GraphVisualization.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import networkx as nx\n",
    "from RobotGraphModel import RobotGraph\n",
    "import gym\n",
    "import numpy as np\n",
    "from pathlib import Path\n",
    "import matplotlib.pyplot as plt\n",
    "from pygraphviz import *\n",
    "from CustomGymEnvs import MujocoGraphNormalWrapper, FetchReachGraphWrapper\n",
    "\n",
    "home = str(Path.home())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# env_name = 'FetchReach-v1'\n",
    "# fig_size=(30, 10)\n",
    "# env = FetchReachGraphWrapper(gym.make(env_name))\n",
    "\n",
    "# env_name = 'Ant-v2'\n",
    "# fig_size = (15, 7)\n",
    "# env = MujocoGraphWrapper(gym.make(env_name))\n",
    "\n",
    "# env_name = 'HalfCheetah-v2'\n",
    "# fig_size = (15, 10)\n",
    "# env = MujocoGraphNormalWrapper(env_name)\n",
    "\n",
    "env_name = 'Walker2d-v2'\n",
    "fig_size = (15, 10)\n",
    "env = MujocoGraphNormalWrapper(env_name)\n",
    "\n",
    "# env_name = 'Hopper-v2'\n",
    "# fig_size = (15, 7)\n",
    "# env = MujocoGraphWrapper(gym.make(env_name))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "rg = env.robot_graph "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def process_joint_name(joint_name):\n",
    "    separated = joint_name.split(':')[1].split('_') if 'robot0' in joint_name else joint_name.split('_')\n",
    "    final_key = ''\n",
    "    for sk in separated:\n",
    "        if len(sk) == 1:\n",
    "            final_key += sk + '-'\n",
    "        else:\n",
    "            final_key += sk + ' '\n",
    "    return final_key\n",
    "\n",
    "joint_names = []\n",
    "for joint_list in rg.edge_list.values():\n",
    "    if len(joint_list) > 0:\n",
    "        joint_names.append(\n",
    "            process_joint_name(joint_list[0].attrib['name'])\n",
    "            if len(joint_list) == 1\n",
    "#             else '\\n'.join([process_joint_name(j.attrib['name']) for j in joint_list])\n",
    "            else 'torso'\n",
    "        )\n",
    "    else:\n",
    "        joint_names.append('w')\n",
    "\n",
    "joint_names = [j.split('joint')[0] for j in joint_names]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0 thigh_left\n",
      "1 thigh\n",
      "2 foot_left\n",
      "3 world\n",
      "4 leg\n",
      "5 torso\n",
      "6 leg_left\n",
      "7 foot\n",
      "(5, 3) torso\n",
      "(1, 5) thigh \n",
      "(4, 1) leg \n",
      "(7, 4) foot \n",
      "(0, 5) thigh left \n",
      "(6, 0) leg left \n",
      "(2, 6) foot left \n"
     ]
    }
   ],
   "source": [
    "for i in range(len(rg.node_list)):\n",
    "    print(i, rg.node_list[i].attrib['name'])\n",
    "\n",
    "for e, j in zip(rg.edge_list.keys(), joint_names):\n",
    "    print(e, j)\n",
    "    \n",
    "nodelist = [i for i in range(len(rg.node_list))]\n",
    "edgelist = rg.edge_list.values()\n",
    "g = nx.Graph()\n",
    "g.add_nodes_from(nodelist)\n",
    "\n",
    "# use PyGraphviz to create the initial graph with lengths\n",
    "g = AGraph()\n",
    "g.add_nodes_from(nodelist)\n",
    "\n",
    "for edge, j in rg.edge_list.items():\n",
    "    if len(g) > 0:\n",
    "        g.add_edge(edge[0], edge[1], length=3)\n",
    "    else:\n",
    "        g.add_edge(edge[0], edge[1], length=1)    \n",
    "        \n",
    "# convert to NetworkX Graph from a PyGraphviz graph\n",
    "g = nx.nx_agraph.from_agraph(g)\n",
    "pos = nx.nx_agraph.graphviz_layout(g)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=fig_size)\n",
    "nx.draw(g,pos,width=2,linewidths=1,\\\n",
    "node_size=1000, with_labels=True, labels={node:node for node in g.nodes()})\n",
    "nx.draw_networkx_edge_labels(g,pos,\n",
    "                edge_labels={(f'{edge[0]}', f'{edge[1]}'): j  \n",
    "                             for edge, j in zip(rg.edge_list.keys(), joint_names)},\n",
    "                             font_color='dodgerblue', \n",
    "                             font_size=18)\n",
    "plt.axis('off')\n",
    "plt.savefig(f'{env_name}.jpg', dpi=300)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "ename": "AttributeError",
     "evalue": "'Walker2dEnv' object has no attribute 'robot_graph'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mAttributeError\u001b[0m                            Traceback (most recent call last)",
      "\u001b[0;32m/tmp/ipykernel_11891/2146546483.py\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[0menv\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mgym\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmake\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0menv_name\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 2\u001b[0;31m \u001b[0mrg\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0menv\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrobot_graph\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      3\u001b[0m \u001b[0mnodelist\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0menv\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msim\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbody_name2id\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mn\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mattrib\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'name'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mn\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mrg\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnode_list\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0medgelist\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0menv\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msim\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbody_name2id\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mrg\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnode_list\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mf\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mattrib\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'name'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0menv\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msim\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbody_name2id\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mrg\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnode_list\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mt\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mattrib\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'name'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mt\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mzip\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mrg\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0medges_from\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mrg\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0medges_to\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m \u001b[0mg\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnx\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mGraph\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Documents/env/lib/python3.8/site-packages/gym/core.py\u001b[0m in \u001b[0;36m__getattr__\u001b[0;34m(self, name)\u001b[0m\n\u001b[1;32m    236\u001b[0m                 \u001b[0;34m\"attempted to get missing private attribute '{}'\"\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mformat\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mname\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    237\u001b[0m             )\n\u001b[0;32m--> 238\u001b[0;31m         \u001b[0;32mreturn\u001b[0m \u001b[0mgetattr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0menv\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mname\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    239\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    240\u001b[0m     \u001b[0;34m@\u001b[0m\u001b[0mproperty\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mAttributeError\u001b[0m: 'Walker2dEnv' object has no attribute 'robot_graph'"
     ]
    }
   ],
   "source": [
    "env = gym.make(env_name)\n",
    "rg = env.robot_graph\n",
    "nodelist = [env.sim.model.body_name2id(n.attrib['name']) for n in rg.node_list]\n",
    "edgelist = [(env.sim.model.body_name2id(rg.node_list[f].attrib['name']), env.sim.model.body_name2id(rg.node_list[t].attrib['name'])) for f, t in zip(rg.edges_from, rg.edges_to)]\n",
    "g = nx.Graph()\n",
    "g.add_nodes_from(nodelist)\n",
    "\n",
    "for edge, j in zip(edgelist, rg.edge_list):\n",
    "    if j is not None:\n",
    "        g.add_edge(edge[0], edge[1], length=5)\n",
    "    else:\n",
    "        g.add_edge(edge[0], edge[1], length=1)    \n",
    "    \n",
    "# g.add_edges_from(edgelist, length=10)\n",
    "pos = nx.spring_layout(g)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.figure(figsize=(10, 15))\n",
    "nx.draw(g,pos,edge_color='black',width=2,linewidths=1,\\\n",
    "node_size=500,node_color='cyan',alpha=0.9,\\\n",
    "labels={node:node for node in g.nodes()})\n",
    "# nx.draw_networkx_edge_labels(g,pos,edge_labels={edge: 'j' if kind is not None else 'w' for edge, kind in zip(edgelist, rg.edge_list)},font_color='red')\n",
    "nx.draw_networkx_edge_labels(g,pos,edge_labels={edge: np.round(rg.edge_features[i, :], 3) for edge, i in zip(edgelist, range(rg.edge_features.shape[0]))},font_color='red')\n",
    "\n",
    "plt.axis('off')\n",
    "plt.savefig(f'{env_name}.jpg', dpi=300)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.figure(figsize=(10, 15))\n",
    "nx.draw(g,pos,edge_color='black',width=2,linewidths=1,\\\n",
    "node_size=500,node_color='pink',alpha=0.9,\\\n",
    "labels={node:node for node in g.nodes()})\n",
    "\n",
    "nx.draw_networkx_edge_labels(g,pos,edge_labels={edge: rg.sim.model.joint_name2id(j.attrib['name']) if j is not None else 'w' for edge, j in zip(edgelist, rg.edge_list)},font_color='red')\n",
    "plt.axis('off')\n",
    "plt.show()\n",
    "plt.savefig(f'{env_name}.jpg', dpi=300)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}