PoC - Optimization - 2D.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Visualization is based on a plotly example that can be found here: https://plot.ly/python/3d-network-graph/\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# maybe put all this in __init__.py ?\n",
    "import gently as mz\n",
    "import gently.physics \n",
    "import gently.simulation \n",
    "import gently.geometry \n",
    "import gently.visualization \n",
    "import gently.metrics \n",
    "import gently.terrain\n",
    "\n",
    "import networkx as nx\n",
    "#Plotly seems very nice but I think it's not very open dand kinda commercial so keep an eye on that!\n",
    "import plotly as py\n",
    "import plotly.graph_objs as go\n",
    "import math\n",
    "import copy\n",
    "import random\n",
    "%matplotlib inline \n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "#import scipy.stats as ss\n",
    "import pyswarms as ps\n",
    "import itertools"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Global Parameters (here so can be played around easily)\n",
    "# Random seed is fixed for reproducibility\n",
    "random.seed(42) # This doesn't fix for scipy.stats so moved everything to random. \n",
    "# scipy.stats wasn't really helping with anything anyway, removed one dependency this way...\n",
    "\n",
    "\n",
    "# Population Paratemeters\n",
    "N = 500\n",
    "N_building_frac = 250/500; # This is the number of nodes that are reserved solely for buildings.\n",
    "                                # Buildings may (likely) end up with more nodes due to reg. nodes falling in buildings\n",
    "N_building_reserved = int(N*N_building_frac)\n",
    "N_node = N-N_building_reserved # excluding building ones, they are generated seperately.\n",
    "N_b = 4 # let's try 6 APs, so optimization is a bit easier...\n",
    "\n",
    "# Sim Parameters\n",
    "d_basestation = 0.25 # AP comm range, normalized\n",
    "d_node = 0.001   # node comm range, normalized\n",
    "pcost_node = 1  # Cost of node-node transmission\n",
    "pcost_basestation = 1.5  # cost of node-AP transmission\n",
    "dim = 5; #3D\n",
    "\n",
    "# Optimization Parameters\n",
    "n_swarm = 100\n",
    "n_iter = 50\n",
    "\n",
    "# about saving and stuff\n",
    "file_name = '2D_Nb' + str(N_b) + '_db' + str(d_basestation) + '_dn' + str(d_node) + '_' + str(n_swarm) + '_' + str(n_iter)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "## Generate buildings first so buildings can be filled accordingly.\n",
    "\n",
    "# @TODO: Automatize this by having a grid and deciding if to place a building at each grid point via a probability \n",
    "# and randomizing the sizes. Fully random can be done by doing 2D rectangle-rectange collusion checks, however\n",
    "# grid is more realistic for US Urban scenario than a random mess. Thats more of a continental thing :)\n",
    "\n",
    "base_origin = mz.geometry.Point(0.2, 0.3, 0)\n",
    "base_size = {'length':0.1, 'width':0.15}\n",
    "height = 0.2\n",
    "\n",
    "#                    {'base':{'x': 0.45, 'y':0.6, 'z':0},'size':{'length':0.2, 'width':0.1},'height':0.05},\n",
    "pert_param_list = [ {'base':mz.geometry.Point(0.2,0.55,0)  ,'size':{'length':0.1, 'width':0.15}    ,'height': 0.085},\n",
    "                    {'base':mz.geometry.Point(0.6,0.35,0)  ,'size':{'length':0.1, 'width':0.1}     ,'height':0.125},\n",
    "                    {'base':mz.geometry.Point(0.3,0.2,0)   ,'size':{'length':0.15, 'width':0.2}    ,'height':0.065},\n",
    "                    {'base':mz.geometry.Point(0.7,0.8,0)   ,'size':{'length':0.075, 'width':0.1}   ,'height':0.1},\n",
    "                    {'base':mz.geometry.Point(0.15,0.35,0) ,'size':{'length':0.075, 'width':0.1}   ,'height':0.06},\n",
    "                    {'base':mz.geometry.Point(0.45,0.65,0) ,'size':{'length':0.125, 'width':0.1}   ,'height':0.060},\n",
    "                    {'base':mz.geometry.Point(0.45,0.45,0) ,'size':{'length':0.085, 'width':0.095} ,'height':0.075}\n",
    "                  ]\n",
    "\n",
    "pert_list = [mz.geometry.RectPrism(p['base'],p['size'],p['height']) for p in pert_param_list]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "## First generate the population\n",
    "\n",
    "# Populate Nodes\n",
    "# Generate the x-y coordinates (@TODO we also need to make sure there are no duplicate 3D coordinates)\n",
    "mu = 0;\n",
    "rho = 5;\n",
    "#x = ss.norm.rvs(loc=mu,scale=rho,size=N)\n",
    "\n",
    "#x = np.array([random.gauss(mu,rho) for n in np.arange(N_node)])\n",
    "#x = [s + abs(min(x)) for s in x]\n",
    "# Trying with uniform dist.\n",
    "x = np.array([random.uniform(0, 1) for n in np.arange(N_node)])\n",
    "\n",
    "\n",
    "#y = ss.norm.rvs(loc=mu,scale=rho,size=N)\n",
    "#y = np.array([random.gauss(mu,rho) for n in np.arange(N_node)])\n",
    "#y = [s + abs(min(y)) for s in y]\n",
    "# Trying with uniform dist.\n",
    "y = np.array([random.uniform(0, 1) for n in np.arange(N_node)])\n",
    "\n",
    "\n",
    "# generate nodes that will be inside the buildings. Can put them on base points of buildings, they will get \n",
    "# mixed into the building when conform function is called.\n",
    "x_building = np.array([])\n",
    "y_building = np.array([])\n",
    "h_building = np.array([])\n",
    "    \n",
    "#sf = max(max(x),max(y))\n",
    "#x = x/sf;\n",
    "#y = y/sf;\n",
    "\n",
    "# to make it work with uniform\n",
    "sf = 25\n",
    "\n",
    "# These are already normalized, so don't do it twice.\n",
    "for idx in np.arange(N_building_reserved):\n",
    "    pert = random.choice(pert_list)\n",
    "    x_building = np.append(x_building,pert.base_origin.x)\n",
    "    y_building = np.append(y_building,pert.base_origin.y)\n",
    "    h_building = np.append(h_building,pert.base_origin.z)\n",
    "# append new building nodes into the lists\n",
    "x = np.append(x,x_building)\n",
    "y = np.append(y,y_building)\n",
    "\n",
    "\n",
    "# Generate z axis(height), heavily around zero.\n",
    "lambd = 10\n",
    "low = 0\n",
    "high = 0.1\n",
    "\n",
    "#h = ss.expon.rvs(loc=0,scale=1/lambd,size=N)\n",
    "h = np.array([random.expovariate(lambd) for n in np.arange(N_node)])\n",
    "h = h/(sf*2)*2 # @TODO: Hack Solution! Fix sampling of height!!\n",
    "# append new building nodes into the lists, here because they are already normalized\n",
    "h = np.append(h,h_building)\n",
    "# Normalize the coordinates to be in a unit cube\n",
    "node_coordinates = (x,y,h)\n",
    "\n",
    "# Populate basestations\n",
    "\n",
    "bs_range = (0.1,0.9) # Outlying box of AP rectangle\n",
    "h_bs = 0.005; # starting height for APs\n",
    "\n",
    "x_list = np.linspace(bs_range[0],bs_range[1],int(math.sqrt(N_b)))\n",
    "y_list = np.linspace(bs_range[0],bs_range[1],int(math.sqrt(N_b)))\n",
    "xx, yy = np.meshgrid(x_list, y_list)\n",
    "\n",
    "x_basestation = [] \n",
    "y_basestation = []\n",
    "for i in range(len(xx)):\n",
    "    for j in range(len(yy)):\n",
    "        x_basestation.append(xx[i,j]) \n",
    "        y_basestation.append(yy[i,j])\n",
    "\n",
    "x_basestation = np.array(x_basestation)\n",
    "y_basestation = np.array(y_basestation)\n",
    "\n",
    "#sf = max(max(x_basestation),max(y_basestation))\n",
    "#x_basestation = x_basestation/sf\n",
    "#y_basestation = y_basestation/sf\n",
    "\n",
    "#h_basestation = np.ones(N_b) * np.mean(node_coordinates[2])\n",
    "h_basestation = np.ones(N_b)*h_bs\n",
    "\n",
    "basestation_coordinates = (x_basestation,y_basestation,h_basestation) # couldn't do this in a better way for some reason.\n",
    "\n",
    "# Convert into dictionary and give every vertex their unique id\n",
    "node_keys = np.arange(N)\n",
    "node_coordinates_dict = {i: (node_coordinates[0][i],\n",
    "                             node_coordinates[1][i],\n",
    "                             node_coordinates[2][i]) for i in node_keys}\n",
    "# Dictionary keys will also be node id's so have to make them unique\n",
    "basestation_keys = np.arange(N,N+N_b)\n",
    "basestation_coordinates_dict = {i: (basestation_coordinates[0][i-N],\n",
    "                                    basestation_coordinates[1][i-N],\n",
    "                                    basestation_coordinates[2][i-N]) for i in basestation_keys}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "## Construct the network and prepare the stage\n",
    "\n",
    "coordinates = {'node': node_coordinates_dict, 'basestation': basestation_coordinates_dict}\n",
    "radii = {'node': d_node, 'basestation': d_basestation}\n",
    "costs = {'node': pcost_node, 'basestation': pcost_basestation}\n",
    "\n",
    "S = mz.simulation.SimStage(coordinates,radii,costs)\n",
    "\n",
    "G = S.G_dict['combined']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "## Decide on the terrain and update the stage accordingly.\n",
    "\n",
    "t_params = {'range': {'low':0.3,'high':0.75},'height':0.1}\n",
    "S.update_terrain(mz.terrain.surface_height_func4,t_params)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Wall time: 5.59 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "## Update the stage with buildings \n",
    "S.update_perturbations(pert_list) #0ns\n",
    "S.conform_node_heights(['node','basestation']) #41ms\n",
    "S.update_connections()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "## Visualize to see what everything looks like\n",
    "# Enter the params for the network to be visualized\n",
    "mz.visualization.visualize_stage(S,figure_name='module_test_' +file_name)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "network is NOT connected\n",
      "percentage connected:0.256\n",
      "k-connectedness k:1\n",
      "average shortest path length:1.9473684210526316\n",
      "average shortest path length(weighted):2.9210526315789473\n"
     ]
    }
   ],
   "source": [
    "## Compute Initial performance metrics\n",
    "G_met = S.G_dict['combined']\n",
    "\n",
    "is_conn = mz.metrics.is_connected(G_met)\n",
    "if is_conn == True:\n",
    "    print('network IS connected')\n",
    "else: \n",
    "    print('network is NOT connected')\n",
    "\n",
    "bs_keys = S.coordinates_dict_base['basestation'].keys()\n",
    "conn_frac = mz.metrics.connected_fraction(G_met,True,bs_keys)\n",
    "print('percentage connected:' + str(conn_frac))\n",
    "\n",
    "conn_k = mz.metrics.k_connectedness(G_met)\n",
    "print('k-connectedness k:' + str(conn_k[1]))\n",
    "\n",
    "avg_path = mz.metrics.avg_shortest_path_length(G_met)\n",
    "print('average shortest path length:' + str(avg_path[1]))\n",
    "\n",
    "avg_path_weighted = mz.metrics.avg_shortest_path_length(G_met,'power_cost')\n",
    "print('average shortest path length(weighted):' + str(avg_path_weighted[1]))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Grid search is not doable, going with Particle Swarm\n",
    "def network_sim_2d(params):\n",
    "    \"\"\"Network Simulation as objective function\n",
    "\n",
    "    This computes a network simulation for given AP locations\n",
    "    and returns the goal, which is the connected percentage\n",
    "\n",
    "    Inputs\n",
    "    ------\n",
    "    params: np.ndarray\n",
    "        Unrolled version of the coordinates for each access point\n",
    "        ((x1,y1),(x2,y2),...,h) last one is the height for all, since\n",
    "        this is 2D\n",
    "\n",
    "    Returns\n",
    "    -------\n",
    "    float\n",
    "        The computed unconnected percentage of the network \n",
    "        (unconnected because expects cost to return)\n",
    "    \"\"\"\n",
    "    # Unroll the received parameters into coordinates\n",
    "    \n",
    "    h = params[-1] # height is the last element\n",
    "    #print(params)\n",
    "    #print(h)\n",
    "    n_axes = 2 # only (x,y)\n",
    "    n_aps = int((len(params)-1)/n_axes)\n",
    "    # excluding last element from reshape\n",
    "    xy_array = params[:-1].reshape((n_aps,n_axes)) \n",
    "    #print(xy_array)\n",
    "    x_array = xy_array[:,0]\n",
    "    y_array = xy_array[:,1]\n",
    "    #print(x_array)\n",
    "    #print(y_array)\n",
    "    h_array = np.empty(n_aps);\n",
    "    h_array.fill(h) # create a list of same height (2D approx.) for all APs\n",
    "    \n",
    "    # Perform the network simulation\n",
    "    \n",
    "    # Here should check if all basestation coordinates are viable, and continue to next coordinate if not.\n",
    "    # this will seriously mess with the optimizer though, I don't know what to do about that. Maybe use discrete\n",
    "    # values.\n",
    "\n",
    "    S.update_node_coordinates(x_array,y_array,h_array,n_type='basestation')\n",
    "    S.update_connections()\n",
    "    G_met = S.G_dict['combined']\n",
    "    # avg length of shortest paths, LINK WEIGHTED \n",
    "    # First check if the network is connected \n",
    "\n",
    "    # Compute the cost\n",
    "    conn_frac = mz.metrics.connected_fraction(G_met,True,S.coordinates_dict_base['basestation'].keys()) \n",
    "    # conn_frac is gain, need to return loss\n",
    "    loss = 1 - conn_frac # fraction of unconnected nodes\n",
    "    #print(loss)\n",
    "    return loss\n",
    "\n",
    "def f_2d(x):\n",
    "    \"\"\"Higher-level method to do the simulation for the\n",
    "    whole swarm.\n",
    "\n",
    "    Inputs\n",
    "    ------\n",
    "    x: numpy.ndarray of shape (n_particles, dimensions)\n",
    "        The swarm that will perform the search\n",
    "\n",
    "    Returns\n",
    "    -------\n",
    "    numpy.ndarray of shape (n_particles, )\n",
    "        The computed loss for each particle\n",
    "    \"\"\"\n",
    "    #print('selams')\n",
    "    n_particles = x.shape[0] # this can be parallelized...\n",
    "    j = [network_sim_2d(x[i]) for i in range(n_particles)]\n",
    "    print(x[0])\n",
    "    #print(x[1])\n",
    "    #print(str(j[0]) + ' ' + str(j[1]) + ' ' + str(j[2]) + ' ' + str(j[3]))\n",
    "    return np.array(j)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.90724183 0.05974785 0.5069715  0.85258387 0.26518138 0.77634418\n",
      " 0.93916216 0.48980086 0.06607206]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:pyswarms.single.global_best:Iteration 1/50, cost: 0.42600000000000005\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.90724183 0.05974785 0.5069715  0.85258387 0.26518138 0.77634418\n",
      " 0.93916216 0.48980086 0.06607206]\n",
      "[0.9239885  0.21423519 0.45702289 0.70491452 0.37757967 0.67497421\n",
      " 0.81018476 0.53769222 0.08066693]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:pyswarms.single.global_best:Iteration 2/50, cost: 0.42600000000000005\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.90724183 0.05974785 0.5069715  0.85258387 0.26518138 0.77634418\n",
      " 0.93916216 0.48980086 0.06607206]\n",
      "[0.89268432 0.25607606 0.43028488 0.68374612 0.42124667 0.62469902\n",
      " 0.704146   0.5411855  0.086291  ]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:pyswarms.single.global_best:Iteration 3/50, cost: 0.406\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.9239885  0.21423519 0.45702289 0.70491452 0.37757967 0.67497421\n",
      " 0.81018476 0.53769222 0.08066693]\n",
      "[0.8646505  0.25834239 0.39849063 0.62947628 0.42649569 0.58057778\n",
      " 0.67931846 0.59917679 0.09591638]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:pyswarms.single.global_best:Iteration 4/50, cost: 0.376\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.89268432 0.25607606 0.43028488 0.68374612 0.42124667 0.62469902\n",
      " 0.704146   0.5411855  0.086291  ]\n",
      "[0.81581992 0.39191255 0.34184765 0.60886769 0.43700458 0.51485702\n",
      " 0.67901535 0.59978606 0.10014008]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:pyswarms.single.global_best:Iteration 5/50, cost: 0.376\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.8646505  0.25834239 0.39849063 0.62947628 0.42649569 0.58057778\n",
      " 0.67931846 0.59917679 0.09591638]\n",
      "[0.76780857 0.52769239 0.29579832 0.60246549 0.45100571 0.49115753\n",
      " 0.67916123 0.59844659 0.10035842]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:pyswarms.single.global_best:Iteration 6/50, cost: 0.376\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.81581992 0.39191255 0.34184765 0.60886769 0.43700458 0.51485702\n",
      " 0.67901535 0.59978606 0.10014008]\n",
      "[0.72404895 0.59974937 0.28609652 0.5853028  0.49683022 0.44311397\n",
      " 0.68845801 0.59679664 0.10030971]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:pyswarms.single.global_best:Iteration 7/50, cost: 0.372\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.81581992 0.39191255 0.34184765 0.60886769 0.43700458 0.51485702\n",
      " 0.67901535 0.59978606 0.10014008]\n",
      "[0.71742958 0.55645793 0.29623808 0.56853256 0.51800286 0.46967975\n",
      " 0.65718785 0.60295772 0.09994256]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:pyswarms.single.global_best:Iteration 8/50, cost: 0.372\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.81581992 0.39191255 0.34184765 0.60886769 0.43700458 0.51485702\n",
      " 0.67901535 0.59978606 0.10014008]\n",
      "[0.74887783 0.57488033 0.28058421 0.55214142 0.52816732 0.48176352\n",
      " 0.65338984 0.61055638 0.09862749]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:pyswarms.single.global_best:Iteration 9/50, cost: 0.37\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.81581992 0.39191255 0.34184765 0.60886769 0.43700458 0.51485702\n",
      " 0.67901535 0.59978606 0.10014008]\n",
      "[0.73501867 0.55595611 0.29921548 0.56995876 0.5179841  0.45693073\n",
      " 0.66400501 0.60274815 0.09838407]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:pyswarms.single.global_best:Iteration 10/50, cost: 0.368\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.81581992 0.39191255 0.34184765 0.60886769 0.43700458 0.51485702\n",
      " 0.67901535 0.59978606 0.10014008]\n",
      "[0.67362896 0.62546174 0.29750798 0.55244024 0.52141572 0.42281673\n",
      " 0.66367164 0.59874752 0.09734121]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:pyswarms.single.global_best:Iteration 11/50, cost: 0.368\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.73501867 0.55595611 0.29921548 0.56995876 0.5179841  0.45693073\n",
      " 0.66400501 0.60274815 0.09838407]\n",
      "[0.61823433 0.63613804 0.29426358 0.54047594 0.52265646 0.40994964\n",
      " 0.66328897 0.5984446  0.09549604]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:pyswarms.single.global_best:Iteration 12/50, cost: 0.368\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.67362896 0.62546174 0.29750798 0.55244024 0.52141572 0.42281673\n",
      " 0.66367164 0.59874752 0.09734121]\n",
      "[0.59545359 0.64229616 0.2782404  0.53430205 0.52527557 0.40441144\n",
      " 0.663174   0.59553307 0.09494615]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:pyswarms.single.global_best:Iteration 13/50, cost: 0.364\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.61823433 0.63613804 0.29426358 0.54047594 0.52265646 0.40994964\n",
      " 0.66328897 0.5984446  0.09549604]\n",
      "[0.59411815 0.65558596 0.28411648 0.52198847 0.5322031  0.38261541\n",
      " 0.66443876 0.59366336 0.09518252]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:pyswarms.single.global_best:Iteration 14/50, cost: 0.35\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.61823433 0.63613804 0.29426358 0.54047594 0.52265646 0.40994964\n",
      " 0.66328897 0.5984446  0.09549604]\n",
      "[0.56997359 0.6616889  0.27874522 0.51962067 0.54145498 0.35020903\n",
      " 0.66312772 0.60246184 0.09478993]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:pyswarms.single.global_best:Iteration 15/50, cost: 0.35\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.59411815 0.65558596 0.28411648 0.52198847 0.5322031  0.38261541\n",
      " 0.66443876 0.59366336 0.09518252]\n",
      "[0.52712726 0.67687665 0.27712136 0.51882394 0.55066004 0.33423413\n",
      " 0.66251634 0.6132038  0.0946054 ]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:pyswarms.single.global_best:Iteration 16/50, cost: 0.33799999999999997\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.56997359 0.6616889  0.27874522 0.51962067 0.54145498 0.35020903\n",
      " 0.66312772 0.60246184 0.09478993]\n",
      "[0.50510678 0.69080524 0.27219651 0.51862551 0.55746416 0.32094585\n",
      " 0.64950217 0.62136456 0.09486022]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:pyswarms.single.global_best:Iteration 17/50, cost: 0.33799999999999997\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.52712726 0.67687665 0.27712136 0.51882394 0.55066004 0.33423413\n",
      " 0.66251634 0.6132038  0.0946054 ]\n",
      "[0.50106857 0.69307097 0.2673487  0.51532243 0.56914344 0.31478846\n",
      " 0.64651059 0.63187195 0.09524333]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:pyswarms.single.global_best:Iteration 18/50, cost: 0.33799999999999997\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.50510678 0.69080524 0.27219651 0.51862551 0.55746416 0.32094585\n",
      " 0.64950217 0.62136456 0.09486022]\n",
      "[0.49725555 0.70451277 0.26006196 0.51485762 0.5806054  0.31107292\n",
      " 0.64047324 0.63725909 0.09533921]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:pyswarms.single.global_best:Iteration 19/50, cost: 0.33599999999999997\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.50106857 0.69307097 0.2673487  0.51532243 0.56914344 0.31478846\n",
      " 0.64651059 0.63187195 0.09524333]\n",
      "[0.49490472 0.71323232 0.25859432 0.51361326 0.58266417 0.30217516\n",
      " 0.64549514 0.63563919 0.09534689]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:pyswarms.single.global_best:Iteration 20/50, cost: 0.32999999999999996\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.50106857 0.69307097 0.2673487  0.51532243 0.56914344 0.31478846\n",
      " 0.64651059 0.63187195 0.09524333]\n",
      "[0.4752452  0.70997515 0.26091    0.50447338 0.57849343 0.30386334\n",
      " 0.64394871 0.63624321 0.09554299]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:pyswarms.single.global_best:Iteration 21/50, cost: 0.32599999999999996\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.50106857 0.69307097 0.2673487  0.51532243 0.56914344 0.31478846\n",
      " 0.64651059 0.63187195 0.09524333]\n",
      "[0.47276815 0.71446635 0.25500311 0.49155272 0.57833038 0.30305715\n",
      " 0.64266182 0.63908726 0.09572855]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:pyswarms.single.global_best:Iteration 22/50, cost: 0.32599999999999996\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.4752452  0.70997515 0.26091    0.50447338 0.57849343 0.30386334\n",
      " 0.64394871 0.63624321 0.09554299]\n",
      "[0.46966667 0.72206344 0.25189121 0.48740381 0.57954914 0.30120685\n",
      " 0.64138369 0.6422237  0.09583524]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:pyswarms.single.global_best:Iteration 23/50, cost: 0.32399999999999995\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.47276815 0.71446635 0.25500311 0.49155272 0.57833038 0.30305715\n",
      " 0.64266182 0.63908726 0.09572855]\n",
      "[0.46728999 0.73306467 0.24558944 0.48365823 0.587573   0.29407781\n",
      " 0.64348193 0.64210871 0.09660181]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:pyswarms.single.global_best:Iteration 24/50, cost: 0.31999999999999995\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.46966667 0.72206344 0.25189121 0.48740381 0.57954914 0.30120685\n",
      " 0.64138369 0.6422237  0.09583524]\n",
      "[0.46539628 0.73524194 0.24369174 0.4781648  0.58970558 0.28883516\n",
      " 0.64267481 0.64238431 0.09668466]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:pyswarms.single.global_best:Iteration 25/50, cost: 0.31999999999999995\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.46728999 0.73306467 0.24558944 0.48365823 0.587573   0.29407781\n",
      " 0.64348193 0.64210871 0.09660181]\n",
      "[0.46414094 0.73910547 0.24180152 0.47572768 0.59082171 0.28451255\n",
      " 0.64225989 0.64317704 0.09669524]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:pyswarms.single.global_best:Iteration 26/50, cost: 0.31999999999999995\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.46539628 0.73524194 0.24369174 0.4781648  0.58970558 0.28883516\n",
      " 0.64267481 0.64238431 0.09668466]\n",
      "[0.45461937 0.74136133 0.24119335 0.47314742 0.59233847 0.28420492\n",
      " 0.64172399 0.64395678 0.0967022 ]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:pyswarms.single.global_best:Iteration 27/50, cost: 0.31999999999999995\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.46414094 0.73910547 0.24180152 0.47572768 0.59082171 0.28451255\n",
      " 0.64225989 0.64317704 0.09669524]\n",
      "[0.45183341 0.74398915 0.24001639 0.47161401 0.59323077 0.28308014\n",
      " 0.64093848 0.64443602 0.09670361]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:pyswarms.single.global_best:Iteration 28/50, cost: 0.31799999999999995\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.45461937 0.74136133 0.24119335 0.47314742 0.59233847 0.28420492\n",
      " 0.64172399 0.64395678 0.0967022 ]\n",
      "[0.44989877 0.74541351 0.2380626  0.46781324 0.59211126 0.28498489\n",
      " 0.64088758 0.64541965 0.09678632]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:pyswarms.single.global_best:Iteration 29/50, cost: 0.31799999999999995\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.45183341 0.74398915 0.24001639 0.47161401 0.59323077 0.28308014\n",
      " 0.64093848 0.64443602 0.09670361]\n",
      "[0.44670343 0.74590991 0.23567414 0.46579741 0.59143024 0.28616774\n",
      " 0.64084176 0.64562124 0.09687167]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:pyswarms.single.global_best:Iteration 30/50, cost: 0.31799999999999995\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.44989877 0.74541351 0.2380626  0.46781324 0.59211126 0.28498489\n",
      " 0.64088758 0.64541965 0.09678632]\n",
      "[0.44258876 0.74679402 0.23238687 0.46334258 0.59119758 0.2871492\n",
      " 0.64080253 0.64591336 0.09697504]\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-11-1787992dea98>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m()\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[0mget_ipython\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mrun_cell_magic\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m'time'\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;34m''\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;34m\"# bounds, not my best but I'm really sleepy...\\nmax_bound = 1 * np.ones(N_b*2)#xy bounds\\nmin_bound = np.zeros(N_b*2)   #xy bounds\\nmax_bound = np.append(max_bound,0.125) #h bound\\nmin_bound = np.append(min_bound,0.025)  #h bound\\n\\nbounds = (min_bound, max_bound)\\n\\n\\n# Initialize swarm, arbitrary for now. This can also be searched for apparently...\\noptions = {'c1': 0.5, 'c2': 0.3, 'w':0.05} #optimizing this didn't work very well for some reason, check that again.\\n\\n# Call instance of PSO, also remember constraints\\ndimensions = ((2 * N_b) + 1)\\noptimizer = ps.single.GlobalBestPSO(n_particles=n_swarm, dimensions=dimensions, options=options, bounds=bounds)\\n\\n# Perform optimization # each iteration appears to be the same, warum?\\ncost, pos = optimizer.optimize(f_2d, print_step=1, iters=n_iter, verbose=3)\"\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[1;32mD:\\Base\\Anaconda3\\lib\\site-packages\\IPython\\core\\interactiveshell.py\u001b[0m in \u001b[0;36mrun_cell_magic\u001b[1;34m(self, magic_name, line, cell)\u001b[0m\n\u001b[0;32m   2113\u001b[0m             \u001b[0mmagic_arg_s\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mvar_expand\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mline\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mstack_depth\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   2114\u001b[0m             \u001b[1;32mwith\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mbuiltin_trap\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 2115\u001b[1;33m                 \u001b[0mresult\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mfn\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mmagic_arg_s\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mcell\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   2116\u001b[0m             \u001b[1;32mreturn\u001b[0m \u001b[0mresult\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   2117\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m<decorator-gen-60>\u001b[0m in \u001b[0;36mtime\u001b[1;34m(self, line, cell, local_ns)\u001b[0m\n",
      "\u001b[1;32mD:\\Base\\Anaconda3\\lib\\site-packages\\IPython\\core\\magic.py\u001b[0m in \u001b[0;36m<lambda>\u001b[1;34m(f, *a, **k)\u001b[0m\n\u001b[0;32m    186\u001b[0m     \u001b[1;31m# but it's overkill for just that one bit of state.\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    187\u001b[0m     \u001b[1;32mdef\u001b[0m \u001b[0mmagic_deco\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0marg\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 188\u001b[1;33m         \u001b[0mcall\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;32mlambda\u001b[0m \u001b[0mf\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m*\u001b[0m\u001b[0ma\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m**\u001b[0m\u001b[0mk\u001b[0m\u001b[1;33m:\u001b[0m \u001b[0mf\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m*\u001b[0m\u001b[0ma\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m**\u001b[0m\u001b[0mk\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    189\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    190\u001b[0m         \u001b[1;32mif\u001b[0m \u001b[0mcallable\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0marg\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mD:\\Base\\Anaconda3\\lib\\site-packages\\IPython\\core\\magics\\execution.py\u001b[0m in \u001b[0;36mtime\u001b[1;34m(self, line, cell, local_ns)\u001b[0m\n\u001b[0;32m   1183\u001b[0m         \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1184\u001b[0m             \u001b[0mst\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mclock2\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 1185\u001b[1;33m             \u001b[0mexec\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mcode\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mglob\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mlocal_ns\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   1186\u001b[0m             \u001b[0mend\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mclock2\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1187\u001b[0m             \u001b[0mout\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;32mNone\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m<timed exec>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m()\u001b[0m\n",
      "\u001b[1;32mD:\\Base\\Anaconda3\\lib\\site-packages\\pyswarms\\single\\global_best.py\u001b[0m in \u001b[0;36moptimize\u001b[1;34m(self, objective_func, iters, print_step, verbose)\u001b[0m\n\u001b[0;32m    132\u001b[0m             \u001b[1;31m# Compute cost for current position and personal best\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    133\u001b[0m             \u001b[0mcurrent_cost\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mobjective_func\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mpos\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 134\u001b[1;33m             \u001b[0mpbest_cost\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mobjective_func\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mpersonal_best_pos\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    135\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    136\u001b[0m             \u001b[1;31m# Update personal bests if the current position is better\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m<ipython-input-10-831e015e2d16>\u001b[0m in \u001b[0;36mf_2d\u001b[1;34m(x)\u001b[0m\n\u001b[0;32m     71\u001b[0m     \u001b[1;31m#print('selams')\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     72\u001b[0m     \u001b[0mn_particles\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mx\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mshape\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;36m0\u001b[0m\u001b[1;33m]\u001b[0m \u001b[1;31m# this can be parallelized...\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 73\u001b[1;33m     \u001b[0mj\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;33m[\u001b[0m\u001b[0mnetwork_sim_2d\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mx\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mi\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;32mfor\u001b[0m \u001b[0mi\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mn_particles\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     74\u001b[0m     \u001b[0mprint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mx\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;36m0\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     75\u001b[0m     \u001b[1;31m#print(x[1])\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m<ipython-input-10-831e015e2d16>\u001b[0m in \u001b[0;36m<listcomp>\u001b[1;34m(.0)\u001b[0m\n\u001b[0;32m     71\u001b[0m     \u001b[1;31m#print('selams')\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     72\u001b[0m     \u001b[0mn_particles\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mx\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mshape\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;36m0\u001b[0m\u001b[1;33m]\u001b[0m \u001b[1;31m# this can be parallelized...\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 73\u001b[1;33m     \u001b[0mj\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;33m[\u001b[0m\u001b[0mnetwork_sim_2d\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mx\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mi\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;32mfor\u001b[0m \u001b[0mi\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mn_particles\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     74\u001b[0m     \u001b[0mprint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mx\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;36m0\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     75\u001b[0m     \u001b[1;31m#print(x[1])\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m<ipython-input-10-831e015e2d16>\u001b[0m in \u001b[0;36mnetwork_sim_2d\u001b[1;34m(params)\u001b[0m\n\u001b[0;32m     43\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     44\u001b[0m     \u001b[0mS\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mupdate_node_coordinates\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mx_array\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0my_array\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0mh_array\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0mn_type\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;34m'basestation'\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 45\u001b[1;33m     \u001b[0mS\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mupdate_connections\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     46\u001b[0m     \u001b[0mG_met\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mS\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mG_dict\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;34m'combined'\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     47\u001b[0m     \u001b[1;31m# avg length of shortest paths, LINK WEIGHTED\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mD:\\Base\\Box Sync\\Python\\Robust Network\\PoC Optimization\\gently\\simulation.py\u001b[0m in \u001b[0;36mupdate_connections\u001b[1;34m(self)\u001b[0m\n\u001b[0;32m     35\u001b[0m         \u001b[0mcost_bs2n\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mcosts_dict\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;34m'basestation'\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     36\u001b[0m         \u001b[1;31m#!WARNING: This is a hacky solution until I implement the arbitrary numbers of node types in the future\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 37\u001b[1;33m         \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mG_dict\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mgen_sim_network\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mn_coordinates\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0mbs_coordinates\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0md_node\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0md_basestation\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mperturbation_list\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mcost_n2n\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0mcost_bs2n\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     38\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     39\u001b[0m     \u001b[1;31m# Think about where to put the movement_model harder\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mD:\\Base\\Box Sync\\Python\\Robust Network\\PoC Optimization\\gently\\network.py\u001b[0m in \u001b[0;36mgen_sim_network\u001b[1;34m(node_coordinates, basestation_coordinates, d_node, d_basestation, pert_list, cost_n2n, cost_bs2n)\u001b[0m\n\u001b[0;32m     42\u001b[0m         \u001b[1;32mfor\u001b[0m \u001b[0medge\u001b[0m \u001b[1;32min\u001b[0m \u001b[0medge_list\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     43\u001b[0m             \u001b[0mp_1\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mcombined_coordinates_list\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0medge\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;36m0\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 44\u001b[1;33m             \u001b[0mp_2\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mcombined_coordinates_list\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0medge\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;36m1\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     45\u001b[0m             \u001b[0mp_1\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mPoint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mp_1\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;36m0\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0mp_1\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;36m1\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0mp_1\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;36m2\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     46\u001b[0m             \u001b[0mp_2\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mPoint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mp_2\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;36m0\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0mp_2\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;36m1\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0mp_2\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;36m2\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mD:\\Base\\Box Sync\\Python\\Robust Network\\PoC Optimization\\gently\\physics.py\u001b[0m in \u001b[0;36mcheck_los\u001b[1;34m(pert_list, p1, p2, epsilon)\u001b[0m\n\u001b[0;32m     19\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     20\u001b[0m             \u001b[1;31m# if surface is not skipped, that means it's in the DANGER ZONE and should be checked.\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 21\u001b[1;33m             \u001b[0mcol_result\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0msurface\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mcheck_collision_line\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mp1\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0mp2\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0mepsilon\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     22\u001b[0m             \u001b[1;32mif\u001b[0m \u001b[0mcol_result\u001b[0m \u001b[1;32mis\u001b[0m \u001b[1;32mnot\u001b[0m \u001b[1;32mNone\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     23\u001b[0m                 \u001b[0mlos_obstruction_flag\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;32mTrue\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mD:\\Base\\Box Sync\\Python\\Robust Network\\PoC Optimization\\gently\\geometry.py\u001b[0m in \u001b[0;36mcheck_collision_line\u001b[1;34m(self, l_p1, l_p2, epsilon)\u001b[0m\n\u001b[0;32m    208\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    209\u001b[0m             \u001b[1;31m#print(c_p)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 210\u001b[1;33m             \u001b[1;31m#print(fac)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    211\u001b[0m             \u001b[1;31m# Check if the point is within the line segment\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    212\u001b[0m             \u001b[1;32mif\u001b[0m \u001b[1;36m0\u001b[0m \u001b[1;33m<=\u001b[0m \u001b[0msI\u001b[0m \u001b[1;33m<=\u001b[0m \u001b[1;36m1\u001b[0m\u001b[1;33m:\u001b[0m \u001b[1;31m#this is not working, figure out why or figure out a different way to implement this...\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mKeyboardInterrupt\u001b[0m: "
     ],
     "output_type": "error"
    }
   ],
   "source": [
    "%%time\n",
    "# bounds, not my best but I'm really sleepy...\n",
    "max_bound = 1 * np.ones(N_b*2)#xy bounds\n",
    "min_bound = np.zeros(N_b*2)   #xy bounds\n",
    "max_bound = np.append(max_bound,0.125) #h bound\n",
    "min_bound = np.append(min_bound,0.025)  #h bound\n",
    "\n",
    "bounds = (min_bound, max_bound)\n",
    "\n",
    "\n",
    "# Initialize swarm, arbitrary for now. This can also be searched for apparently...\n",
    "options = {'c1': 0.5, 'c2': 0.3, 'w':0.05} #optimizing this didn't work very well for some reason, check that again.\n",
    "\n",
    "# Call instance of PSO, also remember constraints\n",
    "dimensions = ((2 * N_b) + 1)\n",
    "optimizer = ps.single.GlobalBestPSO(n_particles=n_swarm, dimensions=dimensions, options=options, bounds=bounds)\n",
    "\n",
    "# Perform optimization # each iteration appears to be the same, warum?\n",
    "cost, pos = optimizer.optimize(f_2d, print_step=1, iters=n_iter, verbose=3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import pickle\n",
    "# Obtain the cost history\n",
    "cost_history_2d = optimizer.get_cost_history\n",
    "# Obtain the position history\n",
    "pos_history_2d = optimizer.get_pos_history\n",
    "# Obtain the velocity history\n",
    "velocity_history = optimizer.get_velocity_history\n",
    "\n",
    "pickle.dump((cost_history_2d,pos_history_2d,velocity_history),open('2d_opt_results_'+ file_name +'.p','wb'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "h = pos[-1] # height is the last element\n",
    "#print(params)\n",
    "#print(h)\n",
    "n_axes = 2 # only (x,y)\n",
    "n_aps = int((len(pos)-1)/n_axes)\n",
    "# excluding last element from reshape\n",
    "xy_array = pos[:-1].reshape((n_aps,n_axes)) \n",
    "#print(xy_array)\n",
    "x_array = xy_array[:,0]\n",
    "y_array = xy_array[:,1]\n",
    "#print(x_array)\n",
    "#print(y_array)\n",
    "h_array = np.empty(n_aps);\n",
    "h_array.fill(h) # create a list of same height (2D approx.) for all APs\n",
    "\n",
    "S.update_node_coordinates(x_array,y_array,h_array,n_type='basestation')\n",
    "S.update_connections()\n",
    "G_met = S.G_dict['combined']\n",
    "# avg length of shortest paths, LINK WEIGHTED \n",
    "# First check if the network is connected \n",
    "\n",
    "# Compute the cost\n",
    "conn_frac = mz.metrics.connected_fraction(G_met,True,S.coordinates_dict_base['basestation'].keys()) \n",
    "\n",
    "mz.visualization.visualize_stage(S,figure_name='2d_opt_result_' +file_name)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.18253968253968256\n",
      "Wall time: 12.2 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "# network_sim_2d appears to be working just fine, I'll reduce the node2node radius and we'll be goldern.\n",
    "foop = np.array([0.82,0.82,0.25,0.17,0.85,0.36,0.31,0.31,0.11]) # hand input 4 bs on 4 corners\n",
    "print(network_sim_2d(foop))\n",
    "mz.visualization.visualize_stage(S,figure_name='opt_test_2d')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(0.2093535478111632, 0.45249999999999996, 0.03413678893973993)\n"
     ]
    }
   ],
   "source": [
    "# Debug, check if p1 and p2 show up as coinciding\n",
    "p1 = (0.1734337,0.3603601,0.0.0425)\n",
    "p2 = (0.6460097,0.418961,0.04932695)\n",
    "n_type = 'node'\n",
    "node_point = mz.geometry.Point(p1[0],p1[1],p1[2])\n",
    "node_contained_flag = False\n",
    "for pert in S.perturbation_list:\n",
    "    node_contained_flag = pert.contains_point(node_point)\n",
    "    if node_contained_flag == True :\n",
    "        if n_type is 'node':\n",
    "            node = S.place_in_building(pert)\n",
    "            print('placing in building')\n",
    "        else: # else is only basestation for now\n",
    "            h_new = pert.base_origin.z + pert.height + 2*0.0025\n",
    "            node = (node[0],node[1],h_new) #this is where node coordinate is updated\n",
    "        break\n",
    "print(node)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Optimize!\n",
    "# First, move the whole AP plane 2x2D plane approximation\n",
    "#S_2d = copy.deepcopy(S) # This line explodes, apparently not deepcopiable...\n",
    "\n",
    "h_limits = (0.067,0.15) #@TODO this should be computed in the future\n",
    "y_limits = (0.01,0.99)\n",
    "x_limits = (0.01,0.99)\n",
    "\n",
    "#h_min = 0.105\n",
    "#h_max = 0.120 #0.10815789\n",
    "\n",
    "h_iter = 10 # actual iter count is much, much higher. Start with a coarse grid, move into a finer grid.\n",
    "xy_iter = 5\n",
    "\n",
    "h_grid = np.linspace(h_limits[0],h_limits[1],h_iter)\n",
    "y_grid = np.linspace(y_limits[0],y_limits[1],xy_iter)\n",
    "x_grid = np.linspace(x_limits[0],x_limits[1],xy_iter)\n",
    "\n",
    "grid_list = [x_grid, y_grid, h_grid ]\n",
    "\n",
    "coord_list = list(itertools.product(*grid_list))\n",
    "len(coord_list)\n",
    "\n",
    "n_bs = len(S.coordinates_dict['basestation'])\n",
    "coord_combinations = list(itertools.combinations(coord_list, n_bs))\n",
    "len(coord_combinations) # 158882750 combinations for K = 4, grid search is officially undoable.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def foo(bar):\n",
    "    print(S)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "best_score = 9999; #very badly practice heh\n",
    "best_h = 0;\n",
    "\n",
    "for x_k in x_grid:\n",
    "    for y_k in y_grid:\n",
    "        for h_k in h_grid:\n",
    "            # updating bs heights\n",
    "            h_k_list=np.empty(n_bs); h_k_list.fill(h_k) # create a list of same height (2D approx.) for all APs\n",
    "            # Here check if all basestation coordinates are viable, and continue to next coordinate if not.\n",
    "\n",
    "            S.update_node_coordinates(x_list=None,y_list=None,h_k_list,n_type='basestation') # this needs to become update node coordinates\n",
    "            S.update_connections()\n",
    "            G_met = S_2d.G_dict['combined']\n",
    "            # avg length of shortest paths, LINK WEIGHTED \n",
    "            # First check if the network is connected \n",
    "            conn_frac = mz.metrics.connected_fraction(G_met)\n",
    "            print('percentage connected:' + str(conn_frac))\n",
    "\n",
    "            if conn_frac > best_score: #not adding the equal here...\n",
    "                best_score = conn_frac\n",
    "                best_coordinates = (x_k,y_k,h_k)\n",
    "                print('Best so far! coordinates: ' + str(best_coordinates) + ' score: ' + str(best_score) )\n",
    "        \n",
    "print('[FINAL] Best coordinates: ' + str(best_coordinates) + ' score: ' + str(best_score) )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Height: 0.067\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 7.165992063492063\n",
      "Best so far! h: 0.067 score: 7.165992063492063\n",
      "Height: 0.07136842105263158\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 7.2168083965371945\n",
      "Height: 0.07573684210526316\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 7.447542313934616\n",
      "Height: 0.08010526315789473\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 7.321648324601807\n",
      "Height: 0.08447368421052631\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 7.960843793584379\n",
      "Height: 0.08884210526315789\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 7.481903765690377\n",
      "Height: 0.09321052631578948\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 6.90295358649789\n",
      "Best so far! h: 0.09321052631578948 score: 6.90295358649789\n",
      "Height: 0.09757894736842106\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 6.177420470262794\n",
      "Best so far! h: 0.09757894736842106 score: 6.177420470262794\n",
      "Height: 0.10194736842105263\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.8939336785432594\n",
      "Best so far! h: 0.10194736842105263 score: 5.8939336785432594\n",
      "Height: 0.10631578947368421\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.841946435307431\n",
      "Best so far! h: 0.10631578947368421 score: 5.841946435307431\n",
      "Height: 0.11068421052631579\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.808739569609135\n",
      "Best so far! h: 0.11068421052631579 score: 5.808739569609135\n",
      "Height: 0.11505263157894738\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.701706506054332\n",
      "Best so far! h: 0.11505263157894738 score: 5.701706506054332\n",
      "Height: 0.11942105263157896\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.662839575883054\n",
      "Best so far! h: 0.11942105263157896 score: 5.662839575883054\n",
      "Height: 0.12378947368421053\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.403086112493778\n",
      "Best so far! h: 0.12378947368421053 score: 5.403086112493778\n",
      "Height: 0.1281578947368421\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.618881699021072\n",
      "Height: 0.1325263157894737\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.631809354530792\n",
      "Height: 0.13689473684210526\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.564436266309802\n",
      "Height: 0.14126315789473684\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.540682502509201\n",
      "Height: 0.14563157894736842\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.643994647039143\n",
      "Height: 0.15\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.581197724991636\n",
      "Best h: 0.12378947368421053 score: 5.403086112493778\n"
     ]
    }
   ],
   "source": [
    "# Optimize!\n",
    "# First, move the whole AP plane 2x2D plane approximation\n",
    "S_2d = copy.deepcopy(S) # make a brand new instance for messing around with optimization\n",
    "\n",
    "h_limits = (0.067,0.15) #@TODO this should be computed in the future\n",
    "y_limits = (0.01,0.99)\n",
    "x_limits = (0.01,0.99)\n",
    "\n",
    "#h_min = 0.105\n",
    "#h_max = 0.120 #0.10815789\n",
    "\n",
    "n_iter = 20 # actual iter count is n_iter^3\n",
    "\n",
    "h_grid = np.linspace(h_limits[0],h_limits[1],n_iter)\n",
    "y_grid = np.linspace(y_limits[0],y_limits[1],n_iter)\n",
    "x_grid = np.linspace(x_limits[0],x_limits[1],n_iter)\n",
    "\n",
    "n_bs = len(S.coordinates_dict['basestation'])\n",
    "\n",
    "best_score = 9999; #very badly practice heh\n",
    "best_h = 0;\n",
    "\n",
    "for h_k in h_grid:\n",
    "    # updating bs heights\n",
    "    h_k_list=np.empty(n_bs); h_k_list.fill(h_k) # create a list of same height (2D approx.) for all APs\n",
    "    # Here check if all basestation coordinates are viable, and continue to next coordinate if not.\n",
    "    \n",
    "    S.update_node_coordinates(x_list=None,y_list=None,h_k_list,n_type='basestation') # this needs to become update node coordinates\n",
    "    S.update_connections()\n",
    "    G_met = S_2d.G_dict['combined']\n",
    "    # avg length of shortest paths, LINK WEIGHTED \n",
    "    # First check if the network is connected \n",
    "    is_conn = mz.metrics.is_connected(G_met)\n",
    "    # If the graph is not connected, check it for the giant component\n",
    "    if is_conn == True:\n",
    "        avg_sp_length = nx.average_shortest_path_length(G_met,'power_cost')\n",
    "        print('Link Cost - Average shortest path length: ' + str(avg_sp_length))\n",
    "    elif is_conn == False: \n",
    "        Gcc=sorted(nx.connected_component_subgraphs(G_met), key = len, reverse=True)\n",
    "        G0=Gcc[0] # Got the giant component\n",
    "        avg_sp_length = nx.average_shortest_path_length(G0,'power_cost')\n",
    "        print('Height: '+ str(h_k))\n",
    "        print('Warning: Graph not connecting, displaying the average shortest path length for the giant component')\n",
    "        print('Link Cost - Average shortest path cost: ' + str(avg_sp_length))\n",
    "    \n",
    "    if avg_sp_length <= best_score:\n",
    "        best_score = avg_sp_length\n",
    "        best_h = h_k\n",
    "        print('Best so far! h: ' + str(best_h) + ' score: ' + str(best_score) )\n",
    "        \n",
    "print('Best h: ' + str(best_h) + ' score: ' + str(best_score) )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'x': 0.2, 'y': 0.55, 'z': 0}\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "'file://D:\\\\Base\\\\Box Sync\\\\Python\\\\Robust Network\\\\PoC Simulation - 2D Planes vs Full 3D\\\\stage_2D_optimization_250.html'"
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "h_winrar=np.empty(n_bs); h_winrar.fill(best_h)\n",
    "S.update_node_heights(h_winrar,n_type='basestation')\n",
    "S.update_connections()\n",
    "G_vis = S.G_dict['combined']\n",
    "\n",
    "# Add these as node properties to G in the future\n",
    "#basestation_keys\n",
    "#node_keys\n",
    "# Params\n",
    "N_vis = len(G_vis)\n",
    "pos = nx.get_node_attributes(G_vis,'pos')\n",
    "\n",
    "pert_list = S.perturbation_list\n",
    "print(pert_list[0].base_origin)\n",
    "\n",
    "layout_dim = 1000 # have it always cubic for now\n",
    "d = 0.015\n",
    "s_col = \"rgb(30, 140, 15)\"\n",
    "\n",
    "Xn=[pos[k][0] for k in node_keys]# x-coordinates of nodes\n",
    "Yn=[pos[k][1] for k in node_keys]# y-coordinates\n",
    "Zn=[pos[k][2] for k in node_keys]# z-coordinates\n",
    "\n",
    "Xbs=[pos[k][0] for k in basestation_keys]# x-coordinates of nodes\n",
    "Ybs=[pos[k][1] for k in basestation_keys]# y-coordinates\n",
    "Zbs=[pos[k][2] for k in basestation_keys]# z-coordinates\n",
    "\n",
    "# Get the edge end point coordinates\n",
    "edge_list = G_vis.edges()\n",
    "Xe=[]\n",
    "Ye=[]\n",
    "Ze=[]\n",
    "for e in edge_list:\n",
    "    Xe+=[pos[e[0]][0],pos[e[1]][0], None]# x-coordinates of edge ends\n",
    "    Ye+=[pos[e[0]][1],pos[e[1]][1], None]\n",
    "    Ze+=[pos[e[0]][2],pos[e[1]][2], None]    \n",
    "\n",
    "\n",
    "node_size = d*layout_dim/math.pi\n",
    "\n",
    "trace_edges=go.Scatter3d(x=Xe,\n",
    "               y=Ye,\n",
    "               z=Ze,\n",
    "               mode='lines',\n",
    "               name='connections',\n",
    "               line=go.Line(color='rgba(125,125,125,0.6)', width=0.3)\n",
    "               )\n",
    "\n",
    "trace_nodes=go.Scatter3d(x=Xn,\n",
    "               y=Yn,\n",
    "               z=Zn,\n",
    "               mode='markers',\n",
    "               name='nodes',\n",
    "               marker=go.Marker(symbol='dot',\n",
    "                             size=node_size,\n",
    "                             line=go.Line(color='rgb(50,50,50)', width=0.5),\n",
    "                             color='rgba(125,0,0,0.9)'\n",
    "                             )\n",
    "               )\n",
    "\n",
    "trace_bs=go.Scatter3d(x=Xbs,\n",
    "               y=Ybs,\n",
    "               z=Zbs,\n",
    "               mode='markers',\n",
    "               name='basestations',\n",
    "               marker=go.Marker(symbol='dot',\n",
    "                             size=node_size*3,\n",
    "                             line=go.Line(color='rgb(50,50,50)', width=0.5),\n",
    "                             color='rgba(0,0,125,1)'\n",
    "                             )\n",
    "               )\n",
    "\n",
    "\n",
    "# This is defining a unit cube mesh, any rectangular prism can be achieved by \n",
    "# arithmetic on x, y, z vectors.\n",
    "# go through a list of cubes here.\n",
    "\n",
    "#mesh_cube_unit = { 'x': np.array([-0.5, -0.5, 0.5, 0.5, -0.5, -0.5, 0.5, 0.5]),\n",
    "#                   'y': np.array([-0.5, 0.5, 0.5, -0.5, -0.5, 0.5, -0.5, 0.5]),\n",
    "#                   'z': np.array([0, 0, 0, 0, 1, 1, 1, 1]),\n",
    "#                   'i': np.array([0, 0, 0, 1, 0, 3, 3, 7, 2, 5, 6, 7]),\n",
    "#                   'j': np.array([1, 2, 4, 4, 4, 4, 2, 6, 1, 7, 5, 5]),\n",
    "#                   'k': np.array([2, 3, 1, 5, 3, 6, 6, 2, 7, 1, 4, 6])\n",
    "#                 }\n",
    "\n",
    "\n",
    "mesh_list = [go.Mesh3d(pert.mesh, color='rgba(125,125,125,1)') for pert in pert_list]\n",
    "#mesh=go.Mesh3d(pert_1.mesh)\n",
    "\n",
    "x_range=np.linspace(0,1,100)\n",
    "y_range=np.linspace(0,1,100)\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "# Looks like X and Y is all confused\n",
    "h_surface = []\n",
    "for x_h in x_range:\n",
    "    h_temp = []\n",
    "    for y_h in y_range:\n",
    "        h_temp.append(S.terrain_function({'x':y_h,'y':x_h},S.terrain_params))\n",
    "    h_surface.append(h_temp)\n",
    "    \n",
    "s_col = \"rgb(30, 140, 15)\"\n",
    "\n",
    "trace_surface = go.Surface(\n",
    "        x=x_range,\n",
    "        y=y_range,\n",
    "        z=h_surface,\n",
    "        showscale = False,\n",
    "        autocolorscale = False, \n",
    "        cauto = False, \n",
    "        cmax = 0, \n",
    "        cmin = -0.5, \n",
    "        colorscale = [ [  0, s_col], [0.35, s_col], [0.5, s_col], \n",
    "                       [0.6, s_col], [0.7, s_col],  [1, s_col] ],\n",
    "        surfacecolor= s_col\n",
    "    )\n",
    "\n",
    "\n",
    "\n",
    "x_range=np.linspace(0,1,100)\n",
    "y_range=np.linspace(0,1,100)\n",
    "\n",
    "s_flat_range = {'low':-0.1,'high':1.1}\n",
    "top_height=0.1\n",
    "\n",
    "h_surface_flat = []\n",
    "for x_h in x_range:\n",
    "    h_temp = []\n",
    "    for y_h in y_range:\n",
    "        h_temp.append(surface_height_func({'x':x_h,'y':y_h},s_flat_range,top_height))\n",
    "    h_surface_flat.append(h_temp)\n",
    "    \n",
    "\n",
    "trace_surface_flat = go.Surface(\n",
    "        x=x_range,\n",
    "        y=y_range,\n",
    "        z=h_surface_flat,\n",
    "        showscale = False,\n",
    "        autocolorscale = False, \n",
    "        cauto = False, \n",
    "        cmax = 0, \n",
    "        cmin = -0.5, \n",
    "        colorscale = [ [  0, s_col], [0.35, s_col], [0.5, s_col], \n",
    "                       [0.6, s_col], [0.7, s_col],  [1, s_col] ],\n",
    "        surfacecolor= s_col\n",
    "        )\n",
    "    \n",
    "# size is a bit arbitrary, figure that out to match d with size\n",
    "axis=dict(autorange=False,\n",
    "          showbackground=False,\n",
    "          showline=True,\n",
    "          zeroline=True,\n",
    "          showgrid=True,\n",
    "          showticklabels=False,\n",
    "          range=[0,1],\n",
    "          title=''\n",
    "          )\n",
    "\n",
    "z_axis=dict(autorange=False,\n",
    "          showbackground=False,\n",
    "          showline=True,\n",
    "          zeroline=True,\n",
    "          showgrid=True,\n",
    "          showticklabels=False,\n",
    "          range=[0,0.2],\n",
    "          title='',\n",
    "          nticks=6,\n",
    "          )\n",
    "\n",
    "#camera = dict(\n",
    "#    up=dict(x=0, y=0, z=1),\n",
    "#    center=dict(x=0, y=0, z=0),\n",
    "#    eye=dict(x=1.75, y=-1.1, z=0.65)\n",
    "#)\n",
    "\n",
    "camera = dict(\n",
    "    up=dict(x=0, y=0, z=1),\n",
    "    center=dict(x=0, y=0, z=0),\n",
    "    eye=dict(x=2.0710389646410197, y=0.6609030190054094, z=0.707922068348940)\n",
    "    \n",
    ")     \n",
    "#        eye=dict(x=2.0710389646410197, y=0.6609030190054094, z=0.707922068348940)\n",
    "\n",
    "# Actualy, size of the thing is given as 1000*1000*1000 so size ends up d*1000,\n",
    "# or whatever the width and height is.\n",
    "aspect_ratio = dict(x=1.2,y=1.2,z=0.5)\n",
    "marg = dict(l=10,r=10,t=10,b=10,pad=0,autoexpand=True)\n",
    "#title=\"2D Deployment without Perturbations\",\n",
    "layout_2d_noper = go.Layout(\n",
    "         autosize=False,\n",
    "         width=layout_dim,\n",
    "         height=layout_dim,\n",
    "         showlegend=False,\n",
    "         scene=go.Scene(\n",
    "         xaxis=go.XAxis(axis),\n",
    "         yaxis=go.YAxis(axis),\n",
    "         zaxis=go.ZAxis(z_axis),\n",
    "         camera=camera,\n",
    "         aspectratio = aspect_ratio,\n",
    "         aspectmode = \"manual\",\n",
    "        ),\n",
    "     margin=marg,\n",
    "    hovermode='closest',\n",
    "   )\n",
    "\n",
    "#         title=\"2D Deployment with Perturbations\",\n",
    "layout_2d_yesper = go.Layout(\n",
    "         autosize=False,\n",
    "         width=layout_dim,\n",
    "         height=layout_dim,\n",
    "         showlegend=False,\n",
    "         scene=go.Scene(\n",
    "         xaxis=go.XAxis(axis),\n",
    "         yaxis=go.YAxis(axis),\n",
    "         zaxis=go.ZAxis(z_axis),\n",
    "         camera=camera,\n",
    "         aspectratio = aspect_ratio,\n",
    "         aspectmode = \"manual\",\n",
    "        ),\n",
    "     margin=marg,\n",
    "    hovermode='closest',\n",
    "    )\n",
    "\n",
    "\n",
    "#data=go.Data([trace_nodes, trace_bs] + mesh_list)\n",
    "data_2d_noper=go.Data([trace_nodes, trace_edges, trace_bs, trace_surface] + mesh_list) # trace_surface]  + mesh_list\n",
    "fig_2d_noper=go.Figure(data=data_2d_noper, layout=layout_2d_noper)\n",
    "\n",
    "py.offline.plot(fig_2d_noper, filename='stage_2D_optimization_'+str(N)+'.html')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5.403086112493778\n",
      "h_iter: 0.12378947368421053\n",
      "Height: 0.12378947368421053\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.403086112493778\n",
      "h_iter: 0.11797500000000001\n",
      "Height: 0.11797500000000001\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.403086112493778\n",
      "h_iter: 0.11216052631578947\n",
      "Height: 0.11216052631578947\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.4002322880371665\n",
      "Best so far! h: 0.11216052631578947 score: 5.4002322880371665\n",
      "h_iter: 0.10634605263157895\n",
      "Height: 0.10634605263157895\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.4002322880371665\n",
      "h_iter: 0.10053157894736843\n",
      "Height: 0.10053157894736843\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.4002322880371665\n",
      "h_iter: 0.09471710526315791\n",
      "Height: 0.09471710526315791\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.407267297162767\n",
      "h_iter: 0.08890263157894737\n",
      "Height: 0.08890263157894737\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.410320225651236\n",
      "h_iter: 0.08308815789473685\n",
      "Height: 0.08308815789473685\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.410320225651236\n",
      "h_iter: 0.07727368421052633\n",
      "Height: 0.07727368421052633\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.339704662352746\n",
      "Best so far! h: 0.07727368421052633 score: 5.339704662352746\n",
      "h_iter: 0.07145921052631579\n",
      "Height: 0.07145921052631579\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.351883192301311\n",
      "h_iter: 0.06564473684210527\n",
      "Height: 0.06564473684210527\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.362402521984404\n",
      "h_iter: 0.05983026315789475\n",
      "Height: 0.05983026315789475\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.3795918367346935\n",
      "h_iter: 0.05401578947368421\n",
      "Height: 0.05401578947368421\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.3795918367346935\n",
      "h_iter: 0.04820131578947369\n",
      "Height: 0.04820131578947369\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.385133565621371\n",
      "h_iter: 0.04238684210526317\n",
      "Height: 0.04238684210526317\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.4023892483822795\n",
      "h_iter: 0.03657236842105263\n",
      "Height: 0.03657236842105263\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.458370665339307\n",
      "h_iter: 0.03075789473684211\n",
      "Height: 0.03075789473684211\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.458370665339307\n",
      "h_iter: 0.024943421052631587\n",
      "Height: 0.024943421052631587\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.458370665339307\n",
      "h_iter: 0.01912894736842105\n",
      "Height: 0.01912894736842105\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.458370665339307\n",
      "h_iter: 0.013314473684210529\n",
      "Height: 0.013314473684210529\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.459233449477352\n",
      "Best h: 0.12378947368421053 score: 5.339704662352746\n",
      "h_iter: 0.12378947368421053\n",
      "Height: 0.12378947368421053\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.339704662352746\n",
      "h_iter: 0.11797500000000001\n",
      "Height: 0.11797500000000001\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.339704662352746\n",
      "h_iter: 0.11216052631578947\n",
      "Height: 0.11216052631578947\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.332205077152812\n",
      "Best so far! h: 0.11216052631578947 score: 5.332205077152812\n",
      "h_iter: 0.10634605263157895\n",
      "Height: 0.10634605263157895\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.332205077152812\n",
      "h_iter: 0.10053157894736843\n",
      "Height: 0.10053157894736843\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.332205077152812\n",
      "h_iter: 0.09471710526315791\n",
      "Height: 0.09471710526315791\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.317229842756775\n",
      "Best so far! h: 0.09471710526315791 score: 5.317229842756775\n",
      "h_iter: 0.08890263157894737\n",
      "Height: 0.08890263157894737\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.317229842756775\n",
      "h_iter: 0.08308815789473685\n",
      "Height: 0.08308815789473685\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.229274004683841\n",
      "Best so far! h: 0.08308815789473685 score: 5.229274004683841\n",
      "h_iter: 0.07727368421052633\n",
      "Height: 0.07727368421052633\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.22485781197725\n",
      "Best so far! h: 0.07727368421052633 score: 5.22485781197725\n",
      "h_iter: 0.07145921052631579\n",
      "Height: 0.07145921052631579\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.22485781197725\n",
      "h_iter: 0.06564473684210527\n",
      "Height: 0.06564473684210527\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.212713282034125\n",
      "Best so far! h: 0.06564473684210527 score: 5.212713282034125\n",
      "h_iter: 0.05983026315789475\n",
      "Height: 0.05983026315789475\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.2207427233188355\n",
      "h_iter: 0.05401578947368421\n",
      "Height: 0.05401578947368421\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.231850117096019\n",
      "h_iter: 0.04820131578947369\n",
      "Height: 0.04820131578947369\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.2119171513110905\n",
      "Best so far! h: 0.04820131578947369 score: 5.2119171513110905\n",
      "h_iter: 0.04238684210526317\n",
      "Height: 0.04238684210526317\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.2119171513110905\n",
      "h_iter: 0.03657236842105263\n",
      "Height: 0.03657236842105263\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.214263850627487\n",
      "h_iter: 0.03075789473684211\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Height: 0.03075789473684211\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.230856280648812\n",
      "h_iter: 0.024943421052631587\n",
      "Height: 0.024943421052631587\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.279757209972223\n",
      "h_iter: 0.01912894736842105\n",
      "Height: 0.01912894736842105\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.289187613593498\n",
      "h_iter: 0.013314473684210529\n",
      "Height: 0.013314473684210529\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.291450910462604\n",
      "Best h: 0.12378947368421053 score: 5.2119171513110905\n",
      "h_iter: 0.12378947368421053\n",
      "Height: 0.12378947368421053\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.2119171513110905\n",
      "h_iter: 0.11797500000000001\n",
      "Height: 0.11797500000000001\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.2119171513110905\n",
      "h_iter: 0.11216052631578947\n",
      "Height: 0.11216052631578947\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.2119171513110905\n",
      "h_iter: 0.10634605263157895\n",
      "Height: 0.10634605263157895\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.2119171513110905\n",
      "h_iter: 0.10053157894736843\n",
      "Height: 0.10053157894736843\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.2119171513110905\n",
      "h_iter: 0.09471710526315791\n",
      "Height: 0.09471710526315791\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.2119171513110905\n",
      "h_iter: 0.08890263157894737\n",
      "Height: 0.08890263157894737\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.2119171513110905\n",
      "h_iter: 0.08308815789473685\n",
      "Height: 0.08308815789473685\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.2119171513110905\n",
      "h_iter: 0.07727368421052633\n",
      "Height: 0.07727368421052633\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.2119171513110905\n",
      "h_iter: 0.07145921052631579\n",
      "Height: 0.07145921052631579\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.2119171513110905\n",
      "h_iter: 0.06564473684210527\n",
      "Skipping\n",
      "h_iter: 0.05983026315789475\n",
      "Skipping\n",
      "h_iter: 0.05401578947368421\n",
      "Skipping\n",
      "h_iter: 0.04820131578947369\n",
      "Skipping\n",
      "h_iter: 0.04238684210526317\n",
      "Skipping\n",
      "h_iter: 0.03657236842105263\n",
      "Skipping\n",
      "h_iter: 0.03075789473684211\n",
      "Skipping\n",
      "h_iter: 0.024943421052631587\n",
      "Skipping\n",
      "h_iter: 0.01912894736842105\n",
      "Skipping\n",
      "h_iter: 0.013314473684210529\n",
      "Skipping\n",
      "Best h: 0.12378947368421053 score: 5.2119171513110905\n",
      "h_iter: 0.12378947368421053\n",
      "Height: 0.12378947368421053\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.2119171513110905\n",
      "h_iter: 0.11797500000000001\n",
      "Height: 0.11797500000000001\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.20613542835765\n",
      "Best so far! h: 0.11797500000000001 score: 5.20613542835765\n",
      "h_iter: 0.11216052631578947\n",
      "Height: 0.11216052631578947\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.18583137775057\n",
      "Best so far! h: 0.11216052631578947 score: 5.18583137775057\n",
      "h_iter: 0.10634605263157895\n",
      "Height: 0.10634605263157895\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.18583137775057\n",
      "h_iter: 0.10053157894736843\n",
      "Height: 0.10053157894736843\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.167227833894501\n",
      "Best so far! h: 0.10053157894736843 score: 5.167227833894501\n",
      "h_iter: 0.09471710526315791\n",
      "Height: 0.09471710526315791\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.178519198721219\n",
      "h_iter: 0.08890263157894737\n",
      "Height: 0.08890263157894737\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.213850786216126\n",
      "h_iter: 0.08308815789473685\n",
      "Height: 0.08308815789473685\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.205720976915356\n",
      "h_iter: 0.07727368421052633\n",
      "Height: 0.07727368421052633\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.198475342373339\n",
      "h_iter: 0.07145921052631579\n",
      "Height: 0.07145921052631579\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.198475342373339\n",
      "h_iter: 0.06564473684210527\n",
      "Height: 0.06564473684210527\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.18535029662906\n",
      "h_iter: 0.05983026315789475\n",
      "Height: 0.05983026315789475\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.18535029662906\n",
      "h_iter: 0.05401578947368421\n",
      "Skipping\n",
      "h_iter: 0.04820131578947369\n",
      "Skipping\n",
      "h_iter: 0.04238684210526317\n",
      "Skipping\n",
      "h_iter: 0.03657236842105263\n",
      "Skipping\n",
      "h_iter: 0.03075789473684211\n",
      "Skipping\n",
      "h_iter: 0.024943421052631587\n",
      "Skipping\n",
      "h_iter: 0.01912894736842105\n",
      "Skipping\n",
      "h_iter: 0.013314473684210529\n",
      "Skipping\n",
      "Best h: 0.12378947368421053 score: 5.167227833894501\n",
      "h_iter: 0.12378947368421053\n",
      "Height: 0.12378947368421053\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.167227833894501\n",
      "h_iter: 0.11797500000000001\n",
      "Height: 0.11797500000000001\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.465638554216867\n",
      "h_iter: 0.11216052631578947\n",
      "Height: 0.11216052631578947\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.626795180722891\n",
      "h_iter: 0.10634605263157895\n",
      "Height: 0.10634605263157895\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.647164305052565\n",
      "h_iter: 0.10053157894736843\n",
      "Height: 0.10053157894736843\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.751344889420203\n",
      "h_iter: 0.09471710526315791\n",
      "Height: 0.09471710526315791\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 6.363220433823772\n",
      "h_iter: 0.08890263157894737\n",
      "Height: 0.08890263157894737\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 6.57292386689049\n",
      "h_iter: 0.08308815789473685\n",
      "Height: 0.08308815789473685\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 6.990224153253679\n",
      "h_iter: 0.07727368421052633\n",
      "Skipping\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "h_iter: 0.07145921052631579\n",
      "Skipping\n",
      "h_iter: 0.06564473684210527\n",
      "Skipping\n",
      "h_iter: 0.05983026315789475\n",
      "Skipping\n",
      "h_iter: 0.05401578947368421\n",
      "Skipping\n",
      "h_iter: 0.04820131578947369\n",
      "Skipping\n",
      "h_iter: 0.04238684210526317\n",
      "Skipping\n",
      "h_iter: 0.03657236842105263\n",
      "Skipping\n",
      "h_iter: 0.03075789473684211\n",
      "Skipping\n",
      "h_iter: 0.024943421052631587\n",
      "Skipping\n",
      "h_iter: 0.01912894736842105\n",
      "Skipping\n",
      "h_iter: 0.013314473684210529\n",
      "Skipping\n",
      "Best h: 0.12378947368421053 score: 5.167227833894501\n",
      "h_iter: 0.12378947368421053\n",
      "Height: 0.12378947368421053\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.167227833894501\n",
      "h_iter: 0.11797500000000001\n",
      "Height: 0.11797500000000001\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.167227833894501\n",
      "h_iter: 0.11216052631578947\n",
      "Height: 0.11216052631578947\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.170866918341666\n",
      "h_iter: 0.10634605263157895\n",
      "Height: 0.10634605263157895\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.170866918341666\n",
      "h_iter: 0.10053157894736843\n",
      "Height: 0.10053157894736843\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.170866918341666\n",
      "h_iter: 0.09471710526315791\n",
      "Height: 0.09471710526315791\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.171309050096929\n",
      "h_iter: 0.08890263157894737\n",
      "Height: 0.08890263157894737\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.171309050096929\n",
      "h_iter: 0.08308815789473685\n",
      "Height: 0.08308815789473685\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.171309050096929\n",
      "h_iter: 0.07727368421052633\n",
      "Height: 0.07727368421052633\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.171309050096929\n",
      "h_iter: 0.07145921052631579\n",
      "Height: 0.07145921052631579\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.162602455531749\n",
      "Best so far! h: 0.07145921052631579 score: 5.162602455531749\n",
      "h_iter: 0.06564473684210527\n",
      "Skipping\n",
      "h_iter: 0.05983026315789475\n",
      "Skipping\n",
      "h_iter: 0.05401578947368421\n",
      "Skipping\n",
      "h_iter: 0.04820131578947369\n",
      "Skipping\n",
      "h_iter: 0.04238684210526317\n",
      "Skipping\n",
      "h_iter: 0.03657236842105263\n",
      "Skipping\n",
      "h_iter: 0.03075789473684211\n",
      "Skipping\n",
      "h_iter: 0.024943421052631587\n",
      "Skipping\n",
      "h_iter: 0.01912894736842105\n",
      "Skipping\n",
      "h_iter: 0.013314473684210529\n",
      "Skipping\n",
      "Best h: 0.12378947368421053 score: 5.162602455531749\n",
      "h_iter: 0.12378947368421053\n",
      "Height: 0.12378947368421053\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.162602455531749\n",
      "h_iter: 0.11797500000000001\n",
      "Height: 0.11797500000000001\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.157602965683774\n",
      "Best so far! h: 0.11797500000000001 score: 5.157602965683774\n",
      "h_iter: 0.11216052631578947\n",
      "Height: 0.11216052631578947\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.156276570417985\n",
      "Best so far! h: 0.11216052631578947 score: 5.156276570417985\n",
      "h_iter: 0.10634605263157895\n",
      "Height: 0.10634605263157895\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.156276570417985\n",
      "h_iter: 0.10053157894736843\n",
      "Height: 0.10053157894736843\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.156276570417985\n",
      "h_iter: 0.09471710526315791\n",
      "Height: 0.09471710526315791\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.156276570417985\n",
      "h_iter: 0.08890263157894737\n",
      "Height: 0.08890263157894737\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.156276570417985\n",
      "h_iter: 0.08308815789473685\n",
      "Height: 0.08308815789473685\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.156276570417985\n",
      "h_iter: 0.07727368421052633\n",
      "Height: 0.07727368421052633\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.156276570417985\n",
      "h_iter: 0.07145921052631579\n",
      "Height: 0.07145921052631579\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.154950175152195\n",
      "Best so far! h: 0.07145921052631579 score: 5.154950175152195\n",
      "h_iter: 0.06564473684210527\n",
      "Height: 0.06564473684210527\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.148114138013128\n",
      "Best so far! h: 0.06564473684210527 score: 5.148114138013128\n",
      "h_iter: 0.05983026315789475\n",
      "Height: 0.05983026315789475\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.148114138013128\n",
      "h_iter: 0.05401578947368421\n",
      "Height: 0.05401578947368421\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.148114138013128\n",
      "h_iter: 0.04820131578947369\n",
      "Height: 0.04820131578947369\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.148114138013128\n",
      "h_iter: 0.04238684210526317\n",
      "Height: 0.04238684210526317\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.148114138013128\n",
      "h_iter: 0.03657236842105263\n",
      "Height: 0.03657236842105263\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.148114138013128\n",
      "h_iter: 0.03075789473684211\n",
      "Height: 0.03075789473684211\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.154950175152195\n",
      "h_iter: 0.024943421052631587\n",
      "Height: 0.024943421052631587\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.154950175152195\n",
      "h_iter: 0.01912894736842105\n",
      "Height: 0.01912894736842105\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.156276570417985\n",
      "h_iter: 0.013314473684210529\n",
      "Height: 0.013314473684210529\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.16576539808863\n",
      "Best h: 0.12378947368421053 score: 5.148114138013128\n",
      "h_iter: 0.12378947368421053\n",
      "Height: 0.12378947368421053\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.148114138013128\n",
      "h_iter: 0.11797500000000001\n",
      "Height: 0.11797500000000001\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.148114138013128\n",
      "h_iter: 0.11216052631578947\n",
      "Height: 0.11216052631578947\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.1493385028738565\n",
      "h_iter: 0.10634605263157895\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Height: 0.10634605263157895\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.147161854232562\n",
      "Best so far! h: 0.10634605263157895 score: 5.147161854232562\n",
      "h_iter: 0.10053157894736843\n",
      "Height: 0.10053157894736843\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.147161854232562\n",
      "h_iter: 0.09471710526315791\n",
      "Height: 0.09471710526315791\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.147161854232562\n",
      "h_iter: 0.08890263157894737\n",
      "Height: 0.08890263157894737\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.147161854232562\n",
      "h_iter: 0.08308815789473685\n",
      "Height: 0.08308815789473685\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.147161854232562\n",
      "h_iter: 0.07727368421052633\n",
      "Height: 0.07727368421052633\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.140631908308676\n",
      "Best so far! h: 0.07727368421052633 score: 5.140631908308676\n",
      "h_iter: 0.07145921052631579\n",
      "Height: 0.07145921052631579\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.140631908308676\n",
      "h_iter: 0.06564473684210527\n",
      "Height: 0.06564473684210527\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.129625184321525\n",
      "Best so far! h: 0.06564473684210527 score: 5.129625184321525\n",
      "h_iter: 0.05983026315789475\n",
      "Height: 0.05983026315789475\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.128424951133363\n",
      "Best so far! h: 0.05983026315789475 score: 5.128424951133363\n",
      "h_iter: 0.05401578947368421\n",
      "Height: 0.05401578947368421\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.127807688350879\n",
      "Best so far! h: 0.05401578947368421 score: 5.127807688350879\n",
      "h_iter: 0.04820131578947369\n",
      "Height: 0.04820131578947369\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.129042213915847\n",
      "h_iter: 0.04238684210526317\n",
      "Height: 0.04238684210526317\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.129042213915847\n",
      "h_iter: 0.03657236842105263\n",
      "Height: 0.03657236842105263\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.130208154727204\n",
      "h_iter: 0.03075789473684211\n",
      "Height: 0.03075789473684211\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.1294537224375025\n",
      "h_iter: 0.024943421052631587\n",
      "Height: 0.024943421052631587\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.15229244538939\n",
      "h_iter: 0.01912894736842105\n",
      "Height: 0.01912894736842105\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.157367717156476\n",
      "h_iter: 0.013314473684210529\n",
      "Height: 0.013314473684210529\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.157744933301327\n",
      "Best h: 0.12378947368421053 score: 5.127807688350879\n",
      "h_iter: 0.12378947368421053\n",
      "Height: 0.12378947368421053\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.127807688350879\n",
      "h_iter: 0.11797500000000001\n",
      "Height: 0.11797500000000001\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.127807688350879\n",
      "h_iter: 0.11216052631578947\n",
      "Height: 0.11216052631578947\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.127807688350879\n",
      "h_iter: 0.10634605263157895\n",
      "Height: 0.10634605263157895\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.127807688350879\n",
      "h_iter: 0.10053157894736843\n",
      "Height: 0.10053157894736843\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.127807688350879\n",
      "h_iter: 0.09471710526315791\n",
      "Height: 0.09471710526315791\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.127807688350879\n",
      "h_iter: 0.08890263157894737\n",
      "Height: 0.08890263157894737\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.127807688350879\n",
      "h_iter: 0.08308815789473685\n",
      "Height: 0.08308815789473685\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.127807688350879\n",
      "h_iter: 0.07727368421052633\n",
      "Height: 0.07727368421052633\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.127807688350879\n",
      "h_iter: 0.07145921052631579\n",
      "Height: 0.07145921052631579\n",
      "Warning: Graph not connecting, displaying the average shortest path length for the giant component\n",
      "Link Cost - Average shortest path cost: 5.127807688350879\n",
      "h_iter: 0.06564473684210527\n",
      "Skipping\n",
      "h_iter: 0.05983026315789475\n",
      "Skipping\n",
      "h_iter: 0.05401578947368421\n",
      "Skipping\n",
      "h_iter: 0.04820131578947369\n",
      "Skipping\n",
      "h_iter: 0.04238684210526317\n",
      "Skipping\n",
      "h_iter: 0.03657236842105263\n",
      "Skipping\n",
      "h_iter: 0.03075789473684211\n",
      "Skipping\n",
      "h_iter: 0.024943421052631587\n",
      "Skipping\n",
      "h_iter: 0.01912894736842105\n",
      "Skipping\n",
      "h_iter: 0.013314473684210529\n",
      "Skipping\n",
      "Best h: 0.12378947368421053 score: 5.127807688350879\n"
     ]
    }
   ],
   "source": [
    "# Second, move the AP heights individually, these are going to slam themselves on the surface most likely\n",
    "# for this one, need conformed heights as starting points.\n",
    "# create the grid for the largest one, than remove points that do not work.\n",
    "\n",
    "S.conform_node_heights(['basestation'])\n",
    "bs_coords = S.coordinates_dict['basestation']\n",
    "\n",
    "h_base = [bs_coords[idx][2] for idx in bs_coords]\n",
    "\n",
    "# Start with 2D solution\n",
    "n_bs = len(h_base)\n",
    "h_init = np.empty(n_bs); \n",
    "h_init.fill(best_h)\n",
    "\n",
    "h_min = min(h_base)\n",
    "h_max = best_h # comes up pretty high so might as well do this.\n",
    "n_iter = 20; #with independent suboptimal solution, this is 9x20 = 180 iters, a bit less because unviable ones are erased.\n",
    "\n",
    "h_step = (h_max - h_min)/n_iter;\n",
    "\n",
    "h_best_list = h_init.copy()\n",
    "print(best_score)\n",
    "for bs_idx in np.arange(n_bs):\n",
    "    for iter_count in np.arange(n_iter):\n",
    "        h_iter = h_init[bs_idx] - np.floor(iter_count)*h_step \n",
    "        print('h_iter: ' +str(h_iter))\n",
    "        if h_iter <= h_base[bs_idx]:\n",
    "            print('Skipping')\n",
    "            continue\n",
    "        \n",
    "        h_iter_list = h_best_list.copy()\n",
    "        h_iter_list[bs_idx] = h_iter;\n",
    "        \n",
    "        \n",
    "        S.update_node_heights(h_iter_list,n_type='basestation')\n",
    "        S.update_connections()\n",
    "        G_met = S.G_dict['combined']\n",
    "        # avg length of shortest paths, LINK WEIGHTED \n",
    "        # First check if the network is connected \n",
    "        is_conn = nx.is_connected(G_met)\n",
    "        # If the graph is not connected, check it for the giant component\n",
    "        if is_conn == True:\n",
    "            avg_sp_length = nx.average_shortest_path_length(G_met,'power_cost')\n",
    "            print('Link Cost - Average shortest path length: ' + str(avg_sp_length))\n",
    "        elif is_conn == False: \n",
    "            Gcc=sorted(nx.connected_component_subgraphs(G_met), key = len, reverse=True)\n",
    "            G0=Gcc[0] # Got the giant component\n",
    "            avg_sp_length = nx.average_shortest_path_length(G0,'power_cost')\n",
    "            print('Height: '+ str(h_iter))\n",
    "            print('Warning: Graph not connecting, displaying the average shortest path length for the giant component')\n",
    "            print('Link Cost - Average shortest path cost: ' + str(avg_sp_length))\n",
    "    \n",
    "        if avg_sp_length < best_score:\n",
    "            best_score = avg_sp_length\n",
    "            h_best_list[bs_idx] = h_iter\n",
    "            print('Best so far! h: ' + str(h_iter) + ' score: ' + str(best_score) )\n",
    "        \n",
    "    print('Best h: ' + str(best_h) + ' score: ' + str(best_score) )\n",
    "        \n",
    "# Third, maybe add a minimum height so more flying."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'x': 0.2, 'y': 0.55, 'z': 0}\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "'file://D:\\\\Base\\\\Box Sync\\\\Python\\\\Robust Network\\\\PoC Simulation - 2D Planes vs Full 3D\\\\stage_3D_optimization_250.html'"
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "S.update_node_heights(h_best_list,n_type='basestation')\n",
    "S.update_connections()\n",
    "G_vis = S.G_dict['combined']\n",
    "\n",
    "# Add these as node properties to G in the future\n",
    "#basestation_keys\n",
    "#node_keys\n",
    "# Params\n",
    "N_vis = len(G_vis)\n",
    "pos = nx.get_node_attributes(G_vis,'pos')\n",
    "\n",
    "pert_list = S.perturbation_list\n",
    "print(pert_list[0].base_origin)\n",
    "\n",
    "layout_dim = 1000 # have it always cubic for now\n",
    "d = 0.015\n",
    "s_col = \"rgb(30, 140, 15)\"\n",
    "\n",
    "Xn=[pos[k][0] for k in node_keys]# x-coordinates of nodes\n",
    "Yn=[pos[k][1] for k in node_keys]# y-coordinates\n",
    "Zn=[pos[k][2] for k in node_keys]# z-coordinates\n",
    "\n",
    "Xbs=[pos[k][0] for k in basestation_keys]# x-coordinates of nodes\n",
    "Ybs=[pos[k][1] for k in basestation_keys]# y-coordinates\n",
    "Zbs=[pos[k][2] for k in basestation_keys]# z-coordinates\n",
    "\n",
    "# Get the edge end point coordinates\n",
    "edge_list = G_vis.edges()\n",
    "Xe=[]\n",
    "Ye=[]\n",
    "Ze=[]\n",
    "for e in edge_list:\n",
    "    Xe+=[pos[e[0]][0],pos[e[1]][0], None]# x-coordinates of edge ends\n",
    "    Ye+=[pos[e[0]][1],pos[e[1]][1], None]\n",
    "    Ze+=[pos[e[0]][2],pos[e[1]][2], None]    \n",
    "\n",
    "\n",
    "node_size = d*layout_dim/math.pi\n",
    "\n",
    "trace_edges=go.Scatter3d(x=Xe,\n",
    "               y=Ye,\n",
    "               z=Ze,\n",
    "               mode='lines',\n",
    "               name='connections',\n",
    "               line=go.Line(color='rgba(125,125,125,0.6)', width=0.3)\n",
    "               )\n",
    "\n",
    "trace_nodes=go.Scatter3d(x=Xn,\n",
    "               y=Yn,\n",
    "               z=Zn,\n",
    "               mode='markers',\n",
    "               name='nodes',\n",
    "               marker=go.Marker(symbol='dot',\n",
    "                             size=node_size,\n",
    "                             line=go.Line(color='rgb(50,50,50)', width=0.5),\n",
    "                             color='rgba(125,0,0,0.9)'\n",
    "                             )\n",
    "               )\n",
    "\n",
    "trace_bs=go.Scatter3d(x=Xbs,\n",
    "               y=Ybs,\n",
    "               z=Zbs,\n",
    "               mode='markers',\n",
    "               name='basestations',\n",
    "               marker=go.Marker(symbol='dot',\n",
    "                             size=node_size*3,\n",
    "                             line=go.Line(color='rgb(50,50,50)', width=0.5),\n",
    "                             color='rgba(0,0,125,1)'\n",
    "                             )\n",
    "               )\n",
    "\n",
    "\n",
    "# This is defining a unit cube mesh, any rectangular prism can be achieved by \n",
    "# arithmetic on x, y, z vectors.\n",
    "# go through a list of cubes here.\n",
    "\n",
    "#mesh_cube_unit = { 'x': np.array([-0.5, -0.5, 0.5, 0.5, -0.5, -0.5, 0.5, 0.5]),\n",
    "#                   'y': np.array([-0.5, 0.5, 0.5, -0.5, -0.5, 0.5, -0.5, 0.5]),\n",
    "#                   'z': np.array([0, 0, 0, 0, 1, 1, 1, 1]),\n",
    "#                   'i': np.array([0, 0, 0, 1, 0, 3, 3, 7, 2, 5, 6, 7]),\n",
    "#                   'j': np.array([1, 2, 4, 4, 4, 4, 2, 6, 1, 7, 5, 5]),\n",
    "#                   'k': np.array([2, 3, 1, 5, 3, 6, 6, 2, 7, 1, 4, 6])\n",
    "#                 }\n",
    "\n",
    "\n",
    "mesh_list = [go.Mesh3d(pert.mesh, color='rgba(125,125,125,1)') for pert in pert_list]\n",
    "#mesh=go.Mesh3d(pert_1.mesh)\n",
    "\n",
    "x_range=np.linspace(0,1,100)\n",
    "y_range=np.linspace(0,1,100)\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "# Looks like X and Y is all confused\n",
    "h_surface = []\n",
    "for x_h in x_range:\n",
    "    h_temp = []\n",
    "    for y_h in y_range:\n",
    "        h_temp.append(S.terrain_function({'x':y_h,'y':x_h},S.terrain_params))\n",
    "    h_surface.append(h_temp)\n",
    "    \n",
    "s_col = \"rgb(30, 140, 15)\"\n",
    "\n",
    "trace_surface = go.Surface(\n",
    "        x=x_range,\n",
    "        y=y_range,\n",
    "        z=h_surface,\n",
    "        showscale = False,\n",
    "        autocolorscale = False, \n",
    "        cauto = False, \n",
    "        cmax = 0, \n",
    "        cmin = -0.5, \n",
    "        colorscale = [ [  0, s_col], [0.35, s_col], [0.5, s_col], \n",
    "                       [0.6, s_col], [0.7, s_col],  [1, s_col] ],\n",
    "        surfacecolor= s_col\n",
    "    )\n",
    "\n",
    "\n",
    "\n",
    "x_range=np.linspace(0,1,100)\n",
    "y_range=np.linspace(0,1,100)\n",
    "\n",
    "s_flat_range = {'low':-0.1,'high':1.1}\n",
    "top_height=0.1\n",
    "\n",
    "h_surface_flat = []\n",
    "for x_h in x_range:\n",
    "    h_temp = []\n",
    "    for y_h in y_range:\n",
    "        h_temp.append(surface_height_func({'x':x_h,'y':y_h},s_flat_range,top_height))\n",
    "    h_surface_flat.append(h_temp)\n",
    "    \n",
    "\n",
    "trace_surface_flat = go.Surface(\n",
    "        x=x_range,\n",
    "        y=y_range,\n",
    "        z=h_surface_flat,\n",
    "        showscale = False,\n",
    "        autocolorscale = False, \n",
    "        cauto = False, \n",
    "        cmax = 0, \n",
    "        cmin = -0.5, \n",
    "        colorscale = [ [  0, s_col], [0.35, s_col], [0.5, s_col], \n",
    "                       [0.6, s_col], [0.7, s_col],  [1, s_col] ],\n",
    "        surfacecolor= s_col\n",
    "        )\n",
    "    \n",
    "# size is a bit arbitrary, figure that out to match d with size\n",
    "axis=dict(autorange=False,\n",
    "          showbackground=False,\n",
    "          showline=True,\n",
    "          zeroline=True,\n",
    "          showgrid=True,\n",
    "          showticklabels=False,\n",
    "          range=[0,1],\n",
    "          title=''\n",
    "          )\n",
    "\n",
    "z_axis=dict(autorange=False,\n",
    "          showbackground=False,\n",
    "          showline=True,\n",
    "          zeroline=True,\n",
    "          showgrid=True,\n",
    "          showticklabels=False,\n",
    "          range=[0,0.2],\n",
    "          title='',\n",
    "          nticks=6,\n",
    "          )\n",
    "\n",
    "#camera = dict(\n",
    "#    up=dict(x=0, y=0, z=1),\n",
    "#    center=dict(x=0, y=0, z=0),\n",
    "#    eye=dict(x=1.75, y=-1.1, z=0.65)\n",
    "#)\n",
    "\n",
    "camera = dict(\n",
    "    up=dict(x=0, y=0, z=1),\n",
    "    center=dict(x=0, y=0, z=0),\n",
    "    eye=dict(x=2.0710389646410197, y=0.6609030190054094, z=0.707922068348940)\n",
    "    \n",
    ")     \n",
    "#        eye=dict(x=2.0710389646410197, y=0.6609030190054094, z=0.707922068348940)\n",
    "\n",
    "# Actualy, size of the thing is given as 1000*1000*1000 so size ends up d*1000,\n",
    "# or whatever the width and height is.\n",
    "aspect_ratio = dict(x=1.2,y=1.2,z=0.5)\n",
    "marg = dict(l=10,r=10,t=10,b=10,pad=0,autoexpand=True)\n",
    "#title=\"2D Deployment without Perturbations\",\n",
    "layout_2d_noper = go.Layout(\n",
    "         autosize=False,\n",
    "         width=layout_dim,\n",
    "         height=layout_dim,\n",
    "         showlegend=False,\n",
    "         scene=go.Scene(\n",
    "         xaxis=go.XAxis(axis),\n",
    "         yaxis=go.YAxis(axis),\n",
    "         zaxis=go.ZAxis(z_axis),\n",
    "         camera=camera,\n",
    "         aspectratio = aspect_ratio,\n",
    "         aspectmode = \"manual\",\n",
    "        ),\n",
    "     margin=marg,\n",
    "    hovermode='closest',\n",
    "   )\n",
    "\n",
    "#         title=\"2D Deployment with Perturbations\",\n",
    "layout_2d_yesper = go.Layout(\n",
    "         autosize=False,\n",
    "         width=layout_dim,\n",
    "         height=layout_dim,\n",
    "         showlegend=False,\n",
    "         scene=go.Scene(\n",
    "         xaxis=go.XAxis(axis),\n",
    "         yaxis=go.YAxis(axis),\n",
    "         zaxis=go.ZAxis(z_axis),\n",
    "         camera=camera,\n",
    "         aspectratio = aspect_ratio,\n",
    "         aspectmode = \"manual\",\n",
    "        ),\n",
    "     margin=marg,\n",
    "    hovermode='closest',\n",
    "    )\n",
    "\n",
    "\n",
    "#data=go.Data([trace_nodes, trace_bs] + mesh_list)\n",
    "data_2d_noper=go.Data([trace_nodes, trace_edges, trace_bs, trace_surface] + mesh_list) # trace_surface]  + mesh_list\n",
    "fig_2d_noper=go.Figure(data=data_2d_noper, layout=layout_2d_noper)\n",
    "\n",
    "py.offline.plot(fig_2d_noper, filename='stage_3D_optimization_'+str(N)+'.html')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Well, I don't know how to put the whole thing inside these functions, thus a raincheck on this one.\n",
    "def network_sim_linkcost(params):\n",
    "\"\"\"\n",
    " Simulates network and returns linkcost as the loss function result. Done this way so pyswarms is happy.\n",
    "\n",
    " Inputs\n",
    "    ------\n",
    "    params: np.ndarray\n",
    "        Unrolled array of basestation heights, h1,h2,h3,h4,...\n",
    "        \n",
    "    Returns\n",
    "    -------\n",
    "    float\n",
    "        The computed metric, tries to minimize so returned as 1-metric\n",
    "\"\"\"\n",
    "\n",
    "def f(x):\n",
    "    \"\"\"Higher-level method to do network sim in the\n",
    "    whole swarm.\n",
    "\n",
    "    Inputs\n",
    "    ------\n",
    "    x: numpy.ndarray of shape (n_particles, dimensions)\n",
    "        The swarm that will perform the search\n",
    "\n",
    "    Returns\n",
    "    -------\n",
    "    numpy.ndarray of shape (n_particles, )\n",
    "        The computed loss for each particle\n",
    "    \"\"\"\n",
    "    n_particles = x.shape[0]\n",
    "    j = [forward_prop(x[i]) for i in range(n_particles)]\n",
    "    return np.array(j)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "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": "3.6.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}