PYGTesting.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "4a8374b6-ac59-4ef9-81c2-a26f00d5888c",
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "import torch_geometric\n",
    "from torch_geometric.data import Data\n",
    "\n",
    "edge_index = torch.tensor([[0, 1, 1, 2],\n",
    "                           [1, 0, 2, 1]], dtype=torch.long)\n",
    "x = torch.tensor([[-1], [0], [1]], dtype=torch.float)\n",
    "\n",
    "data = Data(x=x, edge_index=edge_index)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "7756d99c-5fe0-4772-a1e6-3bdaf0e546f0",
   "metadata": {},
   "outputs": [],
   "source": [
    "from torch_geometric.datasets import TUDataset\n",
    "\n",
    "dataset = TUDataset(root='/tmp/ENZYMES', name='ENZYMES')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "188866a4-28e0-4338-8b87-2c0fefaafa4b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "shapes of list of edges, list of nodes w/ features and of class to predict\n",
      "Data(edge_index=[2, 168], x=[37, 3], y=[1])\n",
      "data is undirected? True\n",
      "data has self loops? False\n"
     ]
    }
   ],
   "source": [
    "data = dataset[0]\n",
    "print(\"shapes of list of edges, list of nodes w/ features and of class to predict\")\n",
    "print(data)\n",
    "print(\"data is undirected?\", data.is_undirected())\n",
    "print(\"data has self loops?\", data.contains_self_loops())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6d2f71d0-8c1f-427e-a329-fd0d22998232",
   "metadata": {},
   "outputs": [],
   "source": [
    "import networkx as nx\n",
    "import torch_geometric\n",
    "g = torch_geometric.utils.to_networkx(data, to_undirected=True)\n",
    "nx.draw(g)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "9b6067a3-f9f0-492a-9d9a-c7261aa4817e",
   "metadata": {},
   "outputs": [],
   "source": [
    "from torch_geometric.datasets import Planetoid\n",
    "\n",
    "dataset = Planetoid(root='/tmp/Cora', name='Cora')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "7792d619-df53-4d18-996b-bc6e56058bf6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import networkx as nx\n",
    "import torch_geometric\n",
    "g = torch_geometric.utils.to_networkx(dataset[0], to_undirected=True)\n",
    "nx.draw(g)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "d98bdc1d-c04d-47fd-aed5-8a9be79f899f",
   "metadata": {},
   "outputs": [],
   "source": [
    "from ocpmodels.preprocessing import AtomsToGraphs\n",
    "from ocpmodels.datasets import SinglePointLmdbDataset, TrajectoryLmdbDataset\n",
    "import ase.io\n",
    "from ase.build import bulk\n",
    "from ase.build import fcc100, add_adsorbate, molecule\n",
    "from ase.constraints import FixAtoms\n",
    "from ase.calculators.emt import EMT\n",
    "from ase.optimize import BFGS\n",
    "import matplotlib.pyplot as plt\n",
    "import lmdb\n",
    "import pickle\n",
    "from tqdm import tqdm\n",
    "import torch\n",
    "import os"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "id": "25f24351-bbf2-4574-b30e-2ca97c560b28",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10000\n"
     ]
    }
   ],
   "source": [
    "# dataset at D:\\is2res_train_val_test_lmdbs\\data\\is2re\\10k\\train\n",
    "\n",
    "dataset = SinglePointLmdbDataset({\"src\": \"D:\\\\is2res_train_val_test_lmdbs\\\\data\\\\is2re\\\\10k\\\\train\\\\data.lmdb\"})\n",
    "print(len(dataset))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "id": "f9099aaf-121a-41d2-b437-7b3021fdb0e3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Data(atomic_numbers=[86], cell=[1, 3, 3], cell_offsets=[2964, 3], distances=[2964], edge_index=[2, 2964], fixed=[86], force=[86, 3], natoms=86, pos=[86, 3], pos_relaxed=[86, 3], sid=2472718, tags=[86], y_init=6.282500615000004, y_relaxed=-0.025550085000020317)\n"
     ]
    }
   ],
   "source": [
    "print(dataset[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "20c5122d-169d-4376-b051-187e2a972073",
   "metadata": {},
   "outputs": [],
   "source": [
    "from mpl_toolkits import mplot3d\n",
    "import numpy as np\n",
    "\n",
    "#print(dataset[0].pos)\n",
    "x = dataset[0].pos[:,0]\n",
    "y = dataset[0].pos[:,1]\n",
    "z = dataset[0].pos[:,2]\n",
    "\n",
    "# Creating figure\n",
    "fig = plt.figure(figsize = (10, 7))\n",
    "ax = plt.axes(projection =\"3d\")\n",
    " \n",
    "# Creating plot\n",
    "ax.scatter3D(x, y, z)\n",
    " \n",
    "# show plot\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "b6fa6c84-0041-4452-8c00-3602e58c85d6",
   "metadata": {},
   "outputs": [],
   "source": [
    "symbols = [\"H\", \"He\", \"Li\", \"Be\", \"B\", \"C\", \"N\", \"O\", \"F\", \"Ne\", \"Na\", \"Mg\", \"Al\", \"Si\", \"P\", \"S\", \"Cl\", \"Ar\", \"K\", \"Ca\", \"Sc\", \"Ti\", \"V\", \"Cr\", \"Mn\", \"Fe\", \"Co\", \"Ni\", \"Cu\", \"Zn\", \"Ga\", \"Ge\", \"As\", \"Se\", \"Br\", \"Kr\", \"Rb\", \"Sr\", \"Y\", \"Zr\", \"Nb\", \"Mo\", \"Tc\", \"Ru\", \"Rh\", \"Pd\", \"Ag\", \"Cd\", \"In\", \"Sn\", \"Sb\", \"Te\", \"I\", \"Xe\", \"Cs\", \"Ba\", \"La\", \"Ce\", \"Pr\", \"Nd\", \"Pm\", \"Sm\", \"Eu\", \"Gd\", \"Tb\", \"Dy\", \"Ho\", \"Er\", \"Tm\", \"Yb\", \"Lu\", \"Hf\", \"Ta\", \"W\", \"Re\", \"Os\", \"Ir\", \"Pt\", \"Au\", \"Hg\", \"Tl\", \"Pb\", \"Bi\", \"Po\", \"At\", \"Rn\", \"Fr\", \"Ra\", \"Ac\", \"Th\", \"Pa\", \"U\", \"Np\", \"Pu\", \"Am\", \"Cm\", \"Bk\", \"Cf\", \"Es\", \"Fm\", \"Md\", \"No\", \"Lr\", \"Rf\", \"Db\", \"Sg\", \"Bh\", \"Hs\", \"Mt\", \"Ds\", \"Rg\", \"Cn\", \"Uut\", \"Fl\", \"Uup\", \"Lv\", \"Uus\", \"Uuo\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "1ef69868-4257-4be6-9eee-5665dbe6e6b1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Data(atomic_numbers=[86], cell=[1, 3, 3], cell_offsets=[2964, 3], distances=[2964], edge_index=[2, 2964], fixed=[86], force=[86, 3], natoms=86, pos=[86, 3], pos_relaxed=[86, 3], sid=2472718, tags=[86], y_init=6.282500615000004, y_relaxed=-0.025550085000020317)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<subprocess.Popen at 0x1d61ab55f10>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from ase.io import read\n",
    "from ase.visualize import view\n",
    "\n",
    "i = 0\n",
    "print(dataset[i])\n",
    "\n",
    "x = dataset[i].pos[:,0]\n",
    "y = dataset[i].pos[:,1]\n",
    "z = dataset[i].pos[:,2]\n",
    "atom = dataset[i].atomic_numbers\n",
    "\n",
    "f = open(\"adsPicture.xyz\", \"w\")\n",
    "f.write(str(len(atom))+\"\\n\")\n",
    "for i in range(len(atom)):\n",
    "    f.write(\"\\n{}\\t {}\\t {}\\t {}\".format(symbols[int(atom[i])-1],x[i],y[i],z[i]))\n",
    "f.close()\n",
    "\n",
    "\n",
    "pic = read(\"adsPicture.xyz\")\n",
    "view(pic)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "b83e9292-a668-422b-a174-8551aaa0d0c1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Data(atomic_numbers=[86], cell=[1, 3, 3], cell_offsets=[2964, 3], distances=[2964], edge_index=[2, 2964], fixed=[86], force=[86, 3], natoms=86, pos=[86, 3], pos_relaxed=[86, 3], sid=2472718, tags=[86], y_init=6.282500615000004, y_relaxed=-0.025550085000020317)\n"
     ]
    }
   ],
   "source": [
    "print(dataset[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 127,
   "id": "5d509b66-020d-41e0-a439-7c7622eaa57e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10000\n"
     ]
    }
   ],
   "source": [
    "from torch_geometric.data import DataLoader\n",
    "print(len(dataset))\n",
    "train_loader = DataLoader(dataset, batch_size=64, shuffle=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "id": "9a391563-3e26-478e-bd15-a2e339904dfa",
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "import torch.nn.functional as F\n",
    "from torch_geometric.nn import GCNConv\n",
    "\n",
    "class GCN(torch.nn.Module):\n",
    "    def __init__(self):\n",
    "        super().__init__()\n",
    "        self.conv1 = GCNConv(3, 16)\n",
    "        self.conv2 = GCNConv(16, 3)\n",
    "\n",
    "    def forward(self, data):\n",
    "        x, edge_index = data.pos, data.edge_index\n",
    "\n",
    "        x = self.conv1(x, edge_index)\n",
    "        x = F.relu(x)\n",
    "        x = F.dropout(x, training=self.training)\n",
    "        x = self.conv2(x, edge_index)\n",
    "\n",
    "        return F.log_softmax(x, dim=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 206,
   "id": "821be721-2275-4ee3-bc9a-edc46120fd50",
   "metadata": {},
   "outputs": [],
   "source": [
    "from torch_geometric.nn import MessagePassing, global_mean_pool\n",
    "from torch_geometric.utils import add_self_loops, degree\n",
    "from ocpmodels.datasets.embeddings import KHOT_EMBEDDINGS, QMOF_KHOT_EMBEDDINGS, ATOMIC_RADII\n",
    "\n",
    "class messageLayer(MessagePassing):\n",
    "    \n",
    "    def __init__(self, my_node_dim, edge_dim):\n",
    "        super(messageLayer, self).__init__(aggr=\"mean\", node_dim=-2)\n",
    "        self.my_node_dim = my_node_dim\n",
    "        self.edge_dim = edge_dim\n",
    "        \n",
    "        self.msg1 = nn.Linear(2*self.my_node_dim + self.edge_dim, self.my_node_dim)\n",
    "        self.msg2 = nn.Linear(self.my_node_dim, self.my_node_dim)\n",
    "        \n",
    "    def forward(self, x_emb, x_rad, edge_index, edge_attr, energy):\n",
    "        num_nodes = x_emb.shape[0]\n",
    "        #print(\"num_nodes =\", num_nodes)\n",
    "        \n",
    "        # Step 1: Add self-loops to the adjacency matrix.\n",
    "        #edge_index, _ = add_self_loops(edge_index, num_nodes=num_nodes)\n",
    "        \n",
    "        #print(\"x.shape =\", x_emb.shape)\n",
    "        #print(\"x.size(0) =\", x_emb.size(0))\n",
    "        #print(\"edge_index =\", edge_index.shape)\n",
    "        #print(\"edge_attr =\", edge_attr.shape)\n",
    "        #print(energy.shape)\n",
    "        x = torch.cat([x_emb, x_rad, energy], dim=1)\n",
    "        return self.propagate(edge_index, x=x, edge_attr=edge_attr, size=(x.size(0), x.size(0)))\n",
    "        \n",
    "    def message(self, x_i, x_j, edge_attr):\n",
    "        #print(x_i.shape)\n",
    "        #print(x_j.shape)\n",
    "        edge_attr = edge_attr.view(edge_attr.size(0), 1)\n",
    "        #print(edge_attr.shape)\n",
    "        z = self.msg1(torch.cat([x_i, x_j, edge_attr], dim=1))\n",
    "        z = F.leaky_relu(z)\n",
    "        z = self.msg2(z)\n",
    "        return z\n",
    "    \n",
    "    def update(self, aggr_out):\n",
    "        return aggr_out\n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 223,
   "id": "9ae8dbc1-edd0-4283-9dd3-493d50461180",
   "metadata": {},
   "outputs": [],
   "source": [
    "from torch import nn\n",
    "\n",
    "class messageGNN(torch.nn.Module):\n",
    "    \n",
    "    def __init__(self, atom_embedding_size, edge_embedding_size):\n",
    "        super(messageGNN, self).__init__()\n",
    "        \n",
    "        self.atom_embedding_size = atom_embedding_size\n",
    "        self.edge_embedding_size = edge_embedding_size\n",
    "        \n",
    "        # atomic radii\n",
    "        radii = ATOMIC_RADII\n",
    "        self.radii = torch.zeros(101, 1)\n",
    "        for i in range(101): \n",
    "            self.radii[i] = torch.tensor(radii[i])\n",
    "        \n",
    "        # get embeddings list\n",
    "        embeddings = KHOT_EMBEDDINGS\n",
    "        # create copy of embeddings as torch tensor\n",
    "        self.embedding = torch.zeros(100, len(embeddings[1]))\n",
    "        for i in range(100):\n",
    "            # don't forget that elements start at 1 \n",
    "            self.embedding[i] = torch.tensor(embeddings[i + 1])\n",
    "        # learnable layer to compress embedding\n",
    "        self.embedding_fc1 = nn.Linear(len(embeddings[1]), self.atom_embedding_size)\n",
    "        self.embedding_fc2 = nn.Linear(self.atom_embedding_size, self.atom_embedding_size)\n",
    "        \n",
    "        #print(self.atom_embedding_size)\n",
    "        #print(self.edge_embedding_size)\n",
    "        self.msg1 = messageLayer(self.atom_embedding_size + 1 + 1, self.edge_embedding_size)\n",
    "        \n",
    "        self.lin1 = nn.Linear(self.atom_embedding_size + 1 + 3 + 1, self.atom_embedding_size)\n",
    "        self.linOut = nn.Linear(self.atom_embedding_size, 3)\n",
    "        \n",
    "        self.eng1 = nn.Linear(self.atom_embedding_size + 1 + 1, self.atom_embedding_size)\n",
    "        self.engOut = nn.Linear(self.atom_embedding_size, 1)\n",
    "        \n",
    "    def forward(self, data):\n",
    "        #print(data)\n",
    "        # get attributes from data\n",
    "        x_atom, x_pos = data.atomic_numbers, data.pos\n",
    "        edge_index, edge_dist = data.edge_index, data.distances\n",
    "        batch = data.batch\n",
    "        energy = data.y_init\n",
    "        #print(\"x_atom.shape =\", x_atom.shape)\n",
    "        #print(\"batch.shape =\", batch.shape)\n",
    "        #print(\"energy.shape =\", energy.shape)\n",
    "        energyCopy = energy.clone().detach().numpy()\n",
    "        energy = torch.zeros(batch.shape[0], 1)\n",
    "        j = 0\n",
    "        for i in batch:\n",
    "            energy[j] = torch.tensor(energyCopy[i])\n",
    "            j += 1\n",
    "        #print(\"energies.shape =\", energies.shape)\n",
    "        \n",
    "        # find condensed embeddings for atoms\n",
    "        # element number -> big embedding -> small embedding\n",
    "        x_emb = self.embedding_fc1(self.embedding[x_atom.long()])\n",
    "        x_emb = F.leaky_relu(x_emb)\n",
    "        x_emb = self.embedding_fc2(x_emb)\n",
    "        \n",
    "        # find radii \n",
    "        x_rad = self.radii[x_atom.long()]\n",
    "        #print(x_emb.shape)\n",
    "        #print(x_rad.shape)\n",
    "        \n",
    "        # do message passing with embeddings \n",
    "        x = self.msg1(x_emb, x_rad, edge_index, edge_dist, energy)\n",
    "        z = global_mean_pool(x, batch)\n",
    "        \n",
    "        # turn result from size of node embedding to size of goal\n",
    "        x = self.lin1(torch.cat([x, x_pos], dim=1))\n",
    "        x = F.leaky_relu(x)\n",
    "        x = self.linOut(x)\n",
    "        \n",
    "        # calculate total energy\n",
    "        #z = global_mean_pool(x, batch)\n",
    "        z = F.dropout(z, p=0.3, training=self.training)\n",
    "        z = self.eng1(z)\n",
    "        z = F.leaky_relu(z)\n",
    "        z = self.engOut(z)\n",
    "        \n",
    "        return x, z\n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 224,
   "id": "f1baae8f-1029-41a0-b2db-bdf8f36a0d33",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\ipunc\\AppData\\Local\\Temp/ipykernel_6600/2137216186.py:52: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).\n",
      "  energy[j] = torch.tensor(energyCopy[i])\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "loss = 4128.027784472656 eng loss = 11664.455554296876\n",
      "loss = 325.5091760742188 eng loss = 9240.173178125\n",
      "loss = 182.26807817382812 eng loss = 9565.845580859375\n",
      "loss = 180.55560192871093 eng loss = 10031.427580859376\n",
      "loss = 179.99999150390624 eng loss = 10216.446145898437\n",
      "loss = 179.78672934570312 eng loss = 10342.21423359375\n",
      "loss = 179.48857747802734 eng loss = 10300.991773046875\n",
      "loss = 179.85551459960936 eng loss = 10364.3096859375\n",
      "loss = 179.6115073486328 eng loss = 10211.41513125\n",
      "loss = 179.5494567138672 eng loss = 10166.88345390625\n",
      "loss = 179.52657722167967 eng loss = 9959.480923046875\n",
      "loss = 178.81321557617187 eng loss = 9749.308491796875\n",
      "loss = 178.84955107421874 eng loss = 9900.646127734375\n",
      "loss = 178.80979702148437 eng loss = 9720.223628125\n",
      "loss = 178.9947372314453 eng loss = 9714.1207578125\n",
      "loss = 179.00353208007812 eng loss = 9301.112981640625\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[1;32m~\\AppData\\Local\\Temp/ipykernel_6600/1994926906.py\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[0;32m     23\u001b[0m         \u001b[1;31m#print(loss.shape)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     24\u001b[0m         \u001b[1;31m#loss = torch.cdist(out, data.pos_relaxed)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 25\u001b[1;33m         \u001b[0mloss\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mbackward\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     26\u001b[0m         \u001b[0moptimizer\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mstep\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     27\u001b[0m     \u001b[0mprint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m\"loss =\"\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mtotalLoss\u001b[0m\u001b[1;33m/\u001b[0m\u001b[0mlen\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mdataset\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;34m\"eng loss =\"\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mtotalEngLoss\u001b[0m\u001b[1;33m/\u001b[0m\u001b[0mlen\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mdataset\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\anaconda3\\envs\\ocp-models\\lib\\site-packages\\torch\\tensor.py\u001b[0m in \u001b[0;36mbackward\u001b[1;34m(self, gradient, retain_graph, create_graph, inputs)\u001b[0m\n\u001b[0;32m    243\u001b[0m                 \u001b[0mcreate_graph\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mcreate_graph\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    244\u001b[0m                 inputs=inputs)\n\u001b[1;32m--> 245\u001b[1;33m         \u001b[0mtorch\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mautograd\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mbackward\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mgradient\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mretain_graph\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mcreate_graph\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0minputs\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0minputs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    246\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    247\u001b[0m     \u001b[1;32mdef\u001b[0m \u001b[0mregister_hook\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mhook\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\anaconda3\\envs\\ocp-models\\lib\\site-packages\\torch\\autograd\\__init__.py\u001b[0m in \u001b[0;36mbackward\u001b[1;34m(tensors, grad_tensors, retain_graph, create_graph, grad_variables, inputs)\u001b[0m\n\u001b[0;32m    143\u001b[0m         \u001b[0mretain_graph\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mcreate_graph\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    144\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 145\u001b[1;33m     Variable._execution_engine.run_backward(\n\u001b[0m\u001b[0;32m    146\u001b[0m         \u001b[0mtensors\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mgrad_tensors_\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mretain_graph\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mcreate_graph\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0minputs\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    147\u001b[0m         allow_unreachable=True, accumulate_grad=True)  # allow_unreachable flag\n",
      "\u001b[1;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "# Message GNN\n",
    "device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')\n",
    "model = messageGNN(16, 1).to(device)\n",
    "data = train_loader\n",
    "optimizer = torch.optim.Adam(model.parameters(), lr=0.002, weight_decay=5e-4)\n",
    "batchSize = 16\n",
    "\n",
    "model.train()\n",
    "for epoch in range(1000):\n",
    "    totalLoss = 0\n",
    "    totalEngLoss = 0\n",
    "    for batch in data:\n",
    "        optimizer.zero_grad()\n",
    "        out, engPred = model(batch)\n",
    "        #print(batch.natoms.sum())\n",
    "        #print(out.shape)\n",
    "        #print(data.pos_relaxed)\n",
    "        loss = ((out-batch.pos_relaxed)**2).sum()\n",
    "        engLoss = ((engPred - batch.y_relaxed)**2).sum() * batchSize\n",
    "        totalEngLoss += engLoss.detach().numpy()\n",
    "        # uncomment below to optimize for both position and global energy\n",
    "        #loss += engLoss\n",
    "        totalLoss += loss.detach().numpy()\n",
    "        #print(loss.shape)\n",
    "        #loss = torch.cdist(out, data.pos_relaxed)\n",
    "        loss.backward()\n",
    "        optimizer.step()\n",
    "    print(\"loss =\", totalLoss/len(data.dataset), \"eng loss =\", totalEngLoss/len(data.dataset))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "829885c5-a22a-4ba2-9170-480b5b8c4158",
   "metadata": {},
   "outputs": [],
   "source": [
    "device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')\n",
    "model = GCN().to(device)\n",
    "data = dataset[0].to(device)\n",
    "optimizer = torch.optim.Adam(model.parameters(), lr=0.01, weight_decay=5e-4)\n",
    "\n",
    "model.train()\n",
    "for epoch in range(20000):\n",
    "    optimizer.zero_grad()\n",
    "    out = model(data)\n",
    "    #print(out.shape)\n",
    "    #print(data.pos_relaxed)\n",
    "    loss = ((out-data.pos_relaxed)**2).sum()\n",
    "    #print(loss.shape)\n",
    "    print(\"loss =\", loss.detach().numpy())\n",
    "    #loss = torch.cdist(out, data.pos_relaxed)\n",
    "    loss.backward()\n",
    "    optimizer.step()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 174,
   "id": "73ab4a42-7e12-4fed-84b5-88a839958ac6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10000\n"
     ]
    }
   ],
   "source": [
    "dataset = SinglePointLmdbDataset({\"src\": \"D:\\\\is2res_train_val_test_lmdbs\\\\data\\\\is2re\\\\10k\\\\train\\\\data.lmdb\"})\n",
    "print(len(dataset))\n",
    "test_loader = DataLoader(dataset, batch_size=1, shuffle=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 176,
   "id": "31c0df31-a959-437c-9dc2-0674b115db8c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Batch(atomic_numbers=[125], batch=[125], cell=[1, 3, 3], cell_offsets=[4022, 3], distances=[4022], edge_index=[2, 4022], fixed=[125], force=[125, 3], natoms=[1], pos=[125, 3], pos_relaxed=[125, 3], ptr=[2], sid=[1], tags=[125], y_init=[1], y_relaxed=[1])\n",
      "125\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<subprocess.Popen at 0x204ca660250>"
      ]
     },
     "execution_count": 176,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from ase.io import read\n",
    "from ase.visualize import view\n",
    "from ase.io import write\n",
    "import time\n",
    "\n",
    "data = next(iter(test_loader))\n",
    "print(data)\n",
    "\n",
    "out = model(data).detach().numpy()\n",
    "#print(out)\n",
    "\n",
    "xInit = data.pos[:,0]\n",
    "yInit = data.pos[:,1]\n",
    "zInit = data.pos[:,2]\n",
    "xTrue = data.pos_relaxed[:,0]\n",
    "yTrue = data.pos_relaxed[:,1]\n",
    "zTrue = data.pos_relaxed[:,2]\n",
    "xPred = out[:,0]\n",
    "yPred = out[:,1]\n",
    "zPred = out[:,2]\n",
    "atom = data.atomic_numbers\n",
    "print(len(atom))\n",
    "\n",
    "f = open(\"adsPictureInit.xyz\", \"w\")\n",
    "f.write(str(len(atom))+\"\\n\")\n",
    "for i in range(len(atom)):\n",
    "    f.write(\"\\n{}\\t {}\\t {}\\t {}\".format(symbols[int(atom[i])-1],xInit[i],yInit[i],zInit[i]))\n",
    "f.close()\n",
    "\n",
    "f = open(\"adsPicturePred.xyz\", \"w\")\n",
    "f.write(str(len(atom))+\"\\n\")\n",
    "for i in range(len(atom)):\n",
    "    f.write(\"\\n{}\\t {}\\t {}\\t {}\".format(symbols[int(atom[i])-1],xPred[i],yPred[i],zPred[i]))\n",
    "f.close()\n",
    "\n",
    "f = open(\"adsPictureTrue.xyz\", \"w\")\n",
    "f.write(str(len(atom))+\"\\n\")\n",
    "for i in range(len(atom)):\n",
    "    f.write(\"\\n{}\\t {}\\t {}\\t {}\".format(symbols[int(atom[i])-1],xTrue[i],yTrue[i],zTrue[i]))\n",
    "f.close()\n",
    "\n",
    "pic = read(\"adsPictureInit.xyz\")\n",
    "view(pic)\n",
    "#write(\"adsPictureInit.png\", pic)\n",
    "time.sleep(3)\n",
    "\n",
    "pic = read(\"adsPicturePred.xyz\")\n",
    "view(pic)\n",
    "time.sleep(3)\n",
    "\n",
    "pic = read(\"adsPictureTrue.xyz\")\n",
    "view(pic)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 152,
   "id": "21434761-12cb-406f-81f5-23e878e3dbc2",
   "metadata": {},
   "outputs": [
    {
     "ename": "AttributeError",
     "evalue": "'Data' object has no attribute 'test_mask'",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mAttributeError\u001b[0m                            Traceback (most recent call last)",
      "\u001b[1;32m~\\AppData\\Local\\Temp/ipykernel_48800/3032524375.py\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[0;32m      1\u001b[0m \u001b[0mmodel\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0meval\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      2\u001b[0m \u001b[0mpred\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mmodel\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0margmax\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mdim\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;36m1\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m----> 3\u001b[1;33m \u001b[0mcorrect\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;33m(\u001b[0m\u001b[0mpred\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mdata\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mtest_mask\u001b[0m\u001b[1;33m]\u001b[0m \u001b[1;33m==\u001b[0m \u001b[0mdata\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0my\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mdata\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mtest_mask\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0msum\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m      4\u001b[0m \u001b[0macc\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mcorrect\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;33m/\u001b[0m \u001b[0mint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mtest_mask\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0msum\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      5\u001b[0m \u001b[0mprint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34mf'Accuracy: {acc:.4f}'\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mAttributeError\u001b[0m: 'Data' object has no attribute 'test_mask'"
     ]
    }
   ],
   "source": [
    "model.eval()\n",
    "pred = model(data).argmax(dim=1)\n",
    "correct = (pred[data.test_mask] == data.y[data.test_mask]).sum()\n",
    "acc = int(correct) / int(data.test_mask.sum())\n",
    "print(f'Accuracy: {acc:.4f}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "39a71a84-0e77-4d25-af8c-49fd5c1c58b2",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "90b996bd-6f57-406d-bb43-7aeff90c2489",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6404548e-6114-4f30-b32d-7b1ac8f88194",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "ocp-models",
   "language": "python",
   "name": "ocp-models"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}