[NC] Visualize X paper.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "fe5aceb2",
   "metadata": {},
   "source": [
    "# Visualize\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "id": "93dd0b8b",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "<ipython-input-75-69899bdc05d0>:6: DeprecationWarning: Importing display from IPython.core.display is deprecated since IPython 7.14, please import from IPython display\n",
      "  from IPython.core.display import display, HTML\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<style>#notebook { padding-top:0px !important; } .container { width:100% !important;ù } .end_space { min-heig§ht:0px !important; } </style>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The autoreload extension is already loaded. To reload it, use:\n",
      "  %reload_ext autoreload\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import networkx as nx\n",
    "import matplotlib.pyplot as plt\n",
    "import os\n",
    "\n",
    "from IPython.core.display import display, HTML\n",
    "display(HTML(\n",
    "    '<style>'\n",
    "        '#notebook { padding-top:0px !important; } ' \n",
    "        '.container { width:100% !important;ù } '\n",
    "        '.end_space { min-heig§ht:0px !important; } '\n",
    "    '</style>'\n",
    "))\n",
    "%load_ext autoreload\n",
    "%autoreload 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "id": "4b48183a",
   "metadata": {},
   "outputs": [],
   "source": [
    "def load_graphs(DATASET,MODEL,EXPL,MODE,verbose=True,lamb=0.001,normalize=True, raw=False):\n",
    "    if EXPL in [\"sal_edge\",\"ig_edge\",\"gnnexpl\",\"pgexplainer\",\"subgraphX\",\"rgexpl\"]:\n",
    "        FOLDER = \"edge_imp\"\n",
    "    else:\n",
    "        FOLDER = \"node_imp\"\n",
    "\n",
    "    path = \"Explanations/NodeClassification/\"+DATASET+\"/\"+MODEL+\"/\"+FOLDER+\"/\"+EXPL+\"/\"+MODE+\"/\"\n",
    "    graphs = dict()\n",
    "\n",
    "    labels = os.listdir(path)\n",
    "    for lab in labels:\n",
    "        graphs[int(lab)] = dict()\n",
    "        for i in os.listdir(path+\"/\"+lab):\n",
    "            if i[-7:] == \"gpickle\":\n",
    "                gid,y = i.split(\".\")[0].split(\"_\")\n",
    "                if y == lab:\n",
    "                    graphs[int(lab)][int(gid)] = g = nx.read_gpickle(path+\"/\"+lab+\"/\"+i)\n",
    "\n",
    "    \n",
    "    # get node/edge attrs\n",
    "    g = graphs[0][list(graphs[0].keys())[0]]\n",
    "    node_impo = nx.get_node_attributes(g,\"node_imp\")\n",
    "    edge_impo = nx.get_edge_attributes(g,\"edge_imp\")\n",
    "\n",
    "    # convert to node_impo\n",
    "    if not edge_impo == {}:\n",
    "        for lab , graph_dict in graphs.items():\n",
    "            for gid,g in graph_dict.items():\n",
    "                graphs[lab][gid] = from_edge_to_nodeExpl(g)    \n",
    "    if raw:\n",
    "        return graphs\n",
    "    \n",
    "    # clean_expl\n",
    "    graphs = get_cleaned_graphs(graphs, explainer_name=EXPL, verbose=verbose, lamb=lamb)    \n",
    "\n",
    "    # normalize \n",
    "    if normalize:\n",
    "        all_node_imp = []\n",
    "        for lab,graph_dict in graphs.items():\n",
    "            if not graph_dict == None:\n",
    "                for gid,g in graph_dict.items():\n",
    "                    all_node_imp.extend(list(dict(nx.get_node_attributes(g,\"node_imp\")).values()))\n",
    "        if not all_node_imp == []:\n",
    "            min_val = np.min(all_node_imp)\n",
    "            max_val = np.max(all_node_imp)\n",
    "\n",
    "            for lab,graph_dict in graphs.items():\n",
    "                if not graph_dict == None:\n",
    "                    for gid,g in graph_dict.items():\n",
    "                        for node in g.nodes():\n",
    "                            orig = g.nodes()[node][\"node_imp\"]\n",
    "                            g.nodes()[node][\"node_imp_norm\"] = (orig - min_val) / (max_val-min_val)\n",
    "\n",
    "    return graphs\n",
    "\n",
    "\n",
    "def from_edge_to_nodeExpl(g):\n",
    "    for n in g.nodes():\n",
    "        nei = list(nx.neighbors(g,n))\n",
    "        scores2 = []\n",
    "        for u in nei:\n",
    "            scores2.append(g.edges()[(u,n)][\"edge_imp\"])\n",
    "        g.nodes()[n][\"node_imp\"] = np.mean(scores2)\n",
    "    return g\n",
    "\n",
    "\n",
    "def clean(graphs, lab, lamb=0.001):\n",
    "    graphs_to_keep = dict()\n",
    "    for gid,g in graphs[lab].items():\n",
    "        a = list(nx.get_node_attributes(g,\"node_imp\").values())\n",
    "        un_a = np.unique(a)\n",
    "        if len(un_a)>1:\n",
    "            diff = un_a[-1] - un_a[0]\n",
    "            if diff > lamb:\n",
    "                graphs_to_keep[gid] = g\n",
    "    return graphs_to_keep\n",
    "\n",
    "\n",
    "def get_cleaned_graphs(graphs, explainer_name, verbose=True, lamb=0.001):\n",
    "    originals , kepts = [] , []\n",
    "    for c in sorted(graphs.keys()):\n",
    "        original = len(graphs[c])\n",
    "        kept = clean(graphs, c, lamb)\n",
    "        originals.append(original)\n",
    "        kepts.append(kept)\n",
    "    \n",
    "    if verbose:\n",
    "        #print(\"{:.3f}\".format(len(kept) / original),\",\",\"{:.3f}\".format(len(res1)/orig_1))\n",
    "        print(\",\".join([\"{:.3f} \".format(len(kepts[i]) / originals[i]) for i in range(len(originals))]))  #f\"{explainer_name:12}: \\t\", \n",
    "    \n",
    "    return {i: kepts[i] if len(kepts[i])/originals[i] >= 0.5 else None for i in range(len(originals))}"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "df898580",
   "metadata": {},
   "source": [
    "# rq1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "e9d63eb0",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\t\t\t\t\t train\n",
      "GraphSAGE\n",
      "cam\n",
      "0.000 ,0.000 ,0.000 ,0.000 \n",
      "grad_cam\n",
      "1.000 ,1.000 ,1.000 ,1.000 \n",
      "guided_bp\n",
      "0.000 ,0.000 ,0.000 ,0.000 \n",
      "ig_node\n",
      "1.000 ,1.000 ,1.000 ,1.000 \n",
      "grad_exp\n",
      "0.283 ,1.000 ,1.000 ,1.000 \n",
      "pgmexplainer\n",
      "0.950 ,0.195 ,0.172 ,0.141 \n",
      "gnnexpl\n",
      "1.000 ,1.000 ,1.000 ,1.000 \n",
      "pgexplainer\n",
      "0.000 ,0.000 ,0.000 ,0.000 \n",
      "subgraphX\n",
      "0.996 ,1.000 ,1.000 ,1.000 \n",
      "rgexpl\n",
      "0.996 ,1.000 ,1.000 ,1.000 \n"
     ]
    }
   ],
   "source": [
    "dataset = \"BAShapes\"\n",
    "MODELS = [\"GraphSAGE\"]\n",
    "GT_len = 9\n",
    "\n",
    "da_plottare = dict()\n",
    "tmp = []\n",
    "for MODE in [\"train\"]:\n",
    "    print(\"\\t\\t\\t\\t\\t\",MODE)\n",
    "    for model_architecture in MODELS:\n",
    "        print(model_architecture)\n",
    "\n",
    "        EXPLs = [\"cam\",\"grad_cam\",\"guided_bp\",\"ig_node\",\"grad_exp\",\"pgmexplainer\",\"gnnexpl\",\"pgexplainer\",\"subgraphX\",\"rgexpl\"]\n",
    "        \n",
    "        for expl in EXPLs:\n",
    "            if expl == \"-\":\n",
    "                print(\"- , -\")\n",
    "            else:\n",
    "                print(expl)\n",
    "                graphs = load_graphs(DATASET=dataset,\n",
    "                                 MODEL=model_architecture,\n",
    "                                 EXPL=expl,\n",
    "                                 MODE=MODE,\n",
    "                                 verbose=True,\n",
    "                                 lamb=0.001,\n",
    "                                 normalize=True)\n",
    "                \n",
    "                da_plottare[expl] = graphs\n",
    "                "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "dc9c5375",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "792f5e39",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "13b77a5e",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "c4112397",
   "metadata": {},
   "outputs": [],
   "source": [
    "p = 520#637 #520 #354 #\n",
    "y = 1 #1   #3\n",
    "radius = 2\n",
    "g1 = da_plottare[\"grad_cam\"][y][p]\n",
    "g1 = nx.ego_graph(g1,p,radius=radius)\n",
    "pos = nx.spring_layout(g1,seed=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "a8607f21",
   "metadata": {},
   "outputs": [],
   "source": [
    "pos[523] = [-0.5,0.2]\n",
    "pos[522] = [-0.5,0.6]\n",
    "pos[520] = [-0.2,0.2]\n",
    "pos[521] = [-0.2,0.6]\n",
    "pos[524] = [0.1 ,0.4]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3aa5bc49",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "ead3dc21",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "nx.draw_networkx(nx.ego_graph(da_plottare[\"grad_cam\"][y][p],p,radius=radius),pos=pos,with_labels=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d931e010",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "5addf820",
   "metadata": {},
   "outputs": [],
   "source": [
    "def my_plot(g,title=None,pos=None,ax=None):\n",
    "    node_impo = nx.get_node_attributes(g,\"node_imp_norm\")\n",
    "    if not ax == None:\n",
    "        ax.set_title(title)\n",
    "    nx.draw(g,pos,ax,node_color=list(node_impo.values()), cmap=plt.cm.Reds,node_size=50,edgecolors=\"gray\")\n",
    "    \n",
    "def my_plot_edge(g,title=None,pos=None,ax=None):\n",
    "    if not ax == None:\n",
    "        ax.set_title(title)\n",
    "    colors = list(nx.get_edge_attributes(g,\"edge_imp\").values())\n",
    "    options = {\n",
    "        \"node_color\":\"#cccccc\",\n",
    "        \"edge_color\": colors,\n",
    "        \"width\": 2,\n",
    "        \"edge_cmap\": plt.cm.Reds,\n",
    "        \"with_labels\": False,\n",
    "        \"node_size\":25    }\n",
    "    #nodes = nx.draw_networkx_nodes(g, pos)\n",
    "    #nodes.set_edgecolor('red')\n",
    "    nx.draw(g,pos,ax,**options)\n",
    "    #plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "id": "8352a19f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "NodeView((3, 520, 233, 521, 523, 524, 522, 140, 16, 87, 60))"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nx.ego_graph(da_plottare[\"rgexpl\"][y][p],p,radius=radius).nodes()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "fb3f2213",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x144 with 10 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "[\"cam\",\"grad_cam\",\"guided_bp\",\"ig_node\",\"grad_exp\",\"pgmexplainer\",\"sal_edge\",\"ig_edge\",\"gnnexpl\",\"pgexpl\",\"subgraphX\",\"rgexpl\"]\n",
    "fig, ax = plt.subplots(1,10,figsize=(20,2))\n",
    "#my_plot(da_plottare[\"cam\"][p],title=\"CAM\",pos=pos,ax=ax[0])\n",
    "\n",
    "\n",
    "nx.draw_networkx_nodes([p],pos,node_size=300,node_color=\"#33FFFC\",ax=ax[1])\n",
    "my_plot(nx.ego_graph(da_plottare[\"grad_cam\"][y][p],p,radius=radius),title=\"Grad.CAM\",pos=pos,ax=ax[1])\n",
    "#my_plot(nx.ego_graph(da_plottare[\"guided_bp\"][y][p],p,radius=3),title=\"Guid.BP\",pos=pos,ax=ax[2])\n",
    "nx.draw_networkx_nodes([p],pos,node_size=300,node_color=\"#33FFFC\",ax=ax[3])\n",
    "my_plot(nx.ego_graph(da_plottare[\"ig_node\"][y][p],p,radius=radius),title=\"IG (n)\",pos=pos,ax=ax[3])\n",
    "nx.draw_networkx_nodes([p],pos,node_size=300,node_color=\"#33FFFC\",ax=ax[4])\n",
    "my_plot(nx.Graph(nx.ego_graph(da_plottare[\"grad_exp\"][y][p],p,radius=radius)),title=\"Grad.Exp\",pos=pos,ax=ax[4])\n",
    "#my_plot(nx.ego_graph(da_plottare[\"pgmexplainer\"][y][p],p,radius=3),title=\"PGMExp\",pos=pos,ax=ax[5])\n",
    "nx.draw_networkx_nodes([p],pos,node_size=300,node_color=\"#33FFFC\",ax=ax[6])\n",
    "my_plot_edge(nx.ego_graph(da_plottare[\"gnnexpl\"][y][p],p,radius=radius),title=\"GNNExp\",pos=pos,ax=ax[6])\n",
    "#my_plot_edge(nx.ego_graph(da_plottare[\"pgexpl\"][y][p],p,radius=3),title=\"PGExp\",pos=pos,ax=ax[7])\n",
    "my_plot_edge(nx.ego_graph(da_plottare[\"subgraphX\"][y][p],p,radius=radius),title=\"subgraphX\",pos=pos,ax=ax[8])\n",
    "nx.draw_networkx_nodes([p],pos,node_size=300,node_color=\"#33FFFC\",ax=ax[8])\n",
    "my_plot_edge(nx.ego_graph(da_plottare[\"rgexpl\"][y][p],p,radius=radius),title=\"rgexpl\",pos=pos,ax=ax[9])\n",
    "nx.draw_networkx_nodes([p],pos,node_size=300,node_color=\"#33FFFC\",ax=ax[9])\n",
    "\n",
    "plt.savefig(\"Explanations/images/RQ11/node/BAshapes_graphSAGE_1.pdf\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "83226f09",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "904149d1",
   "metadata": {},
   "source": [
    "# gabri"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "2e564d15",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/antonio/anaconda3/envs/geo_new/lib/python3.9/site-packages/tqdm/auto.py:22: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
      "  from .autonotebook import tqdm as notebook_tqdm\n"
     ]
    }
   ],
   "source": [
    "from metrics.utils_nc import set_seeds\n",
    "from metrics.utils_nc import load_graphs as load_graphs_gabri\n",
    "from Datasets.synthetics import Infection\n",
    "set_seeds()\n",
    "dataset = Infection()\n",
    "y = dataset.data.y.numpy()\n",
    "infected_nodes = np.argwhere(y == 0)[:, 0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "3ef9c00f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\t\t\t\t\t train\n",
      "GraphSAGE\n",
      " 1.000  1.000  1.000\tcam\n",
      " 0.975  1.000  1.000\tgrad_cam\n",
      " 0.900  0.902  0.891\tguided_bp\n",
      " 1.000  1.000  1.000\tig_node\n",
      " 0.975  0.641  0.969\tgrad_exp\n",
      " 0.325  0.490  0.383\tpgmexplainer\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/antonio/anaconda3/envs/geo_new/lib/python3.9/site-packages/numpy/core/fromnumeric.py:3440: RuntimeWarning: Mean of empty slice.\n",
      "  return _methods._mean(a, axis=axis, dtype=dtype,\n",
      "/home/antonio/anaconda3/envs/geo_new/lib/python3.9/site-packages/numpy/core/_methods.py:189: RuntimeWarning: invalid value encountered in double_scalars\n",
      "  ret = ret.dtype.type(ret / rcount)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " 0.975  1.000  0.969\tgnnexpl\n",
      " 0.000  0.000  0.000\tpgexplainer\n",
      " 1.000  1.000  1.000\tsubgraphX\n"
     ]
    },
    {
     "ename": "FileNotFoundError",
     "evalue": "[Errno 2] No such file or directory: 'Explanations/NodeClassification/Infection/GraphSAGE/node_imp/rgexpl/train/'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mFileNotFoundError\u001b[0m                         Traceback (most recent call last)",
      "Input \u001b[0;32mIn [49]\u001b[0m, in \u001b[0;36m<cell line: 8>\u001b[0;34m()\u001b[0m\n\u001b[1;32m     15\u001b[0m     \u001b[38;5;28mprint\u001b[39m(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m- , -\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[1;32m     16\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m---> 17\u001b[0m     graphs \u001b[38;5;241m=\u001b[39m \u001b[43mload_graphs_gabri\u001b[49m\u001b[43m(\u001b[49m\u001b[43mDATASET\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mdataset\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m     18\u001b[0m \u001b[43m                     \u001b[49m\u001b[43mMODEL\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mmodel_architecture\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m     19\u001b[0m \u001b[43m                     \u001b[49m\u001b[43mEXPL\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mexpl\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m     20\u001b[0m \u001b[43m                     \u001b[49m\u001b[43mMODE\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mMODE\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m     21\u001b[0m \u001b[43m                     \u001b[49m\u001b[43mverbose\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mTrue\u001b[39;49;00m\u001b[43m,\u001b[49m\n\u001b[1;32m     22\u001b[0m \u001b[43m                     \u001b[49m\u001b[43mlamb\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;241;43m0.001\u001b[39;49m\u001b[43m,\u001b[49m\n\u001b[1;32m     23\u001b[0m \u001b[43m                     \u001b[49m\u001b[43mnormalize\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mTrue\u001b[39;49;00m\u001b[43m)\u001b[49m\n\u001b[1;32m     25\u001b[0m     da_plottare[expl] \u001b[38;5;241m=\u001b[39m graphs\n",
      "File \u001b[0;32m~/Desktop/Dottorato/LIO/repository/Explaining-the-Explainers-in-Graph-Neural-Networks/metrics/utils_nc.py:32\u001b[0m, in \u001b[0;36mload_graphs\u001b[0;34m(DATASET, MODEL, EXPL, MODE, verbose, lamb, normalize, raw)\u001b[0m\n\u001b[1;32m     30\u001b[0m graphs \u001b[38;5;241m=\u001b[39m {}\n\u001b[1;32m     31\u001b[0m \u001b[38;5;66;03m# Get the labels\u001b[39;00m\n\u001b[0;32m---> 32\u001b[0m labels \u001b[38;5;241m=\u001b[39m \u001b[43mos\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mlistdir\u001b[49m\u001b[43m(\u001b[49m\u001b[43mpath\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m     33\u001b[0m \u001b[38;5;66;03m# Loop over the labels\u001b[39;00m\n\u001b[1;32m     34\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m label \u001b[38;5;129;01min\u001b[39;00m labels:\n\u001b[1;32m     35\u001b[0m     \u001b[38;5;66;03m# Initialize a dictionary to collect all the nodes with this label\u001b[39;00m\n",
      "\u001b[0;31mFileNotFoundError\u001b[0m: [Errno 2] No such file or directory: 'Explanations/NodeClassification/Infection/GraphSAGE/node_imp/rgexpl/train/'"
     ]
    }
   ],
   "source": [
    "dataset = \"Infection\"\n",
    "MODELS = [\"GraphSAGE\"]\n",
    "\n",
    "EXPLs = [\"cam\",\"grad_cam\",\"guided_bp\",\"ig_node\",\"grad_exp\",\"pgmexplainer\",\"gnnexpl\",\"pgexplainer\",\"subgraphX\"]\n",
    "\n",
    "da_plottare = dict()\n",
    "tmp = []\n",
    "for MODE in [\"train\"]:\n",
    "    print(\"\\t\\t\\t\\t\\t\",MODE)\n",
    "    for model_architecture in MODELS:\n",
    "        print(model_architecture)\n",
    "            \n",
    "        for expl in EXPLs:\n",
    "            if expl == \"-\":\n",
    "                print(\"- , -\")\n",
    "            else:\n",
    "                graphs = load_graphs_gabri(DATASET=dataset,\n",
    "                                 MODEL=model_architecture,\n",
    "                                 EXPL=expl,\n",
    "                                 MODE=MODE,\n",
    "                                 verbose=True,\n",
    "                                 lamb=0.001,\n",
    "                                 normalize=True)\n",
    "                \n",
    "                da_plottare[expl] = graphs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b6397614",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "ab57833a",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "def get_inf_node_to_plot(g1,infected_nodes):\n",
    "    tmp = []\n",
    "    for i in infected_nodes:\n",
    "        if i in list(g1.nodes()):\n",
    "            tmp.append(i)\n",
    "    return tmp"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8ec0cdd0",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "772c3f89",
   "metadata": {},
   "outputs": [],
   "source": [
    "p = 891  #637 #520 #354 #\n",
    "y = 1 #1   #3\n",
    "radius = 2\n",
    "\n",
    "\n",
    "g1 = da_plottare[\"cam\"][y][p]\n",
    "g1 = nx.ego_graph(g1,p,radius=radius,undirected=True)\n",
    "g11 = nx.ego_graph(g1,p,radius=radius,undirected=False)\n",
    "\n",
    "g111 = g1.copy()\n",
    "\n",
    "for e1 in g1.edges():\n",
    "    \n",
    "    if e1 in list(g11.edges()):\n",
    "        \n",
    "        g111.remove_edge(e1[0],e1[1])\n",
    "\n",
    "g111.remove_nodes_from(list(nx.isolates(g111)))\n",
    "g1 = g111.copy()\n",
    "\n",
    "for n in g111:\n",
    "    if not nx.has_path(g111,n,p):\n",
    "        g1.remove_node(n)\n",
    "        \n",
    "        \n",
    "\n",
    "infet = get_inf_node_to_plot(g1,infected_nodes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "2701fca9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "pos = nx.spring_layout(g1,seed=11)\n",
    "nx.draw_networkx_nodes(infet,pos,node_size=300,node_color=\"#B2FF33\")\n",
    "nx.draw_networkx_nodes([p],pos,node_size=300,node_color=\"red\")\n",
    "nx.draw_networkx(g1,pos,node_size=100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "61248246",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2542f888",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1b006b0a",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "e9e12ab7",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "    \n",
    "def plot_gabri_compress(da_plottare,name,y,p,radius,infected_nodes,pos,title,ax):\n",
    "\n",
    "    g1 = da_plottare[name][y][p]\n",
    "    g1 = nx.ego_graph(g1,p,radius=radius,undirected=True)\n",
    "    g11 = nx.ego_graph(g1,p,radius=radius,undirected=False)\n",
    "\n",
    "    g111 = g1.copy()\n",
    "\n",
    "    for e1 in g1.edges():\n",
    "\n",
    "        if e1 in list(g11.edges()):\n",
    "\n",
    "            g111.remove_edge(e1[0],e1[1])\n",
    "\n",
    "    g111.remove_nodes_from(list(nx.isolates(g111)))\n",
    "    g1 = g111.copy()\n",
    "\n",
    "    for n in g111:\n",
    "        if not nx.has_path(g111,n,p):\n",
    "            g1.remove_node(n)\n",
    "\n",
    "\n",
    "\n",
    "    infet = get_inf_node_to_plot(g1,infected_nodes)\n",
    "\n",
    "    pos = nx.spring_layout(g1,seed=11)\n",
    "    nx.draw_networkx_nodes(infet,pos,node_size=300,node_color=\"#B2FF33\",ax=ax)\n",
    "    nx.draw_networkx_nodes([p],pos,node_size=300,node_color=\"#33FFFC\",ax=ax)\n",
    "\n",
    "    my_plot(g1,pos=pos,title=title,ax=ax)\n",
    "    \n",
    "def plot_gabri_compress_edge(da_plottare,name,y,p,radius,infected_nodes,pos,title,ax):\n",
    "\n",
    "    g1 = da_plottare[name][y][p]\n",
    "    g1 = nx.ego_graph(g1,p,radius=radius,undirected=True)\n",
    "    g11 = nx.ego_graph(g1,p,radius=radius,undirected=False)\n",
    "\n",
    "    g111 = g1.copy()\n",
    "\n",
    "    for e1 in g1.edges():\n",
    "\n",
    "        if e1 in list(g11.edges()):\n",
    "\n",
    "            g111.remove_edge(e1[0],e1[1])\n",
    "\n",
    "    g111.remove_nodes_from(list(nx.isolates(g111)))\n",
    "    g1 = g111.copy()\n",
    "\n",
    "    for n in g111:\n",
    "        if not nx.has_path(g111,n,p):\n",
    "            g1.remove_node(n)\n",
    "\n",
    "\n",
    "\n",
    "    infet = get_inf_node_to_plot(g1,infected_nodes)\n",
    "\n",
    "    pos = nx.spring_layout(g1,seed=11)\n",
    "    nx.draw_networkx_nodes(infet,pos,node_size=300,node_color=\"#B2FF33\",ax=ax)\n",
    "    nx.draw_networkx_nodes([p],pos,node_size=300,node_color=\"#33FFFC\",ax=ax)\n",
    "\n",
    "    my_plot_edge(g1,title=title,pos=pos,ax=ax)\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1ae2aa13",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "dd823f87",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x144 with 8 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "fig, ax = plt.subplots(1,8,figsize=(20,2))\n",
    "\n",
    "plot_gabri_compress(da_plottare,\"cam\",y,p,radius,infected_nodes,pos,\"Cam\",ax[0])\n",
    "plot_gabri_compress(da_plottare,\"grad_cam\",y,p,radius,infected_nodes,pos,\"GradCam\",ax[1])\n",
    "plot_gabri_compress(da_plottare,\"guided_bp\",y,p,radius,infected_nodes,pos,\"guided_bp\",ax[2])\n",
    "plot_gabri_compress(da_plottare,\"ig_node\",y,p,radius,infected_nodes,pos,\"IGnode\",ax[3])\n",
    "#plot_gabri_compress(da_plottare,\"pgmexplainer\",y,p,radius,infected_nodes,pos,\"pgme\",ax[4])\n",
    "plot_gabri_compress_edge(da_plottare,\"gnnexpl\",y,p,radius,infected_nodes,pos,\"gnnexpl\",ax[5])\n",
    "#plot_gabri_compress_edge(da_plottare,\"pgexplainer\",y,p,radius,infected_nodes,pos,\"pgexpl\",ax[6])\n",
    "plot_gabri_compress_edge(da_plottare,\"subgraphX\",y,p,radius,infected_nodes,pos,\"subgraphx\",ax[7])\n",
    "\n",
    "plt.savefig(\"Explanations/images/RQ11/node/infectopn_graphSAGE_1_v1.pdf\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7d996ce9",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "8e7adb45",
   "metadata": {},
   "source": [
    "# rq2 \n",
    "## bashapes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "id": "c10cd3d6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\t\t\t\t\t train\n",
      "GCN\n",
      "gnnexpl\n",
      "1.000 ,1.000 ,1.000 ,1.000 \n",
      "GIN\n",
      "gnnexpl\n",
      "1.000 ,1.000 ,1.000 ,1.000 \n",
      "GraphSAGE\n",
      "gnnexpl\n",
      "1.000 ,1.000 ,1.000 ,1.000 \n"
     ]
    }
   ],
   "source": [
    "dataset = \"BAShapes\"\n",
    "MODELS = [\"GCN\",\"GIN\",\"GraphSAGE\"]\n",
    "GT_len = 9\n",
    "\n",
    "da_plottare = dict()\n",
    "tmp = []\n",
    "for MODE in [\"train\"]:\n",
    "    print(\"\\t\\t\\t\\t\\t\",MODE)\n",
    "    for model_architecture in MODELS:\n",
    "        print(model_architecture)\n",
    "\n",
    "        EXPLs = [\"gnnexpl\"]\n",
    "        \n",
    "        for expl in EXPLs:\n",
    "            if expl == \"-\":\n",
    "                print(\"- , -\")\n",
    "            else:\n",
    "                print(expl)\n",
    "                graphs = load_graphs(DATASET=dataset,\n",
    "                                 MODEL=model_architecture,\n",
    "                                 EXPL=expl,\n",
    "                                 MODE=MODE,\n",
    "                                 verbose=True,\n",
    "                                 lamb=0.001,\n",
    "                                 normalize=True)\n",
    "                \n",
    "                da_plottare[model_architecture] = graphs[1]\n",
    "                "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e5f3ca2f",
   "metadata": {},
   "outputs": [],
   "source": [
    "p = 526#637 #520 #354 #\n",
    "y = 1 #1   #3\n",
    "radius = 2\n",
    "g1 = da_plottare[\"GCN\"][p]\n",
    "g1 = nx.ego_graph(g1,p,radius=radius)\n",
    "pos = nx.spring_layout(g1,seed=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a68c0cba",
   "metadata": {},
   "outputs": [],
   "source": [
    "nx.draw(g1,pos)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3309d698",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ba9307be",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "fig, ax = plt.subplots(1,8,figsize=(20,2))\n",
    "#my_plot(da_plottare[\"cam\"][p],title=\"CAM\",pos=pos,ax=ax[0])\n",
    "\n",
    "\n",
    "nx.draw_networkx_nodes([p],pos,node_size=300,node_color=\"#33FFFC\",ax=ax[1])\n",
    "my_plot(nx.ego_graph(da_plottare[\"GCN\"][p],p,radius=radius),title=\"GCN\",pos=pos,ax=ax[1])\n",
    "#my_plot(nx.ego_graph(da_plottare[\"guided_bp\"][y][p],p,radius=3),title=\"Guid.BP\",pos=pos,ax=ax[2])\n",
    "nx.draw_networkx_nodes([p],pos,node_size=300,node_color=\"#33FFFC\",ax=ax[3])\n",
    "my_plot(nx.ego_graph(da_plottare[\"GIN\"][p],p,radius=radius),title=\"GIN\",pos=pos,ax=ax[3])\n",
    "nx.draw_networkx_nodes([p],pos,node_size=300,node_color=\"#33FFFC\",ax=ax[4])\n",
    "my_plot(nx.Graph(nx.ego_graph(da_plottare[\"GraphSAGE\"][p],p,radius=radius)),title=\"SAGE\",pos=pos,ax=ax[4])\n",
    "#my_plot(nx.ego_graph(da_plottare[\"pgmexplainer\"][y][p],p,radius=3),title=\"PGMExp\",pos=pos,ax=ax[5])\n",
    "#my_plot_edge(nx.ego_graph(da_plottare[\"pgexpl\"][y][p],p,radius=3),title=\"PGExp\",pos=pos,ax=ax[7])\n",
    "\n",
    "plt.savefig(\"paper/TABELLINE_X_PAPER/nodi/BAshapes_graphSAGE_rq2.pdf\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "105b7edc",
   "metadata": {},
   "source": [
    "## Gabri"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "dbc5757c",
   "metadata": {},
   "outputs": [],
   "source": [
    "from utils import set_seeds\n",
    "from utils import load_graphs as load_graphs_gabri\n",
    "from Datasets.synthetics import Infection\n",
    "set_seeds()\n",
    "dataset = Infection()\n",
    "y = dataset.data.y.numpy()\n",
    "infected_nodes = np.argwhere(y == 0)[:, 0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "1739d465",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\t\t\t\t\t train\n",
      "GCN\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/antonio/anaconda3/envs/geometric_new/lib/python3.9/site-packages/numpy/core/fromnumeric.py:3474: RuntimeWarning: Mean of empty slice.\n",
      "  return _methods._mean(a, axis=axis, dtype=dtype,\n",
      "/home/antonio/anaconda3/envs/geometric_new/lib/python3.9/site-packages/numpy/core/_methods.py:189: RuntimeWarning: invalid value encountered in double_scalars\n",
      "  ret = ret.dtype.type(ret / rcount)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " 0.000  1.000  0.969\tgnnexpl\n",
      "GAT\n",
      " 1.000  1.000  0.969\tgnnexpl\n",
      "GraphSAGE\n",
      " 0.975  1.000  0.969\tgnnexpl\n"
     ]
    }
   ],
   "source": [
    "dataset = \"Infection\"\n",
    "MODELS = [\"GCN\",\"GAT\",\"GraphSAGE\"]\n",
    "\n",
    "EXPLs = [\"gnnexpl\"]\n",
    "\n",
    "da_plottare = dict()\n",
    "tmp = []\n",
    "for MODE in [\"train\"]:\n",
    "    print(\"\\t\\t\\t\\t\\t\",MODE)\n",
    "    for model_architecture in MODELS:\n",
    "        print(model_architecture)\n",
    "            \n",
    "        for expl in EXPLs:\n",
    "            if expl == \"-\":\n",
    "                print(\"- , -\")\n",
    "            else:\n",
    "                graphs = load_graphs_gabri(DATASET=dataset,\n",
    "                                 MODEL=model_architecture,\n",
    "                                 EXPL=expl,\n",
    "                                 MODE=MODE,\n",
    "                                 verbose=True,\n",
    "                                 lamb=0.001,\n",
    "                                 normalize=True)\n",
    "                \n",
    "                da_plottare[model_architecture] = graphs[1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "b0239a0a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{817: <networkx.classes.digraph.DiGraph at 0x7eff67a481c0>,\n",
       " 143: <networkx.classes.digraph.DiGraph at 0x7eff67a48040>,\n",
       " 835: <networkx.classes.digraph.DiGraph at 0x7eff67a48310>,\n",
       " 72: <networkx.classes.digraph.DiGraph at 0x7eff67a48100>,\n",
       " 914: <networkx.classes.digraph.DiGraph at 0x7eff67a480d0>,\n",
       " 944: <networkx.classes.digraph.DiGraph at 0x7eff67a48070>,\n",
       " 47: <networkx.classes.digraph.DiGraph at 0x7eff67a48160>,\n",
       " 723: <networkx.classes.digraph.DiGraph at 0x7eff67a483a0>,\n",
       " 930: <networkx.classes.digraph.DiGraph at 0x7eff67a48220>,\n",
       " 412: <networkx.classes.digraph.DiGraph at 0x7eff67a48280>,\n",
       " 625: <networkx.classes.digraph.DiGraph at 0x7eff67a48880>,\n",
       " 865: <networkx.classes.digraph.DiGraph at 0x7eff67a48b20>,\n",
       " 102: <networkx.classes.digraph.DiGraph at 0x7eff67a488e0>,\n",
       " 608: <networkx.classes.digraph.DiGraph at 0x7eff67a48910>,\n",
       " 206: <networkx.classes.digraph.DiGraph at 0x7eff67a487c0>,\n",
       " 105: <networkx.classes.digraph.DiGraph at 0x7eff67a48a60>,\n",
       " 837: <networkx.classes.digraph.DiGraph at 0x7eff67a48850>,\n",
       " 202: <networkx.classes.digraph.DiGraph at 0x7eff67a48820>,\n",
       " 188: <networkx.classes.digraph.DiGraph at 0x7eff67a486a0>,\n",
       " 741: <networkx.classes.digraph.DiGraph at 0x7eff67a487f0>,\n",
       " 371: <networkx.classes.digraph.DiGraph at 0x7eff67a488b0>,\n",
       " 519: <networkx.classes.digraph.DiGraph at 0x7eff67a486d0>,\n",
       " 413: <networkx.classes.digraph.DiGraph at 0x7eff67a48700>,\n",
       " 213: <networkx.classes.digraph.DiGraph at 0x7eff67a48580>,\n",
       " 953: <networkx.classes.digraph.DiGraph at 0x7eff67a48af0>,\n",
       " 760: <networkx.classes.digraph.DiGraph at 0x7eff67a485b0>,\n",
       " 181: <networkx.classes.digraph.DiGraph at 0x7eff67a485e0>,\n",
       " 597: <networkx.classes.digraph.DiGraph at 0x7eff67a48460>,\n",
       " 502: <networkx.classes.digraph.DiGraph at 0x7eff67a48970>,\n",
       " 385: <networkx.classes.digraph.DiGraph at 0x7eff67a48490>,\n",
       " 866: <networkx.classes.digraph.DiGraph at 0x7eff67a484c0>,\n",
       " 797: <networkx.classes.digraph.DiGraph at 0x7eff67a48400>,\n",
       " 620: <networkx.classes.digraph.DiGraph at 0x7eff67a48430>,\n",
       " 341: <networkx.classes.digraph.DiGraph at 0x7eff67a489d0>,\n",
       " 946: <networkx.classes.digraph.DiGraph at 0x7eff67a48fd0>,\n",
       " 34: <networkx.classes.digraph.DiGraph at 0x7eff67a48d90>,\n",
       " 365: <networkx.classes.digraph.DiGraph at 0x7eff67a48f10>,\n",
       " 184: <networkx.classes.digraph.DiGraph at 0x7eff67a48d60>,\n",
       " 836: <networkx.classes.digraph.DiGraph at 0x7eff67a48fa0>,\n",
       " 239: <networkx.classes.digraph.DiGraph at 0x7eff67a48f70>,\n",
       " 52: <networkx.classes.digraph.DiGraph at 0x7eff67a48df0>,\n",
       " 456: <networkx.classes.digraph.DiGraph at 0x7eff67a48f40>,\n",
       " 449: <networkx.classes.digraph.DiGraph at 0x7eff67a48c70>,\n",
       " 39: <networkx.classes.digraph.DiGraph at 0x7eff67a48e20>,\n",
       " 827: <networkx.classes.digraph.DiGraph at 0x7eff67a48e50>,\n",
       " 224: <networkx.classes.digraph.DiGraph at 0x7eff67a48cd0>,\n",
       " 112: <networkx.classes.digraph.DiGraph at 0x7eff67a48c40>,\n",
       " 345: <networkx.classes.digraph.DiGraph at 0x7eff67a48d00>,\n",
       " 512: <networkx.classes.digraph.DiGraph at 0x7eff67a48d30>,\n",
       " 766: <networkx.classes.digraph.DiGraph at 0x7eff67a48bb0>,\n",
       " 343: <networkx.classes.digraph.DiGraph at 0x7eff67a48dc0>,\n",
       " 335: <networkx.classes.digraph.DiGraph at 0x7eff67a48be0>,\n",
       " 790: <networkx.classes.digraph.DiGraph at 0x7eff67a48c10>,\n",
       " 187: <networkx.classes.digraph.DiGraph at 0x7eff67a48b50>,\n",
       " 794: <networkx.classes.digraph.DiGraph at 0x7eff67a48b80>,\n",
       " 291: <networkx.classes.digraph.DiGraph at 0x7eff67a48e80>,\n",
       " 931: <networkx.classes.digraph.DiGraph at 0x7eff67a482e0>,\n",
       " 635: <networkx.classes.digraph.DiGraph at 0x7eff67a4ee20>,\n",
       " 870: <networkx.classes.digraph.DiGraph at 0x7eff67a4edc0>,\n",
       " 695: <networkx.classes.digraph.DiGraph at 0x7eff67a4ed00>,\n",
       " 178: <networkx.classes.digraph.DiGraph at 0x7eff67a4ed60>,\n",
       " 73: <networkx.classes.digraph.DiGraph at 0x7eff67a4eca0>,\n",
       " 894: <networkx.classes.digraph.DiGraph at 0x7eff67a4eeb0>,\n",
       " 171: <networkx.classes.digraph.DiGraph at 0x7eff67a4ec70>,\n",
       " 358: <networkx.classes.digraph.DiGraph at 0x7eff67a4ee80>,\n",
       " 133: <networkx.classes.digraph.DiGraph at 0x7eff67a4edf0>,\n",
       " 366: <networkx.classes.digraph.DiGraph at 0x7eff67a4eee0>,\n",
       " 983: <networkx.classes.digraph.DiGraph at 0x7eff67a4ed30>,\n",
       " 383: <networkx.classes.digraph.DiGraph at 0x7eff67a4ed90>,\n",
       " 212: <networkx.classes.digraph.DiGraph at 0x7eff67a4ecd0>,\n",
       " 416: <networkx.classes.digraph.DiGraph at 0x7eff67a4efa0>,\n",
       " 541: <networkx.classes.digraph.DiGraph at 0x7eff67a4ec40>,\n",
       " 686: <networkx.classes.digraph.DiGraph at 0x7eff67a4ea00>,\n",
       " 475: <networkx.classes.digraph.DiGraph at 0x7eff67a4e850>,\n",
       " 978: <networkx.classes.digraph.DiGraph at 0x7eff67a4e160>,\n",
       " 54: <networkx.classes.digraph.DiGraph at 0x7eff67a4e820>,\n",
       " 3: <networkx.classes.digraph.DiGraph at 0x7eff67a4e6d0>,\n",
       " 855: <networkx.classes.digraph.DiGraph at 0x7eff67a4eb50>,\n",
       " 907: <networkx.classes.digraph.DiGraph at 0x7eff67a4eb80>,\n",
       " 829: <networkx.classes.digraph.DiGraph at 0x7eff67a4e3a0>,\n",
       " 326: <networkx.classes.digraph.DiGraph at 0x7eff67a4ef70>,\n",
       " 309: <networkx.classes.digraph.DiGraph at 0x7eff67a4efd0>,\n",
       " 636: <networkx.classes.digraph.DiGraph at 0x7eff67a4ec10>,\n",
       " 450: <networkx.classes.digraph.DiGraph at 0x7eff67a4e610>,\n",
       " 977: <networkx.classes.digraph.DiGraph at 0x7eff67a4e400>,\n",
       " 833: <networkx.classes.digraph.DiGraph at 0x7eff67a4e430>,\n",
       " 847: <networkx.classes.digraph.DiGraph at 0x7eff67a4e700>,\n",
       " 700: <networkx.classes.digraph.DiGraph at 0x7eff67a4e5e0>,\n",
       " 850: <networkx.classes.digraph.DiGraph at 0x7eff67a4e5b0>,\n",
       " 18: <networkx.classes.digraph.DiGraph at 0x7eff67a4eac0>,\n",
       " 318: <networkx.classes.digraph.DiGraph at 0x7eff67a4e730>,\n",
       " 912: <networkx.classes.digraph.DiGraph at 0x7eff67a4e310>,\n",
       " 841: <networkx.classes.digraph.DiGraph at 0x7eff67a4eaf0>,\n",
       " 973: <networkx.classes.digraph.DiGraph at 0x7eff67a4e7f0>,\n",
       " 971: <networkx.classes.digraph.DiGraph at 0x7eff67a4e370>,\n",
       " 742: <networkx.classes.digraph.DiGraph at 0x7eff67a4ea90>,\n",
       " 997: <networkx.classes.digraph.DiGraph at 0x7eff67a4ea60>,\n",
       " 613: <networkx.classes.digraph.DiGraph at 0x7eff67a4ef40>,\n",
       " 671: <networkx.classes.digraph.DiGraph at 0x7eff67a4e2b0>,\n",
       " 891: <networkx.classes.digraph.DiGraph at 0x7eff67a4e490>,\n",
       " 778: <networkx.classes.digraph.DiGraph at 0x7eff67a4e040>,\n",
       " 472: <networkx.classes.digraph.DiGraph at 0x7eff67a4e1f0>,\n",
       " 899: <networkx.classes.digraph.DiGraph at 0x7eff67a4e3d0>,\n",
       " 292: <networkx.classes.digraph.DiGraph at 0x7eff67a4e100>,\n",
       " 476: <networkx.classes.digraph.DiGraph at 0x7eff67a4e280>,\n",
       " 642: <networkx.classes.digraph.DiGraph at 0x7eff67a4e460>,\n",
       " 996: <networkx.classes.digraph.DiGraph at 0x7eff67a4e2e0>,\n",
       " 605: <networkx.classes.digraph.DiGraph at 0x7eff67a4e340>,\n",
       " 730: <networkx.classes.digraph.DiGraph at 0x7eff67a4e940>,\n",
       " 378: <networkx.classes.digraph.DiGraph at 0x7eff67a4ebe0>,\n",
       " 537: <networkx.classes.digraph.DiGraph at 0x7eff67a4e9a0>,\n",
       " 419: <networkx.classes.digraph.DiGraph at 0x7eff67a4e9d0>,\n",
       " 28: <networkx.classes.digraph.DiGraph at 0x7eff67a4e880>,\n",
       " 148: <networkx.classes.digraph.DiGraph at 0x7eff67a4eb20>,\n",
       " 375: <networkx.classes.digraph.DiGraph at 0x7eff67a4e910>,\n",
       " 410: <networkx.classes.digraph.DiGraph at 0x7eff67a4e8e0>,\n",
       " 661: <networkx.classes.digraph.DiGraph at 0x7eff67a4e760>,\n",
       " 972: <networkx.classes.digraph.DiGraph at 0x7eff67a4e8b0>,\n",
       " 583: <networkx.classes.digraph.DiGraph at 0x7eff67a4e970>,\n",
       " 928: <networkx.classes.digraph.DiGraph at 0x7eff67a4e790>,\n",
       " 278: <networkx.classes.digraph.DiGraph at 0x7eff67a4e7c0>,\n",
       " 426: <networkx.classes.digraph.DiGraph at 0x7eff67a4e640>,\n",
       " 736: <networkx.classes.digraph.DiGraph at 0x7eff67a4ebb0>,\n",
       " 552: <networkx.classes.digraph.DiGraph at 0x7eff67a4e670>,\n",
       " 182: <networkx.classes.digraph.DiGraph at 0x7eff67a4e6a0>,\n",
       " 830: <networkx.classes.digraph.DiGraph at 0x7eff67a4e520>,\n",
       " 658: <networkx.classes.digraph.DiGraph at 0x7eff67a4ea30>,\n",
       " 917: <networkx.classes.digraph.DiGraph at 0x7eff67a4e550>,\n",
       " 382: <networkx.classes.digraph.DiGraph at 0x7eff67a4e580>,\n",
       " 376: <networkx.classes.digraph.DiGraph at 0x7eff67a4e4c0>,\n",
       " 474: <networkx.classes.digraph.DiGraph at 0x7eff67a4e4f0>,\n",
       " 313: <networkx.classes.digraph.DiGraph at 0x7eff67a4e070>,\n",
       " 863: <networkx.classes.digraph.DiGraph at 0x7eff67a4e0a0>,\n",
       " 130: <networkx.classes.digraph.DiGraph at 0x7eff67a4ef10>,\n",
       " 96: <networkx.classes.digraph.DiGraph at 0x7eff67a4ee50>,\n",
       " 515: <networkx.classes.digraph.DiGraph at 0x7eff66fa23d0>,\n",
       " 508: <networkx.classes.digraph.DiGraph at 0x7eff66fa2490>,\n",
       " 480: <networkx.classes.digraph.DiGraph at 0x7eff66fa2640>,\n",
       " 356: <networkx.classes.digraph.DiGraph at 0x7eff66fa2370>,\n",
       " 170: <networkx.classes.digraph.DiGraph at 0x7eff66fa2790>,\n",
       " 372: <networkx.classes.digraph.DiGraph at 0x7eff66fa2880>,\n",
       " 958: <networkx.classes.digraph.DiGraph at 0x7eff66fa2730>,\n",
       " 68: <networkx.classes.digraph.DiGraph at 0x7eff66fa2850>,\n",
       " 570: <networkx.classes.digraph.DiGraph at 0x7eff66fa2160>,\n",
       " 998: <networkx.classes.digraph.DiGraph at 0x7eff66fa2310>,\n",
       " 329: <networkx.classes.digraph.DiGraph at 0x7eff66fa2460>,\n",
       " 246: <networkx.classes.digraph.DiGraph at 0x7eff66fa2100>,\n",
       " 310: <networkx.classes.digraph.DiGraph at 0x7eff66fa2670>,\n",
       " 785: <networkx.classes.digraph.DiGraph at 0x7eff66fa28b0>,\n",
       " 337: <networkx.classes.digraph.DiGraph at 0x7eff66fa2610>,\n",
       " 544: <networkx.classes.digraph.DiGraph at 0x7eff66fa2760>,\n",
       " 690: <networkx.classes.digraph.DiGraph at 0x7eff66cedc40>,\n",
       " 377: <networkx.classes.digraph.DiGraph at 0x7eff66fa2040>,\n",
       " 274: <networkx.classes.digraph.DiGraph at 0x7eff66fa2b50>,\n",
       " 880: <networkx.classes.digraph.DiGraph at 0x7eff66fa2fd0>,\n",
       " 66: <networkx.classes.digraph.DiGraph at 0x7eff66fa2dc0>,\n",
       " 57: <networkx.classes.digraph.DiGraph at 0x7eff66fa2d90>,\n",
       " 91: <networkx.classes.digraph.DiGraph at 0x7eff66fa2fa0>,\n",
       " 370: <networkx.classes.digraph.DiGraph at 0x7eff66fa2b20>,\n",
       " 887: <networkx.classes.digraph.DiGraph at 0x7eff66fa2400>,\n",
       " 56: <networkx.classes.digraph.DiGraph at 0x7eff66fa2250>,\n",
       " 922: <networkx.classes.digraph.DiGraph at 0x7eff66fa24f0>,\n",
       " 120: <networkx.classes.digraph.DiGraph at 0x7eff66fa22b0>,\n",
       " 814: <networkx.classes.digraph.DiGraph at 0x7eff66fa22e0>,\n",
       " 883: <networkx.classes.digraph.DiGraph at 0x7eff66fa2190>,\n",
       " 99: <networkx.classes.digraph.DiGraph at 0x7eff66fa2430>,\n",
       " 448: <networkx.classes.digraph.DiGraph at 0x7eff66fa2220>,\n",
       " 436: <networkx.classes.digraph.DiGraph at 0x7eff66fa21f0>,\n",
       " 414: <networkx.classes.digraph.DiGraph at 0x7eff66fa2070>,\n",
       " 42: <networkx.classes.digraph.DiGraph at 0x7eff66fa21c0>,\n",
       " 210: <networkx.classes.digraph.DiGraph at 0x7eff66fa2280>,\n",
       " 129: <networkx.classes.digraph.DiGraph at 0x7eff66fa20a0>,\n",
       " 115: <networkx.classes.digraph.DiGraph at 0x7eff66fa20d0>,\n",
       " 788: <networkx.classes.digraph.DiGraph at 0x7eff66fa24c0>,\n",
       " 780: <networkx.classes.digraph.DiGraph at 0x7eff66fa2340>,\n",
       " 796: <networkx.classes.digraph.DiGraph at 0x7eff66fa23a0>,\n",
       " 394: <networkx.classes.digraph.DiGraph at 0x7eff66fa29a0>,\n",
       " 652: <networkx.classes.digraph.DiGraph at 0x7eff66fa2c40>,\n",
       " 298: <networkx.classes.digraph.DiGraph at 0x7eff66fa2a00>,\n",
       " 407: <networkx.classes.digraph.DiGraph at 0x7eff66fa2a30>,\n",
       " 854: <networkx.classes.digraph.DiGraph at 0x7eff66fa28e0>,\n",
       " 875: <networkx.classes.digraph.DiGraph at 0x7eff66fa2b80>,\n",
       " 164: <networkx.classes.digraph.DiGraph at 0x7eff66fa2970>,\n",
       " 982: <networkx.classes.digraph.DiGraph at 0x7eff66fa2940>,\n",
       " 248: <networkx.classes.digraph.DiGraph at 0x7eff66fa27c0>,\n",
       " 65: <networkx.classes.digraph.DiGraph at 0x7eff66fa2910>,\n",
       " 561: <networkx.classes.digraph.DiGraph at 0x7eff66fa29d0>,\n",
       " 920: <networkx.classes.digraph.DiGraph at 0x7eff66fa27f0>,\n",
       " 437: <networkx.classes.digraph.DiGraph at 0x7eff66fa2820>,\n",
       " 670: <networkx.classes.digraph.DiGraph at 0x7eff66fa26a0>,\n",
       " 981: <networkx.classes.digraph.DiGraph at 0x7eff66fa2c10>,\n",
       " 873: <networkx.classes.digraph.DiGraph at 0x7eff66fa26d0>,\n",
       " 237: <networkx.classes.digraph.DiGraph at 0x7eff66fa2700>,\n",
       " 275: <networkx.classes.digraph.DiGraph at 0x7eff66fa2580>,\n",
       " 197: <networkx.classes.digraph.DiGraph at 0x7eff66fa2a90>,\n",
       " 316: <networkx.classes.digraph.DiGraph at 0x7eff66fa25b0>,\n",
       " 205: <networkx.classes.digraph.DiGraph at 0x7eff66fa25e0>,\n",
       " 76: <networkx.classes.digraph.DiGraph at 0x7eff66fa2520>,\n",
       " 116: <networkx.classes.digraph.DiGraph at 0x7eff66fa2550>,\n",
       " 784: <networkx.classes.digraph.DiGraph at 0x7eff66fa2af0>,\n",
       " 247: <networkx.classes.digraph.DiGraph at 0x7eff66fa2f10>,\n",
       " 716: <networkx.classes.digraph.DiGraph at 0x7eff66fa2f40>,\n",
       " 712: <networkx.classes.digraph.DiGraph at 0x7eff66fa2f70>,\n",
       " 755: <networkx.classes.digraph.DiGraph at 0x7eff66fa2df0>,\n",
       " 560: <networkx.classes.digraph.DiGraph at 0x7eff66fa2e20>,\n",
       " 368: <networkx.classes.digraph.DiGraph at 0x7eff66fa2e50>,\n",
       " 468: <networkx.classes.digraph.DiGraph at 0x7eff66fa2cd0>,\n",
       " 834: <networkx.classes.digraph.DiGraph at 0x7eff66fa2d00>,\n",
       " 315: <networkx.classes.digraph.DiGraph at 0x7eff66fa2d30>,\n",
       " 761: <networkx.classes.digraph.DiGraph at 0x7eff66fa2c70>,\n",
       " 425: <networkx.classes.digraph.DiGraph at 0x7eff66fa2ca0>,\n",
       " 411: <networkx.classes.digraph.DiGraph at 0x7eff66fa2bb0>,\n",
       " 126: <networkx.classes.digraph.DiGraph at 0x7eff66fa2d60>,\n",
       " 720: <networkx.classes.digraph.DiGraph at 0x7eff66fa2be0>,\n",
       " 967: <networkx.classes.digraph.DiGraph at 0x7eff66fa2e80>,\n",
       " 77: <networkx.classes.digraph.DiGraph at 0x7eff66fa2ac0>,\n",
       " 792: <networkx.classes.digraph.DiGraph at 0x7eff66fb8e80>,\n",
       " 709: <networkx.classes.digraph.DiGraph at 0x7eff66fb8670>,\n",
       " 763: <networkx.classes.digraph.DiGraph at 0x7eff66fb8340>,\n",
       " 390: <networkx.classes.digraph.DiGraph at 0x7eff66fb8b50>,\n",
       " 585: <networkx.classes.digraph.DiGraph at 0x7eff66fb8ac0>,\n",
       " 155: <networkx.classes.digraph.DiGraph at 0x7eff66fb8af0>,\n",
       " 632: <networkx.classes.digraph.DiGraph at 0x7eff66fb8760>,\n",
       " 338: <networkx.classes.digraph.DiGraph at 0x7eff66fb8a00>,\n",
       " 234: <networkx.classes.digraph.DiGraph at 0x7eff66fb8f70>,\n",
       " 43: <networkx.classes.digraph.DiGraph at 0x7eff66fb8f40>,\n",
       " 299: <networkx.classes.digraph.DiGraph at 0x7eff66fb84f0>,\n",
       " 514: <networkx.classes.digraph.DiGraph at 0x7eff66fb8790>,\n",
       " 487: <networkx.classes.digraph.DiGraph at 0x7eff66fb8640>,\n",
       " 244: <networkx.classes.digraph.DiGraph at 0x7eff66fb8e20>,\n",
       " 280: <networkx.classes.digraph.DiGraph at 0x7eff66fb8d30>,\n",
       " 497: <networkx.classes.digraph.DiGraph at 0x7eff66fb8d00>,\n",
       " 2: <networkx.classes.digraph.DiGraph at 0x7eff66fb8730>,\n",
       " 264: <networkx.classes.digraph.DiGraph at 0x7eff66fb8520>,\n",
       " 668: <networkx.classes.digraph.DiGraph at 0x7eff66fb8d60>,\n",
       " 122: <networkx.classes.digraph.DiGraph at 0x7eff66fb8fa0>,\n",
       " 422: <networkx.classes.digraph.DiGraph at 0x7eff66fb8e50>,\n",
       " 858: <networkx.classes.digraph.DiGraph at 0x7eff66fb8a60>,\n",
       " 260: <networkx.classes.digraph.DiGraph at 0x7eff66fb8940>,\n",
       " 851: <networkx.classes.digraph.DiGraph at 0x7eff66fb8b80>,\n",
       " 903: <networkx.classes.digraph.DiGraph at 0x7eff66fb8970>,\n",
       " 505: <networkx.classes.digraph.DiGraph at 0x7eff66fb8be0>,\n",
       " 649: <networkx.classes.digraph.DiGraph at 0x7eff66fb89a0>,\n",
       " 138: <networkx.classes.digraph.DiGraph at 0x7eff66fb89d0>,\n",
       " 21: <networkx.classes.digraph.DiGraph at 0x7eff66fb81f0>,\n",
       " 44: <networkx.classes.digraph.DiGraph at 0x7eff66fb8b20>,\n",
       " 757: <networkx.classes.digraph.DiGraph at 0x7eff66fb8040>,\n",
       " 198: <networkx.classes.digraph.DiGraph at 0x7eff66fb82b0>,\n",
       " 200: <networkx.classes.digraph.DiGraph at 0x7eff66fb8250>,\n",
       " 400: <networkx.classes.digraph.DiGraph at 0x7eff66fb82e0>,\n",
       " 683: <networkx.classes.digraph.DiGraph at 0x7eff66fb8370>,\n",
       " 403: <networkx.classes.digraph.DiGraph at 0x7eff66fb8bb0>,\n",
       " 615: <networkx.classes.digraph.DiGraph at 0x7eff66fb8a30>,\n",
       " 303: <networkx.classes.digraph.DiGraph at 0x7eff66fb8a90>,\n",
       " 862: <networkx.classes.digraph.DiGraph at 0x7eff66fb8eb0>,\n",
       " 744: <networkx.classes.digraph.DiGraph at 0x7eff66fb8fd0>,\n",
       " 22: <networkx.classes.digraph.DiGraph at 0x7eff66fb8ee0>,\n",
       " 820: <networkx.classes.digraph.DiGraph at 0x7eff66fb8f10>,\n",
       " 58: <networkx.classes.digraph.DiGraph at 0x7eff66fb8d90>,\n",
       " 752: <networkx.classes.digraph.DiGraph at 0x7eff66fb8dc0>,\n",
       " 373: <networkx.classes.digraph.DiGraph at 0x7eff66fb8df0>,\n",
       " 271: <networkx.classes.digraph.DiGraph at 0x7eff66fb8c70>,\n",
       " 529: <networkx.classes.digraph.DiGraph at 0x7eff66fb8ca0>,\n",
       " 638: <networkx.classes.digraph.DiGraph at 0x7eff66fb8cd0>,\n",
       " 696: <networkx.classes.digraph.DiGraph at 0x7eff66fb8c10>,\n",
       " 805: <networkx.classes.digraph.DiGraph at 0x7eff66fb8c40>,\n",
       " 518: <networkx.classes.digraph.DiGraph at 0x7eff66fb8130>,\n",
       " 255: <networkx.classes.digraph.DiGraph at 0x7eff66fb83d0>,\n",
       " 549: <networkx.classes.digraph.DiGraph at 0x7eff66fb8190>,\n",
       " 732: <networkx.classes.digraph.DiGraph at 0x7eff66fb81c0>,\n",
       " 799: <networkx.classes.digraph.DiGraph at 0x7eff66fb8070>,\n",
       " 762: <networkx.classes.digraph.DiGraph at 0x7eff66fb8310>,\n",
       " 626: <networkx.classes.digraph.DiGraph at 0x7eff66fb8100>,\n",
       " 587: <networkx.classes.digraph.DiGraph at 0x7eff66fb80d0>,\n",
       " 108: <networkx.classes.digraph.DiGraph at 0x7eff66fb80a0>,\n",
       " 492: <networkx.classes.digraph.DiGraph at 0x7eff66fb8160>,\n",
       " 405: <networkx.classes.digraph.DiGraph at 0x7eff66fb83a0>,\n",
       " 111: <networkx.classes.digraph.DiGraph at 0x7eff66fb8220>,\n",
       " 793: <networkx.classes.digraph.DiGraph at 0x7eff66fb8280>,\n",
       " 327: <networkx.classes.digraph.DiGraph at 0x7eff66fb8880>,\n",
       " 282: <networkx.classes.digraph.DiGraph at 0x7eff66fb88e0>,\n",
       " 336: <networkx.classes.digraph.DiGraph at 0x7eff66fb8910>,\n",
       " 789: <networkx.classes.digraph.DiGraph at 0x7eff66fb87c0>,\n",
       " 672: <networkx.classes.digraph.DiGraph at 0x7eff66fb8850>,\n",
       " 611: <networkx.classes.digraph.DiGraph at 0x7eff66fb8820>,\n",
       " 995: <networkx.classes.digraph.DiGraph at 0x7eff66fb86a0>,\n",
       " 618: <networkx.classes.digraph.DiGraph at 0x7eff66fb87f0>,\n",
       " 657: <networkx.classes.digraph.DiGraph at 0x7eff66fb86d0>,\n",
       " 1: <networkx.classes.digraph.DiGraph at 0x7eff66fb8700>,\n",
       " 360: <networkx.classes.digraph.DiGraph at 0x7eff66fb8580>,\n",
       " 848: <networkx.classes.digraph.DiGraph at 0x7eff66fb85b0>,\n",
       " 331: <networkx.classes.digraph.DiGraph at 0x7eff66fb85e0>,\n",
       " 268: <networkx.classes.digraph.DiGraph at 0x7eff66fb8460>,\n",
       " 493: <networkx.classes.digraph.DiGraph at 0x7eff66fb8490>,\n",
       " 536: <networkx.classes.digraph.DiGraph at 0x7eff66fb84c0>,\n",
       " 650: <networkx.classes.digraph.DiGraph at 0x7eff66fb8400>,\n",
       " 564: <networkx.classes.digraph.DiGraph at 0x7eff66fb8430>,\n",
       " 937: <networkx.classes.digraph.DiGraph at 0x7eff679d6550>,\n",
       " 167: <networkx.classes.digraph.DiGraph at 0x7eff679d6bb0>,\n",
       " 194: <networkx.classes.digraph.DiGraph at 0x7eff679d6cd0>,\n",
       " 199: <networkx.classes.digraph.DiGraph at 0x7eff679d69a0>,\n",
       " 594: <networkx.classes.digraph.DiGraph at 0x7eff679d6790>,\n",
       " 571: <networkx.classes.digraph.DiGraph at 0x7eff679d6a30>,\n",
       " 738: <networkx.classes.digraph.DiGraph at 0x7eff679d67f0>,\n",
       " 775: <networkx.classes.digraph.DiGraph at 0x7eff679d6820>,\n",
       " 804: <networkx.classes.digraph.DiGraph at 0x7eff679d6730>,\n",
       " 697: <networkx.classes.digraph.DiGraph at 0x7eff679d6970>,\n",
       " 53: <networkx.classes.digraph.DiGraph at 0x7eff679d6760>,\n",
       " 75: <networkx.classes.digraph.DiGraph at 0x7eff679d6700>,\n",
       " 219: <networkx.classes.digraph.DiGraph at 0x7eff679d65b0>,\n",
       " 388: <networkx.classes.digraph.DiGraph at 0x7eff679d6670>,\n",
       " 110: <networkx.classes.digraph.DiGraph at 0x7eff679d67c0>,\n",
       " 641: <networkx.classes.digraph.DiGraph at 0x7eff679d62b0>,\n",
       " 660: <networkx.classes.digraph.DiGraph at 0x7eff679d6640>,\n",
       " 705: <networkx.classes.digraph.DiGraph at 0x7eff679d6580>,\n",
       " 677: <networkx.classes.digraph.DiGraph at 0x7eff679d6a00>,\n",
       " 207: <networkx.classes.digraph.DiGraph at 0x7eff679d6430>,\n",
       " 687: <networkx.classes.digraph.DiGraph at 0x7eff679d6610>,\n",
       " 38: <networkx.classes.digraph.DiGraph at 0x7eff679d6880>,\n",
       " 190: <networkx.classes.digraph.DiGraph at 0x7eff679d68e0>,\n",
       " 128: <networkx.classes.digraph.DiGraph at 0x7eff679d6ee0>,\n",
       " 593: <networkx.classes.digraph.DiGraph at 0x7eff679d6f10>,\n",
       " 510: <networkx.classes.digraph.DiGraph at 0x7eff679d6f40>,\n",
       " 243: <networkx.classes.digraph.DiGraph at 0x7eff679d6f70>,\n",
       " 637: <networkx.classes.digraph.DiGraph at 0x7eff679d6e20>,\n",
       " 144: <networkx.classes.digraph.DiGraph at 0x7eff679d6eb0>,\n",
       " 29: <networkx.classes.digraph.DiGraph at 0x7eff679d6e80>,\n",
       " 152: <networkx.classes.digraph.DiGraph at 0x7eff679d6d00>,\n",
       " 815: <networkx.classes.digraph.DiGraph at 0x7eff679d6e50>,\n",
       " 525: <networkx.classes.digraph.DiGraph at 0x7eff679d6d30>,\n",
       " 769: <networkx.classes.digraph.DiGraph at 0x7eff679d6d60>,\n",
       " 555: <networkx.classes.digraph.DiGraph at 0x7eff679d6be0>,\n",
       " 666: <networkx.classes.digraph.DiGraph at 0x7eff679d6c10>,\n",
       " 550: <networkx.classes.digraph.DiGraph at 0x7eff679d6c40>,\n",
       " 351: <networkx.classes.digraph.DiGraph at 0x7eff679d6ac0>,\n",
       " 791: <networkx.classes.digraph.DiGraph at 0x7eff679d6fd0>,\n",
       " 435: <networkx.classes.digraph.DiGraph at 0x7eff679d6af0>,\n",
       " 85: <networkx.classes.digraph.DiGraph at 0x7eff679d6b20>,\n",
       " 455: <networkx.classes.digraph.DiGraph at 0x7eff679d6a60>,\n",
       " 64: <networkx.classes.digraph.DiGraph at 0x7eff679d6a90>,\n",
       " 218: <networkx.classes.digraph.DiGraph at 0x7eff679d6d90>,\n",
       " 633: <networkx.classes.digraph.DiGraph at 0x7eff679d6910>,\n",
       " 137: <networkx.classes.digraph.DiGraph at 0x7eff679d6dc0>,\n",
       " 9: <networkx.classes.digraph.DiGraph at 0x7eff679d6df0>,\n",
       " 161: <networkx.classes.digraph.DiGraph at 0x7eff679d68b0>,\n",
       " 297: <networkx.classes.digraph.DiGraph at 0x7eff679d6fa0>,\n",
       " 869: <networkx.classes.digraph.DiGraph at 0x7eff679d6850>,\n",
       " 860: <networkx.classes.digraph.DiGraph at 0x7eff679d6ca0>,\n",
       " 513: <networkx.classes.digraph.DiGraph at 0x7eff679d6b50>,\n",
       " 20: <networkx.classes.digraph.DiGraph at 0x7eff679d64f0>,\n",
       " 896: <networkx.classes.digraph.DiGraph at 0x7eff679d6220>,\n",
       " 37: <networkx.classes.digraph.DiGraph at 0x7eff679d66d0>,\n",
       " 882: <networkx.classes.digraph.DiGraph at 0x7eff679d65e0>,\n",
       " 895: <networkx.classes.digraph.DiGraph at 0x7eff679d6940>,\n",
       " 352: <networkx.classes.digraph.DiGraph at 0x7eff679d66a0>,\n",
       " 864: <networkx.classes.digraph.DiGraph at 0x7eff679d6130>,\n",
       " 902: <networkx.classes.digraph.DiGraph at 0x7eff679d6040>,\n",
       " 294: <networkx.classes.digraph.DiGraph at 0x7eff679d60a0>,\n",
       " 905: <networkx.classes.digraph.DiGraph at 0x7eff679d6070>,\n",
       " 486: <networkx.classes.digraph.DiGraph at 0x7eff679e5f70>,\n",
       " 740: <networkx.classes.digraph.DiGraph at 0x7eff679e5e80>,\n",
       " 324: <networkx.classes.digraph.DiGraph at 0x7eff679e5ee0>,\n",
       " 885: <networkx.classes.digraph.DiGraph at 0x7eff679e5e20>,\n",
       " 339: <networkx.classes.digraph.DiGraph at 0x7eff679e5fa0>,\n",
       " 710: <networkx.classes.digraph.DiGraph at 0x7eff679e5f10>,\n",
       " 288: <networkx.classes.digraph.DiGraph at 0x7eff679e50d0>,\n",
       " 69: <networkx.classes.digraph.DiGraph at 0x7eff679e5040>,\n",
       " 384: <networkx.classes.digraph.DiGraph at 0x7eff679e5070>,\n",
       " 442: <networkx.classes.digraph.DiGraph at 0x7eff679e5400>,\n",
       " 627: <networkx.classes.digraph.DiGraph at 0x7eff679e52e0>,\n",
       " 776: <networkx.classes.digraph.DiGraph at 0x7eff679e52b0>,\n",
       " 300: <networkx.classes.digraph.DiGraph at 0x7eff679e57c0>,\n",
       " 663: <networkx.classes.digraph.DiGraph at 0x7eff679e5280>,\n",
       " 963: <networkx.classes.digraph.DiGraph at 0x7eff679e54c0>,\n",
       " 471: <networkx.classes.digraph.DiGraph at 0x7eff679e54f0>,\n",
       " 13: <networkx.classes.digraph.DiGraph at 0x7eff679e5a30>,\n",
       " 964: <networkx.classes.digraph.DiGraph at 0x7eff679e5100>,\n",
       " 868: <networkx.classes.digraph.DiGraph at 0x7eff679e5730>,\n",
       " 507: <networkx.classes.digraph.DiGraph at 0x7eff679e5820>,\n",
       " 396: <networkx.classes.digraph.DiGraph at 0x7eff679e5c70>,\n",
       " 307: <networkx.classes.digraph.DiGraph at 0x7eff679e5af0>,\n",
       " 567: <networkx.classes.digraph.DiGraph at 0x7eff679e5b20>,\n",
       " 929: <networkx.classes.digraph.DiGraph at 0x7eff679e5c10>,\n",
       " 193: <networkx.classes.digraph.DiGraph at 0x7eff679e5c40>,\n",
       " 664: <networkx.classes.digraph.DiGraph at 0x7eff679e5130>,\n",
       " 286: <networkx.classes.digraph.DiGraph at 0x7eff679e5160>,\n",
       " 787: <networkx.classes.digraph.DiGraph at 0x7eff679e50a0>,\n",
       " 773: <networkx.classes.digraph.DiGraph at 0x7eff679e5610>,\n",
       " 874: <networkx.classes.digraph.DiGraph at 0x7eff679e58b0>,\n",
       " 943: <networkx.classes.digraph.DiGraph at 0x7eff679e5670>,\n",
       " 813: <networkx.classes.digraph.DiGraph at 0x7eff679e56a0>,\n",
       " 319: <networkx.classes.digraph.DiGraph at 0x7eff679e5550>,\n",
       " 843: <networkx.classes.digraph.DiGraph at 0x7eff679e57f0>,\n",
       " 362: <networkx.classes.digraph.DiGraph at 0x7eff679e55e0>,\n",
       " 495: <networkx.classes.digraph.DiGraph at 0x7eff679e55b0>,\n",
       " 588: <networkx.classes.digraph.DiGraph at 0x7eff679e5430>,\n",
       " 494: <networkx.classes.digraph.DiGraph at 0x7eff679e5580>,\n",
       " 915: <networkx.classes.digraph.DiGraph at 0x7eff679e5640>,\n",
       " 598: <networkx.classes.digraph.DiGraph at 0x7eff679e5460>,\n",
       " 241: <networkx.classes.digraph.DiGraph at 0x7eff679e5490>,\n",
       " 932: <networkx.classes.digraph.DiGraph at 0x7eff679e5310>,\n",
       " 612: <networkx.classes.digraph.DiGraph at 0x7eff679e5880>,\n",
       " 179: <networkx.classes.digraph.DiGraph at 0x7eff679e5340>,\n",
       " 725: <networkx.classes.digraph.DiGraph at 0x7eff679e5370>,\n",
       " 906: <networkx.classes.digraph.DiGraph at 0x7eff679e51f0>,\n",
       " 501: <networkx.classes.digraph.DiGraph at 0x7eff679e5700>,\n",
       " 844: <networkx.classes.digraph.DiGraph at 0x7eff679e5220>,\n",
       " 408: <networkx.classes.digraph.DiGraph at 0x7eff679e5250>,\n",
       " 227: <networkx.classes.digraph.DiGraph at 0x7eff679e5190>,\n",
       " 669: <networkx.classes.digraph.DiGraph at 0x7eff679e51c0>}"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "da_plottare[\"GAT\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d8939aca",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "090b326b",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "4c3d229f",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "    \n",
    "def plot_gabri_compress_edge_rq2(da_plottare,name,y,p,radius,infected_nodes,pos,title,ax,seed=11):\n",
    "\n",
    "    g1 = da_plottare[name][p]\n",
    "    g1 = nx.ego_graph(g1,p,radius=radius,undirected=True)\n",
    "    g11 = nx.ego_graph(g1,p,radius=radius,undirected=False)\n",
    "\n",
    "    g111 = g1.copy()\n",
    "\n",
    "    for e1 in g1.edges():\n",
    "\n",
    "        if e1 in list(g11.edges()):\n",
    "\n",
    "            g111.remove_edge(e1[0],e1[1])\n",
    "\n",
    "    g111.remove_nodes_from(list(nx.isolates(g111)))\n",
    "    g1 = g111.copy()\n",
    "\n",
    "    for n in g111:\n",
    "        if not nx.has_path(g111,n,p):\n",
    "            g1.remove_node(n)\n",
    "\n",
    "\n",
    "\n",
    "    infet = get_inf_node_to_plot(g1,infected_nodes)\n",
    "\n",
    "    pos = nx.spring_layout(g1,seed=seed)\n",
    "    nx.draw_networkx_nodes(infet,pos,node_size=300,node_color=\"#B2FF33\",ax=ax)\n",
    "    nx.draw_networkx_nodes([p],pos,node_size=300,node_color=\"#33FFFC\",ax=ax)\n",
    "\n",
    "    my_plot_edge(g1,title=title,pos=pos,ax=ax)\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "72ff192f",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "34209b3f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 288x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "p = 184  #637 #520 #354 #\n",
    "y = 1 #1   #3\n",
    "radius = 2\n",
    "g1 = da_plottare[\"GAT\"][p]\n",
    "g1 = nx.ego_graph(g1,p,radius=radius)\n",
    "pos = nx.spring_layout(g1,seed=355)\n",
    "seed = 1310\n",
    "fig, ax = plt.subplots(1,1,figsize=(4,4))\n",
    "plot_gabri_compress_edge_rq2(da_plottare,\"GCN\",y,p,radius,infected_nodes,pos,\"GCN\",ax,seed=seed)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5881538d",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "3b830603",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "NodeDataView({419: {'node_imp': 0.0, 'node_imp_norm': nan}, 163: {'node_imp': 0.0, 'node_imp_norm': nan}, 454: {'node_imp': 0.0, 'node_imp_norm': nan}, 472: {'node_imp': 0.0, 'node_imp_norm': nan}, 296: {'node_imp': 0.0, 'node_imp_norm': nan}, 520: {'node_imp': 0.0, 'node_imp_norm': nan}, 616: {'node_imp': 0.0, 'node_imp_norm': nan}, 106: {'node_imp': 0.0, 'node_imp_norm': nan}, 77: {'node_imp': 0.0, 'node_imp_norm': nan}, 750: {'node_imp': 0.0, 'node_imp_norm': nan}, 398: {'node_imp': 0.0, 'node_imp_norm': nan}, 48: {'node_imp': 0.0, 'node_imp_norm': nan}, 117: {'node_imp': 0.0, 'node_imp_norm': nan}, 87: {'node_imp': 0.0, 'node_imp_norm': nan}, 184: {'node_imp': 0.0, 'node_imp_norm': nan}, 508: {'node_imp': 0.0, 'node_imp_norm': nan}, 893: {'node_imp': nan, 'node_imp_norm': nan}})"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "g1.nodes(data=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "6e5b30ec",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x144 with 7 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "fig, ax = plt.subplots(1,7,figsize=(20,2))\n",
    "plot_gabri_compress_edge_rq2(da_plottare,\"GCN\",y,p,radius,infected_nodes,pos,\"GCN\",ax[0],seed=seed)\n",
    "plot_gabri_compress_edge_rq2(da_plottare,\"GAT\",y,p,radius,infected_nodes,pos,\"GAT\",ax[1],seed=seed)\n",
    "plot_gabri_compress_edge_rq2(da_plottare,\"GraphSAGE\",y,p,radius,infected_nodes,pos,\"GraphSAGE\",ax[2],seed=seed)\n",
    "#plot_gabri_compress_edge(da_plottare,\"pgexplainer\",y,p,radius,infected_nodes,pos,\"pgexpl\",ax[6])\n",
    "\n",
    "plt.savefig(\"paper/TABELLINE_X_PAPER/nodi/infectopn_graphSAGE_rq2_v2.pdf\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "64c81825",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1e87934d",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "1784f7d5",
   "metadata": {},
   "source": [
    "# singole classi \n",
    "# shapes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 166,
   "id": "c55519a1",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\t\t\t\t\t train\n",
      "GCN pgexplainer\n",
      "1.000 ,1.000 ,1.000 ,1.000 \n",
      "GIN grad_cam\n",
      "0.921 ,1.000 ,1.000 ,1.000 \n",
      "Cheb grad_cam\n",
      "1.000 ,1.000 ,1.000 ,1.000 \n",
      "GraphSAGE grad_cam\n",
      "1.000 ,1.000 ,1.000 ,1.000 \n"
     ]
    }
   ],
   "source": [
    "dataset = \"BAShapes\"\n",
    "MODELS = [(\"GCN\",\"guided_bp\"),(\"GIN\",\"guided_bp\"),(\"Cheb\",\"grad_cam\"),(\"GraphSAGE\",\"grad_cam\")] # 1\n",
    "MODELS = [(\"GCN\",\"pgexplainer\"),(\"GIN\",\"grad_cam\"),(\"Cheb\",\"grad_cam\"),(\"GraphSAGE\",\"grad_cam\")] # 2\n",
    "MODELS = [(\"GCN\",\"pgexplainer\"),(\"GIN\",\"gnnexpl\"),(\"Cheb\",\"grad_cam\"),(\"GraphSAGE\",\"gnnexpl\")] # 3\n",
    "\n",
    "\n",
    "\n",
    "MODELS = [(\"GCN\",\"guided_bp\"),(\"GIN\",\"guided_bp\"),(\"Cheb\",\"rgexpl\"),(\"GraphSAGE\",\"gnnexpl\")] # 1\n",
    "MODELS = [(\"GCN\",\"pgexplainer\"),(\"GIN\",\"grad_cam\"),(\"Cheb\",\"grad_cam\"),(\"GraphSAGE\",\"grad_cam\")] # 2\n",
    "#MODELS = [(\"GCN\",\"pgexplainer\"),(\"GIN\",\"rgexpl\"),(\"Cheb\",\"grad_cam\"),(\"GraphSAGE\",\"gnnexpl\")] # 3 # p = 614\n",
    "GT_len = 9\n",
    "\n",
    "da_plottare = dict()\n",
    "tmp = []\n",
    "for MODE in [\"train\"]:\n",
    "    print(\"\\t\\t\\t\\t\\t\",MODE)\n",
    "    for model_architecture,expl in MODELS:\n",
    "        print(model_architecture, expl)\n",
    "        graphs = load_graphs(DATASET=dataset,\n",
    "                         MODEL=model_architecture,\n",
    "                         EXPL=expl,\n",
    "                         MODE=MODE,\n",
    "                         verbose=True,\n",
    "                         lamb=0.001,\n",
    "                         normalize=True)\n",
    "\n",
    "        da_plottare[model_architecture] = (expl,graphs[2])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d2443538",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "id": "9ba6b0a5",
   "metadata": {},
   "outputs": [
    {
     "ename": "KeyError",
     "evalue": "614",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyError\u001b[0m                                  Traceback (most recent call last)",
      "Input \u001b[0;32mIn [113]\u001b[0m, in \u001b[0;36m<cell line: 4>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[38;5;66;03m# class = 3\u001b[39;00m\n\u001b[1;32m      3\u001b[0m p \u001b[38;5;241m=\u001b[39m \u001b[38;5;241m614\u001b[39m \u001b[38;5;66;03m# 382#620\u001b[39;00m\n\u001b[0;32m----> 4\u001b[0m a \u001b[38;5;241m=\u001b[39m nx\u001b[38;5;241m.\u001b[39mGraph(\u001b[43mda_plottare\u001b[49m\u001b[43m[\u001b[49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;124;43mGCN\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[43m]\u001b[49m\u001b[43m[\u001b[49m\u001b[38;5;241;43m1\u001b[39;49m\u001b[43m]\u001b[49m\u001b[43m[\u001b[49m\u001b[43mp\u001b[49m\u001b[43m]\u001b[49m)\n\u001b[1;32m      5\u001b[0m b \u001b[38;5;241m=\u001b[39m nx\u001b[38;5;241m.\u001b[39mGraph(da_plottare[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mGIN\u001b[39m\u001b[38;5;124m\"\u001b[39m][\u001b[38;5;241m1\u001b[39m][p])\n\u001b[1;32m      6\u001b[0m c \u001b[38;5;241m=\u001b[39m nx\u001b[38;5;241m.\u001b[39mGraph(da_plottare[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mCheb\u001b[39m\u001b[38;5;124m\"\u001b[39m][\u001b[38;5;241m1\u001b[39m][p])\n",
      "\u001b[0;31mKeyError\u001b[0m: 614"
     ]
    }
   ],
   "source": [
    "# class = 3\n",
    "\n",
    "p = 614 # 382#620\n",
    "a = nx.Graph(da_plottare[\"GCN\"][1][p])\n",
    "b = nx.Graph(da_plottare[\"GIN\"][1][p])\n",
    "c = nx.Graph(da_plottare[\"Cheb\"][1][p])\n",
    "d = nx.Graph(da_plottare[\"GraphSAGE\"][1][p])\n",
    "\n",
    "g1 = nx.ego_graph(a,p,radius=radius)\n",
    "pos = nx.spring_layout(g1,seed=10)\n",
    "\n",
    "pos[610] = [0.3,0]\n",
    "pos[613] = [0.3,-0.5]\n",
    "pos[612] = [0.0,-0.5]\n",
    "pos[611] = [0.0,-0]\n",
    "pos[614] = [0.15,0.35]\n",
    "pos[115] = [0.55,-0.25]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 167,
   "id": "3b7772f2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# class = 2\n",
    "\n",
    "p = 382 #620\n",
    "a = nx.Graph(da_plottare[\"GCN\"][1][p])\n",
    "b = nx.Graph(da_plottare[\"GIN\"][1][p])\n",
    "c = nx.Graph(da_plottare[\"Cheb\"][1][p])\n",
    "d = nx.Graph(da_plottare[\"GraphSAGE\"][1][p])\n",
    "\n",
    "g1 = nx.ego_graph(a,p,radius=radius)\n",
    "pos = nx.spring_layout(g1,seed=10)\n",
    "pos[381] = [0.3,0]\n",
    "pos[382] = [0.3,-2]\n",
    "pos[383] = [0.0,-2]\n",
    "pos[380] = [0.0,-0]\n",
    "pos[384] = [0.15,2]\n",
    "nx.draw_networkx(g1,pos)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 162,
   "id": "f8506203",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# class = 1\n",
    "\n",
    "p = 620\n",
    "a = nx.Graph(da_plottare[\"GCN\"][1][p])\n",
    "b = nx.Graph(da_plottare[\"GIN\"][1][p])\n",
    "c = nx.Graph(da_plottare[\"Cheb\"][1][p])\n",
    "d = nx.Graph(da_plottare[\"GraphSAGE\"][1][p])\n",
    "\n",
    "g1 = nx.ego_graph(a,p,radius=radius)\n",
    "pos = nx.spring_layout(g1,seed=10)\n",
    "\n",
    "pos[620] = [-0.3,0]\n",
    "pos[624] = [-0.5,0.3]\n",
    "pos[621] = [-0.7  ,0]\n",
    "pos[622] = [-0.7  ,-0.3]\n",
    "pos[623] = [-0.3  ,-0.3]\n",
    "nx.draw_networkx(g1,pos)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 163,
   "id": "d97196d1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[-0.5, 0.3]"
      ]
     },
     "execution_count": 163,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pos[624]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 169,
   "id": "ad824eed",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x72 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1,4,figsize=(10,1))\n",
    "#my_plot(da_plottare[\"cam\"][p],title=\"CAM\",pos=pos,ax=ax[0])\n",
    "\n",
    "nx.draw_networkx_nodes([p],pos,node_size=300,node_color=\"#33FFFC\",ax=ax[0])\n",
    "my_plot_edge(nx.ego_graph(a,p,radius=radius),title=\"GCN\",pos=pos,ax=ax[0])\n",
    "\n",
    "nx.draw_networkx_nodes([p],pos,node_size=300,node_color=\"#33FFFC\",ax=ax[1])\n",
    "my_plot(nx.ego_graph(b,p,radius=radius),title=\"GIN\",pos=pos,ax=ax[1])\n",
    "\n",
    "nx.draw_networkx_nodes([p],pos,node_size=300,node_color=\"#33FFFC\",ax=ax[2])\n",
    "my_plot(nx.ego_graph(c,p,radius=radius),title=\"Cheb\",pos=pos,ax=ax[2])\n",
    "\n",
    "nx.draw_networkx_nodes([p],pos,node_size=300,node_color=\"#33FFFC\",ax=ax[3])\n",
    "my_plot(nx.ego_graph(d,p,radius=radius),title=\"Sage\",pos=pos,ax=ax[3])\n",
    "\n",
    "plt.savefig(\"Explanations/images/RQ13/node/shapes_rq3_2.pdf\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f000bc7a",
   "metadata": {},
   "source": [
    "# Infections"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "97cb74fc",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_inf_node_to_plot(g1,infected_nodes):\n",
    "    tmp = []\n",
    "    for i in infected_nodes:\n",
    "        if i in list(g1.nodes()):\n",
    "            tmp.append(i)\n",
    "    return tmp"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "f3f61791",
   "metadata": {},
   "outputs": [],
   "source": [
    "from utils import set_seeds\n",
    "from utils import load_graphs as load_graphs_gabri\n",
    "from Datasets.synthetics import Infection\n",
    "set_seeds()\n",
    "dataset = Infection()\n",
    "y = dataset.data.y.numpy()\n",
    "infected_nodes = np.argwhere(y == 0)[:, 0]\n",
    "\n",
    "    \n",
    "def plot_gabri_compress_rq3(da_plottare,name,y,p,radius,infected_nodes,pos,title,ax):\n",
    "    g1 = da_plottare[name][1][p]\n",
    "    g1 = nx.ego_graph(g1,p,radius=radius)\n",
    "    infet = get_inf_node_to_plot(g1,infected_nodes)\n",
    "    nx.draw_networkx_nodes(infet,pos,node_size=300,node_color=\"#B2FF33\",ax=ax)\n",
    "    nx.draw_networkx_nodes([p],pos,node_size=300,node_color=\"#33FFFC\",ax=ax)\n",
    "    my_plot(g1,title=title,pos=pos,ax=ax)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "a97bf59a",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "    \n",
    "def plot_gabri_compress_rq3(da_plottare,name,y,p,radius,infected_nodes,pos,title,ax,seed=11):\n",
    "    \n",
    "    \n",
    "    g1 = da_plottare[name][1][p]\n",
    "    g1 = nx.ego_graph(g1,p,radius=radius,undirected=True)\n",
    "    g11 = nx.ego_graph(g1,p,radius=radius,undirected=False)\n",
    "\n",
    "    g111 = g1.copy()\n",
    "\n",
    "    for e1 in g1.edges():\n",
    "\n",
    "        if e1 in list(g11.edges()):\n",
    "\n",
    "            g111.remove_edge(e1[0],e1[1])\n",
    "\n",
    "    g111.remove_nodes_from(list(nx.isolates(g111)))\n",
    "    g1 = g111.copy()\n",
    "\n",
    "    for n in g111:\n",
    "        if not nx.has_path(g111,n,p):\n",
    "            g1.remove_node(n)\n",
    "\n",
    "\n",
    "\n",
    "    infet = get_inf_node_to_plot(g1,infected_nodes)\n",
    "\n",
    "    pos = nx.spring_layout(g1,seed=seed)\n",
    "    nx.draw_networkx_nodes(infet,pos,node_size=300,node_color=\"#B2FF33\",ax=ax)\n",
    "    nx.draw_networkx_nodes([p],pos,node_size=300,node_color=\"#33FFFC\",ax=ax)\n",
    "\n",
    "    my_plot(g1,title=title,pos=pos,ax=ax)\n",
    "    \n",
    "def plot_gabri_compress_edge_rq3(da_plottare,name,y,p,radius,infected_nodes,pos,title,ax,seed=11):\n",
    "\n",
    "    g1 = da_plottare[name][1][p]\n",
    "    g1 = nx.ego_graph(g1,p,radius=radius,undirected=True)\n",
    "    g11 = nx.ego_graph(g1,p,radius=radius,undirected=False)\n",
    "\n",
    "    g111 = g1.copy()\n",
    "\n",
    "    for e1 in g1.edges():\n",
    "\n",
    "        if e1 in list(g11.edges()):\n",
    "\n",
    "            g111.remove_edge(e1[0],e1[1])\n",
    "\n",
    "    g111.remove_nodes_from(list(nx.isolates(g111)))\n",
    "    g1 = g111.copy()\n",
    "\n",
    "    for n in g111:\n",
    "        if not nx.has_path(g111,n,p):\n",
    "            g1.remove_node(n)\n",
    "\n",
    "\n",
    "\n",
    "    infet = get_inf_node_to_plot(g1,infected_nodes)\n",
    "\n",
    "    pos = nx.spring_layout(g1,seed=seed)\n",
    "    nx.draw_networkx_nodes(infet,pos,node_size=300,node_color=\"#B2FF33\",ax=ax)\n",
    "    nx.draw_networkx_nodes([p],pos,node_size=300,node_color=\"#33FFFC\",ax=ax)\n",
    "\n",
    "    my_plot_edge(g1,title=title,pos=pos,ax=ax)\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "19e485e4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\t\t\t\t\t train\n",
      "GCN grad_cam\n",
      " 1.000  1.000  0.997\tgrad_cam\n",
      "GAT guided_bp\n",
      " 0.409  0.507  0.000\tguided_bp\n",
      "GIN guided_bp\n",
      " 0.500  0.990  0.937\tguided_bp\n",
      "Cheb cam\n",
      " 1.000  1.000  1.000\tcam\n",
      "GraphSAGE cam\n",
      " 1.000  1.000  1.000\tcam\n"
     ]
    }
   ],
   "source": [
    "dataset = \"Infection\"\n",
    "MODELS = [\"GCN\",\"GAT\",\"GraphSAGE\"]\n",
    "\n",
    "MODELS = [(\"GCN\",\"grad_cam\"),(\"GAT\",\"guided_bp\"),(\"GIN\",\"guided_bp\"),(\"Cheb\",\"cam\"),(\"GraphSAGE\",\"cam\")] # 1\n",
    "#MODELS = [(\"GCN\",\"grad_cam\"),(\"GAT\",\"cam\"),(\"GIN\",\"grad_cam\"),(\"Cheb\",\"cam\"),(\"GraphSAGE\",\"grad_cam\")] # 2\n",
    "#MODELS = [(\"GCN\",\"pgexplainer\"),(\"GAT\",\"guided_bp\"),(\"GIN\",\"gnnexpl\"),(\"Cheb\",\"grad_cam\"),(\"GraphSAGE\",\"gnnexpl\")] # 3\n",
    "\n",
    "da_plottare = dict()\n",
    "tmp = []\n",
    "for MODE in [\"train\"]:\n",
    "    print(\"\\t\\t\\t\\t\\t\",MODE)\n",
    "    for model_architecture,expl in MODELS:\n",
    "        print(model_architecture, expl)\n",
    "        graphs = load_graphs_gabri(DATASET=dataset,\n",
    "                         MODEL=model_architecture,\n",
    "                         EXPL=expl,\n",
    "                         MODE=MODE,\n",
    "                         verbose=True,\n",
    "                         lamb=0.001,\n",
    "                         normalize=True)\n",
    "\n",
    "        da_plottare[model_architecture] = (expl,graphs[1])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "061cd8a4",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "835 106 79\n",
      "914 20 20\n",
      "983 0 1\n"
     ]
    }
   ],
   "source": [
    "minimo = 10000\n",
    "for i,j in da_plottare[\"GAT\"][1].items():\n",
    "    g1 = nx.ego_graph(j,i,radius=3)\n",
    "    \n",
    "    \n",
    "    if len(g1.edges())<minimo:\n",
    "        print(i,len(g1.edges()),len(g1.nodes()))\n",
    "        minimo = len(g1.edges())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "2599fbc2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "p = 914 # 483 #797 #\n",
    "radius = 2\n",
    "fig, ax = plt.subplots(1,1,figsize=(10,10))\n",
    "pos = nx.spring_layout(da_plottare[\"GAT\"][1][p])\n",
    "plot_gabri_compress_rq3(da_plottare,\"GCN\",y,p,radius,infected_nodes,pos,\"GCN\",ax,seed=14)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "id": "fdad9526",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "48"
      ]
     },
     "execution_count": 95,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(g1.edges())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a363c62a",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "d63c9766",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x144 with 5 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "seed = 14\n",
    "fig, ax = plt.subplots(1,5,figsize=(16,2))\n",
    "plot_gabri_compress_rq3(da_plottare,\"GCN\",y,p,radius,infected_nodes,pos,\"GCN\",ax[0],seed=seed)\n",
    "plot_gabri_compress_rq3(da_plottare,\"GAT\",y,p,radius,infected_nodes,pos,\"GAT\",ax[1],seed=seed)\n",
    "plot_gabri_compress_rq3(da_plottare,\"GIN\",y,p,radius,infected_nodes,pos,\"GIN\",ax[2],seed=seed)\n",
    "plot_gabri_compress_rq3(da_plottare,\"Cheb\",y,p,radius,infected_nodes,pos,\"Cheb\",ax[3],seed=seed)\n",
    "plot_gabri_compress_rq3(da_plottare,\"GraphSAGE\",y,p,radius,infected_nodes,pos,\"GraphSAGE\",ax[4],seed=seed)\n",
    "\n",
    "plt.savefig(\"paper/TABELLINE_X_PAPER/nodi/infection_rq3_1_v2.pdf\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "532ddb03",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "62b0c6ab",
   "metadata": {},
   "source": [
    "# eliminami"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "95b32c37",
   "metadata": {},
   "outputs": [],
   "source": [
    "g1 = nx.read_gpickle(\"Explanations/GraphClassification/BA_2grid/GCN/node_imp/cam/train/0/1001_0.gpickle\")\n",
    "g2 = nx.read_gpickle(\"Explanations/GraphClassification/BA_2grid/GCN/node_imp/cam/train/1/1003_1.gpickle\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "4b629cf9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "fig, ax = plt.subplots(1,2,figsize=(8,4))\n",
    "nx.draw(g1,ax=ax[0],node_color=\"Blue\",alpha=0.8,node_size=30)\n",
    "nx.draw(g2,ax=ax[1],node_color=\"Red\",alpha=0.8,node_size=30)\n",
    "plt.savefig(\"eliminami_talk/tmp_grid.png\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "74d6cad1",
   "metadata": {},
   "outputs": [],
   "source": [
    "g1 = nx.read_gpickle(\"Explanations/GraphClassification/BA_2grid_house/GCN/node_imp/cam/train/0/1001_0.gpickle\")\n",
    "g2 = nx.read_gpickle(\"Explanations/GraphClassification/BA_2grid_house/GCN/node_imp/cam/train/0/1004_0.gpickle\")\n",
    "g3 = nx.read_gpickle(\"Explanations/GraphClassification/BA_2grid_house/GCN/node_imp/cam/train/1/507_1.gpickle\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "5718a6d2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x288 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "fig, ax = plt.subplots(1,3,figsize=(12,4))\n",
    "nx.draw(g1,ax=ax[0],pos=nx.spring_layout(g1,seed=11),node_color=\"Blue\",alpha=0.8,node_size=30)\n",
    "nx.draw(g2,ax=ax[1],pos=nx.spring_layout(g2,seed=11),node_color=\"Blue\",alpha=0.8,node_size=30)\n",
    "nx.draw(g3,ax=ax[2],pos=nx.spring_layout(g3,seed=18),node_color=\"Red\",alpha=0.8,node_size=30)\n",
    "plt.savefig(\"eliminami_talk/tmp_grid_HOUSE.png\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "af3eeaed",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "c62bba29",
   "metadata": {},
   "outputs": [],
   "source": [
    "g1 = nx.read_gpickle(\"Explanations/GraphClassification/ER_nb_stars2/GCN/node_imp/cam/train/0/1001_0.gpickle\")\n",
    "g2 = nx.read_gpickle(\"Explanations/GraphClassification/ER_nb_stars2/GCN/node_imp/cam/train/1/307_1.gpickle\")\n",
    "g3 = nx.read_gpickle(\"Explanations/GraphClassification/ER_nb_stars2/GCN/node_imp/cam/train/2/1015_2.gpickle\")\n",
    "g4 = nx.read_gpickle(\"Explanations/GraphClassification/ER_nb_stars2/GCN/node_imp/cam/train/2/505_2.gpickle\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "12fbefec",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x288 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "fig, ax = plt.subplots(1,4,figsize=(15,4))\n",
    "nx.draw(g1,ax=ax[0],pos=nx.spring_layout(g1,seed=11),node_color=\"Blue\",alpha=0.8,node_size=30)\n",
    "nx.draw(g2,ax=ax[1],pos=nx.spring_layout(g2,seed=15),node_color=\"Red\",alpha=0.8,node_size=30)\n",
    "nx.draw(g3,ax=ax[2],pos=nx.spring_layout(g3,seed=18),node_color=\"Green\",alpha=0.8,node_size=30)\n",
    "nx.draw(g4,ax=ax[3],pos=nx.spring_layout(g4,seed=18),node_color=\"Green\",alpha=0.8,node_size=30)\n",
    "plt.savefig(\"eliminami_talk/tmp_stars.png\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2dabf43f",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}