august-systems-gaps-summary.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from scipy.stats import rv_discrete\n",
    "from matplotlib import pyplot as plt\n",
    "import cvxpy as cp\n",
    "import pickle\n",
    "# import gurobipy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import utils\n",
    "def run_extremes(N, W, pmf_cust, pmf_serv, gurobi_flag = 1):\n",
    "    Ex = round (np.sum(np.outer(pmf_cust,pmf_serv)*W), 3) \n",
    "    print(\"Expected cost for c = 0 case:\", Ex )\n",
    "    if gurobi_flag == 1:\n",
    "        fluid_sol = utils.fluid_solution_gurobi(N, W, pmf_cust, pmf_serv)\n",
    "    else:\n",
    "        fluid_sol = utils.fluid_solution(N, W, pmf_cust, pmf_serv)\n",
    "#     print(\"\\nFluid Solution:\\n\")\n",
    "#     print(fluid_sol)\n",
    "    return([Ex,fluid_sol])\n",
    "def plot_from_sim_paths(file_name, save_file= 0, frac = 0):\n",
    "    with open(file_name,'rb') as f:\n",
    "        M_res,B_res, N, W, pmf_cust, pmf_serv = pickle.load(f)\n",
    "    Q_paths_M = M_res[\"Q_paths\"]\n",
    "    C_paths_M = M_res[\"C_paths\"]\n",
    "    Q_paths_B = B_res[\"Q_paths\"]\n",
    "    C_paths_B = B_res[\"C_paths\"]\n",
    "    (average_cost_M, average_queue_M) = utils.average_paths(C_paths_M,Q_paths_M, frac)\n",
    "    (average_cost_B, average_queue_B) = utils.average_paths(C_paths_B,Q_paths_B, frac)\n",
    "    [Ex,fluid_sol] = run_extremes(N,W,pmf_cust,pmf_serv)\n",
    "    utils.qc_plot(Ex,fluid_sol,average_queue_B,average_cost_B,average_queue_M,average_cost_M, save_file)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_gaps(file_names, gaps, x_vals):\n",
    "    for file_name in file_names:\n",
    "        with open(file_name,'rb') as f:\n",
    "            M_res,B_res, N, W, pmf_cust, pmf_serv = pickle.load(f)\n",
    "        Q_paths_M = M_res[\"Q_paths\"]\n",
    "        C_paths_M = M_res[\"C_paths\"]\n",
    "        Q_paths_B = B_res[\"Q_paths\"]\n",
    "        C_paths_B = B_res[\"C_paths\"]\n",
    "\n",
    "        # zero_probs[i] = round (np.sum(np.outer(pmf_cust,pmf_serv)*W), 3) \n",
    "        (average_cost_M, average_queue_M) = utils.average_paths(C_paths_M,Q_paths_M)\n",
    "        (average_cost_B, average_queue_B) = utils.average_paths(C_paths_B,Q_paths_B)\n",
    "        # [Ex,fluid_sol] = run_extremes(N,W,pmf_cust,pmf_serv)\n",
    "        cM, qM, cB, qB = utils.minmax(average_cost_M, average_queue_M, average_cost_B, average_queue_B)  \n",
    "        gaps = np.append( gaps, [cB - cM] )\n",
    "        x_vals = np.append(x_vals, [np.var(W)/np.mean(W)])\n",
    "        return ( [gaps, x_vals] )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_gaps_nonU(file_names, gaps, x_vals):\n",
    "    for file_name in file_names:\n",
    "        with open(file_name,'rb') as f:\n",
    "            M_res,B_res, N, W, pmf_cust, pmf_serv = pickle.load(f)\n",
    "        cost_pmf = np.outer(pmf_cust,pmf_serv).ravel()\n",
    "        weights = W.ravel()\n",
    "\n",
    "        mean_cost = sum(cost_pmf*weights)\n",
    "        variance_cost = sum(cost_pmf*(weights**2)) - mean_cost**2\n",
    "        \n",
    "        Q_paths_M = M_res[\"Q_paths\"]\n",
    "        C_paths_M = M_res[\"C_paths\"]\n",
    "        Q_paths_B = B_res[\"Q_paths\"]\n",
    "        C_paths_B = B_res[\"C_paths\"]\n",
    "\n",
    "        # zero_probs[i] = round (np.sum(np.outer(pmf_cust,pmf_serv)*W), 3) \n",
    "        (average_cost_M, average_queue_M) = utils.average_paths(C_paths_M,Q_paths_M)\n",
    "        (average_cost_B, average_queue_B) = utils.average_paths(C_paths_B,Q_paths_B)\n",
    "        # [Ex,fluid_sol] = run_extremes(N,W,pmf_cust,pmf_serv)\n",
    "        cM, qM, cB, qB = utils.minmax(average_cost_M, average_queue_M, average_cost_B, average_queue_B)  \n",
    "        gaps = np.append( gaps, [cB - cM] )\n",
    "        x_vals = np.append(x_vals, [variance_cost/mean_cost])\n",
    "    return ( [gaps, x_vals] )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Gather all file names"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "N = 16"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [],
   "source": [
    "file_names = []\n",
    "indicator = []\n",
    "weights = [0.1,0.5,1.0,1.25,1.414,2.0]\n",
    "for w in weights:\n",
    "    file_names.append(\"../sim_paths_spatial_uniform_16_W_\"+ str(w) + '_new.pkl' )\n",
    "    indicator.append(0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(file_names) == len(indicator)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Original files used for this plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Increased diagonal weights\n",
    "weights = [0.1,0.5,1.0,1.25,1.414,2.0]\n",
    "for w in weights:\n",
    "    file_names.append( \"../sim_paths_spatial_uniform_9_W_\" + str(w) + '.pkl' )\n",
    "    indicator.append(1)\n",
    "for w in weights:\n",
    "    file_names.append( \"../sim_paths_spatial_uniform_4_W_\" + str(w) + '.pkl' )\n",
    "    indicator.append(2)\n",
    "\n",
    "# Also including (not included as part of original files before) longest spatial\n",
    "file_names.append(\"longest_spatial_uniform_4.pkl\")\n",
    "indicator.append(3)\n",
    "file_names.append(\"longest_spatial_uniform_9.pkl\")\n",
    "indicator.append(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(file_names) == len(indicator)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "June"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "# June results\n",
    "# Probab curve simulations so all have weird PMFs\n",
    "for i in range(9):\n",
    "    file_names.append( \"../probab_curve_50k/sim_paths_prob_curve\" + str(i) + '.pkl' )\n",
    "    indicator.append(4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(file_names) == len(indicator)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Spring 2024"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "file_names = file_names + [\"../sim_paths_spatial_uniform_4.pkl\", \"../sim_paths_spatial_uniform_9.pkl\"]\n",
    "indicator = indicator + [5,5]\n",
    "file_names = file_names + [\"../sim_paths_random_4.pkl\", \"../sim_paths_random_2.pkl\", \"../sim_paths_random_5.pkl\", \"../sim_paths_random_4_random_pmf.pkl\"]\n",
    "indicator = indicator + [6,6,6,6]\n",
    "file_names = file_names + [\"../sim_paths_spatial_scale_10_rerun.pkl\", \"../sim_paths_spatial_scale_16.pkl\"]\n",
    "indicator = indicator + [7,7]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Get Gaps\n",
    "Lazy so I am just calling get_gap_nonU for everything"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "gaps = np.array([])\n",
    "x_vals = np.array([])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(file_names) == len(indicator)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [],
   "source": [
    "[gaps, x_vals] = get_gaps_nonU(file_names, gaps, x_vals)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['N=16,D_w', 'N=16,D_w', 'N=16,D_w', 'N=16,D_w', 'N=16,D_w', 'N=16,D_w', 'N=9,D_w', 'N=9,D_w', 'N=9,D_w', 'N=9,D_w', 'N=9,D_w', 'N=9,D_w', 'N=4,D_w', 'N=4,D_w', 'N=4,D_w', 'N=4,D_w', 'N=4,D_w', 'N=4,D_w', 'longest', 'longest', 'PC', 'PC', 'PC', 'PC', 'PC', 'PC', 'PC', 'PC', 'PC', 'Base_N=4,9', 'Base_N=4,9', 'Rwts_N', 'Rwts_N', 'Rwts_N', 'Rwts_N', 'scale', 'scale']\n"
     ]
    }
   ],
   "source": [
    "values = [\"N=16,D_w\", \"N=9,D_w\", \"N=4,D_w\",\"longest\",\"PC\",\"Base_N=4,9\",\"Rwts_N\",\"scale\"]\n",
    "new_ind = [values[i] for i in indicator]\n",
    "print(new_ind)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure( figsize=(12,8) )\n",
    "\n",
    "scatter = plt.scatter(x_vals, gaps, c=indicator, cmap='viridis', alpha=0.7)\n",
    "plt.ylabel(\"Cost Gap\", weight='bold',fontsize=12)\n",
    "plt.xlabel(\"Variance/Mean\", weight='bold',fontsize=12)\n",
    "plt.title(\"Cost Gap vs scaled variances\", weight='bold',fontsize=15)\n",
    "# plt.legend(loc=\"lower left\",ncol=1)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "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>x_vals</th>\n",
       "      <th>Cost_Gap</th>\n",
       "      <th>class</th>\n",
       "      <th>indicator</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.468340</td>\n",
       "      <td>0.457256</td>\n",
       "      <td>N=16,D_w</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.420261</td>\n",
       "      <td>0.405708</td>\n",
       "      <td>N=16,D_w</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.374529</td>\n",
       "      <td>0.340720</td>\n",
       "      <td>N=16,D_w</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.357443</td>\n",
       "      <td>0.262804</td>\n",
       "      <td>N=16,D_w</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.348271</td>\n",
       "      <td>0.274566</td>\n",
       "      <td>N=16,D_w</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     x_vals  Cost_Gap     class  indicator\n",
       "0  0.468340  0.457256  N=16,D_w          0\n",
       "1  0.420261  0.405708  N=16,D_w          0\n",
       "2  0.374529  0.340720  N=16,D_w          0\n",
       "3  0.357443  0.262804  N=16,D_w          0\n",
       "4  0.348271  0.274566  N=16,D_w          0"
      ]
     },
     "execution_count": 105,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import seaborn as sns\n",
    "import pandas as pd\n",
    "data = pd.DataFrame({'x_vals':x_vals,'Cost_Gap':gaps,'class':new_ind,'indicator':indicator})\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "37"
      ]
     },
     "execution_count": 106,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(12, 8))\n",
    "\n",
    "sns.scatterplot(data=data, x='x_vals', y='Cost_Gap', hue=new_ind, palette='bright', alpha=0.7)\n",
    "plt.ylabel(\"Cost Gap\", weight='bold',fontsize=12)\n",
    "plt.xlabel(\"Variance/Mean\", weight='bold',fontsize=12)\n",
    "plt.title(\"Cost Gap vs scaled variances\", weight='bold',fontsize=15)\n",
    "# plt.legend(loc=\"lower left\",ncol=1)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "lval = 35\n",
    "data = pd.DataFrame({'x_vals':x_vals[:lval],'Cost_Gap':gaps[:lval],'class':new_ind[:lval]})\n",
    "data.head()\n",
    "plt.figure(figsize=(12, 8))\n",
    "\n",
    "sns.scatterplot(data=data, x='x_vals', y='Cost_Gap', hue=new_ind[:lval], hue_norm=(0,255), palette='bright', alpha=0.8, s=100)\n",
    "plt.ylabel(\"Cost Gap\", weight='bold',fontsize=12)\n",
    "plt.xlabel(\"Variance/Mean\", weight='bold',fontsize=12)\n",
    "plt.title(\"Cost Gap vs scaled variances\", weight='bold',fontsize=15)\n",
    "# plt.legend(loc=\"lower left\",ncol=1)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Cost Gap vs scaled variances')"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "lval = 35\n",
    "plt.figure( figsize=(10,6) )\n",
    "plt.scatter(x_vals[:lval],gaps[:lval], c=indicator[:lval], cmap='viridis', alpha=0.7)\n",
    "plt.ylabel(\"Cost Gap\", weight='bold',fontsize=12)\n",
    "plt.xlabel(\"Variance/Mean\", weight='bold',fontsize=12)\n",
    "plt.title(\"Cost Gap vs scaled variances\", weight='bold',fontsize=15)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle as pkl\n",
    "with open(\"../gaps_data.pkl\",\"wb\") as f:\n",
    "    pkl.dump([gaps,x_vals,indicator],f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open(\"../gaps_data.pkl\",\"rb\") as f:\n",
    "   [gaps_loaded,x_vals_loaded,indicator_loaded] = pkl.load(f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Cost Gap vs scaled variances')"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure( figsize=(10,6) )\n",
    "plt.scatter(gaps_loaded, x_vals_loaded, c=indicator_loaded, cmap='viridis', alpha=0.7)\n",
    "plt.ylabel(\"Cost Gap\", weight='bold',fontsize=12)\n",
    "plt.xlabel(\"Variance/Mean\", weight='bold',fontsize=12)\n",
    "plt.title(\"Cost Gap vs scaled variances\", weight='bold',fontsize=15)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I think the scaled plots are the ones with high gaps"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Cost Gap vs scaled variances')"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "lval = 33\n",
    "plt.figure( figsize=(10,6) )\n",
    "plt.scatter(gaps_loaded[:lval], x_vals_loaded[:lval], c=indicator_loaded[:lval], cmap='viridis', alpha=0.7)\n",
    "plt.ylabel(\"Cost Gap\", weight='bold',fontsize=12)\n",
    "plt.xlabel(\"Variance/Mean\", weight='bold',fontsize=12)\n",
    "plt.title(\"Cost Gap vs scaled variances\", weight='bold',fontsize=15)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "38"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(indicator)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "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>x_vals</th>\n",
       "      <th>Cost_Gap</th>\n",
       "      <th>class</th>\n",
       "      <th>indicator</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>5.142131</td>\n",
       "      <td>2.955938</td>\n",
       "      <td>scale</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>4.886308</td>\n",
       "      <td>3.546097</td>\n",
       "      <td>scale</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      x_vals  Cost_Gap  class  indicator\n",
       "35  5.142131  2.955938  scale          7\n",
       "36  4.886308  3.546097  scale          7"
      ]
     },
     "execution_count": 108,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data[-2:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "metadata": {},
   "outputs": [],
   "source": [
    "file_name = \"../sim_paths_spatial_scale_10_rerun.pkl\" \n",
    "\n",
    "# \"../sim_paths_spatial_scale_16.pkl\"\n",
    "with open(file_name,'rb') as f:\n",
    "    M_res,B_res, N, W, pmf_cust, pmf_serv = pickle.load(f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[ 0.         10.         20.         30.         10.         14.14213562\n",
      "  22.36067977 31.6227766  20.         22.36067977 28.28427125 36.05551275\n",
      "  30.         31.6227766  36.05551275 42.42640687]\n",
      " [10.          0.         10.         20.         14.14213562 10.\n",
      "  14.14213562 22.36067977 22.36067977 20.         22.36067977 28.28427125\n",
      "  31.6227766  30.         31.6227766  36.05551275]\n",
      " [20.         10.          0.         10.         22.36067977 14.14213562\n",
      "  10.         14.14213562 28.28427125 22.36067977 20.         22.36067977\n",
      "  36.05551275 31.6227766  30.         31.6227766 ]\n",
      " [30.         20.         10.          0.         31.6227766  22.36067977\n",
      "  14.14213562 10.         36.05551275 28.28427125 22.36067977 20.\n",
      "  42.42640687 36.05551275 31.6227766  30.        ]\n",
      " [10.         14.14213562 22.36067977 31.6227766   0.         10.\n",
      "  20.         30.         10.         14.14213562 22.36067977 31.6227766\n",
      "  20.         22.36067977 28.28427125 36.05551275]\n",
      " [14.14213562 10.         14.14213562 22.36067977 10.          0.\n",
      "  10.         20.         14.14213562 10.         14.14213562 22.36067977\n",
      "  22.36067977 20.         22.36067977 28.28427125]\n",
      " [22.36067977 14.14213562 10.         14.14213562 20.         10.\n",
      "   0.         10.         22.36067977 14.14213562 10.         14.14213562\n",
      "  28.28427125 22.36067977 20.         22.36067977]\n",
      " [31.6227766  22.36067977 14.14213562 10.         30.         20.\n",
      "  10.          0.         31.6227766  22.36067977 14.14213562 10.\n",
      "  36.05551275 28.28427125 22.36067977 20.        ]\n",
      " [20.         22.36067977 28.28427125 36.05551275 10.         14.14213562\n",
      "  22.36067977 31.6227766   0.         10.         20.         30.\n",
      "  10.         14.14213562 22.36067977 31.6227766 ]\n",
      " [22.36067977 20.         22.36067977 28.28427125 14.14213562 10.\n",
      "  14.14213562 22.36067977 10.          0.         10.         20.\n",
      "  14.14213562 10.         14.14213562 22.36067977]\n",
      " [28.28427125 22.36067977 20.         22.36067977 22.36067977 14.14213562\n",
      "  10.         14.14213562 20.         10.          0.         10.\n",
      "  22.36067977 14.14213562 10.         14.14213562]\n",
      " [36.05551275 28.28427125 22.36067977 20.         31.6227766  22.36067977\n",
      "  14.14213562 10.         30.         20.         10.          0.\n",
      "  31.6227766  22.36067977 14.14213562 10.        ]\n",
      " [30.         31.6227766  36.05551275 42.42640687 20.         22.36067977\n",
      "  28.28427125 36.05551275 10.         14.14213562 22.36067977 31.6227766\n",
      "   0.         10.         20.         30.        ]\n",
      " [31.6227766  30.         31.6227766  36.05551275 22.36067977 20.\n",
      "  22.36067977 28.28427125 14.14213562 10.         14.14213562 22.36067977\n",
      "  10.          0.         10.         20.        ]\n",
      " [36.05551275 31.6227766  30.         31.6227766  28.28427125 22.36067977\n",
      "  20.         22.36067977 22.36067977 14.14213562 10.         14.14213562\n",
      "  20.         10.          0.         10.        ]\n",
      " [42.42640687 36.05551275 31.6227766  30.         36.05551275 28.28427125\n",
      "  22.36067977 20.         31.6227766  22.36067977 14.14213562 10.\n",
      "  30.         20.         10.          0.        ]]\n"
     ]
    }
   ],
   "source": [
    "print(W)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {},
   "outputs": [],
   "source": [
    "file_name = \"../sim_paths_spatial_scale_16.pkl\"\n",
    "with open(file_name,'rb') as f:\n",
    "    M_res,B_res, N2, W2, pmf_cust2, pmf_serv2 = pickle.load(f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[ 0.         10.         20.         30.         10.         14.14213562\n",
      "  22.36067977 31.6227766  20.         22.36067977 28.28427125 36.05551275\n",
      "  30.         31.6227766  36.05551275 42.42640687]\n",
      " [10.          0.         10.         20.         14.14213562 10.\n",
      "  14.14213562 22.36067977 22.36067977 20.         22.36067977 28.28427125\n",
      "  31.6227766  30.         31.6227766  36.05551275]\n",
      " [20.         10.          0.         10.         22.36067977 14.14213562\n",
      "  10.         14.14213562 28.28427125 22.36067977 20.         22.36067977\n",
      "  36.05551275 31.6227766  30.         31.6227766 ]\n",
      " [30.         20.         10.          0.         31.6227766  22.36067977\n",
      "  14.14213562 10.         36.05551275 28.28427125 22.36067977 20.\n",
      "  42.42640687 36.05551275 31.6227766  30.        ]\n",
      " [10.         14.14213562 22.36067977 31.6227766   0.         10.\n",
      "  20.         30.         10.         14.14213562 22.36067977 31.6227766\n",
      "  20.         22.36067977 28.28427125 36.05551275]\n",
      " [14.14213562 10.         14.14213562 22.36067977 10.          0.\n",
      "  10.         20.         14.14213562 10.         14.14213562 22.36067977\n",
      "  22.36067977 20.         22.36067977 28.28427125]\n",
      " [22.36067977 14.14213562 10.         14.14213562 20.         10.\n",
      "   0.         10.         22.36067977 14.14213562 10.         14.14213562\n",
      "  28.28427125 22.36067977 20.         22.36067977]\n",
      " [31.6227766  22.36067977 14.14213562 10.         30.         20.\n",
      "  10.          0.         31.6227766  22.36067977 14.14213562 10.\n",
      "  36.05551275 28.28427125 22.36067977 20.        ]\n",
      " [20.         22.36067977 28.28427125 36.05551275 10.         14.14213562\n",
      "  22.36067977 31.6227766   0.         10.         20.         30.\n",
      "  10.         14.14213562 22.36067977 31.6227766 ]\n",
      " [22.36067977 20.         22.36067977 28.28427125 14.14213562 10.\n",
      "  14.14213562 22.36067977 10.          0.         10.         20.\n",
      "  14.14213562 10.         14.14213562 22.36067977]\n",
      " [28.28427125 22.36067977 20.         22.36067977 22.36067977 14.14213562\n",
      "  10.         14.14213562 20.         10.          0.         10.\n",
      "  22.36067977 14.14213562 10.         14.14213562]\n",
      " [36.05551275 28.28427125 22.36067977 20.         31.6227766  22.36067977\n",
      "  14.14213562 10.         30.         20.         10.          0.\n",
      "  31.6227766  22.36067977 14.14213562 10.        ]\n",
      " [30.         31.6227766  36.05551275 42.42640687 20.         22.36067977\n",
      "  28.28427125 36.05551275 10.         14.14213562 22.36067977 31.6227766\n",
      "   0.         10.         20.         30.        ]\n",
      " [31.6227766  30.         31.6227766  36.05551275 22.36067977 20.\n",
      "  22.36067977 28.28427125 14.14213562 10.         14.14213562 22.36067977\n",
      "  10.          0.         10.         20.        ]\n",
      " [36.05551275 31.6227766  30.         31.6227766  28.28427125 22.36067977\n",
      "  20.         22.36067977 22.36067977 14.14213562 10.         14.14213562\n",
      "  20.         10.          0.         10.        ]\n",
      " [42.42640687 36.05551275 31.6227766  30.         36.05551275 28.28427125\n",
      "  22.36067977 20.         31.6227766  22.36067977 14.14213562 10.\n",
      "  30.         20.         10.          0.        ]]\n"
     ]
    }
   ],
   "source": [
    "print(W)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
       "       [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
       "       [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
       "       [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
       "       [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
       "       [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
       "       [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
       "       [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
       "       [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
       "       [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
       "       [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
       "       [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
       "       [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
       "       [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
       "       [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
       "       [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.]])"
      ]
     },
     "execution_count": 117,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "W2 - W"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "16"
      ]
     },
     "execution_count": 118,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "N2 "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.095886  , 0.064833  , 0.02011467, 0.10579828, 0.06332473,\n",
       "       0.06126731, 0.03189242, 0.02230265, 0.05465714, 0.09789766,\n",
       "       0.00486615, 0.07277437, 0.09431652, 0.04608176, 0.1115095 ,\n",
       "       0.05247784])"
      ]
     },
     "execution_count": 119,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pmf_cust"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.07632258, 0.04416851, 0.06035114, 0.09632552, 0.04445993,\n",
       "       0.00653104, 0.10674206, 0.03118331, 0.09051541, 0.04553928,\n",
       "       0.09873315, 0.00951475, 0.07374131, 0.04462872, 0.09551354,\n",
       "       0.07572974])"
      ]
     },
     "execution_count": 120,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pmf_cust2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "base",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}