxL237CsF/NKuXTuTEBFwdLn1zyZwJBjfsBnjGzZjfMN2No3xpBQwOnTooKysLIVCIWVlZSkcDmvv3r3q1KlTo/3S0tJUWVmZeF1ZWdmooHGw99u3b3/YcweDQWVkZBylK2k+JSUl/1bc4TalCgUCrrxmeMu/O8YBN2B8w2aMb9iM8Q3buXWMl5WVHXR7UtYd9OvXT2vWrJExRrt27VJ1dbU6dOhwwH6nn3663n33XdXU1Ki8vFyffvqpevfunXg/LS1NKSkp+vzzz2WM0dq1a9W/f/9khOxaUcfQwBMAAAAAYL2kzMDIyclRUVGRRo8eLWOMCgoKFAgEDtjvuOOOU15ensaPHy9jjH72s58pHA5r/fr1evfddzV9+nTdfvvtmjlzphzHUXZ2ts4444xkhOxaNbE4DTwBAAAAANZLSgFDkvLz8w+6fcaMGY1ejxkzRmPGjGm0bdCgQRo0aJAkqW/fvnrhhReSE6QFok5cYWZgAAAAAAAsR+brcpFYXCkBX0uHAQAAAABAUlHAcLmowxISAAAAAID9yHxdLurEaeIJAAAAALAema/L0cQTAAAAAOAFZL4uxwwMAAAAAIAXkPm6XCQWV4gZGAAAAAAAy5H5ulzUMdyFBAAAAABgPQoYLscSEgAAAACAF5D5ulyEJp4AAAAAAA8g83W5CDMwAAAAAAAeQObrcjTxBAAAAAB4AZmvy0UdlpAAAAAAAOxH5utyUcewhAQAAAAAYD0yX5ejiScAAAAAwAvIfF3MGEMTTwAAAACAJ5D5ulgsbiRJoYCvhSMBAAAAACC5KGC4WCQWlySWkAAAAAAArEfm62JRp7aAwRISAAAAAIDtmsx8lyxZ0uj1r3/966QFg2+GGRgAAAAAAK8IHuqN5cuXa9WqVSosLNRbb70lSXIcR5988omuuOKKZgsQhxZhBgYAAAAAwCMOWcA4++yzddxxx6msrExjx46VJPn9fmVmZjZbcDi8qFPfxJMCBgAAAADAbocsYBxzzDEaOHCgBg4cqD179qimpkZS7SwMtA4sIQEAAAAAeMUhCxj1br/9dv3lL39Rly5dZIyRz+fT888/3xyxoQk08QQAAAAAeEWTBYwPPvhAr732mvx+kuTWpiYxA8PXwpEAAAAAAJBcTVYlTjjhhMTyEbQuzMAAAAAAAHhFkzMwSkpKlJOToxNOOEGSWELSiiQKGPTAAAAAAABYrskCxkMPPdQcceDfQBNPAAAAAIBXNFnA+N3vfnfAtunTpyclGHwzLCEBAAAAAHhFkwWMzp07S5KMMfroo48Uj8eTHhSOTA0zMAAAAAAAHtFkAWPcuHGNXl9zzTVJCwbfTNQxkqQwMzAAAAAAAJZrsoCxdevWxPPdu3erpKQkqQHhyNUvIWEGBgAAAADAdk0WMAoKChLPw+Gw8vPzkxoQjtz+Jp6+Fo4EAAAAAIDkarKAsXDhQpWWlmr79u3q1q2bOnbs2Bxx4QjQxBMAAAAA4BVNZr5/+tOfNG7cOM2bN09jx47VsmXLmiMuHAGaeAIAAAAAvKLJGRjPPvusXnrpJbVr104VFRW68sorNXLkyOaIDU1IzMCggAEAAAAAsFyTma/P51O7du0kSWlpaQqHw0kPCkcm6sQV9Pvk99MDAwAAAABgtyZnYHTv3l333nuv+vfvr3feeUfdu3dvjrhwBCKxOMtHAAAAAACe0GT2O3fuXGVmZurNN99UZmam7rrrruaIC0cg6hgaeAIAAAAAPKHJ7Le6ulrHH3+8+vfvr2OPPVZ//vOfmyMuHIEaZmAAAAAAADyiySUkkyZN0oknnqj09HRJtT0xLrzwwqQHhqZFnbjCzMAAAAAAAHhAkwWM9PR03XPPPc0RC76hqBNXSoAGngAAAAAA+zVZwMjOztbixYt14oknJrZ973vfS2pQODI08QQAAAAAeEWTBYx33nlHkUhERUVFkmqXkFDAaB2iTpwmngAAAAAAT2iygFFVVaVnn322GULBN0UTTwAAAACAVzRZwOjVq5eWL1+uU045RT5fbb+FHj16JD0wNI0ZGAAAAAAAr2iygLF582Zt3rxZPp9PpaWl2rZtmzZu3NgcsaEJUceoTUqgpcMAAAAAACDpmvz5fuHChZo1a5aOP/547d27V6NHj26OuHAEapt4chcSAAAAAID9DjkDIxKJ6OWXX9Zzzz2nlJQUVVRUaOXKlUpNTW3O+HAYLCEBAAAAAHjFIbPfoUOHasuWLXrggQf03HPPqUuXLhQvWhluowoAAAAA8IpDzsC44oortHz5cn3xxRcaPXq0jDHf6MDz58/XqlWrFI1GlZubqwEDBmj27Nny+Xzq1auX5syZo7Vr1+qXv/ylJMkYo3fffVfLly9Xz549E8d55plntHTpUnXs2FGSdPvttysrK+vfuVbrRJiBAQAAAADwiEMWMCZPnqzJkyfr7bff1pIlS/TXv/5VDzzwgEaOHKnevXsf9qCFhYXasGGDFi9erOrqai1YsED33HOPrr/+eg0cOFAFBQVauXKlzjvvPA0ePFiS9PTTT+u73/1uo+KFJG3atEn33XefTjvttKNwuXaJOnGFmIEBAAAAAPAAnznCqRVfffWVli1bphdffFG///3vD7vvQw89JJ/Pp08++UQVFRXKz8/Xj3/8Y61evVo+n0+vvfaa1q1bpzlz5kiSdu7cqWuvvVYvvviiQqFQo2NdcMEF6tWrl3bv3q0hQ4bouuuuO+y5N27cqM6dOx/JJbne+fM/0A96H6tZOd1bOhQAAAAAAI6KsrIynXzyyQdsb/I2qvXat2+vvLw85eXlNblvaWmpduzYoXnz5qm4uFhTp06VMUY+X+0dM9q1a6fy8vLE/s8884wmTpx4QPFCki666CKNHz9eaWlpmj59ul5//XXl5OQc8tzBYFAZGRlHelmtRklJyTeO2zEf6tj26a68XnjPvzPGAbdgfMNmjG/YjPEN27l1jJeVlR10e1LWH3To0EHZ2dkKhULKyspSOBxuVLCorKxU+/btJUnxeFxvvPGGLrroogOOY4zRlVdeqY4dOyoUCumcc87RRx99lIyQXYkmngAAAAAAr0hK9tuvXz+tWbNGxhjt2rVL1dXVGjRokAoLCyVJq1evVv/+/SVJf/vb39SjR4+D3uGkoqJCw4cPV2VlpYwxKiwspBdGHWMMTTwBAAAAAJ5xxEtIvomcnBwVFRUl7l5SUFCgbt266dZbb9XDDz+srKwsDRs2TJK0detWZWZmNvr8H//4R1VVVWns2LH62c9+piuuuEKhUEiDBg3SOeeck4yQXScWr21dEgr4WjgSAAAAAACSLykFDEnKz88/YNuiRYsO2HbBBRfoggsuaLRtxIgRieeXXHKJLrnkkqMfoMtFYnFJYgkJAAAAAMATyH5dKurUFjBYQgIAAAAA8AKyX5diBgYAAAAAwEvIfl0qwgwMAAAAAICHkP26VNSpb+LJVwgAAAAAsB/Zr0uxhAQAAAAA4CVkvy5FE08AAAAAgJeQ/bpUTWIGhq+FIwEAAAAAIPkoYLgUMzAAAAAAAF5C9utSiQIGPTAAAAAAAB5A9utSNPEEAAAAAHgJ2a9LsYQEAAAAAOAlZL8uVcMMDAAAAACAh5D9ulTUMZKkMDMwAAAAAAAeQPbrUvVLSJiBAQAAAADwArJfl9rfxNPXwpEAAAAAAJB8FDBciiaeAAAAAAAvIft1KZp4AgAAAAC8hOzXpRIzMChgAAAAAAA8gOzXpaJOXEG/T34/PTAAAAAAAPajgOFSkVic5SMAAAAAAM8gA3apqGNo4AkAAAAA8AwyYJeqYQYGAAAAAMBDyIBdKurEFWYGBgAAAADAI8iAXSrqxJUSoIEnAAAAAMAbKGC4FE08AQAAAABeQgbsUlEnThNPAAAAAIBnkAG7FE08AQAAAABeQgbsUszAAAAAAAB4CRmwS0UdoxAzMAAAAAAAHkEG7FK1TTy5CwkAAAAAwBsoYLgUS0gAAAAAAF5CBuxS3EYVAAAAAOAlZMAuFWEGBgAAAADAQ8iAXSrqxGniCQAAAADwDDJgl2IJCQAAAADAS8iAXSrqGJaQAAAAAAA8gwzYpZiBAQAAAADwEjJgFzLG0MQTAAAAAOApZMAuFIsbSVIo4GvhSAAAAAAAaB4UMFwoEotLEktIAAAAAACeQQbsQlGntoDBEhIAAAAAgFeQAbsQMzAAAAAAAF5DBuxCkfoZGBQwAAAAAAAeQQbsQlGnroknS0gAAAAAAB5BBuxCLCEBAAAAAHgNGbAL0cQTAAAAAOA1ZMAuVN8DIyXga+FIAAAAAABoHhQwXKh+CQlNPAEAAAAAXhFM1oHnz5+vVatWKRqNKjc3VwMGDNDs2bPl8/nUq1cvzZkzR36/X1OmTFFZWZlSUlIUDof19NNPNzrOqlWr9MQTTygYDGrUqFEaM2ZMskJ2DZaQAAAAAAC8JikFjMLCQm3YsEGLFy9WdXW1FixYoHvuuUfXX3+9Bg4cqIKCAq1cuVLnnXeePv/8c7388svy+Q5cDhGNRnXPPfdo6dKlatOmjXJzc5WTk6PjjjsuGWG7Bk08AQAAAABek5QMeO3aterdu7emTZumKVOmaMiQIdq0aZMGDBggSRo8eLDefPNN/etf/9JXX32lKVOmKDc3V6+//nqj43z66afq3r27jjnmGIVCIfXr10/vvPNOMkJ2FWZgAAAAAAC8JikzMEpLS7Vjxw7NmzdPxcXFmjp1qowxiVkW7dq1U3l5uaLRqCZNmqQrrrhCX375pXJzc3X66aerU6dOkqSKigqlp6cnjtuuXTtVVFQc9tyxWEwlJSXJuKykO9K4//mvvZKksr17VKLKZIYEHFVu/bMJHAnGN2zG+IbNGN+wnU1jPCkFjA4dOigrK0uhUEhZWVkKh8PauXNn4v3Kykq1b99enTt31rhx4xQMBtWpUyedfPLJ2rp1a6KAkZaWpsrKykafa1jQOOgFBYPKyMhIxmUlVUlJyRHH3XaHI0n61vFdldGpbTLDAo6abzLGAbdhfMNmjG/YjPEN27l1jJeVlR10e1LWIPTr109r1qyRMUa7du1SdXW1Bg0apMLCQknS6tWr1b9/f7355pu6/vrrJdUWJz755BNlZWUljtOzZ0999tlnKisrUyQS0TvvvKMzzzwzGSG7CktIAAAAAABek5QZGDk5OSoqKtLo0aNljFFBQYG6deumW2+9VQ8//LCysrI0bNgwBQIBrV27VmPGjJHf79cNN9ygjh076o9//KOqqqo0duxYzZ49W1dffbWMMRo1apS6du2ajJBdZX8TzwMbnwIAAAAAYKOk3UY1Pz//gG2LFi06YNvNN998wLYRI0Ykng8dOlRDhw49usG5HDMwAAAAAABeQwbsQhGH26gCAAAAALyFDNiF6peQhChgAAAAAAA8ggzYhaJOXEG/T34/PTAAAAAAAN5AAcOFIrE4y0cAAAAAAJ5CFuxCUcfQwBMAAAAA4ClkwS4UcZiBAQAAAADwFrJgF4rE4goF6H8BAAAAAPAOChguFHXiLCEBAAAAAHgKWbAL0cQTAAAAAOA1ZMEuxAwMAAAAAIDXkAW7UMQxzMAAAAAAAHgKWbALRWKOQhQwAAAAAAAeQhbsQlHHsIQEAAAAAOApZMEuVNvEk9uoAgAAAAC8gwKGC9HEEwAAAADgNWTBLhRxuI0qAAAAAMBbyIJdKBKL08QTAAAAAOApZMEuxBISAAAAAIDXkAW7UG0TT746AAAAAIB3kAW7ELdRBQAAAAB4DVmwC9HEEwAAAADgNWTBLmOMqWvi6WvpUAAAAAAAaDYUMFwmFjeSxBISAAAAAICnkAW7TCQWlySWkAAAAAAAPIUs2GWiTm0BgxkYAAAAAAAvIQt2mYjDDAwAAAAAgPeQBbtM/RKSEAUMAAAAAICHkAW7TNShiScAAAAAwHvIgl2GJp4AAAAAAC8iC3YZmngCAAAAALyILNhl9jfx9LVwJAAAAAAANB8KGC5DE08AAAAAgBeRBbsMS0gAAAAAAF5EFuwyNPEEAAAAAHgRWbDLMAMDAAAAAOBFZMEuE3GMJGZgAAAAAAC8hSzYZWjiCQAAAADwIrJgl2EJCQAAAADAi8iCXWZ/E09fC0cCAAAAAEDzoYDhMszAAAAAAAB4EVmwy0QcbqMKAAAAAPAesmCXoYknAAAAAMCLyIJdJurEFfT75PfTAwMAAAAA4B0UMFwmEouzfAQAAAAA4Dlkwi4TdQwNPAEAAAAAnkMm7DIRhxkYAAAAAADvIRN2mUgsrlCA/hcAAAAAAG+hgOEyUSfOEhIAAAAAgOeQCbsMTTwBAAAAAF4UTNaB58+fr1WrVikajSo3N1cDBgzQ7Nmz5fP51KtXL82ZM0d+v1/33Xef3nvvPcViMY0dO1ZjxoxpdJw///nPuv/++5WRkSFJmjFjhgYMGJCssFs9ZmAAAAAAALwoKQWMwsJCbdiwQYsXL1Z1dbUWLFige+65R9dff70GDhyogoICrVy5Uunp6fr888/129/+VpFIRBdddJGGDRumY445JnGsTZs2adasWRo2bFgyQnWdiGOYgQEAAAAA8JykZMJr165V7969NW3aNE2ZMkVDhgzRpk2bEjMnBg8erDfffFNnnnmm5s6dm/ic4zgKBhvXVDZt2qQXX3xR48eP17333qtYLJaMkF0jEnMUooABAAAAAPCYpMzAKC0t1Y4dOzRv3jwVFxdr6tSpMsbI56u9e0a7du1UXl6ucDiscDisaDSq2bNna+zYsWrXrl2jY5111lk699xz1a1bN82ZM0fPP/+8JkyYcMhzx2IxlZSUJOOyku5I4q6srlFq0O/aa4S3MW5hM8Y3bMb4hs0Y37CdTWM8KQWMDh06KCsrS6FQSFlZWQqHw9q5c2fi/crKSrVv316S9OWXX+onP/mJBgwYoOuuu+6AY40aNSqx7w9+8AO9+uqrhz13MBhM9Mtwk5KSkiOL2/+p0tqGXHmN8LYjHuOACzG+YTPGN2zG+Ibt3DrGy8rKDro9KWsR+vXrpzVr1sgYo127dqm6ulqDBg1SYWGhJGn16tXq37+/9u3bp4kTJ2rUqFGaNm3aAccxxujiiy9OFD/Wr1+vU089NRkhuwZNPAEAAAAAXpSUGRg5OTkqKirS6NGjZYxRQUGBunXrpltvvVUPP/ywsrKyNGzYMC1cuFDbt2/XkiVLtGTJEknS3LlzVVxcrHfffVfTp0/XXXfdpenTpys1NVU9e/Y84C4lXhNxuI0qAAAAAMB7knYb1fz8/AO2LVq0qNHriRMnauLEiQfsl5mZqUGDBkmSsrOzlZ2dnZQY3SgSi9PEEwAAAADgOWTCLsMSEgAAAACAF5EJu0wkxhISAAAAAID3kAm7TNQxzMAAAAAAAHgOmbDL0MQTAAAAAOBFZMIuYoypa+Lpa+lQAAAAAABoVhQwXCQWN5LEEhIAAAAAgOeQCbtIJBaXJJaQAAAAAAA8h0zYRaJObQGDGRgAAAAAAK8hE3aRiMMMDAAAAACAN5EJu0j9EpIQBQwAAAAAgMeQCbtI1KGJJwAAAADAm8iEXYQmngAAAAAAryITdhGaeAIAAAAAvIpM2EX2N/H0tXAkAAAAAAA0LwoYLkITTwAAAACAV5EJuwhLSAAAAAAAXkUm7CI08QQAAAAAeBWZsIswAwMAAAAA4FVkwi4ScYwkZmAAAAAAALyHTNhFaOIJAAAAAPAqMmEXYQkJAAAAAMCryIRdpL6AkRLwtXAkAAAAAAA0LwoY/7+9+4+psu7/OP664AhNURlbOhlaHoWN1vjKj5FuRLrWTLf+SZdaw5z9o6MpiQphKAwmEs5q6u5M52wUii7Xqm+1FUyRcvgjrcmcblZ4CwfDn3GOyPlxXfcf3XLf3lHdN4rnus55Pv4758LrvHFvHe8Xn8/ncpCBLSSswAAAAAAARBkmYQfxh3iMKgAAAAAgOjEJOwiHeAIAAAAAohWTsIMEQqZcMYZiYjgDAwAAAAAQXQgwHCQQstg+AgAAAACISkzDDuIPmhzgCQAAAACISkzDDuIPmazAAAAAAABEJaZhB/EHTcXFcv4FAAAAACD6EGA4SCDEFhIAAAAAQHRiGnaQAFtIAAAAAABRimnYQTjEEwAAAAAQrZiGHcTPY1QBAAAAAFGKadhB/MGQ4ggwAAAAAABRiGnYQQIhiy0kAAAAAICoxDTsIL8d4sljVAEAAAAA0YcAw0E4xBMAAAAAEK2Yhh3Ez2NUAQAAAABRimnYQfxBk0M8AQAAAABRiWnYQQIhtpAAAAAAAKIT07CDBEIWW0gAAAAAAFGJadhBOMQTAAAAABCtmIYdhEM8AQAAAADRimnYISzL+uchnka4SwEAAAAA4IEjwHCIoGlJEltIAAAAAABRiWnYIQIhU5LYQgIAAAAAiEpMww7hD/4WYLACAwAAAAAQjVzDdeMdO3aoublZgUBAixYtUm5urkpLS2UYhlJTU7VhwwbFxMRo27ZtOnTokFwul8rKypSRkXHXfZqbm7V9+3a5XC7NmzdPL7zwwnCVbGt+VmAAAAAAAKLYsEzDbW1tOnXqlPbu3av6+np1d3erpqZGRUVFamhokGVZampqUnt7u44dO6YDBw5oy5YtqqysvOs+gUBANTU12r17t+rr69XY2Kienp7hKNn2BlZgEGAAAAAAAKLQsKzAaG1tVVpamgoLC+X1erV27Vrt379fubm5kqT8/Hx98803mjx5svLy8mQYhpKTkxUKhXTt2jUlJSVJki5cuKBJkyZp7NixkqTs7GydOHFCc+bMGY6yw8Zzs09l//+jFPv3P/yaW/6QJLaQAAAAAACi07AEGNevX1dXV5feffddXbp0ScuXL5dlWTKM3x4BOmrUKPX29srr9SoxMXHgz915/06A4fV6NXr06Luue73eP/3sYDAoj8czDN/V8Om82a/rfUH1B/x/+nX/lzxKyfF+x31/wB30LiIZ/Y1IRn8jktHfiHSR1OPDEmAkJibK7XYrLi5Obrdb8fHx6u7uHrju8/k0ZswYJSQkyOfz3fX+vwcWf3V9MC6XSxMmTLiP383wmzBB+tvYeMfVDfwvPB4PPY6IRX8jktHfiGT0NyKdU3v8xo0bg74/LPsRsrOzdeTIEVmWpcuXL6uvr08zZsxQW1ubJKmlpUU5OTnKyspSa2urTNNUV1eXTNMcWH0hSVOmTFFHR4du3Lghv9+vEydOKDMzczhKBgAAAAAANjYsKzBmzZql48ePa/78+bIsS+vXr1dKSorKy8u1ZcsWud1uzZ49W7GxscrJydGCBQtkmqbWr18vSfr0009169YtLViwQKWlpXrllVdkWZbmzZun8ePHD0fJAAAAAADAxgzLsqxwF3E/nT17Vunp6eEu43/m1KU9wH+LHkcko78RyehvRDL6JJMQKQAACFNJREFUG5HOqT3+R3M9j7QAAAAAAAC2R4ABAAAAAABsjwADAAAAAADYHgEGAAAAAACwPQIMAAAAAABgewQYAAAAAADA9ggwAAAAAACA7RFgAAAAAAAA2yPAAAAAAAAAtkeAAQAAAAAAbI8AAwAAAAAA2B4BBgAAAAAAsD0CDAAAAAAAYHsEGAAAAAAAwPYMy7KscBdxP50+fVrx8fHhLgMAAAAAAAxBf3+/pk2b9rv3Iy7AAAAAAAAAkYctJAAAAAAAwPYIMAAAAAAAgO0RYAAAAAAAANsjwAAAAAAAALZHgAEAAAAAAGzPFe4Cop1pmqqoqNC5c+cUFxen6upqPfLII+EuCxiyQCCgsrIydXZ2yu/3a/ny5Zo6dapKS0tlGIZSU1O1YcMGxcSQn8K5rl69queff167d++Wy+WivxFRduzYoebmZgUCAS1atEi5ubn0OCJCIBBQaWmpOjs7FRMTo6qqKv4PR0T4/vvvtXnzZtXX16ujo2PQnt62bZsOHTokl8ulsrIyZWRkhLvsIeFfZ5h9/fXX8vv9amxsVHFxsTZt2hTukoB78sknnygxMVENDQ3auXOnqqqqVFNTo6KiIjU0NMiyLDU1NYW7TGDIAoGA1q9fr4ceekiS6G9ElLa2Np06dUp79+5VfX29uru76XFEjMOHDysYDGrfvn0qLCzU22+/TX/D8Xbu3Kk33nhD/f39kgb/uaS9vV3Hjh3TgQMHtGXLFlVWVoa56qEjwAizkydP6sknn5QkTZs2TWfOnAlzRcC9efbZZ7Vy5cqB17GxsWpvb1dubq4kKT8/X99++224ygPuWW1trRYuXKhx48ZJEv2NiNLa2qq0tDQVFhZq2bJlmjlzJj2OiDF58mSFQiGZpimv1yuXy0V/w/EmTZqkrVu3DrwerKdPnjypvLw8GYah5ORkhUIhXbt2LVwl3xMCjDDzer1KSEgYeB0bG6tgMBjGioB7M2rUKCUkJMjr9WrFihUqKiqSZVkyDGPgem9vb5irBIbm4MGDSkpKGgieJdHfiCjXr1/XmTNn9M4776iyslKrV6+mxxExRo4cqc7OTs2ZM0fl5eUqKCigv+F4s2fPlsv1r5MhBuvp/5w5ndzrnIERZgkJCfL5fAOvTdO8qwEBJ/J4PCosLNSLL76o5557TnV1dQPXfD6fxowZE8bqgKH76KOPZBiGjh49qrNnz6qkpOSu32DQ33C6xMREud1uxcXFye12Kz4+Xt3d3QPX6XE42Z49e5SXl6fi4mJ5PB69/PLLCgQCA9fpb0SCfz/D5U5P/+fM6fP5NHr06HCUd89YgRFmWVlZamlpkSSdPn1aaWlpYa4IuDdXrlzR0qVLtWbNGs2fP1+S9Nhjj6mtrU2S1NLSopycnHCWCAzZhx9+qA8++ED19fVKT09XbW2t8vPz6W9EjOzsbB05ckSWZeny5cvq6+vTjBkz6HFEhDFjxgwMbWPHjlUwGORnFEScwXo6KytLra2tMk1TXV1dMk1TSUlJYa50aAzLsqxwFxHN7jyF5Pz587IsSxs3btSUKVPCXRYwZNXV1friiy/kdrsH3lu3bp2qq6sVCATkdrtVXV2t2NjYMFYJ3LuCggJVVFQoJiZG5eXl9Dcixptvvqm2tjZZlqXXXntNKSkp9Dgigs/nU1lZmXp6ehQIBLR48WI9/vjj9Dcc79KlS1q1apX279+vn376adCe3rp1q1paWmSapl5//XXHhnUEGAAAAAAAwPbYQgIAAAAAAGyPAAMAAAAAANgeAQYAAAAAALA9AgwAAAAAAGB7BBgAAAAAAMD2XOEuAAAARJ5Nmzapvb1dPT09un37tiZOnCiXy6Xs7Gy9+uqr9/WzvvrqK2VkZGj8+PH39b4AAMBeeIwqAAAYNgcPHtSPP/6o1atXD9tnFBQUqKKiQlOmTBm2zwAAAOHHCgwAAPBAtLW1ad++fXrrrbf0zDPPKDMzUx0dHZo+fbp6e3v1ww8/aPLkyaqrq5PH41F5ebn6+/sVHx+vqqoqJSUlaeXKlfJ6vbp9+7bWrFmjvr4+nT17ViUlJWpoaFBjY6M+++wzGYahuXPnavHixSotLZVlWfJ4PLp165Zqa2uVkpLyu3s98cQT4f4rAgAAf4IAAwAAPHCdnZ16//339fDDDys3N1cHDhxQeXm5nn76af3666+qra1VQUGBnnrqKR09elSbN2/WsmXLdOXKFe3Zs0dXr17Vzz//rJkzZyo9PV0VFRW6ePGiPv/8czU0NMgwDC1ZskR5eXmSpIkTJ6q2tlaHDx9WXV2diouLf3cvAABgbwQYAADggUtMTFRycrIkaeTIkZo6daokafTo0erv79f58+e1Y8cO7dq1S5ZlacSIEUpNTdVLL72kVatWKRgMqqCg4K57nj9/Xl1dXVqyZIkk6ebNm7p48aIkafr06ZKkzMxMbdy48S/vBQAA7IcAAwAAPHCGYfzpdbfbraVLlyorK0sXLlzQ8ePHde7cOfl8Pr333nv65ZdftHDhQs2aNUuGYciyLLndbk2dOlW7du2SYRjas2eP0tLS9OWXX6q9vV05OTn67rvvlJqa+of3AgAA9kWAAQAAbKekpEQVFRXq7+/X7du3tW7dOj366KPavn27Pv74Y40YMUIrVqyQ9NuqirVr12r37t2aMWOGFi1aJL/ff9eTSVpaWtTU1CTTNFVTU6Nx48YNei8AAGBfPIUEAABEtNLSUs2dO1f5+fnhLgUAANyDmHAXAAAAAAAA8FdYgQEAAAAAAGyPFRgAAAAAAMD2CDAAAAAAAIDtEWAAAAAAAADbI8AAAAAAAAC2R4ABAAAAAABsjwADAAAAAADY3j8Aps49v0GDJT0AAAAASUVORK5CYII=\n",
+      "text/plain": [
+       "<Figure size 1080x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1080x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "experiments = experiments.sort_values(by =['rules_price']).reset_index(drop=True)\n",
+    "# df = experiments\n",
+    "cols = 1\n",
+    "rows = 1\n",
+    "cc_idx = 0\n",
+    "\n",
+    "# len(experiments)\n",
+    "# df = experiments.copy()\n",
+    "# print(df.head())\n",
+    "\n",
+    "while cc_idx<len(experiments):\n",
+    "    cc = experiments.iloc[cc_idx]['rules_price']\n",
+    "    cc_label = experiments.iloc[cc_idx]['rules_price']\n",
+    "#     secondary_label = experiments.iloc[cc_idx]['string']\n",
+    "    sub_experiments = experiments[experiments['rules_price']==cc]\n",
+    "    cc_idx += len(sub_experiments)\n",
+    "    fig, axs = plt.subplots(ncols=cols, nrows=rows, figsize=(15*cols,7*rows))\n",
+    "    sub_experiments.reset_index(drop=False)\n",
+    "    for i, experiment in sub_experiments.iterrows():\n",
+    "        df = experiment['dataset'].copy()\n",
+    "        # filter out starting point\n",
+    "        df = df[df.timestep > 0] \n",
+    "        # FIRST RUN ONLY\n",
+    "        df = df[df.run == 1]  \n",
+    "\n",
+    "        df = df.groupby('timestep').agg({'supply': ['min', 'mean', 'max']}).reset_index()\n",
+    "\n",
+    "#         print(df.head())\n",
+    "        plot_label = experiment['rules_price']\n",
+    "        ax = axs\n",
+    "        title = 'supply' + '\\n' + 'Scenario: ' + str(cc_label)  + ' rules_price'\n",
+    "        ax.set_title(title)\n",
+    "        ax.set_ylabel('Amount')\n",
+    "        colors = ['b','orange', 'g', 'magenta', 'r']\n",
+    "        \n",
+    "#         ax.plot(df.timestep, df['price'],color = colors[0], label='Price')\n",
+    "        df.plot(x='timestep', y=('supply','mean'), label='Mean', ax=ax, legend=True)\n",
+    "#         ax.fill_between(df.timestep, df[('supply','min')], df[('supply','max')], supply=0.3) \n",
+    "\n",
+    "\n",
+    "        ax.legend()\n",
+    "\n",
+    "        ax.set_xlabel('Timesteps')\n",
+    "        ax.grid(color='0.9', linestyle='-', linewidth=1)\n",
+    "\n",
+    "        plt.tight_layout()\n",
+    "        \n",
+    "fig.tight_layout(rect=[0, 0, 1, .97])\n",
+    "fig.patch.set_alpha(1)\n",
+    "display(fig)\n",
+    "plt.close()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### INTO AGENT LEVEL across AGENTS"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### This plot is across agents of the same variable (this case is agent_attestations_1)\n",
+    "### Can be repeated for all other columns in agents dataframe"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 32,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1080x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1080x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "experiments = experiments.sort_values(by =['rules_price']).reset_index(drop=True)\n",
+    "\n",
+    "cols = 1\n",
+    "rows = 1\n",
+    "cc_idx = 0\n",
+    "# config_labels = ['RULE 1,'RULE 2']\n",
+    "\n",
+    "\n",
+    "while cc_idx<len(experiments):\n",
+    "    cc = experiments.iloc[cc_idx]['rules_price']\n",
+    "    cc_label = experiments.iloc[cc_idx]['rules_price']\n",
+    "#     secondary_label = experiments.iloc[cc_idx]['string']\n",
+    "    sub_experiments = experiments[experiments['rules_price']==cc]\n",
+    "    cc_idx += len(sub_experiments)\n",
+    "    fig, axs = plt.subplots(ncols=cols, nrows=rows, figsize=(15*cols,7*rows))\n",
+    "    sub_experiments.reset_index(drop=False)\n",
+    "    for i, experiment in sub_experiments.iterrows():\n",
+    "        df = experiment['dataset'].copy()\n",
+    "        # filter out starting point\n",
+    "        df = df[df.timestep > 0] \n",
+    "        # FIRST RUN ONLY\n",
+    "        df = df[df.run == 1]  \n",
+    "        \n",
+    "\n",
+    "# MAKE A FOR LOOP or FUNCTION FOR ALL AGENTS\n",
+    "        df['agent_1_attest_1'] = df.agents.apply(lambda x: np.array(x['agent_attestations_1'][0]))\n",
+    "        df['agent_2_attest_1'] = df.agents.apply(lambda x: np.array(x['agent_attestations_1'][1]))\n",
+    "        df['agent_3_attest_1'] = df.agents.apply(lambda x: np.array(x['agent_attestations_1'][2]))\n",
+    "        df['agent_4_attest_1'] = df.agents.apply(lambda x: np.array(x['agent_attestations_1'][3]))\n",
+    "        df['agent_5_attest_1'] = df.agents.apply(lambda x: np.array(x['agent_attestations_1'][4]))\n",
+    "    \n",
+    "#         df = df.groupby('timestep').agg({'supply': ['min', 'mean', 'max']}).reset_index()\n",
+    "    \n",
+    "#         df['agent_attest_1'] = np.array(df.agent_attest_1,dtype = float)\n",
+    "#         print(df['agent_attest_1'])\n",
+    "#         print(df['agent_attest_1'][10])\n",
+    "        plot_label = experiment['rules_price']\n",
+    "        ax = axs\n",
+    "        title = 'agent_attest_1' + '\\n' + 'Scenario: ' + str(cc_label)  + ' rules_price'\n",
+    "        ax.set_title(title)\n",
+    "        ax.set_ylabel('Amount')\n",
+    "        colors = ['b','orange', 'g', 'magenta', 'r']\n",
+    "        \n",
+    "#         ax.plot(df.timestep, df['price'],color = colors[0], label='Price')\n",
+    "        df.plot(x='timestep', y='agent_1_attest_1', label='agent_1', ax=ax, legend=True)\n",
+    "        df.plot(x='timestep', y='agent_2_attest_1', label='agent_2', ax=ax, legend=True)\n",
+    "        df.plot(x='timestep', y='agent_3_attest_1', label='agent_3', ax=ax, legend=True)\n",
+    "        df.plot(x='timestep', y='agent_4_attest_1', label='agent_4', ax=ax, legend=True)\n",
+    "        df.plot(x='timestep', y='agent_5_attest_1', label='agent_5', ax=ax, legend=True)\n",
+    "\n",
+    "#         ax.fill_between(df.timestep, df[('supply','min')], df[('supply','max')], supply=0.3) \n",
+    "\n",
+    "        ax.legend()\n",
+    "\n",
+    "        ax.set_xlabel('Timesteps')\n",
+    "        ax.grid(color='0.9', linestyle='-', linewidth=1)\n",
+    "\n",
+    "        plt.tight_layout()\n",
+    "        \n",
+    "fig.tight_layout(rect=[0, 0, 1, .97])\n",
+    "fig.patch.set_alpha(1)\n",
+    "display(fig)\n",
+    "plt.close()\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### INTO AGENT LEVEL across columns in AGENTS dataframe\n",
+    "\n",
+    "### THIS CASE FIRST AGENT\n",
+    "## CAN ADD MORE VARS and REPEAT FOR OTHER AGENTS"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 37,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1080x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1080x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "experiments = experiments.sort_values(by =['rules_price']).reset_index(drop=True)\n",
+    "\n",
+    "cols = 1\n",
+    "rows = 1\n",
+    "cc_idx = 0\n",
+    "# config_labels = ['RULE 1,'RULE 2']\n",
+    "\n",
+    "\n",
+    "while cc_idx<len(experiments):\n",
+    "    cc = experiments.iloc[cc_idx]['rules_price']\n",
+    "    cc_label = experiments.iloc[cc_idx]['rules_price']\n",
+    "#     secondary_label = experiments.iloc[cc_idx]['string']\n",
+    "    sub_experiments = experiments[experiments['rules_price']==cc]\n",
+    "    cc_idx += len(sub_experiments)\n",
+    "    fig, axs = plt.subplots(ncols=cols, nrows=rows, figsize=(15*cols,7*rows))\n",
+    "    sub_experiments.reset_index(drop=False)\n",
+    "    for i, experiment in sub_experiments.iterrows():\n",
+    "        df = experiment['dataset'].copy()\n",
+    "        # filter out starting point\n",
+    "        df = df[df.timestep > 0] \n",
+    "        # FIRST RUN ONLY\n",
+    "        df = df[df.run == 1]  \n",
+    "\n",
+    "# MAKE A FOR LOOP or FUNCTION FOR ALL AGENTS\n",
+    "        df['agent_1_reserve'] = df.agents.apply(lambda x: np.array(x['agent_reserve'][0]))\n",
+    "        df['agent_1_supply_1'] = df.agents.apply(lambda x: np.array(x['agent_supply_1'][0]))\n",
+    "        df['agent_1_supply_0'] = df.agents.apply(lambda x: np.array(x['agent_supply_0'][0]))\n",
+    "        df['agent_1_supply_free'] = df.agents.apply(lambda x: np.array(x['agent_supply_free'][0]))\n",
+    "        df['agent_1_attest_1'] = df.agents.apply(lambda x: np.array(x['agent_attestations_1'][0]))\n",
+    "        df['agent_1_attest_0'] = df.agents.apply(lambda x: np.array(x['agent_attestations_0'][0]))\n",
+    "\n",
+    "    \n",
+    "#         df = df.groupby('timestep').agg({'supply': ['min', 'mean', 'max']}).reset_index()\n",
+    "    \n",
+    "#         df['agent_attest_1'] = np.array(df.agent_attest_1,dtype = float)\n",
+    "#         print(df['agent_attest_1'])\n",
+    "#         print(df['agent_attest_1'][10])\n",
+    "        plot_label = experiment['rules_price']\n",
+    "        ax = axs\n",
+    "        title = 'agent_attest_1' + '\\n' + 'Scenario: ' + str(cc_label)  + ' rules_price'\n",
+    "        ax.set_title(title)\n",
+    "        ax.set_ylabel('Amount')\n",
+    "        colors = ['b','orange', 'g', 'magenta', 'r']\n",
+    "        \n",
+    "#         ax.plot(df.timestep, df['price'],color = colors[0], label='Price')\n",
+    "        df.plot(x='timestep', y='agent_1_reserve', label='agent_reserve', ax=ax, legend=True)\n",
+    "        df.plot(x='timestep', y='agent_1_supply_1', label='agent_supply_1', ax=ax, legend=True)\n",
+    "        df.plot(x='timestep', y='agent_1_supply_0', label='agent_supply_0', ax=ax, legend=True)\n",
+    "        df.plot(x='timestep', y='agent_1_supply_free', label='agent_supply_free', ax=ax, legend=True)\n",
+    "        df.plot(x='timestep', y='agent_1_attest_1', label='agent_attestations_1', ax=ax, legend=True)\n",
+    "        df.plot(x='timestep', y='agent_1_attest_0', label='agent_attestations_0', ax=ax, legend=True)\n",
+    "\n",
+    "#         ax.fill_between(df.timestep, df[('supply','min')], df[('supply','max')], supply=0.3) \n",
+    "\n",
+    "        ax.legend()\n",
+    "\n",
+    "        ax.set_xlabel('Timesteps')\n",
+    "        ax.grid(color='0.9', linestyle='-', linewidth=1)\n",
+    "\n",
+    "        plt.tight_layout()\n",
+    "        \n",
+    "fig.tight_layout(rect=[0, 0, 1, .97])\n",
+    "fig.patch.set_alpha(1)\n",
+    "display(fig)\n",
+    "plt.close()\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.6.5"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}
diff --git a/run2.py b/run2.py
index b9ca822..9f6fdc5 100644
--- a/run2.py
+++ b/run2.py
@@ -9,6 +9,7 @@
 pd.set_option('display.width', None)
 pd.set_option('display.max_colwidth', None)
 
+from src.sim import config 
 
 def run(drop_midsteps=True):
     print()
diff --git a/src/sim/config.py b/src/sim/config.py
index 9fd9fb8..d8012aa 100644
--- a/src/sim/config.py
+++ b/src/sim/config.py
@@ -11,8 +11,8 @@
 import pandas as pd
 import itertools
 
-time_periods_per_run = 10
-monte_carlo_runs = 1
+time_periods_per_run = 100
+monte_carlo_runs = 3
 E = 0.45  # to be reviewed
 
 KAPPA = [2]
diff --git a/src/sim/model/parts/private_beliefs.py b/src/sim/model/parts/private_beliefs.py
index 146e8fa..d701d05 100644
--- a/src/sim/model/parts/private_beliefs.py
+++ b/src/sim/model/parts/private_beliefs.py
@@ -25,7 +25,7 @@ def update_private_price(params, substep, state_history, prev_state, policy_inpu
         # 'period': ['N/A', 2000, 2000, 2000]
         'dP': P0[0]/4,
         'period': 2000,
-        'sigma': [.005],  # , 'N/A', 'N/A', 'N/A']
+        'sigma': .005,  # , 'N/A', 'N/A', 'N/A']
     }
 
     #params = params[0]
diff --git a/src/sim/run.py b/src/sim/run.py
index 5d621cf..2912d92 100644
--- a/src/sim/run.py
+++ b/src/sim/run.py
@@ -1,29 +1,101 @@
 
+# # The following imports NEED to be in the exact order
+# from cadCAD.engine import ExecutionMode, ExecutionContext, Executor
+# from cadCAD import configs
+# import pandas as pd
+
+
+# def run(drop_midsteps=True):
+#     exec_mode = ExecutionMode()
+#     multi_proc_ctx = ExecutionContext(context=exec_mode.multi_proc)
+#     run = Executor(exec_context=multi_proc_ctx, configs=configs)
+#     results = pd.DataFrame()
+#     i = 0
+#     for raw_result, _ in run.execute():
+#         params = configs[i].sim_config['M']
+#         result_record = pd.DataFrame.from_records(
+#             [tuple([i for i in params.values()])], columns=list(params.keys()))
+
+#         df = pd.DataFrame(raw_result)
+#         # keep only last substep of each timestep
+#         if drop_midsteps:
+#             max_substep = max(df.substep)
+#             is_droppable = (df.substep != max_substep) & (df.substep != 0)
+#             df.drop(df[is_droppable].index, inplace=True)
+
+#         result_record['dataset'] = [df]
+#         results = results.append(result_record)
+#         i += 1
+#     return results.reset_index()
+
+
+
 # The following imports NEED to be in the exact order
-from cadCAD.engine import ExecutionMode, ExecutionContext, Executor
 from cadCAD import configs
-import pandas as pd
+######### ADD FOR PRINTING CONFIG
+from cadCAD.configuration.utils import *
+from cadCAD.engine import ExecutionMode, ExecutionContext, Executor
+
+from src.sim import config
 
+exec_mode = ExecutionMode()
+exec_ctx = ExecutionContext(context=exec_mode.multi_proc)
+simulation = Executor(exec_context=exec_ctx, configs=configs)
+raw_system_events, tensor_field, session = simulation.execute()
+df = pd.DataFrame(raw_system_events)
+
+def get_M(k, v):
+    if k == 'sim_config':
+        k, v = 'M', v['M']
+    return k, v
+config_ids = [
+    dict(
+        get_M(k, v) for k, v in config.__dict__.items() if k in ['simulation_id', 'run_id', 'sim_config']
+    ) for config in configs
+]
 
 def run(drop_midsteps=True):
-    exec_mode = ExecutionMode()
-    multi_proc_ctx = ExecutionContext(context=exec_mode.multi_proc)
-    run = Executor(exec_context=multi_proc_ctx, configs=configs)
+    print('config_ids = ', config_ids)
+    result_records_list, sim_id_records = [], []
     results = pd.DataFrame()
-    i = 0
-    for raw_result, _ in run.execute():
-        params = configs[i].sim_config['M']
-        result_record = pd.DataFrame.from_records(
-            [tuple([i for i in params.values()])], columns=list(params.keys()))
+    sim_ids = list(set([_id['simulation_id'] for _id in config_ids]))
+
+    # print(sim_ids)
+    sim_dfs = {_id: [] for _id in sim_ids}
+    for i, config_id in enumerate(config_ids):
+        sim_id, run_id = config_id['simulation_id'], config_id['run_id']
+        params = config_id['M']
+        result_record = pd.DataFrame.from_records([tuple([i for i in params.values()])], columns=list(params.keys()))
 
-        df = pd.DataFrame(raw_result)
+        mod_record = {'sim_id': sim_id, 'meta': result_record}
+        if sim_id not in sim_id_records:
+            sim_id_records.append(sim_id)
+            result_records_list.append(mod_record)
+
+        sim_id = config_id['simulation_id']
+        # print('sim id first loop = ',sim_id)
+
+        sub_df = df[df.simulation == config_id['simulation_id']][df.run == config_id['run_id'] + 1]
+        sim_dfs[sim_id].append(sub_df)
+        # print(sub_df[['simulation', 'run', 'substep', 'timestep']].tail(5))
+        # print(sub_df.tail(5))
+    
+    for sim_id in sim_ids:
+        result_record = [d for d in result_records_list if d['sim_id'] == sim_id][0]['meta']
+        sim_dfs[sim_id] = pd.concat(sim_dfs[sim_id])
+        sub_df = sim_dfs[sim_id]
+
+        # print('sim id second loop = ',sim_id)
         # keep only last substep of each timestep
         if drop_midsteps:
-            max_substep = max(df.substep)
-            is_droppable = (df.substep != max_substep) & (df.substep != 0)
-            df.drop(df[is_droppable].index, inplace=True)
+            max_substep = max(sub_df.substep)
+            is_droppable = (sub_df.substep != max_substep) & (sub_df.substep != 0)
+            sub_df.drop(sub_df[is_droppable].index, inplace=True)
 
-        result_record['dataset'] = [df]
+        # print(sub_df.head(3))
+        # print(sub_df.tail(3))
+        result_record['dataset'] = [sub_df]
         results = results.append(result_record)
-        i += 1
-    return results.reset_index()
+        # print(sub_df[['simulation', 'run', 'substep', 'timestep']].tail(5))
+
+    return results.reset_index()
\ No newline at end of file
diff --git a/tests/mutliscale_ABM.ipynb b/tests/mutliscale_ABM.ipynb
index 5acc53b..7bbce2a 100644
--- a/tests/mutliscale_ABM.ipynb
+++ b/tests/mutliscale_ABM.ipynb
@@ -6,31 +6,13 @@
    "metadata": {
     "scrolled": false
    },
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\n",
-      "<class 'list'>\n",
-      "[1000]\n",
-      "\n",
-      "Initial Conditions (config.py) :  {'reserve': 300, 'price': 1, 'realized_price': 0, 'spot_price': 1, 'private_price': 0, 'private_alpha': 0, 'kappa': 0, 'supply': 600.0, 'alpha': [0.5], 'supply_0': 200, 'supply_1': 200, 'supply_free': 600.0, 'attestations': 40, 'attestations_0': 20, 'attestations_1': 20, 'invariant_V': 1200.0, 'invariant_I': 650.0, 'agent_attestations_1': 0, 'agent_attestations_0': 0, 'agent_reserve': 50, 'agent_supply': 50, 'agent_supply_1': 10, 'agent_supply_0': 10, 'agent_supply_free': 30, 'agents':    agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                     0                     0              0             0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0               0               0                  0  , 'chosen_agent': 0}\n",
-      "Params (config.py) :  {'starting_kappa': 2, 'starting_alpha': 0.5, 'money_raised': 1000, 'C': 700, 'f': 0.03, 'm': 0.15, 'period': 2000, 'rules_price': 'martin'}\n",
-      "Params (config.py) :  {'starting_kappa': 2, 'starting_alpha': 0.5, 'money_raised': 1000, 'C': 700, 'f': 0.03, 'm': 0.15, 'period': 2000, 'rules_price': 'step'}\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "import sys\n",
     "sys.path.append('../')\n",
     "\n",
     "import matplotlib.pyplot as plt\n",
-    "import run2"
+    "import run2\n"
    ]
   },
   {
@@ -54,373 +36,117 @@
       "by cadCAD\n",
       "\n",
       "Execution Mode: multi_proc\n",
-      "Configuration Count: 2\n",
-      "Dimensions of the first simulation: (Timesteps, Params, Runs, Vars) = (400, 8, 1, 26)\n",
+      "Configuration Count: 4\n",
+      "Dimensions of the first simulation: (Timesteps, Params, Runs, Vars) = (1000, 7, 1, 3)\n",
       "Execution Method: parallelize_simulations\n",
       "Execution Mode: parallelized\n",
-      "Total execution time: 16.77s\n",
+      "Total execution time: 2.11s\n",
       "<class 'pandas.core.frame.DataFrame'>\n",
-      "[                                                                                                  act  \\\n",
-      "0  <bound method Identity.p_identity of <cadCAD.configuration.Identity object at 0x000001A1EE26A6D8>>   \n",
-      "1                                                         <function set_action at 0x000001A1EE3A3BF8>   \n",
-      "2                                                         <function set_action at 0x000001A1EE3A3BF8>   \n",
-      "3                                                         <function set_action at 0x000001A1EE3A3BF8>   \n",
-      "4  <bound method Identity.p_identity of <cadCAD.configuration.Identity object at 0x000001A1EE26A6D8>>   \n",
-      "\n",
-      "                                                                 chosen_agent  \\\n",
-      "0                               <function choose_agent at 0x000001A1EE3A3C80>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE3AA2F0>   \n",
-      "2  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE3AA2F0>   \n",
-      "3  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE3AA2F0>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE3AA2F0>   \n",
-      "\n",
-      "                                                                private_price  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4147B8>   \n",
-      "1                       <function update_private_price at 0x000001A1EE399730>   \n",
-      "2  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4147B8>   \n",
-      "3  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4147B8>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4147B8>   \n",
-      "\n",
-      "                                                                private_alpha  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414840>   \n",
-      "1                       <function update_private_alpha at 0x000001A1EE399A60>   \n",
-      "2  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414840>   \n",
-      "3  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414840>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414840>   \n",
-      "\n",
-      "                                                                      reserve  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4148C8>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4148C8>   \n",
-      "2                                   <function update_R at 0x000001A1EE399D08>   \n",
-      "3  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4148C8>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4148C8>   \n",
+      "[                                                                                       marketing_rate  \\\n",
+      "0                                                   <function p_marketing_rate at 0x000001A1EA142510>   \n",
+      "1  <bound method Identity.p_identity of <cadCAD.configuration.Identity object at 0x000001A1EA006748>>   \n",
       "\n",
-      "                                                                       supply  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414950>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414950>   \n",
-      "2                                   <function update_S at 0x000001A1EE399D90>   \n",
-      "3  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414950>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414950>   \n",
+      "                                                                                    p_marketing_shock  \\\n",
+      "0                                                  <function p_marketing_shock at 0x000001A1EA142620>   \n",
+      "1  <bound method Identity.p_identity of <cadCAD.configuration.Identity object at 0x000001A1EA006748>>   \n",
       "\n",
-      "                                                                agent_reserve  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4149D8>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4149D8>   \n",
-      "2                                   <function update_r at 0x000001A1EE399E18>   \n",
-      "3  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4149D8>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4149D8>   \n",
+      "                                                                                           reputation  \\\n",
+      "0  <bound method Identity.p_identity of <cadCAD.configuration.Identity object at 0x000001A1EA006748>>   \n",
+      "1                                                       <function p_reputation at 0x000001A1EA142730>   \n",
       "\n",
-      "                                                                 agent_supply  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414A60>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414A60>   \n",
-      "2                     <function update_s_free_bondburn at 0x000001A1EE399EA0>   \n",
-      "3  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414A60>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414A60>   \n",
+      "                                                                                           experience  \\\n",
+      "0  <bound method Identity.p_identity of <cadCAD.configuration.Identity object at 0x000001A1EA006748>>   \n",
+      "1                                                       <function p_experience at 0x000001A1EA1427B8>   \n",
       "\n",
-      "                                                                   spot_price  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414AE8>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414AE8>   \n",
-      "2                          <function update_P_bondburn at 0x000001A1EE399F28>   \n",
-      "3                            <function update_P_attest at 0x000001A1EE3A3AE8>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414AE8>   \n",
+      "                                                                       signal  \\\n",
+      "0                                   <function s_signal at 0x000001A1EA1426A8>   \n",
+      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EBE042F0>   \n",
       "\n",
-      "                                                                        price  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414B70>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414B70>   \n",
-      "2                                <function update_pbar at 0x000001A1EE3A3048>   \n",
-      "3  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414B70>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414B70>   \n",
-      "\n",
-      "                                                                  invariant_I  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414BF8>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414BF8>   \n",
-      "2                          <function update_I_bondburn at 0x000001A1EE3A30D0>   \n",
-      "3                            <function update_I_attest at 0x000001A1EE3A3A60>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414BF8>   \n",
-      "\n",
-      "                                                                 attestations  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414C80>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414C80>   \n",
-      "2  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414C80>   \n",
-      "3                                   <function update_Q at 0x000001A1EE3A3400>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414C80>   \n",
-      "\n",
-      "                                                               attestations_1  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414D08>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414D08>   \n",
-      "2  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414D08>   \n",
-      "3                                  <function update_Q1 at 0x000001A1EE3A3488>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414D08>   \n",
-      "\n",
-      "                                                               attestations_0  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414D90>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414D90>   \n",
-      "2  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414D90>   \n",
-      "3                                  <function update_Q0 at 0x000001A1EE3A3510>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414D90>   \n",
-      "\n",
-      "                                                                     supply_1  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414E18>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414E18>   \n",
-      "2  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414E18>   \n",
-      "3                                  <function update_S1 at 0x000001A1EE3A3598>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414E18>   \n",
-      "\n",
-      "                                                                     supply_0  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414EA0>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414EA0>   \n",
-      "2  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414EA0>   \n",
-      "3                                  <function update_S0 at 0x000001A1EE3A3620>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414EA0>   \n",
-      "\n",
-      "                                                         agent_attestations_1  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414F28>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414F28>   \n",
-      "2  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414F28>   \n",
-      "3                                  <function update_q1 at 0x000001A1EE3A36A8>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE414F28>   \n",
-      "\n",
-      "                                                         agent_attestations_0  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432048>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432048>   \n",
-      "2  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432048>   \n",
-      "3                                  <function update_q0 at 0x000001A1EE3A3730>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432048>   \n",
-      "\n",
-      "                                                            agent_supply_free  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4320D0>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4320D0>   \n",
-      "2  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4320D0>   \n",
-      "3                              <function update_s_free at 0x000001A1EE3A37B8>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4320D0>   \n",
-      "\n",
-      "                                                               agent_supply_1  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432158>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432158>   \n",
-      "2  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432158>   \n",
-      "3                                  <function update_s1 at 0x000001A1EE3A3840>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432158>   \n",
-      "\n",
-      "                                                               agent_supply_0  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4321E0>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4321E0>   \n",
-      "2  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4321E0>   \n",
-      "3                                  <function update_s0 at 0x000001A1EE3A38C8>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4321E0>   \n",
-      "\n",
-      "                                                                        alpha  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432268>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432268>   \n",
-      "2  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432268>   \n",
-      "3                               <function update_alpha at 0x000001A1EE3A3950>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432268>   \n",
-      "\n",
-      "                                                                        kappa  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4322F0>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4322F0>   \n",
-      "2  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4322F0>   \n",
-      "3                               <function update_kappa at 0x000001A1EE3A39D8>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4322F0>   \n",
-      "\n",
-      "                                                                  invariant_V  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432378>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432378>   \n",
-      "2  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432378>   \n",
-      "3                                   <function update_V at 0x000001A1EE3A3B70>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432378>   \n",
-      "\n",
-      "                                                                                                   agents  \\\n",
-      "0                              <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432400>   \n",
-      "1                              <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432400>   \n",
-      "2                              <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432400>   \n",
-      "3                              <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432400>   \n",
-      "4  <function putting_the_updated_agent_back_to_its_god_given_home_in_the_dataframe at 0x000001A1EE3A3D90>   \n",
+      "                                                                         pool  \\\n",
+      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EBE2D6A8>   \n",
+      "1                                     <function s_pool at 0x000001A1EA1428C8>   \n",
       "\n",
       "   m  \n",
       "0  1  \n",
-      "1  2  \n",
-      "2  3  \n",
-      "3  4  \n",
-      "4  5  ,                                                                                                   act  \\\n",
-      "0  <bound method Identity.p_identity of <cadCAD.configuration.Identity object at 0x000001A1EE26A6D8>>   \n",
-      "1                                                         <function set_action at 0x000001A1EE3A3BF8>   \n",
-      "2                                                         <function set_action at 0x000001A1EE3A3BF8>   \n",
-      "3                                                         <function set_action at 0x000001A1EE3A3BF8>   \n",
-      "4  <bound method Identity.p_identity of <cadCAD.configuration.Identity object at 0x000001A1EE26A6D8>>   \n",
-      "\n",
-      "                                                                 chosen_agent  \\\n",
-      "0                               <function choose_agent at 0x000001A1EE3A3C80>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432510>   \n",
-      "2  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432510>   \n",
-      "3  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432510>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432510>   \n",
+      "1  2  ,                                                                                        marketing_rate  \\\n",
+      "0                                                   <function p_marketing_rate at 0x000001A1EA142510>   \n",
+      "1  <bound method Identity.p_identity of <cadCAD.configuration.Identity object at 0x000001A1EA006748>>   \n",
       "\n",
-      "                                                                private_price  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432488>   \n",
-      "1                       <function update_private_price at 0x000001A1EE399730>   \n",
-      "2  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432488>   \n",
-      "3  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432488>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432488>   \n",
+      "                                                                                    p_marketing_shock  \\\n",
+      "0                                                  <function p_marketing_shock at 0x000001A1EA142620>   \n",
+      "1  <bound method Identity.p_identity of <cadCAD.configuration.Identity object at 0x000001A1EA006748>>   \n",
       "\n",
-      "                                                                private_alpha  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432598>   \n",
-      "1                       <function update_private_alpha at 0x000001A1EE399A60>   \n",
-      "2  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432598>   \n",
-      "3  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432598>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432598>   \n",
+      "                                                                                           reputation  \\\n",
+      "0  <bound method Identity.p_identity of <cadCAD.configuration.Identity object at 0x000001A1EA006748>>   \n",
+      "1                                                       <function p_reputation at 0x000001A1EA142730>   \n",
       "\n",
-      "                                                                      reserve  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432620>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432620>   \n",
-      "2                                   <function update_R at 0x000001A1EE399D08>   \n",
-      "3  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432620>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432620>   \n",
+      "                                                                                           experience  \\\n",
+      "0  <bound method Identity.p_identity of <cadCAD.configuration.Identity object at 0x000001A1EA006748>>   \n",
+      "1                                                       <function p_experience at 0x000001A1EA1427B8>   \n",
       "\n",
-      "                                                                       supply  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4326A8>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4326A8>   \n",
-      "2                                   <function update_S at 0x000001A1EE399D90>   \n",
-      "3  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4326A8>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4326A8>   \n",
+      "                                                                       signal  \\\n",
+      "0                                   <function s_signal at 0x000001A1EA1426A8>   \n",
+      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EBE2D730>   \n",
       "\n",
-      "                                                                agent_reserve  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432730>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432730>   \n",
-      "2                                   <function update_r at 0x000001A1EE399E18>   \n",
-      "3  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432730>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432730>   \n",
+      "                                                                         pool  \\\n",
+      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EBE2D7B8>   \n",
+      "1                                     <function s_pool at 0x000001A1EA1428C8>   \n",
       "\n",
-      "                                                                 agent_supply  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4327B8>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4327B8>   \n",
-      "2                     <function update_s_free_bondburn at 0x000001A1EE399EA0>   \n",
-      "3  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4327B8>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4327B8>   \n",
-      "\n",
-      "                                                                   spot_price  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432840>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432840>   \n",
-      "2                          <function update_P_bondburn at 0x000001A1EE399F28>   \n",
-      "3                            <function update_P_attest at 0x000001A1EE3A3AE8>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432840>   \n",
-      "\n",
-      "                                                                        price  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4328C8>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4328C8>   \n",
-      "2                                <function update_pbar at 0x000001A1EE3A3048>   \n",
-      "3  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4328C8>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4328C8>   \n",
-      "\n",
-      "                                                                  invariant_I  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432950>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432950>   \n",
-      "2                          <function update_I_bondburn at 0x000001A1EE3A30D0>   \n",
-      "3                            <function update_I_attest at 0x000001A1EE3A3A60>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432950>   \n",
-      "\n",
-      "                                                                 attestations  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4329D8>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4329D8>   \n",
-      "2  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4329D8>   \n",
-      "3                                   <function update_Q at 0x000001A1EE3A3400>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE4329D8>   \n",
-      "\n",
-      "                                                               attestations_1  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432A60>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432A60>   \n",
-      "2  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432A60>   \n",
-      "3                                  <function update_Q1 at 0x000001A1EE3A3488>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432A60>   \n",
-      "\n",
-      "                                                               attestations_0  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432AE8>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432AE8>   \n",
-      "2  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432AE8>   \n",
-      "3                                  <function update_Q0 at 0x000001A1EE3A3510>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432AE8>   \n",
+      "   m  \n",
+      "0  1  \n",
+      "1  2  ,                                                                                        marketing_rate  \\\n",
+      "0                                                   <function p_marketing_rate at 0x000001A1EA142510>   \n",
+      "1  <bound method Identity.p_identity of <cadCAD.configuration.Identity object at 0x000001A1EA006748>>   \n",
       "\n",
-      "                                                                     supply_1  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432B70>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432B70>   \n",
-      "2  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432B70>   \n",
-      "3                                  <function update_S1 at 0x000001A1EE3A3598>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432B70>   \n",
+      "                                                                                    p_marketing_shock  \\\n",
+      "0                                                  <function p_marketing_shock at 0x000001A1EA142620>   \n",
+      "1  <bound method Identity.p_identity of <cadCAD.configuration.Identity object at 0x000001A1EA006748>>   \n",
       "\n",
-      "                                                                     supply_0  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432BF8>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432BF8>   \n",
-      "2  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432BF8>   \n",
-      "3                                  <function update_S0 at 0x000001A1EE3A3620>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432BF8>   \n",
+      "                                                                                           reputation  \\\n",
+      "0  <bound method Identity.p_identity of <cadCAD.configuration.Identity object at 0x000001A1EA006748>>   \n",
+      "1                                                       <function p_reputation at 0x000001A1EA142730>   \n",
       "\n",
-      "                                                         agent_attestations_1  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432C80>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432C80>   \n",
-      "2  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432C80>   \n",
-      "3                                  <function update_q1 at 0x000001A1EE3A36A8>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432C80>   \n",
+      "                                                                                           experience  \\\n",
+      "0  <bound method Identity.p_identity of <cadCAD.configuration.Identity object at 0x000001A1EA006748>>   \n",
+      "1                                                       <function p_experience at 0x000001A1EA1427B8>   \n",
       "\n",
-      "                                                         agent_attestations_0  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432D08>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432D08>   \n",
-      "2  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432D08>   \n",
-      "3                                  <function update_q0 at 0x000001A1EE3A3730>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432D08>   \n",
+      "                                                                       signal  \\\n",
+      "0                                   <function s_signal at 0x000001A1EA1426A8>   \n",
+      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EBE2D840>   \n",
       "\n",
-      "                                                            agent_supply_free  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432D90>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432D90>   \n",
-      "2  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432D90>   \n",
-      "3                              <function update_s_free at 0x000001A1EE3A37B8>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432D90>   \n",
+      "                                                                         pool  \\\n",
+      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EBE2D8C8>   \n",
+      "1                                     <function s_pool at 0x000001A1EA1428C8>   \n",
       "\n",
-      "                                                               agent_supply_1  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432E18>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432E18>   \n",
-      "2  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432E18>   \n",
-      "3                                  <function update_s1 at 0x000001A1EE3A3840>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432E18>   \n",
+      "   m  \n",
+      "0  1  \n",
+      "1  2  ,                                                                                        marketing_rate  \\\n",
+      "0                                                   <function p_marketing_rate at 0x000001A1EA142510>   \n",
+      "1  <bound method Identity.p_identity of <cadCAD.configuration.Identity object at 0x000001A1EA006748>>   \n",
       "\n",
-      "                                                               agent_supply_0  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432EA0>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432EA0>   \n",
-      "2  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432EA0>   \n",
-      "3                                  <function update_s0 at 0x000001A1EE3A38C8>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432EA0>   \n",
+      "                                                                                    p_marketing_shock  \\\n",
+      "0                                                  <function p_marketing_shock at 0x000001A1EA142620>   \n",
+      "1  <bound method Identity.p_identity of <cadCAD.configuration.Identity object at 0x000001A1EA006748>>   \n",
       "\n",
-      "                                                                        alpha  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432F28>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432F28>   \n",
-      "2  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432F28>   \n",
-      "3                               <function update_alpha at 0x000001A1EE3A3950>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EE432F28>   \n",
+      "                                                                                           reputation  \\\n",
+      "0  <bound method Identity.p_identity of <cadCAD.configuration.Identity object at 0x000001A1EA006748>>   \n",
+      "1                                                       <function p_reputation at 0x000001A1EA142730>   \n",
       "\n",
-      "                                                                        kappa  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EF8E2048>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EF8E2048>   \n",
-      "2  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EF8E2048>   \n",
-      "3                               <function update_kappa at 0x000001A1EE3A39D8>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EF8E2048>   \n",
+      "                                                                                           experience  \\\n",
+      "0  <bound method Identity.p_identity of <cadCAD.configuration.Identity object at 0x000001A1EA006748>>   \n",
+      "1                                                       <function p_experience at 0x000001A1EA1427B8>   \n",
       "\n",
-      "                                                                  invariant_V  \\\n",
-      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EF8E20D0>   \n",
-      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EF8E20D0>   \n",
-      "2  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EF8E20D0>   \n",
-      "3                                   <function update_V at 0x000001A1EE3A3B70>   \n",
-      "4  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EF8E20D0>   \n",
+      "                                                                       signal  \\\n",
+      "0                                   <function s_signal at 0x000001A1EA1426A8>   \n",
+      "1  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EBE2D950>   \n",
       "\n",
-      "                                                                                                   agents  \\\n",
-      "0                              <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EF8E2158>   \n",
-      "1                              <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EF8E2158>   \n",
-      "2                              <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EF8E2158>   \n",
-      "3                              <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EF8E2158>   \n",
-      "4  <function putting_the_updated_agent_back_to_its_god_given_home_in_the_dataframe at 0x000001A1EE3A3D90>   \n",
+      "                                                                         pool  \\\n",
+      "0  <function Identity.state_identity.<locals>.<lambda> at 0x000001A1EBE2D9D8>   \n",
+      "1                                     <function s_pool at 0x000001A1EA1428C8>   \n",
       "\n",
       "   m  \n",
       "0  1  \n",
-      "1  2  \n",
-      "2  3  \n",
-      "3  4  \n",
-      "4  5  ]\n",
+      "1  2  ]\n",
       "\n"
      ]
     }
@@ -431,21 +157,263 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 3,
    "metadata": {},
-   "outputs": [],
-   "source": []
+   "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>index</th>\n",
+       "      <th>timestamp</th>\n",
+       "      <th>signal</th>\n",
+       "      <th>pool</th>\n",
+       "      <th>simulation</th>\n",
+       "      <th>run</th>\n",
+       "      <th>substep</th>\n",
+       "      <th>timestep</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>0</td>\n",
+       "      <td>2020-01-01</td>\n",
+       "      <td>0</td>\n",
+       "      <td>&lt;class 'src.sim.model.utils.Adoption_Pool'&gt;: {'state': {'unaware': {'pool': 100000, 'reputation': None}, 'aware': {'pool': 0, 'reputation': 0}, 'adopted': {'pool': 0, 'reputation': 0}, 'loyal': {'pool': 0, 'reputation': 0}, 'churned': {'pool': 0, 'reputation': 0}}, 'threshold': 0.5}</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>1</td>\n",
+       "      <td>2020-01-01</td>\n",
+       "      <td>500</td>\n",
+       "      <td>&lt;class 'src.sim.model.utils.Adoption_Pool'&gt;: {'state': {'unaware': {'pool': 100000, 'reputation': None}, 'aware': {'pool': 0, 'reputation': 0}, 'adopted': {'pool': 0, 'reputation': 0}, 'loyal': {'pool': 0, 'reputation': 0}, 'churned': {'pool': 0, 'reputation': 0}}, 'threshold': 0.5}</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>2</td>\n",
+       "      <td>2020-01-01</td>\n",
+       "      <td>500</td>\n",
+       "      <td>&lt;class 'src.sim.model.utils.Adoption_Pool'&gt;: {'state': {'unaware': {'pool': 99100.0, 'reputation': 1.0089909182643795, 'drip': 0}, 'aware': {'pool': 900.0, 'reputation': 0.01, 'neg_drip': -0.0}, 'adopted': {'pool': 0.0, 'reputation': 0.0, 'neg_drip': 0}, 'loyal': {'pool': 0.0, 'reputation': 0, 'neg_drip': 0}, 'churned': {'pool': 0.0, 'reputation': 0.0, 'neg_drip': -0.0}}, 'threshold': 0.1}</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>2</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>3</td>\n",
+       "      <td>2020-01-01</td>\n",
+       "      <td>500</td>\n",
+       "      <td>&lt;class 'src.sim.model.utils.Adoption_Pool'&gt;: {'state': {'unaware': {'pool': 99100.0, 'reputation': 1.0089909182643795, 'drip': 0}, 'aware': {'pool': 900.0, 'reputation': 0.01, 'neg_drip': -0.0}, 'adopted': {'pool': 0.0, 'reputation': 0.0, 'neg_drip': 0}, 'loyal': {'pool': 0.0, 'reputation': 0, 'neg_drip': 0}, 'churned': {'pool': 0.0, 'reputation': 0.0, 'neg_drip': -0.0}}, 'threshold': 0.1}</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>2</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>4</td>\n",
+       "      <td>2020-01-01</td>\n",
+       "      <td>500</td>\n",
+       "      <td>&lt;class 'src.sim.model.utils.Adoption_Pool'&gt;: {'state': {'unaware': {'pool': 98158.32824688208, 'reputation': 2.02807123405062, 'drip': 0}, 'aware': {'pool': 1841.6717531179208, 'reputation': 0.015159107453722467, 'neg_drip': 0.81}, 'adopted': {'pool': 0.0, 'reputation': 0.0, 'neg_drip': 0}, 'loyal': {'pool': 0.0, 'reputation': 0, 'neg_drip': 0}, 'churned': {'pool': 0.0, 'reputation': 0.0, 'neg_drip': -0.0}}, 'threshold': 0.1}</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>2</td>\n",
+       "      <td>2</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   index  timestamp  signal  \\\n",
+       "0      0 2020-01-01       0   \n",
+       "1      1 2020-01-01     500   \n",
+       "2      2 2020-01-01     500   \n",
+       "3      3 2020-01-01     500   \n",
+       "4      4 2020-01-01     500   \n",
+       "\n",
+       "                                                                                                                                                                                                                                                                                                                                                                                                                                            pool  \\\n",
+       "0                                                                                                                                                    <class 'src.sim.model.utils.Adoption_Pool'>: {'state': {'unaware': {'pool': 100000, 'reputation': None}, 'aware': {'pool': 0, 'reputation': 0}, 'adopted': {'pool': 0, 'reputation': 0}, 'loyal': {'pool': 0, 'reputation': 0}, 'churned': {'pool': 0, 'reputation': 0}}, 'threshold': 0.5}   \n",
+       "1                                                                                                                                                    <class 'src.sim.model.utils.Adoption_Pool'>: {'state': {'unaware': {'pool': 100000, 'reputation': None}, 'aware': {'pool': 0, 'reputation': 0}, 'adopted': {'pool': 0, 'reputation': 0}, 'loyal': {'pool': 0, 'reputation': 0}, 'churned': {'pool': 0, 'reputation': 0}}, 'threshold': 0.5}   \n",
+       "2                                       <class 'src.sim.model.utils.Adoption_Pool'>: {'state': {'unaware': {'pool': 99100.0, 'reputation': 1.0089909182643795, 'drip': 0}, 'aware': {'pool': 900.0, 'reputation': 0.01, 'neg_drip': -0.0}, 'adopted': {'pool': 0.0, 'reputation': 0.0, 'neg_drip': 0}, 'loyal': {'pool': 0.0, 'reputation': 0, 'neg_drip': 0}, 'churned': {'pool': 0.0, 'reputation': 0.0, 'neg_drip': -0.0}}, 'threshold': 0.1}   \n",
+       "3                                       <class 'src.sim.model.utils.Adoption_Pool'>: {'state': {'unaware': {'pool': 99100.0, 'reputation': 1.0089909182643795, 'drip': 0}, 'aware': {'pool': 900.0, 'reputation': 0.01, 'neg_drip': -0.0}, 'adopted': {'pool': 0.0, 'reputation': 0.0, 'neg_drip': 0}, 'loyal': {'pool': 0.0, 'reputation': 0, 'neg_drip': 0}, 'churned': {'pool': 0.0, 'reputation': 0.0, 'neg_drip': -0.0}}, 'threshold': 0.1}   \n",
+       "4  <class 'src.sim.model.utils.Adoption_Pool'>: {'state': {'unaware': {'pool': 98158.32824688208, 'reputation': 2.02807123405062, 'drip': 0}, 'aware': {'pool': 1841.6717531179208, 'reputation': 0.015159107453722467, 'neg_drip': 0.81}, 'adopted': {'pool': 0.0, 'reputation': 0.0, 'neg_drip': 0}, 'loyal': {'pool': 0.0, 'reputation': 0, 'neg_drip': 0}, 'churned': {'pool': 0.0, 'reputation': 0.0, 'neg_drip': -0.0}}, 'threshold': 0.1}   \n",
+       "\n",
+       "   simulation  run  substep  timestep  \n",
+       "0           0    1        0         0  \n",
+       "1           0    1        1         1  \n",
+       "2           0    1        2         1  \n",
+       "3           0    1        1         2  \n",
+       "4           0    1        2         2  "
+      ]
+     },
+     "execution_count": 3,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "exp.head()"
+   ]
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 4,
    "metadata": {},
-   "outputs": [],
-   "source": []
+   "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>index</th>\n",
+       "      <th>timestamp</th>\n",
+       "      <th>signal</th>\n",
+       "      <th>pool</th>\n",
+       "      <th>simulation</th>\n",
+       "      <th>run</th>\n",
+       "      <th>substep</th>\n",
+       "      <th>timestep</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>7999</th>\n",
+       "      <td>7999</td>\n",
+       "      <td>2020-01-01</td>\n",
+       "      <td>500</td>\n",
+       "      <td>&lt;class 'src.sim.model.utils.Adoption_Pool'&gt;: {'state': {'unaware': {'pool': 4.644386578855553, 'reputation': 2054319.0805727735, 'drip': 0}, 'aware': {'pool': 91.75365542761358, 'reputation': 9343.102482312266, 'neg_drip': 5.5465325903151825, 'drip': 0}, 'adopted': {'pool': 6321.873656344102, 'reputation': 7.704058077467613, 'neg_drip': 0, 'drip': 0}, 'loyal': {'pool': 4556.218123429476, 'reputation': 0.044968190288030827, 'neg_drip': 0}, 'churned': {'pool': 89025.51017821995, 'reputation': -8.572421427650248, 'neg_drip': 8453.307238979878}}, 'threshold': 1}</td>\n",
+       "      <td>1</td>\n",
+       "      <td>2</td>\n",
+       "      <td>2</td>\n",
+       "      <td>998</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>8000</th>\n",
+       "      <td>8000</td>\n",
+       "      <td>2020-01-01</td>\n",
+       "      <td>1734</td>\n",
+       "      <td>&lt;class 'src.sim.model.utils.Adoption_Pool'&gt;: {'state': {'unaware': {'pool': 4.644386578855553, 'reputation': 2054319.0805727735, 'drip': 0}, 'aware': {'pool': 91.75365542761358, 'reputation': 9343.102482312266, 'neg_drip': 5.5465325903151825, 'drip': 0}, 'adopted': {'pool': 6321.873656344102, 'reputation': 7.704058077467613, 'neg_drip': 0, 'drip': 0}, 'loyal': {'pool': 4556.218123429476, 'reputation': 0.044968190288030827, 'neg_drip': 0}, 'churned': {'pool': 89025.51017821995, 'reputation': -8.572421427650248, 'neg_drip': 8453.307238979878}}, 'threshold': 1}</td>\n",
+       "      <td>1</td>\n",
+       "      <td>2</td>\n",
+       "      <td>1</td>\n",
+       "      <td>999</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>8001</th>\n",
+       "      <td>8001</td>\n",
+       "      <td>2020-01-01</td>\n",
+       "      <td>1734</td>\n",
+       "      <td>&lt;class 'src.sim.model.utils.Adoption_Pool'&gt;: {'state': {'unaware': {'pool': 4.5979427356748985, 'reputation': 2074863.2712774912, 'drip': 0}, 'aware': {'pool': 90.88266092121573, 'reputation': 9442.198688885383, 'neg_drip': 5.5465325903151825, 'drip': 0}, 'adopted': {'pool': 6311.291591357707, 'reputation': 7.783102932488861, 'neg_drip': 0, 'drip': 0}, 'loyal': {'pool': 4567.717626765449, 'reputation': 0.04535230479794032, 'neg_drip': 0}, 'churned': {'pool': 89025.51017821995, 'reputation': -8.650256992205616, 'neg_drip': 8521.89701237488}}, 'threshold': 1}</td>\n",
+       "      <td>1</td>\n",
+       "      <td>2</td>\n",
+       "      <td>2</td>\n",
+       "      <td>999</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>8002</th>\n",
+       "      <td>8002</td>\n",
+       "      <td>2020-01-01</td>\n",
+       "      <td>500</td>\n",
+       "      <td>&lt;class 'src.sim.model.utils.Adoption_Pool'&gt;: {'state': {'unaware': {'pool': 4.5979427356748985, 'reputation': 2074863.2712774912, 'drip': 0}, 'aware': {'pool': 90.88266092121573, 'reputation': 9442.198688885383, 'neg_drip': 5.5465325903151825, 'drip': 0}, 'adopted': {'pool': 6311.291591357707, 'reputation': 7.783102932488861, 'neg_drip': 0, 'drip': 0}, 'loyal': {'pool': 4567.717626765449, 'reputation': 0.04535230479794032, 'neg_drip': 0}, 'churned': {'pool': 89025.51017821995, 'reputation': -8.650256992205616, 'neg_drip': 8521.89701237488}}, 'threshold': 1}</td>\n",
+       "      <td>1</td>\n",
+       "      <td>2</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>8003</th>\n",
+       "      <td>8003</td>\n",
+       "      <td>2020-01-01</td>\n",
+       "      <td>500</td>\n",
+       "      <td>&lt;class 'src.sim.model.utils.Adoption_Pool'&gt;: {'state': {'unaware': {'pool': 4.551963330478359, 'reputation': 2095612.9038892563, 'drip': 0}, 'aware': {'pool': 90.01990996878718, 'reputation': 9542.336759286367, 'neg_drip': 5.5465325903151825, 'drip': 0}, 'adopted': {'pool': 6300.801982702477, 'reputation': 7.862968542900826, 'neg_drip': 0, 'drip': 0}, 'loyal': {'pool': 4579.115965778304, 'reputation': 0.04574024147590787, 'neg_drip': 0}, 'churned': {'pool': 89025.51017821995, 'reputation': -8.728891251658771, 'neg_drip': 8591.190520820393}}, 'threshold': 1}</td>\n",
+       "      <td>1</td>\n",
+       "      <td>2</td>\n",
+       "      <td>2</td>\n",
+       "      <td>1000</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "      index  timestamp  signal  \\\n",
+       "7999   7999 2020-01-01     500   \n",
+       "8000   8000 2020-01-01    1734   \n",
+       "8001   8001 2020-01-01    1734   \n",
+       "8002   8002 2020-01-01     500   \n",
+       "8003   8003 2020-01-01     500   \n",
+       "\n",
+       "                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                      pool  \\\n",
+       "7999  <class 'src.sim.model.utils.Adoption_Pool'>: {'state': {'unaware': {'pool': 4.644386578855553, 'reputation': 2054319.0805727735, 'drip': 0}, 'aware': {'pool': 91.75365542761358, 'reputation': 9343.102482312266, 'neg_drip': 5.5465325903151825, 'drip': 0}, 'adopted': {'pool': 6321.873656344102, 'reputation': 7.704058077467613, 'neg_drip': 0, 'drip': 0}, 'loyal': {'pool': 4556.218123429476, 'reputation': 0.044968190288030827, 'neg_drip': 0}, 'churned': {'pool': 89025.51017821995, 'reputation': -8.572421427650248, 'neg_drip': 8453.307238979878}}, 'threshold': 1}   \n",
+       "8000  <class 'src.sim.model.utils.Adoption_Pool'>: {'state': {'unaware': {'pool': 4.644386578855553, 'reputation': 2054319.0805727735, 'drip': 0}, 'aware': {'pool': 91.75365542761358, 'reputation': 9343.102482312266, 'neg_drip': 5.5465325903151825, 'drip': 0}, 'adopted': {'pool': 6321.873656344102, 'reputation': 7.704058077467613, 'neg_drip': 0, 'drip': 0}, 'loyal': {'pool': 4556.218123429476, 'reputation': 0.044968190288030827, 'neg_drip': 0}, 'churned': {'pool': 89025.51017821995, 'reputation': -8.572421427650248, 'neg_drip': 8453.307238979878}}, 'threshold': 1}   \n",
+       "8001   <class 'src.sim.model.utils.Adoption_Pool'>: {'state': {'unaware': {'pool': 4.5979427356748985, 'reputation': 2074863.2712774912, 'drip': 0}, 'aware': {'pool': 90.88266092121573, 'reputation': 9442.198688885383, 'neg_drip': 5.5465325903151825, 'drip': 0}, 'adopted': {'pool': 6311.291591357707, 'reputation': 7.783102932488861, 'neg_drip': 0, 'drip': 0}, 'loyal': {'pool': 4567.717626765449, 'reputation': 0.04535230479794032, 'neg_drip': 0}, 'churned': {'pool': 89025.51017821995, 'reputation': -8.650256992205616, 'neg_drip': 8521.89701237488}}, 'threshold': 1}   \n",
+       "8002   <class 'src.sim.model.utils.Adoption_Pool'>: {'state': {'unaware': {'pool': 4.5979427356748985, 'reputation': 2074863.2712774912, 'drip': 0}, 'aware': {'pool': 90.88266092121573, 'reputation': 9442.198688885383, 'neg_drip': 5.5465325903151825, 'drip': 0}, 'adopted': {'pool': 6311.291591357707, 'reputation': 7.783102932488861, 'neg_drip': 0, 'drip': 0}, 'loyal': {'pool': 4567.717626765449, 'reputation': 0.04535230479794032, 'neg_drip': 0}, 'churned': {'pool': 89025.51017821995, 'reputation': -8.650256992205616, 'neg_drip': 8521.89701237488}}, 'threshold': 1}   \n",
+       "8003   <class 'src.sim.model.utils.Adoption_Pool'>: {'state': {'unaware': {'pool': 4.551963330478359, 'reputation': 2095612.9038892563, 'drip': 0}, 'aware': {'pool': 90.01990996878718, 'reputation': 9542.336759286367, 'neg_drip': 5.5465325903151825, 'drip': 0}, 'adopted': {'pool': 6300.801982702477, 'reputation': 7.862968542900826, 'neg_drip': 0, 'drip': 0}, 'loyal': {'pool': 4579.115965778304, 'reputation': 0.04574024147590787, 'neg_drip': 0}, 'churned': {'pool': 89025.51017821995, 'reputation': -8.728891251658771, 'neg_drip': 8591.190520820393}}, 'threshold': 1}   \n",
+       "\n",
+       "      simulation  run  substep  timestep  \n",
+       "7999           1    2        2       998  \n",
+       "8000           1    2        1       999  \n",
+       "8001           1    2        2       999  \n",
+       "8002           1    2        1      1000  \n",
+       "8003           1    2        2      1000  "
+      ]
+     },
+     "execution_count": 4,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "exp.tail()"
+   ]
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 5,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -454,7 +422,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 6,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -469,7 +437,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 7,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -478,7 +446,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": 8,
    "metadata": {},
    "outputs": [
     {
@@ -534,7 +502,7 @@
        "0               0               0                  0  "
       ]
      },
-     "execution_count": 6,
+     "execution_count": 8,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -545,7 +513,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": 9,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -568,7 +536,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 10,
    "metadata": {},
    "outputs": [
     {
@@ -624,7 +592,7 @@
        "0               0               0                  0  "
       ]
      },
-     "execution_count": 8,
+     "execution_count": 10,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -635,7 +603,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": 11,
    "metadata": {},
    "outputs": [
     {
@@ -644,7 +612,7 @@
        "Int64Index([0], dtype='int64')"
       ]
      },
-     "execution_count": 9,
+     "execution_count": 11,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -655,7 +623,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": 12,
    "metadata": {},
    "outputs": [
     {
@@ -664,7 +632,7 @@
        "array([[0, 0, 0, 0, 0, 0, 0]], dtype=int64)"
       ]
      },
-     "execution_count": 10,
+     "execution_count": 12,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -675,7 +643,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 11,
+   "execution_count": 13,
    "metadata": {},
    "outputs": [
     {
@@ -684,7 +652,7 @@
        "array([[0, 0, 0, 0, 0, 0, 0]], dtype=int64)"
       ]
      },
-     "execution_count": 11,
+     "execution_count": 13,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -695,7 +663,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 12,
+   "execution_count": 14,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -709,7 +677,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 13,
+   "execution_count": 15,
    "metadata": {},
    "outputs": [
     {
@@ -727,7 +695,7 @@
        " 0               0               0                  0  }"
       ]
      },
-     "execution_count": 13,
+     "execution_count": 15,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -738,7 +706,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 14,
+   "execution_count": 16,
    "metadata": {},
    "outputs": [
     {
@@ -794,7 +762,7 @@
        "0               0               0                  0  "
       ]
      },
-     "execution_count": 14,
+     "execution_count": 16,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -805,7 +773,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 15,
+   "execution_count": 17,
    "metadata": {},
    "outputs": [
     {
@@ -814,7 +782,7 @@
        "0"
       ]
      },
-     "execution_count": 15,
+     "execution_count": 17,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -825,1806 +793,103 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 1,
-   "metadata": {
-    "scrolled": false
-   },
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\n",
-      "<class 'list'>\n",
-      "[1000]\n",
-      "\n",
-      "CHECKPOINT 1\n",
-      "CHECKPOINT 2\n",
-      "Initial Conditions (config.py) :  {'reserve': 300, 'price': 1, 'realized_price': 0, 'spot_price': 1, 'private_price': 0, 'private_alpha': 0, 'kappa': 0, 'supply': 600.0, 'alpha': [0.5], 'supply_0': 200, 'supply_1': 200, 'supply_free': 600.0, 'attestations': 40, 'attestations_0': 20, 'attestations_1': 20, 'invariant_V': 1200.0, 'invariant_I': 650.0, 'agents':    agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply_1  \\\n",
-      "0                     0                     0             50              10   \n",
-      "1                     0                     0             50              10   \n",
-      "2                     0                     0             50              10   \n",
-      "3                     0                     0             50              10   \n",
-      "4                     0                     0             50              10   \n",
-      "\n",
-      "   agent_supply_0  agent_supply_free  \n",
-      "0              10                 30  \n",
-      "1              10                 30  \n",
-      "2              10                 30  \n",
-      "3              10                 30  \n",
-      "4              10                 30  , 'chosen_agent': 0}\n",
-      "Params (config.py) :  {'starting_kappa': 2, 'starting_alpha': 0.5, 'money_raised': 1000, 'C': 700, 'f': 0.03, 'm': 0.15, 'period': 2000, 'rules_price': 'martin'}\n"
-     ]
-    }
-   ],
-   "source": [
-    "import sys\n",
-    "sys.path.append('../')\n",
-    "\n",
-    "import matplotlib.pyplot as plt\n",
-    "import run2"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 2,
-   "metadata": {
-    "scrolled": false
-   },
+   "execution_count": 18,
+   "metadata": {},
    "outputs": [
     {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\n",
-      "Configurations Length: 1\n",
-      "Execution Method: single_proc_exec\n",
-      "Execution Mode: single_threaded\n",
-      "   agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply_1  \\\n",
-      "0                     0                     0             50              10   \n",
-      "1                     0                     0             50              10   \n",
-      "2                     0                     0             50              10   \n",
-      "3                     0                     0             50              10   \n",
-      "4                     0                     0             50              10   \n",
-      "\n",
-      "   agent_supply_0  agent_supply_free  \n",
-      "0              10                 30  \n",
-      "1              10                 30  \n",
-      "2              10                 30  \n",
-      "3              10                 30  \n",
-      "4              10                 30  \n",
-      "----------------------------\n",
-      "CHOSEN AGENT =     agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply_1  \\\n",
-      "1                     0                     0             50              10   \n",
-      "\n",
-      "   agent_supply_0  agent_supply_free  \n",
-      "1              10                 30   Time  0\n",
-      "Chosen Agent:\n",
-      "{'agent_attestations_1': 0, 'agent_attestations_0': 0, 'agent_reserve': 50, 'agent_supply_1': 10, 'agent_supply_0': 10, 'agent_supply_free': 30, 'id': 1}\n",
-      "Choose Action\n",
-      "AGENT RESERVE =  50\n"
-     ]
-    },
-    {
-     "ename": "TypeError",
-     "evalue": "list indices must be integers or slices, not str",
+     "ename": "AttributeError",
+     "evalue": "'DataFrame' object has no attribute 'private_alpha'",
      "output_type": "error",
      "traceback": [
-      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
-      "\u001b[0;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
-      "\u001b[0;32m<ipython-input-2-1dc3f75a6cbb>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mexp\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mrun2\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrun\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
-      "\u001b[0;32m~/Documents/Github/ICF_Internal/run2.py\u001b[0m in \u001b[0;36mrun\u001b[0;34m(drop_midsteps)\u001b[0m\n\u001b[1;32m     20\u001b[0m     \u001b[0;31m# pprint(configs)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     21\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 22\u001b[0;31m     \u001b[0mraw_system_events\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtensor_field\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msessions\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mrunner\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mexecute\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     23\u001b[0m     \u001b[0msimulation_result\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpd\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mDataFrame\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mraw_system_events\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     24\u001b[0m     \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtype\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msimulation_result\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
-      "\u001b[0;32m/usr/local/lib/python3.7/site-packages/cadCAD/engine/__init__.py\u001b[0m in \u001b[0;36mexecute\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    128\u001b[0m             \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"Execution Method: \"\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mexec_method\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__name__\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    129\u001b[0m             simulations_results = self.exec_method(\n\u001b[0;32m--> 130\u001b[0;31m                 \u001b[0msim_executors\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mvar_dict_list\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstates_lists\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mconfigs_structs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0menv_processes_list\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mTs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mSimIDs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mRunIDs\u001b[0m \u001b[0;31m#Ns\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    131\u001b[0m             )\n\u001b[1;32m    132\u001b[0m             \u001b[0;32mreturn\u001b[0m \u001b[0mget_final_results\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msimulations_results\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mpartial_state_updates\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0meps\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msessions\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mremote_threshold\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
-      "\u001b[0;32m/usr/local/lib/python3.7/site-packages/cadCAD/engine/execution.py\u001b[0m in \u001b[0;36msingle_proc_exec\u001b[0;34m(simulation_execs, var_dict_list, states_lists, configs_structs, env_processes_list, Ts, SimIDs, Ns)\u001b[0m\n\u001b[1;32m     25\u001b[0m     \u001b[0mparams\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0msimulation_execs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstates_lists\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mconfigs_structs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0menv_processes_list\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mTs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mSimIDs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mNs\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     26\u001b[0m     \u001b[0msimulation_exec\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstates_list\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mconfig\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0menv_processes\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mT\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mSimID\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mN\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mlist\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmap\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;32mlambda\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpop\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mparams\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 27\u001b[0;31m     \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0msimulation_exec\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mvar_dict_list\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstates_list\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mconfig\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0menv_processes\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mT\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mSimID\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mN\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     28\u001b[0m     \u001b[0;32mreturn\u001b[0m \u001b[0mflatten\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mresult\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     29\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
-      "\u001b[0;32m/usr/local/lib/python3.7/site-packages/cadCAD/engine/simulation.py\u001b[0m in \u001b[0;36msimulation\u001b[0;34m(self, sweep_dict, states_list, configs, env_processes, time_seq, simulation_id, run, additional_objs)\u001b[0m\n\u001b[1;32m    234\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    235\u001b[0m         pipe_run = flatten(\n\u001b[0;32m--> 236\u001b[0;31m             \u001b[0;34m[\u001b[0m\u001b[0mexecute_run\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msweep_dict\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstates_list\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mconfigs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0menv_processes\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtime_seq\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mrun\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    237\u001b[0m         )\n\u001b[1;32m    238\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
-      "\u001b[0;32m/usr/local/lib/python3.7/site-packages/cadCAD/engine/simulation.py\u001b[0m in \u001b[0;36mexecute_run\u001b[0;34m(sweep_dict, states_list, configs, env_processes, time_seq, run)\u001b[0m\n\u001b[1;32m    227\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    228\u001b[0m             first_timestep_per_run: List[Dict[str, Any]] = self.run_pipeline(\n\u001b[0;32m--> 229\u001b[0;31m                 \u001b[0msweep_dict\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstates_list_copy\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mconfigs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0menv_processes\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtime_seq\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mrun\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0madditional_objs\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    230\u001b[0m             )\n\u001b[1;32m    231\u001b[0m             \u001b[0;32mdel\u001b[0m \u001b[0mstates_list_copy\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
-      "\u001b[0;32m/usr/local/lib/python3.7/site-packages/cadCAD/engine/simulation.py\u001b[0m in \u001b[0;36mrun_pipeline\u001b[0;34m(self, sweep_dict, states_list, configs, env_processes, time_seq, run, additional_objs)\u001b[0m\n\u001b[1;32m    195\u001b[0m         \u001b[0;32mfor\u001b[0m \u001b[0mtime_step\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mtime_seq\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    196\u001b[0m             pipe_run: List[Dict[str, Any]] = self.state_update_pipeline(\n\u001b[0;32m--> 197\u001b[0;31m                 \u001b[0msweep_dict\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msimulation_list\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mconfigs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0menv_processes\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtime_step\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mrun\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0madditional_objs\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    198\u001b[0m             )\n\u001b[1;32m    199\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
-      "\u001b[0;32m/usr/local/lib/python3.7/site-packages/cadCAD/engine/simulation.py\u001b[0m in \u001b[0;36mstate_update_pipeline\u001b[0;34m(self, sweep_dict, simulation_list, configs, env_processes, time_step, run, additional_objs)\u001b[0m\n\u001b[1;32m    171\u001b[0m         \u001b[0;32mfor\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0ms_conf\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mp_conf\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mconfigs\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    172\u001b[0m             states_list: List[Dict[str, Any]] = self.partial_state_update(\n\u001b[0;32m--> 173\u001b[0;31m                 \u001b[0msweep_dict\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msub_step\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstates_list\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msimulation_list\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0ms_conf\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mp_conf\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0menv_processes\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtime_step\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mrun\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0madditional_objs\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    174\u001b[0m             )\n\u001b[1;32m    175\u001b[0m             \u001b[0msub_step\u001b[0m \u001b[0;34m+=\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
-      "\u001b[0;32m/usr/local/lib/python3.7/site-packages/cadCAD/engine/simulation.py\u001b[0m in \u001b[0;36mpartial_state_update\u001b[0;34m(self, sweep_dict, sub_step, sL, sH, state_funcs, policy_funcs, env_processes, time_step, run, additional_objs)\u001b[0m\n\u001b[1;32m    117\u001b[0m         \u001b[0mlast_in_obj\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mDict\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mstr\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mAny\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mdeepcopy\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msL\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    118\u001b[0m         _input: Dict[str, Any] = self.policy_update_exception(\n\u001b[0;32m--> 119\u001b[0;31m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget_policy_input\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msweep_dict\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msub_step\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msH\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlast_in_obj\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mpolicy_funcs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0madditional_objs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    120\u001b[0m         )\n\u001b[1;32m    121\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
-      "\u001b[0;32m/usr/local/lib/python3.7/site-packages/cadCAD/engine/simulation.py\u001b[0m in \u001b[0;36mget_policy_input\u001b[0;34m(self, sweep_dict, sub_step, sL, s, funcs, additional_objs)\u001b[0m\n\u001b[1;32m     53\u001b[0m             \u001b[0;32mreturn\u001b[0m \u001b[0mresult\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     54\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 55\u001b[0;31m         \u001b[0mcol_results\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mget_col_results\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msweep_dict\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msub_step\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msL\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0ms\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfuncs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     56\u001b[0m         \u001b[0mkey_set\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mlist\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mset\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlist\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mreduce\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;32mlambda\u001b[0m \u001b[0ma\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mb\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0ma\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0mb\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlist\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmap\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;32mlambda\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mlist\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mkeys\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcol_results\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     57\u001b[0m         \u001b[0mnew_dict\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m{\u001b[0m\u001b[0mk\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mk\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mkey_set\u001b[0m\u001b[0;34m}\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
-      "\u001b[0;32m/usr/local/lib/python3.7/site-packages/cadCAD/engine/simulation.py\u001b[0m in \u001b[0;36mget_col_results\u001b[0;34m(sweep_dict, sub_step, sL, s, funcs)\u001b[0m\n\u001b[1;32m     40\u001b[0m                 \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     41\u001b[0m                     \u001b[0;32mreturn\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msweep_dict\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msub_step\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msL\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0ms\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0madditional_objs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 42\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mlist\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmap\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;32mlambda\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mpolicy_scope_tuner\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0madditional_objs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfuncs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     43\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     44\u001b[0m         \u001b[0;32mdef\u001b[0m \u001b[0mcompose\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minit_reduction_funct\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfunct_list\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mval_list\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
-      "\u001b[0;32m/usr/local/lib/python3.7/site-packages/cadCAD/engine/simulation.py\u001b[0m in \u001b[0;36m<lambda>\u001b[0;34m(f)\u001b[0m\n\u001b[1;32m     40\u001b[0m                 \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     41\u001b[0m                     \u001b[0;32mreturn\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msweep_dict\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msub_step\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msL\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0ms\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0madditional_objs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 42\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mlist\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmap\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;32mlambda\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mpolicy_scope_tuner\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0madditional_objs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfuncs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     43\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     44\u001b[0m         \u001b[0;32mdef\u001b[0m \u001b[0mcompose\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minit_reduction_funct\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfunct_list\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mval_list\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
-      "\u001b[0;32m/usr/local/lib/python3.7/site-packages/cadCAD/engine/simulation.py\u001b[0m in \u001b[0;36mpolicy_scope_tuner\u001b[0;34m(additional_objs, f)\u001b[0m\n\u001b[1;32m     37\u001b[0m             \u001b[0;32mdef\u001b[0m \u001b[0mpolicy_scope_tuner\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0madditional_objs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     38\u001b[0m                 \u001b[0;32mif\u001b[0m \u001b[0madditional_objs\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 39\u001b[0;31m                     \u001b[0;32mreturn\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msweep_dict\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msub_step\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msL\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0ms\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     40\u001b[0m                 \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     41\u001b[0m                     \u001b[0;32mreturn\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msweep_dict\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msub_step\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msL\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0ms\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0madditional_objs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
-      "\u001b[0;32m~/Documents/Github/ICF_Internal/src/sim/model/parts/choose_action.py\u001b[0m in \u001b[0;36mset_action\u001b[0;34m(params, substep, state_history, prev_state)\u001b[0m\n\u001b[1;32m     32\u001b[0m     \u001b[0mQ1\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mprev_state\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'attestations_1'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     33\u001b[0m     \u001b[0mQ0\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mprev_state\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'attestations_0'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 34\u001b[0;31m     \u001b[0mstart_kappa\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mparams\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'starting_kappa'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     35\u001b[0m     \u001b[0mstart_alpha\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mparams\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'starting_alpha'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     36\u001b[0m     \u001b[0malpha\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mprev_state\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'alpha'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
-      "\u001b[0;31mTypeError\u001b[0m: list indices must be integers or slices, not str"
+      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[1;31mAttributeError\u001b[0m                            Traceback (most recent call last)",
+      "\u001b[1;32m<ipython-input-18-ec4f6f1987e6>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[0mplt\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mplot\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mexp\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mprivate_alpha\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m",
+      "\u001b[1;32m~\\AppData\\Local\\Continuum\\anaconda3\\lib\\site-packages\\pandas\\core\\generic.py\u001b[0m in \u001b[0;36m__getattr__\u001b[1;34m(self, name)\u001b[0m\n\u001b[0;32m   5272\u001b[0m             \u001b[1;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_info_axis\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_can_hold_identifiers_and_holds_name\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mname\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   5273\u001b[0m                 \u001b[1;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mname\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 5274\u001b[1;33m             \u001b[1;32mreturn\u001b[0m \u001b[0mobject\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m__getattribute__\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mname\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   5275\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   5276\u001b[0m     \u001b[1;32mdef\u001b[0m \u001b[0m__setattr__\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mname\u001b[0m\u001b[1;33m:\u001b[0m \u001b[0mstr\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mvalue\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;33m->\u001b[0m \u001b[1;32mNone\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
+      "\u001b[1;31mAttributeError\u001b[0m: 'DataFrame' object has no attribute 'private_alpha'"
      ]
     }
    ],
    "source": [
-    "exp = run2.run()"
+    "plt.plot(exp.private_alpha)"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "[<matplotlib.lines.Line2D at 0x116af2490>]"
-      ]
-     },
-     "execution_count": 3,
-     "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"
-    }
-   ],
+   "outputs": [],
    "source": [
-    "plt.plot(exp.attestations_1)"
+    "plt.plot(exp.alpha)"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "[<matplotlib.lines.Line2D at 0x1166fc390>]"
-      ]
-     },
-     "execution_count": 6,
-     "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(exp.attestations_0)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 7,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "[<matplotlib.lines.Line2D at 0x116817250>]"
-      ]
-     },
-     "execution_count": 7,
-     "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(exp.private_alpha)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 5,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "[<matplotlib.lines.Line2D at 0x11662de50>]"
-      ]
-     },
-     "execution_count": 5,
-     "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(exp.alpha)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "[<matplotlib.lines.Line2D at 0x11a4fa6d0>]"
-      ]
-     },
-     "execution_count": 3,
-     "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"
-    }
-   ],
+   "outputs": [],
    "source": [
     "plt.plot(exp.private_price)"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "[<matplotlib.lines.Line2D at 0x11a62e810>]"
-      ]
-     },
-     "execution_count": 4,
-     "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"
-    }
-   ],
+   "outputs": [],
    "source": [
     "plt.plot(exp.price)"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "[<matplotlib.lines.Line2D at 0x11a69b3d0>]"
-      ]
-     },
-     "execution_count": 5,
-     "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"
-    }
-   ],
+   "outputs": [],
    "source": [
     "plt.plot(exp.spot_price)"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "[<matplotlib.lines.Line2D at 0x11a4a00d0>]"
-      ]
-     },
-     "execution_count": 6,
-     "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"
-    }
-   ],
+   "outputs": [],
    "source": [
     "plt.plot(exp.reserve)"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "[<matplotlib.lines.Line2D at 0x11a3accd0>]"
-      ]
-     },
-     "execution_count": 7,
-     "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"
-    }
-   ],
+   "outputs": [],
    "source": [
     "plt.plot(exp.supply)"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "[<matplotlib.lines.Line2D at 0x11db29450>]"
-      ]
-     },
-     "execution_count": 3,
-     "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"
-    }
-   ],
+   "outputs": [],
    "source": [
     "plt.plot(exp.private_alpha)"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "[<matplotlib.lines.Line2D at 0x11e3f4310>]"
-      ]
-     },
-     "execution_count": 4,
-     "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"
-    }
-   ],
+   "outputs": [],
    "source": [
     "plt.plot(exp.private_price)"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "0         agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                     0                     0              0             0   \n",
-      "1                     0                     0              0             0   \n",
-      "2                     0                     0              0             0   \n",
-      "3                     0                     0              0             0   \n",
-      "4                     0                     0              0             0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0               0               0                  0  \n",
-      "1               0               0                  0  \n",
-      "2               0               0                  0  \n",
-      "3               0               0                  0  \n",
-      "4               0               0                  0  \n",
-      "1         agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                     0                     0              0             0   \n",
-      "1                     0                     0              0             0   \n",
-      "2                     0                     0              0             0   \n",
-      "3                     0                     0              0             0   \n",
-      "4                     0                     0              0             0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0               0               0                  0  \n",
-      "1               0               0                  0  \n",
-      "2               0               0                  0  \n",
-      "3               0               0                  0  \n",
-      "4               0               0                  0  \n",
-      "2         agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                     0                     0              0             0   \n",
-      "1                     0                     0              0             0   \n",
-      "2                     0                     0              0             0   \n",
-      "3                     0                     0              0             0   \n",
-      "4                     0                     0              0             0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0               0               0                  0  \n",
-      "1               0               0                  0  \n",
-      "2               0               0                  0  \n",
-      "3               0               0                  0  \n",
-      "4               0               0                  0  \n",
-      "3         agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                     0                     0              0             0   \n",
-      "1                     0                     0              0             0   \n",
-      "2                     0                     0              0             0   \n",
-      "3                     0                     0              0             0   \n",
-      "4                     0                     0              0             0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0               0               0                  0  \n",
-      "1               0               0                  0  \n",
-      "2               0               0                  0  \n",
-      "3               0               0                  0  \n",
-      "4               0               0                  0  \n",
-      "4         agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                     0                     0              0             0   \n",
-      "1                     0                     0              0             0   \n",
-      "2                     0                     0              0             0   \n",
-      "3                     0                     0              0             0   \n",
-      "4                     0                     0              0             0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0               0               0                  0  \n",
-      "1               0               0                  0  \n",
-      "2               0               0                  0  \n",
-      "3               0               0                  0  \n",
-      "4               0               0                  0  \n",
-      "5         agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             0.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "6         agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             0.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "7         agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             0.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "8         agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             0.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "9         agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             0.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "10        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             0.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "11        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             0.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "12        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             0.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "13        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             0.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "14        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             0.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "15        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "16        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "17        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "18        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "19        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "20        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "21        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "22        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "23        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "24        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "25        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             9.0             0.0                0.0  \n",
-      "26        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             9.0             0.0                0.0  \n",
-      "27        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             9.0             0.0                0.0  \n",
-      "28        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             9.0             0.0                0.0  \n",
-      "29        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             9.0             0.0                0.0  \n",
-      "30        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             9.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             9.0             0.0                0.0  \n",
-      "31        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             9.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             9.0             0.0                0.0  \n",
-      "32        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             9.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             9.0             0.0                0.0  \n",
-      "33        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             9.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             9.0             0.0                0.0  \n",
-      "34        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             9.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             9.0             0.0                0.0  \n",
-      "35        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             9.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             9.0             0.0                0.0  \n",
-      "36        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             9.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             9.0             0.0                0.0  \n",
-      "37        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             9.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             9.0             0.0                0.0  \n",
-      "38        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             9.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             9.0             0.0                0.0  \n",
-      "39        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             9.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             9.0             0.0                0.0  \n",
-      "40        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             9.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             9.0             0.0                0.0  \n",
-      "41        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             9.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             9.0             0.0                0.0  \n",
-      "42        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             9.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             9.0             0.0                0.0  \n",
-      "43        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             9.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             9.0             0.0                0.0  \n",
-      "44        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             9.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             9.0             0.0                0.0  \n",
-      "45        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             9.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             9.0             0.0                0.0  \n",
-      "46        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             9.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             9.0             0.0                0.0  \n",
-      "47        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             9.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             9.0             0.0                0.0  \n",
-      "48        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             9.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             9.0             0.0                0.0  \n",
-      "49        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             9.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             9.0             0.0                0.0  \n",
-      "50        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             9.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             9.0             0.0                0.0  \n",
-      "51        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                     0                     0              0             0   \n",
-      "1                     0                     0              0             0   \n",
-      "2                     0                     0              0             0   \n",
-      "3                     0                     0              0             0   \n",
-      "4                     0                     0              0             0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0               0               0                  0  \n",
-      "1               0               0                  0  \n",
-      "2               0               0                  0  \n",
-      "3               0               0                  0  \n",
-      "4               0               0                  0  \n",
-      "52        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                     0                     0              0             0   \n",
-      "1                     0                     0              0             0   \n",
-      "2                     0                     0              0             0   \n",
-      "3                     0                     0              0             0   \n",
-      "4                     0                     0              0             0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0               0               0                  0  \n",
-      "1               0               0                  0  \n",
-      "2               0               0                  0  \n",
-      "3               0               0                  0  \n",
-      "4               0               0                  0  \n",
-      "53        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                     0                     0              0             0   \n",
-      "1                     0                     0              0             0   \n",
-      "2                     0                     0              0             0   \n",
-      "3                     0                     0              0             0   \n",
-      "4                     0                     0              0             0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0               0               0                  0  \n",
-      "1               0               0                  0  \n",
-      "2               0               0                  0  \n",
-      "3               0               0                  0  \n",
-      "4               0               0                  0  \n",
-      "54        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                     0                     0              0             0   \n",
-      "1                     0                     0              0             0   \n",
-      "2                     0                     0              0             0   \n",
-      "3                     0                     0              0             0   \n",
-      "4                     0                     0              0             0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0               0               0                  0  \n",
-      "1               0               0                  0  \n",
-      "2               0               0                  0  \n",
-      "3               0               0                  0  \n",
-      "4               0               0                  0  \n",
-      "55        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                     0                     0              0             0   \n",
-      "1                     0                     0              0             0   \n",
-      "2                     0                     0              0             0   \n",
-      "3                     0                     0              0             0   \n",
-      "4                     0                     0              0             0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0               0               0                  0  \n",
-      "1               0               0                  0  \n",
-      "2               0               0                  0  \n",
-      "3               0               0                  0  \n",
-      "4               0               0                  0  \n",
-      "56        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             0.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "57        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             0.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "58        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             0.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "59        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             0.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "60        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             0.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             0.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "61        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             0.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "62        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             0.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "63        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             0.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "64        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             0.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "65        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             0.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "66        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "67        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "68        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "69        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "70        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "71        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "72        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "73        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "74        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "75        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "76        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "77        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "78        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "79        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "80        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "81        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "82        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "83        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "84        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "85        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "86        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "87        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "88        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "89        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "90        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             0.0             0.0                0.0  \n",
-      "91        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             9.0             0.0                0.0  \n",
-      "92        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             9.0             0.0                0.0  \n",
-      "93        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             9.0             0.0                0.0  \n",
-      "94        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             9.0             0.0                0.0  \n",
-      "95        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             9.0             0.0                0.0  \n",
-      "96        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             9.0             0.0                0.0  \n",
-      "97        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             9.0             0.0                0.0  \n",
-      "98        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             9.0             0.0                0.0  \n",
-      "99        agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             9.0             0.0                0.0  \n",
-      "100       agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             0.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             9.0             0.0                0.0  \n",
-      "101       agent_attestations_1  agent_attestations_0  agent_reserve  agent_supply  \\\n",
-      "0                   0.0                   0.0            0.0           0.0   \n",
-      "1                   0.0                   0.0            0.0           0.0   \n",
-      "2                   0.0                   0.0            0.0           0.0   \n",
-      "3                   0.0                   0.0            0.0           0.0   \n",
-      "4                   0.0                   0.0            0.0           0.0   \n",
-      "\n",
-      "   agent_supply_1  agent_supply_0  agent_supply_free  \n",
-      "0             9.0             0.0                0.0  \n",
-      "1             9.0             0.0                0.0  \n",
-      "2             9.0             0.0                0.0  \n",
-      "3             9.0             0.0                0.0  \n",
-      "4             9.0             0.0                0.0  \n",
-      "Name: agents, dtype: object\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "print(exp.agents)"
    ]
@@ -2653,7 +918,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.7.4"
+   "version": "3.6.5"
   }
  },
  "nbformat": 4,