bond_gan_mk1.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This notebook addresses the question, \"Can we represent a molecule as a graph via a 1D column vector or a 2D matrix of fixed length, with maximum number of atoms n_rows?\" Then, can we use this representation to learn neural fingerprints? E.g., can we make an aromatic ring detector? \n",
    "\n",
    "Scheme:\n",
    "feature_matrix = X\n",
    "for each ligand:\n",
    "    choose a central atom. this can be the atom (node) that minimizes distance to furthest heavy atom in graph.\n",
    "    set first row of X to be this central atom\n",
    "    set next four rows to be the atoms bonded to that centrl atom\n",
    "        set zeros for rows where row ind > n_bonds of atom\n",
    "    for each of those atoms:\n",
    "        repeat. find their neighbors. add to matrix.\n",
    "\n",
    "algorithm: breadth-first search:\n",
    "1. create networkx graph based on molecule\n",
    "2. find \"central\" atom (different strategies)\n",
    "3. define atom matrix of size (1+4+4*3^(L-1)) x (n_features_per_atom)\n",
    "4. start atom queue q\n",
    "5. central_atom.layer = 0; central_atom.row_idx = 0;\n",
    "6. q.enqueue(central_atom)\n",
    "7. define adjacency matrix of size (1+4+4*3^(L-1)) x 4\n",
    "\n",
    "def get_row_idx(curr_layer, prev_row_idx, curr_neighbor_idx):\n",
    "    if curr_layer == 0:\n",
    "        return(0)\n",
    "    if curr_layer == 1:\n",
    "        row_idx = 1 + curr_neighbor_idx\n",
    "    if layer == 2:\n",
    "        last_max = 5\n",
    "        row_idx = last_max + (3*(prev_row_idx-last_max)) + curr_neighbor_idx\n",
    "    if layer > 2:\n",
    "        last_max = 5 + 4*3^(curr_layer-2) \n",
    "        row_idx = last_max + 3*(prev_row_idx-last_max) + curr_neighbor_idx\n",
    "    return(row_idx)\n",
    "    \n",
    "\n",
    "while q.is_not_empty():\n",
    "    a = q.dequeue()\n",
    "    a.visited = True\n",
    "    for n_idx, n in enumerate(a.neighbors()):\n",
    "        if not n.visited:\n",
    "            row_idx = c\n",
    "            n.layer = a.layer + 1\n",
    "            row_idx = get_row_idx(n.layer, a.row_idx, n_idx)\n",
    "            n.row_idx = row_idx\n",
    "            adj_matrix[a.row_idx][n_idx] = n.row_idx\n",
    "            atom_matrix[row_idx][elem_to_idx[n.elem]] = 1\n",
    "\n",
    "input_matrix = tf.concat([atom_matrix, atom_matrix[adj_matrix[:,0]], atom_matrix[adj_matrix[:,1]], atom_matrix[adj_matrix[:,2]], atom_matrix[adj_matrix[:,3]]\n",
    "\n",
    "neural net:\n",
    "h1 = relu([tf.zeros([n_features_per_atom, 4]) * input_matrix + bias))\n",
    "h1_conc = tf.concat([h1, h1[adj_matrix[:,0], ..., h1[adj_matrix[:,3])\n",
    "\n",
    "repeat h1 to get h2\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "dihedral predictor pseudocode:\n",
    "\n",
    "get bonds for molecule\n",
    "create networkx graph out of molecule (use atom indices)\n",
    "\n",
    "for each edge:\n",
    "   for neighbor_i in atom_i.neighbors():\n",
    "       if neighbor_i == atom_j: continue\n",
    "       for neighbor_j in atom_j.neighbors():\n",
    "           if neighbor_j == atom_i: continue\n",
    "           dihedrals.append((neighbor_i, atom_i, neighbor_j, atom_j))\n",
    "           check to make sure (atom_j, neighbor_j, atom_i, neighbor_i)) not already in list\n",
    "\n",
    "for dihedral in dihedrals:\n",
    "    angle =  rdMolTransforms.GetDihedralDeg(c, 0,1,2,3)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "\n",
    "import matplotlib\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from rdkit import Chem\n",
    "from rdkit.Chem import AllChem, rdMolTransforms\n",
    "import os\n",
    "import fnmatch\n",
    "import numpy as np\n",
    "import deepchem as dc\n",
    "from scipy.sparse import csr_matrix\n",
    "from sklearn.preprocessing import OneHotEncoder\n",
    "from sklearn.mixture import GMM\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def get_torsions_angles(mol):\n",
    "    c = mol.GetConformer(0)\n",
    "\n",
    "    torsions = []\n",
    "    torsion_tuples = []\n",
    "    \n",
    "    torsion_matrix = np.zeros((100,3))\n",
    "    torsion_indices = np.zeros((100,200,12)).astype(np.uint8)\n",
    "    bond_idx = 0\n",
    "    for idx, bond in enumerate(mol.GetBonds()):\n",
    "        atom_i = bond.GetBeginAtom()\n",
    "        atom_j = bond.GetEndAtom()\n",
    "        if bond.IsInRing(): \n",
    "            continue\n",
    "        \n",
    "        exist_dihed = False\n",
    "        for neighbor_j in atom_j.GetNeighbors():\n",
    "            if neighbor_j.GetIdx() == atom_i.GetIdx():\n",
    "                continue\n",
    "            \n",
    "            dihed_idx = 0\n",
    "            for neighbor_i in atom_i.GetNeighbors():\n",
    "                if neighbor_i.GetIdx() == atom_j.GetIdx():\n",
    "                    continue\n",
    "                \n",
    "                exist_dihed=True\n",
    "                torsion_tuple = (neighbor_i.GetIdx(), atom_i.GetIdx(), atom_j.GetIdx(), neighbor_j.GetIdx())\n",
    "\n",
    "                torsion_matrix[bond_idx][dihed_idx] = np.cos(rdMolTransforms.GetDihedralRad(c, *torsion_tuple))\n",
    "                torsion_indices[bond_idx][torsion_tuple[0]][dihed_idx*4] = 1\n",
    "                torsion_indices[bond_idx][torsion_tuple[1]][dihed_idx*4+1] = 1\n",
    "                torsion_indices[bond_idx][torsion_tuple[2]][dihed_idx*4+2] = 1\n",
    "                torsion_indices[bond_idx][torsion_tuple[3]][dihed_idx*4+3] = 1\n",
    "                dihed_idx += 1\n",
    "            #torsion_matrix[bond_idx][3] = 3.14159 - np.abs(np.abs(torsion_matrix[bond_idx][2]-torsion_matrix[bond_idx][1])-3.14159)\n",
    "            #torsion_matrix[bond_idx][4] = 3.14159 - np.abs(np.abs(torsion_matrix[bond_idx][2]-torsion_matrix[bond_idx][0])-3.14159)\n",
    "            #torsion_matrix[bond_idx][5] = 3.14159 - np.abs(np.abs(torsion_matrix[bond_idx][1]-torsion_matrix[bond_idx][0])-3.14159)\n",
    "            break\n",
    "        if exist_dihed:\n",
    "            bond_idx += 1\n",
    "            break\n",
    "    return((torsion_indices, csr_matrix(torsion_matrix)))\n",
    "                "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def featurize_mols(mol_files):\n",
    "    featurizer = AdjacencyFingerprint(max_n_atoms=200)\n",
    "    features = []\n",
    "    for idx, mol_file in enumerate(mol_files):\n",
    "        if idx % 100 == 0:\n",
    "            print(idx)\n",
    "        if 1==1:\n",
    "            mol = Chem.MolFromMol2File(mol_file)\n",
    "            if mol is None:\n",
    "                features.append(None)\n",
    "                continue\n",
    "            torsions = get_torsions_angles(mol)\n",
    "            graph_feat = featurizer.featurize([mol])[0]\n",
    "            features.append((mol_file, torsions, graph_feat))\n",
    "        else:\n",
    "            features.append(None)\n",
    "    return(features)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from deepchem.feat.graph_features import ConvMolFeaturizer\n",
    "from deepchem.feat.adjacency_fingerprints import AdjacencyFingerprint"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle\n",
    "feature_file = \"./dihedral_features_pdbbind.pkl\"\n",
    "if not os.path.exists(feature_file):\n",
    "#if 1== 1:\n",
    "    pdbbind_dir = \"/home/evan/Documents/deep_docking/datasets/v2015/\"\n",
    "    def find_files(directory, pattern):\n",
    "        for root, dirs, files in os.walk(directory):\n",
    "            for basename in files:\n",
    "                if fnmatch.fnmatch(basename, pattern):\n",
    "                    filename = os.path.join(root, basename)\n",
    "                    yield filename\n",
    "    ligand_files = [f for f in find_files(pdbbind_dir, \"*ligand.mol2\")][:1]\n",
    "    features = featurize_mols(ligand_files)\n",
    "    with open(feature_file, \"wb\") as f:\n",
    "        pickle.dump(features, f, protocol=2)\n",
    "else:\n",
    "    with open(feature_file, \"rb\") as f:\n",
    "        features = pickle.load(f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "features = [f for f in features if f is not None]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 0.61143659, -0.99900387,  0.48252666],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ]])"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "features[0][1][1].toarray()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(features)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "alpha = 0.01\n",
    "n_layers = 0\n",
    "S = 64\n",
    "B = 200\n",
    "L_list = [50, 50, 50, 50]\n",
    "p = 75\n",
    "n_bonds=100\n",
    "\n",
    "x = tf.placeholder(tf.float32, shape=[S, B, p])\n",
    "\n",
    "non_zero_inds = tf.placeholder(tf.int32, shape=[None, S*n_bonds])\n",
    "\n",
    "adj_matrix = tf.placeholder(tf.float32, shape=[S, B, B])\n",
    "dihed_indices = tf.placeholder(tf.float32, shape=[S, n_bonds, B, 12])\n",
    "\n",
    "label_placeholder = tf.placeholder(\n",
    "    dtype='float32', shape=[None,3], name=\"label_placeholder\")\n",
    "\n",
    "phase = tf.placeholder(dtype='bool', name='phase')\n",
    "\n",
    "z = tf.placeholder(tf.float32,\n",
    "                          [None, 200, 75], name='z')\n",
    "\n",
    "z_ini = tf.placeholder(tf.float32,\n",
    "                          [None, 200, 75], name='z_ini')\n",
    "\n",
    "ones_like = tf.placeholder(tf.float32, shape=[None,1])\n",
    "zeros_like = tf.placeholder(tf.float32, shape=[None,1])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "#DISCRIMINATOR\n",
    "\n",
    "\"\"\"\n",
    "D_W_list = [None for i in range(n_layers)]\n",
    "D_b_list = [None for i in range(n_layers)]\n",
    "D_h_list = [None for i in range(n_layers)]\n",
    "\n",
    "\n",
    "\n",
    "for layer_idx in range(n_layers):\n",
    "    if layer_idx == 0:\n",
    "      L_in = p\n",
    "      L_out = L_list[0]\n",
    "      atom_matrix = x\n",
    "    else:\n",
    "      L_in = L_list[layer_idx-1]\n",
    "      L_out = L_list[layer_idx]\n",
    "      atom_matrix = D_h_list[layer_idx-1]\n",
    "\n",
    "    D_W_list[layer_idx] = tf.Variable(tf.truncated_normal([L_in, L_out], seed=2017), name=\"D_W_list%d\" %layer_idx)\n",
    "    D_b_list[layer_idx] = tf.Variable(tf.zeros([1, L_out]))\n",
    "\n",
    "    \n",
    "D_W2 = tf.Variable(tf.truncated_normal([L_list[-1]*12+3, 100]))\n",
    "D_b2 = tf.Variable(tf.zeros([1, 100]))\n",
    "\n",
    "D_W3 = tf.Variable(tf.truncated_normal([100, 3]))\n",
    "D_b3 = tf.Variable(tf.zeros([1, 3]))\n",
    "\n",
    "D_W4 = tf.Variable(tf.truncated_normal([3, 10]))\n",
    "D_b4 = tf.Variable(tf.zeros([1, 10]))\n",
    "\n",
    "D_W5 = tf.Variable(tf.truncated_normal([10, 1]))\n",
    "D_b5 = tf.Variable(tf.zeros([1, 1]))\n",
    "\n",
    "\"\"\"\n",
    "\n",
    "D_W0 = tf.Variable(tf.truncated_normal([3, 100]))\n",
    "D_b0 = tf.Variable(tf.zeros([1,100]))\n",
    "\n",
    "D_W1 = tf.Variable(tf.truncated_normal([100, 100]))\n",
    "D_b1 = tf.Variable(tf.zeros([1,100]))\n",
    "\n",
    "D_W2 = tf.Variable(tf.truncated_normal([100, 1]))\n",
    "D_b2 = tf.Variable(tf.zeros([1,1]))\n",
    "\n",
    "def adjacency_conv_layer(atom_matrix, W, b, L_in, L_out, layer_idx, is_training=True):\n",
    "    print(\"layer_idx: %d\" %(layer_idx))\n",
    "    h = tf.matmul(adj_matrix, atom_matrix, name=\"adj_conv1\")\n",
    "    h = tf.reshape(h, shape=(S*B, L_in))\n",
    "\n",
    "    h = tf.matmul(h, W) + b\n",
    "    h = tf.maximum(alpha*h, h)\n",
    "    \n",
    "    h = tf.reshape(h, (S, B, L_out))\n",
    "    #h = tf.contrib.layers.batch_norm(h, \n",
    "    #                                    center=True, scale=False, \n",
    "    #                                    is_training=is_training)\n",
    "\n",
    "    return(h)\n",
    "\n",
    "def discriminator(angles, is_training=True):\n",
    "    D_h0 = tf.matmul(angles, D_W0) + D_b0\n",
    "    D_h0 = tf.maximum(alpha*D_h0, D_h0)\n",
    "\n",
    "    D_h1 = tf.matmul(D_h0, D_W1) + D_b1\n",
    "    D_h1 = tf.maximum(alpha*D_h1, D_h1)\n",
    "    \n",
    "    D_h2 = tf.matmul(D_h1, D_W2) + D_b2\n",
    "    #D_h2 = tf.maximum(alpha*D_h2, D_h2)\n",
    "    \n",
    "    return(D_h2)\n",
    "    \n",
    "    \"\"\"\n",
    "    \n",
    "    for layer_idx in range(n_layers):\n",
    "        if layer_idx == 0:\n",
    "          L_in = p\n",
    "          L_out = L_list[0]\n",
    "          atom_matrix = x\n",
    "        else:\n",
    "          L_in = L_list[layer_idx-1]\n",
    "          L_out = L_list[layer_idx]\n",
    "          atom_matrix = D_h_list[layer_idx-1]\n",
    "        \n",
    "        D_h_list[layer_idx] = adjacency_conv_layer(atom_matrix, D_W_list[layer_idx], D_b_list[layer_idx], L_in, L_out, layer_idx)\n",
    "\n",
    "    L_final = L_list[n_layers-1]\n",
    "    h_final = tf.reshape(D_h_list[layer_idx], (S, B, L_final))\n",
    "\n",
    "    #add dihedral regressor layers\n",
    "\n",
    "    d0 = []\n",
    "    for i in range(0, S):\n",
    "        mol_tuple = []\n",
    "        for j in range(0, 12):\n",
    "          entry = h_final[i]\n",
    "          indices = dihed_indices[i][:,:,j]\n",
    "          atom_list = tf.matmul(indices, entry, name=\"disc1\")\n",
    "          atom_list = tf.reshape(atom_list, (n_bonds, L_final))\n",
    "          mol_tuple.append(atom_list)\n",
    "        mol_tuple = tf.reshape(tf.stack(mol_tuple, axis=1), (n_bonds, L_final*12))\n",
    "        d0.append(mol_tuple)\n",
    "\n",
    "    d0 = tf.concat(d0, axis=0)\n",
    "    d0 = tf.concat([d0, angles], axis=1)\n",
    "    d0 = tf.matmul(tf.cast(non_zero_inds, tf.float32), d0)\n",
    "\n",
    "\n",
    "    d2 = tf.matmul(d0, D_W2, name=\"disc2\") + D_b2\n",
    "    d2 = tf.maximum(d2, alpha*d2)\n",
    "\n",
    "    #d2 = tf.contrib.layers.batch_norm(d2, \n",
    "    #                                    center=True, scale=False, \n",
    "    #                                    is_training=is_training)\n",
    "    \n",
    "    d3 = tf.matmul(d2, D_W3, name=\"disc3\") + D_b3\n",
    "    d3 = tf.maximum(d3, alpha*d3)\n",
    "    \n",
    "    #d3 = tf.nn.tanh(d3)\n",
    "\n",
    "    #d3 = tf.contrib.layers.batch_norm(d3, \n",
    "    #                                    center=True, scale=False, \n",
    "    #                                    is_training=is_training)\n",
    "    \n",
    "                                                                                                                                                               \n",
    "    d4 = tf.matmul(d3, D_W4, name=\"disc3\") + D_b4\n",
    "    \n",
    "    #\n",
    "    output = tf.matmul(d4, D_W5, name=\"disc4\") + D_b5\n",
    "    \n",
    "    #D_prob = tf.nn.sigmoid(D_logit)\n",
    "    return(output)\n",
    "    \"\"\"\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "def atan2(y, x, epsilon=1.0e-12):\n",
    "  # Add a small number to all zeros, to avoid division by zero:\n",
    "  x = tf.where(tf.equal(x, 0.0), x+epsilon, x)\n",
    "  y = tf.where(tf.equal(y, 0.0), y+epsilon, y)\n",
    "\n",
    "  angle = tf.where(tf.greater(x,0.0), tf.atan(y/x), tf.zeros_like(x))\n",
    "  angle = tf.where(tf.logical_and(tf.less(x,0.0),  tf.greater_equal(y,0.0)), tf.atan(y/x) + np.pi, angle)\n",
    "  angle = tf.where(tf.logical_and(tf.less(x,0.0),  tf.less(y,0.0)), tf.atan(y/x) - np.pi, angle)\n",
    "  angle = tf.where(tf.logical_and(tf.equal(x,0.0), tf.greater(y,0.0)), 0.5*np.pi * tf.ones_like(x), angle)\n",
    "  angle = tf.where(tf.logical_and(tf.equal(x,0.0), tf.less(y,0.0)), -0.5*np.pi * tf.ones_like(x), angle)\n",
    "  angle = tf.where(tf.logical_and(tf.equal(x,0.0), tf.equal(y,0.0)), tf.zeros_like(x), angle)\n",
    "  return angle\n",
    "\n",
    "#GENERATOR\n",
    "\n",
    "G_W_list = [None for i in range(n_layers+1)]\n",
    "G_b_list = [None for i in range(n_layers+1)]\n",
    "G_h_list = [None for i in range(n_layers+1)]\n",
    "\n",
    "\n",
    "L_final = p\n",
    "for layer_idx in range(n_layers):\n",
    "    if layer_idx == 0:\n",
    "      L_in = p\n",
    "      L_out = L_list[0]\n",
    "      atom_matrix = x\n",
    "    else:\n",
    "      L_in = L_list[layer_idx-1]\n",
    "      L_out = L_list[layer_idx]\n",
    "      atom_matrix = G_h_list[layer_idx-1]\n",
    "\n",
    "    G_W_list[layer_idx] = tf.Variable(tf.truncated_normal([L_in, L_out], seed=2017), name=\"G_W_list%d\" %layer_idx)\n",
    "    G_b_list[layer_idx] = tf.Variable(tf.zeros([1, L_out]))\n",
    "\n",
    "    \n",
    "G_W2 = tf.Variable(tf.truncated_normal([L_final*12, 3]))\n",
    "G_b2 = tf.Variable(tf.zeros([1, 3]))\n",
    "\n",
    "#G_W3 = tf.Variable(tf.truncated_normal([100, 3]))\n",
    "#G_b3 = tf.Variable(tf.zeros([1, 3]))\n",
    "\n",
    "#G_W4 = tf.Variable(tf.truncated_normal([100, 3]))\n",
    "#G_b4 = tf.Variable(tf.zeros([1, 3]))\n",
    "\n",
    "\n",
    "def gen_adjacency_conv_layer(atom_matrix, W, b, L_in, L_out, layer_idx, z, is_training):\n",
    "    print(\"layer_idx: %d\" %(layer_idx))\n",
    "    h = tf.matmul(adj_matrix, atom_matrix, name=\"gen0_%d\" %layer_idx)\n",
    "    h = tf.reshape(h, shape=(S*B, L_in))\n",
    "\n",
    "    h = tf.matmul(h, W, name=\"gen1\") + b\n",
    "    #h = tf.maximum(alpha*h, h)\n",
    "    h = tf.nn.tanh(h)\n",
    "    \n",
    "    h = tf.reshape(h, (S, B, L_out))\n",
    "    #h = tf.contrib.layers.batch_norm(h, \n",
    "    #                                    center=True, scale=False, \n",
    "    #                                    is_training=is_training)\n",
    "\n",
    "    return(h)\n",
    "\n",
    "def generator(z, is_training=True):\n",
    "    layer_idx = 0\n",
    "    atom_matrix = tf.add(x, z)\n",
    "    G_h_list[0] = atom_matrix\n",
    "    \n",
    "    L_final = p\n",
    "    \n",
    "    for layer_idx in range(n_layers):\n",
    "        if layer_idx == 0:\n",
    "          L_in = p\n",
    "          L_out = L_list[0]\n",
    "\n",
    "        else:\n",
    "          L_in = L_list[layer_idx-1]\n",
    "          L_out = L_list[layer_idx]\n",
    "          atom_matrix = G_h_list[layer_idx-1]\n",
    "        \n",
    "        G_h_list[layer_idx+1] = gen_adjacency_conv_layer(atom_matrix, G_W_list[layer_idx], G_b_list[layer_idx], L_in, L_out, layer_idx, z, is_training)\n",
    "    \n",
    "    g_h_final = tf.reshape(G_h_list[layer_idx], (S, B, L_final))\n",
    "\n",
    "    #add dihedral regressor layers\n",
    "\n",
    "    def mult(idx):\n",
    "        return(tf.matmul(dihed_indices[:,:,:,tf.cast(idx, tf.int32)], g_h_final))\n",
    "\n",
    "    elems = np.array([0., 1., 2., 3., 4., 5., 6., 7., 8., 9., 10., 11.]).astype(np.float32)\n",
    "    g_d0 = tf.concat(tf.map_fn(mult, elems), axis=2)\n",
    "    g_d0 = tf.reshape(g_d0, (S*n_bonds, L_final*12))\n",
    "    \n",
    "    \n",
    "    #g_d0 = []\n",
    "    #for l in range(0,12):\n",
    "    #    indices = dihed_indices[:,:,:,l]\n",
    "    #    g_d0.append(tf.matmul(indices, g_h_final))\n",
    "    \n",
    "    #for i in range(0, S):\n",
    "    #    mol_tuple = []\n",
    "    #    for j in range(0, 12):\n",
    "    #      entry = g_h_final[i]\n",
    "    #      indices = dihed_indices[i][:,:,j]\n",
    "    #      atom_list = tf.matmul(indices, entry, name='gen2')\n",
    "    #      atom_list = tf.reshape(atom_list, (n_bonds, L_final))\n",
    "    #      mol_tuple.append(atom_list)\n",
    "    #    mol_tuple = tf.reshape(tf.stack(mol_tuple, axis=1), (n_bonds, L_final*12))\n",
    "    #    g_d0.append(mol_tuple)\n",
    "\n",
    "    #g_d0 = tf.concat(g_d0, axis=0)\n",
    "    \n",
    "    g_d2 = tf.matmul(g_d0, G_W2, name='gen3') + G_b2\n",
    "    #g_d2 = tf.tanh(g_d2)\n",
    "    #g_d2 = tf.maximum(alpha*g_d2, g_d2)\n",
    "\n",
    "    \n",
    "    #g_d2 = tf.contrib.layers.batch_norm(g_d2, \n",
    "    #                                    center=True, scale=False, \n",
    "    #                                    is_training=is_training)\n",
    "    \n",
    "    #g_d3 = tf.matmul(g_d2, G_W3) + G_b3\n",
    "    #g_d3 = tf.nn.tanh(g_d3)\n",
    "\n",
    "    #g_d3 = tf.contrib.layers.batch_norm(g_d3, \n",
    "    #                                    center=True, scale=False, \n",
    "    #                                    is_training=is_training)\n",
    "                                                                                                                                                               \n",
    "    #g_d4 = tf.matmul(g_d3, G_W4) + G_b4\n",
    "    \n",
    "    output_i = g_d2\n",
    "    output = tf.matmul(tf.cast(non_zero_inds, tf.float32), output_i)\n",
    "\n",
    "    #d3_cos = tf.cos(g_d4)\n",
    "    #d3_sin = tf.sin(g_d4)\n",
    "    #output = atan2(d3_sin, d3_cos)\n",
    "    \n",
    "    #output = tf.concat([output,\n",
    "    #                    tf.reshape(3.14159 - tf.abs(tf.abs(output[:,2]-output[:,1])-3.14159), (-1,1)),\n",
    "    #                    tf.reshape(3.14159 - tf.abs(tf.abs(output[:,2]-output[:,0])-3.14159), (-1,1)),\n",
    "    #                    tf.reshape(3.14159 - tf.abs(tf.abs(output[:,1]-output[:,0])-3.14159), (-1,1))],\n",
    "    #                   axis=1)\n",
    "    print(\"output\")\n",
    "    print(output)\n",
    "    \n",
    "    G_logit = tf.nn.sigmoid(output)\n",
    "\n",
    "    return(output, output_i)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "theta_D = [D_W0, D_W1, D_W2, D_b0, D_b1, D_b2]\n",
    "#theta_D = [D_W_list[0], D_W2, D_W3, D_W4, D_W5, D_b_list[0], D_b2, D_b3, D_b4, D_b5]\n",
    "\n",
    "theta_G = [G_W2, G_b2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "gmm = GMM(2, n_iter=1., covariance_type=\"spherical\")\n",
    "\n",
    "gmm.means_ = np.array([[-10], [10]])\n",
    "gmm.covars_ = np.array([[0.0001], [0.0001]])\n",
    "\n",
    "def construct_feed_dict(X, start=None,\n",
    "                      stop=None, y=None,\n",
    "                      keep_prob=1.0, train=False):\n",
    "    if start is None:\n",
    "      start = 0\n",
    "      stop = len(X)\n",
    "\n",
    "    adj = [X[idx][2][0].toarray().astype(np.float32) for idx in range(start, stop)]\n",
    "    A_batch = [X[idx][2][1].toarray() for idx in range(start, stop)]\n",
    "    D_batch = [X[idx][1][0] for idx in range(start, stop)]\n",
    "    y_batch = [X[idx][1][1].toarray() for idx in range(start, stop)]\n",
    "\n",
    "    y_batch = np.concatenate(y_batch, axis=0)\n",
    "    non_zero_batch = np.where(y_batch[:,0] != 0.)[0]\n",
    "    y_batch = y_batch[non_zero_batch]\n",
    "    \n",
    "    ones_batch = np.random.uniform(0.7, 1.2, size=(len(non_zero_batch),1))\n",
    "    zeros_batch = np.random.uniform(0., 0.3, size=(len(non_zero_batch),1))\n",
    "\n",
    "    #y_batch = gmm.sample(len(y_batch))\n",
    "    \n",
    "    y_batch = y_batch + np.random.normal(0, 0.03, size=y_batch.shape)\n",
    "    \n",
    "    #y_batch = np.squeeze(gmm.sample(len(y_batch)))\n",
    "    \n",
    "    #adj = [np.random.uniform(-1., 1., size=a.shape) for a in adj]\n",
    "    #A_batch = [np.random.uniform(-1., 1., size=a.shape) for a in A_batch]\n",
    "    #D_batch = [np.random.uniform(-1., 1., size=a.shape) for a in D_batch]\n",
    "\n",
    "\n",
    "    onehotter = OneHotEncoder(n_values = S*X[0][1][1].shape[0])\n",
    "    non_zero_onehot = onehotter.fit_transform(non_zero_batch).toarray().reshape((len(non_zero_batch),S*X[0][1][1].shape[0]))\n",
    "    \n",
    "    #z_batch = np.zeros((S, adj[0].shape[0], 75))\n",
    "    z_batch = np.random.uniform(-0.1, 0.1, size=(S, adj[0].shape[0],75))\n",
    "    #y_batch = np.random.random(size=(S*250))\n",
    "    \n",
    "    feed_dict = {x: A_batch,\n",
    "                 adj_matrix: adj,\n",
    "                 phase: train,\n",
    "                 label_placeholder: y_batch,\n",
    "                 non_zero_inds: non_zero_onehot,\n",
    "                 dihed_indices: D_batch,\n",
    "                 z:z_batch,\n",
    "                 ones_like: ones_batch,\n",
    "                 zeros_like: zeros_batch\n",
    "                }\n",
    "    return(feed_dict)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "output\n",
      "Tensor(\"MatMul:0\", shape=(?, 3), dtype=float32)\n",
      "output\n",
      "Tensor(\"MatMul_1:0\", shape=(?, 3), dtype=float32)\n",
      "Training epoch 0\n",
      "Iter: 0\n",
      "D loss: -0.0001181\n",
      "G_loss: -0.0001256\n",
      "()\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fe3bce610>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nZwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fe00acfd0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fd87691d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[-1.30132711  0.18542333  1.07594001]\n",
      "Training epoch 100\n",
      "Iter: 100\n",
      "D loss: 0.01982\n",
      "G_loss: 0.01766\n",
      "()\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\niiDoAFAEQQeAIgg6ABRB0AGgiP8B/f/j3zWhmJoAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fd87aa990>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAD8CAYAAACMwORRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n0bRe170QXQQAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fb803cf10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fb32c5710>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[-1.31420338  0.0982305   0.91644073]\n",
      "Training epoch 200\n",
      "Iter: 200\n",
      "D loss: 0.01734\n",
      "G_loss: 0.01518\n",
      "()\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fb8047850>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nf/CP1DMgACM8AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fb8033590>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nrkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9df9af10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[-1.03601849 -0.09766396  0.80952001]\n",
      "Training epoch 300\n",
      "Iter: 300\n",
      "D loss: 0.01421\n",
      "G_loss: 0.01188\n",
      "()\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9dfd2450>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\naa2EAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9dd8aad0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nXZKaMOiS1IRBl6QmDLokNfF/QndmESuo/hUAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9dcd1e90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[-0.71485353 -0.51566565  0.5694415 ]\n",
      "Training epoch 400\n",
      "Iter: 400\n",
      "D loss: 0.01283\n",
      "G_loss: 0.009909\n",
      "()\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAADBRJREFUeJzt3W2MXnWdxvHrkmpclCya3usDZXZ4QUgIomwmPpGsLrCb\nriV212gCioqSzJtVwZCQYhN5sdmkBuPuJm7WTBYWk21QgxiNVWlFDTEBYlsrFgaFYMUi2BISRTTB\nxssXc7Opk3ZmzsPMmfvX7ydpOuee0/v/66T95vTMOadOIgDA5HvR0AMAAPpB0AGgCIIOAEUQdAAo\ngqADQBEEHQCKIOgAUARBB4AiCDoAFLFhLRfbuHFjpqen13JJAJh4+/btezrJaLn91jTo09PT2rt3\n71ouCQATz/bPV7Ifp1wAoAiCDgBFEHQAKIKgA0ARBB0AiiDoAFAEQQeAIgg6ABRB0AGgiDW9UxTA\n+jG9bddgax/asWWwtSvjCB0AiiDoAFAEQQeAIgg6ABRB0AGgCIIOAEUQdAAogqADQBEEHQCKIOgA\nUARBB4Ailg267VttH7F98LjXbrb9sO0HbH/F9pmrOyYAYDkrOUK/TdLmRa/tkXRBkgsl/VTSjT3P\nBQBoaNmgJ7lH0jOLXtud5Nh48z5Jm1ZhNgBAA32cQ/+wpG/28D4AgA46Bd32dknHJO1cYp9Z23tt\n7z169GiX5QAAS2gddNtXS7pc0vuS5GT7JZlLMpNkZjQatV0OALCMVv9jke3Nkm6Q9LYkv+t3JABA\nGyu5bPF2SfdKOs/2YdvXSPqspDMk7bF9wPbnVnlOAMAylj1CT3LlCV6+ZRVmAQB0wJ2iAFAEQQeA\nIgg6ABRB0AGgCIIOAEUQdAAogqADQBEEHQCKIOgAUARBB4AiCDoAFEHQAaAIgg4ARRB0ACiCoANA\nEQQdAIog6ABQBEEHgCIIOgAUQdABoAiCDgBFEHQAKIKgA0ARywbd9q22j9g+eNxrr7S9x/Yj459f\nsbpjAgCWs5Ij9NskbV702jZJdyc5V9Ld420AwICWDXqSeyQ9s+jlrZI+P/7485L+qee5AAANtT2H\n/qokT44/fkrSq3qaBwDQ0oaub5AktnOyz9uelTQrSVNTU12XG8T0tl2DrHtox5ZB1gVWG3+nVkfb\nI/Rf2X6NJI1/PnKyHZPMJZlJMjMajVouBwBYTtugf03SB8cff1DSV/sZBwDQ1kouW7xd0r2SzrN9\n2PY1knZI+nvbj0i6bLwNABjQsufQk1x5kk9d2vMsAIAOuFMUAIog6ABQBEEHgCIIOgAUQdABoAiC\nDgBFEHQAKIKgA0ARBB0AiiDoAFAEQQeAIgg6ABRB0AGgCIIOAEUQdAAogqADQBEEHQCKIOgAUARB\nB4AiCDoAFEHQAaAIgg4ARRB0ACiiU9Btf9z2g7YP2r7d9kv7GgwA0EzroNs+S9LHJM0kuUDSaZKu\n6GswAEAzXU+5bJD0F7Y3SDpd0i+7jwQAaGND21+Y5Anbn5b0uKTfS9qdZPfi/WzPSpqVpKmpqbbL\n4RQxvW3XIOse2rFlkHWBPnU55fIKSVslnSPptZJeZvuqxfslmUsyk2RmNBq1nxQAsKQup1wuk/Sz\nJEeT/EHSnZLe2s9YAICmugT9cUlvtn26bUu6VNJ8P2MBAJpqHfQk90u6Q9J+ST8ev9dcT3MBABpq\n/U1RSUpyk6SbepoFANABd4oCQBEEHQCKIOgAUARBB4AiCDoAFEHQAaAIgg4ARRB0ACiCoANAEQQd\nAIog6ABQBEEHgCIIOgAUQdABoAiCDgBFEHQAKIKgA0ARBB0AiiDoAFAEQQeAIgg6ABRB0AGgCIIO\nAEV0CrrtM23fYfth2/O239LXYACAZjZ0/PX/KelbSd5t+yWSTu9hJgBAC62DbvsvJf2tpKslKcnz\nkp7vZywAQFNdjtDPkXRU0v/afr2kfZKuTfLc8TvZnpU0K0lTU1OtF5vetqv9pABwCuhyDn2DpL+R\n9N9JLpL0nKRti3dKMpdkJsnMaDTqsBwAYCldgn5Y0uEk94+379BC4AEAA2gd9CRPSfqF7fPGL10q\n6aFepgIANNb1KpePSto5vsLlMUkf6j4SAKCNTkFPckDSTE+zAAA64E5RACiCoANAEQQdAIog6ABQ\nBEEHgCIIOgAUQdABoAiCDgBFEHQAKIKgA0ARXZ/lAqAjnvW/dob8Wh/asWXV1+AIHQCKIOgAUARB\nB4AiCDoAFEHQAaAIgg4ARRB0ACiCoANAEQQdAIog6ABQBEEHgCI6B932abZ/aPvrfQwEAGinjyP0\nayXN9/A+AIAOOgXd9iZJWyT9Tz/jAADa6nqE/h+SbpD0xx5mAQB00Pp56LYvl3QkyT7bb19iv1lJ\ns5I0NTXVdjlgVfFMclTQ5Qj9YknvtH1I0hckXWL7/xbvlGQuyUySmdFo1GE5AMBSWgc9yY1JNiWZ\nlnSFpO8kuaq3yQAAjXAdOgAU0cv/KZrke5K+18d7AQDa4QgdAIog6ABQBEEHgCIIOgAUQdABoAiC\nDgBFEHQAKIKgA0ARBB0AiiDoAFAEQQeAInp5lgtq4dngwGTiCB0AiiDoAFAEQQeAIgg6ABRB0AGg\nCIIOAEUQdAAogqADQBEEHQCKIOgAUARBB4AiWgfd9tm2v2v7IdsP2r62z8EAAM10eTjXMUnXJ9lv\n+wxJ+2zvSfJQT7MBABpofYSe5Mkk+8cfPytpXtJZfQ0GAGiml3PotqclXSTp/j7eDwDQXOeg2365\npC9Lui7Jb07w+Vnbe23vPXr0aNflAAAn0Snotl+shZjvTHLnifZJMpdkJsnMaDTqshwAYAldrnKx\npFskzSf5TH8jAQDa6HKEfrGk90u6xPaB8Y939DQXAKCh1pctJvm+JPc4CwCgA+4UBYAiCDoAFEHQ\nAaAIgg4ARRB0ACiCoANAEQQdAIog6ABQBEEHgCIIOgAUQdABoAiCDgBFEHQAKIKgA0ARBB0AiiDo\nAFAEQQeAIgg6ABRB0AGgCIIOAEUQdAAogqADQBEEHQCK6BR025tt/8T2o7a39TUUAKC51kG3fZqk\n/5L0j5LOl3Sl7fP7GgwA0EyXI/Q3Sno0yWNJnpf0BUlb+xkLANBUl6CfJekXx20fHr8GABjAhtVe\nwPaspNnx5m9t/6ThW2yU9HS/U62p1vP7Uz1P0s4p+/VfJ5h/WL3N3/Hv81+vZKcuQX9C0tnHbW8a\nv/ZnksxJmmu7iO29SWba/vqhMf+wmH9YzL+2upxy+YGkc22fY/slkq6Q9LV+xgIANNX6CD3JMdsf\nkXSXpNMk3Zrkwd4mAwA00ukcepJvSPpGT7OcTOvTNesE8w+L+YfF/GvISYaeAQDQA279B4AiJiro\ntq+3Hdsbh56lCdv/avsB2wds77b92qFnasL2zbYfHv8evmL7zKFnasL2e2w/aPuPtifiioVJf6yG\n7VttH7F9cOhZmrJ9tu3v2n5o/Ofm2qFnWqmJCbrtsyX9g6THh56lhZuTXJjkDZK+LumTQw/U0B5J\nFyS5UNJPJd048DxNHZT0Lkn3DD3IShR5rMZtkjYPPURLxyRdn+R8SW+W9C+T8vWfmKBL+ndJN0ia\nuJP+SX5z3ObLNGG/hyS7kxwbb96nhXsOJkaS+SRNb2gb0sQ/ViPJPZKeGXqONpI8mWT/+ONnJc1r\nQu6CX/U7Rftge6ukJ5L8yPbQ47Ri+98kfUDSryX93cDjdPFhSV8ceojiTvRYjTcNNMspzfa0pIsk\n3T/sJCuzboJu+9uSXn2CT22X9AktnG5Zt5aaP8lXk2yXtN32jZI+IummNR1wGcvNP95nuxb+Obpz\nLWdbiZXMDzRh++WSvizpukX/yl631k3Qk1x2otdtv07SOZJeODrfJGm/7TcmeWoNR1zSyeY/gZ1a\nuHZ/XQV9ufltXy3pckmXZh1e69rg6z8JVvRYDawe2y/WQsx3Jrlz6HlWat0E/WSS/FjSX72wbfuQ\npJkkE/PAH9vnJnlkvLlV0sNDztOU7c1a+P7F25L8buh5TgH//1gNLYT8CknvHXakU4cXjhxvkTSf\n5DNDz9PEJH1TdJLtsH3Q9gNaOHU0MZdBjX1W0hmS9owvvfzc0AM1YfufbR+W9BZJu2zfNfRMSxl/\nA/qFx2rMS/rSpD1Ww/btku6VdJ7tw7avGXqmBi6W9H5Jl4z/vB+w/Y6hh1oJ7hQFgCI4QgeAIgg6\nABRB0AGgCIIOAEUQdAAogqADQBEEHQCKIOgAUMSfAKZMo8TKusQCAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9df46ad0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nasKgS1ITBl2Smvgv7gRY+TrVdPUAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fb168fe90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9ddab4d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[-0.47331899 -0.68668354  0.41492677]\n",
      "Training epoch 500\n",
      "Iter: 500\n",
      "D loss: 0.01002\n",
      "G_loss: 0.006708\n",
      "()\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nERa6JDXCQpekRljoktQIC12SGvH/H2VT6evAA04AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9db18790>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9ddab7d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ndIAhBB1gCEEHGELQAYb4f5IMQEin8VMFAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9da90150>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[-0.11978109 -0.95677364  0.36146411]\n",
      "Training epoch 600\n",
      "Iter: 600\n",
      "D loss: 0.008184\n",
      "G_loss: 0.005165\n",
      "()\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9dc28e10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX4AAAD8CAYAAABw1c+bAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nqfYGg3EAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fd870f410>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n/wMVLonq7MHFmgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9deb2a90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 0.07549416 -1.35025263  0.37304261]\n",
      "Training epoch 700\n",
      "Iter: 700\n",
      "D loss: 0.006084\n",
      "G_loss: 0.00324\n",
      "()\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nwqBLUhP/By5Vn68NrLL/AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fb169f790>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXAAAAD8CAYAAABuHP8oAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9dba3990>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ntTI8AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fb3224b90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 0.54156828 -1.60223603  0.50121295]\n",
      "Training epoch 800\n",
      "Iter: 800\n",
      "D loss: 0.00491\n",
      "G_loss: 0.002202\n",
      "()\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nIv4XxL8KJYyAHGYAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fb3382490>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nKRzqhwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fd877ce10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fb80fe550>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 0.71857083 -1.5971539   0.43989319]\n",
      "Training epoch 900\n",
      "Iter: 900\n",
      "D loss: 0.004552\n",
      "G_loss: 0.003993\n",
      "()\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fb327e650>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nCAtdkhrxPxTeXexiUPeuAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fb166fad0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAC0VJREFUeJzt3X2IZXUdx/HPp11FTUnDi5nrNP4RgkhqDKEJFT7A5kpW\nFCgpWsL804OGICv+If23UUhBUQy5KSRK+IDiWrqZIoFtzW6brY6m2KZr2o5I+FBkm5/+2Bvszj7M\n3HPO7Nn9zvsFw869c+b+vntW3xzO3HPGSQQAOPS9p+8BAADdIOgAUARBB4AiCDoAFEHQAaAIgg4A\nRRB0ACiCoANAEQQdAIpYfiAXO/744zM+Pn4glwSAQ97GjRtfSzKYb7sDGvTx8XFNT08fyCUB4JBn\n+68L2Y5TLgBQBEEHgCIIOgAUQdABoAiCDgBFEHQAKIKgA0ARBB0AiiDoAFDEAb1SFKMZX72ul3W3\nrlnVy7oA2uEIHQCKIOgAUARBB4AiCDoAFEHQAaAIgg4ARRB0ACiCoANAEQQdAIog6ABQBEEHgCLm\nDbrttba3296yy3Pfsf2M7Sdt32v72MUdEwAwn4Ucod8qaeWc59ZLOj3JRyT9WdINHc8FABjRvEFP\n8rik1+c893CSHcOHv5W0YhFmAwCMoItz6F+R9IsOXgcA0EKroNu+UdIOSbfvZ5tJ29O2p2dnZ9ss\nBwDYj8ZBt32VpIslfSlJ9rVdkqkkE0kmBoNB0+UAAPNo9BuLbK+UdL2kTyb5Z7cjAQCaWMjbFu+Q\n9ISkU21vs321pB9IOkbSetubbf94kecEAMxj3iP0JJft5elbFmEWAEALXCkKAEUQdAAogqADQBEE\nHQCKIOgAUARBB4AiCDoAFEHQAaAIgg4ARRB0ACiCoANAEQQdAIog6ABQBEEHgCIIOgAUQdABoAiC\nDgBFEHQAKIKgA0ARBB0AiiDoAFAEQQeAIgg6ABQxb9Btr7W93faWXZ57v+31tp8b/nnc4o4JAJjP\nQo7Qb5W0cs5zqyU9kuTDkh4ZPgYA9GjeoCd5XNLrc56+RNJtw89vk/TZjucCAIyo6Tn0E5K8Mvz8\nVUkndDQPAKCh1j8UTRJJ2dfXbU/anrY9PTs723Y5AMA+NA36322fKEnDP7fva8MkU0kmkkwMBoOG\nywEA5tM06PdLunL4+ZWS7utmHABAUwt52+Idkp6QdKrtbbavlrRG0oW2n5N0wfAxAKBHy+fbIMll\n+/jS+R3PAgBogStFAaAIgg4ARRB0ACiCoANAEQQdAIog6ABQBEEHgCIIOgAUQdABoAiCDgBFzHvp\nP4DFNb56XS/rbl2zqpd1sXg4QgeAIgg6ABRB0AGgCIIOAEUQdAAogqADQBEEHQCKIOgAUARBB4Ai\nCDoAFEHQAaAIgg4ARbQKuu1v2n7K9hbbd9g+oqvBAACjaRx02ydJ+oakiSSnS1om6dKuBgMAjKbt\nKZflko60vVzSUZL+1n4kAEATje+HnuRl29+V9KKkf0l6OMnDc7ezPSlpUpLGxsaaLtervu5X3Zc+\n/7593aN7qf0bo6Y2p1yOk3SJpFMkfVDSe21fPne7JFNJJpJMDAaD5pMCAParzSmXCyT9Jclskv9I\nukfSx7sZCwAwqjZBf1HS2baPsm1J50ua6WYsAMCoGgc9yQZJd0naJOlPw9ea6mguAMCIWv2S6CQ3\nSbqpo1kAAC1wpSgAFEHQAaAIgg4ARRB0ACiCoANAEQQdAIog6ABQBEEHgCIIOgAUQdABoAiCDgBF\nEHQAKIKgA0ARBB0AiiDoAFAEQQeAIgg6ABRB0AGgCIIOAEUQdAAogqADQBEEHQCKIOgAUESroNs+\n1vZdtp+xPWP7nK4GAwCMZnnL7/++pF8m+YLtwyUd1cFMAIAGGgfd9vskfULSVZKU5B1J73QzFgBg\nVG2O0E+RNCvpp7bPkLRR0jVJ3t51I9uTkiYlaWxsrMVyWArGV6/re4Qlo899vXXNqt7WrqzNOfTl\nkj4q6UdJzpL0tqTVczdKMpVkIsnEYDBosRwAYH/aBH2bpG1JNgwf36WdgQcA9KBx0JO8Kukl26cO\nnzpf0tOdTAUAGFnbd7l8XdLtw3e4vCDpy+1HAgA00SroSTZLmuhoFgBAC1wpCgBFEHQAKIKgA0AR\nBB0AiiDoAFAEQQeAIgg6ABRB0AGgCIIOAEUQdAAogqADQBEEHQCKIOgAUARBB4AiCDoAFEHQAaAI\ngg4ARRB0ACiCoANAEQQdAIog6ABQBEEHgCJaB932Mtt/sP1AFwMBAJrp4gj9GkkzHbwOAKCFVkG3\nvULSKkk/6WYcAEBTbY/QvyfpeknvdjALAKCF5U2/0fbFkrYn2Wj7U/vZblLSpCSNjY01XU7jq9c1\n/l4AWAraHKGfK+kztrdKulPSebZ/NnejJFNJJpJMDAaDFssBAPancdCT3JBkRZJxSZdK+nWSyzub\nDAAwEt6HDgBFND6Hvqskj0l6rIvXAgA0wxE6ABRB0AGgCIIOAEUQdAAogqADQBEEHQCKIOgAUARB\nB4AiCDoAFEHQAaAIgg4ARRB0ACiCoANAEQQdAIog6ABQBEEHgCIIOgAUQdABoAiCDgBFEHQAKIKg\nA0ARBB0AiiDoAFBE46DbPtn2o7aftv2U7Wu6HAwAMJrlLb53h6TrkmyyfYykjbbXJ3m6o9kAACNo\nfISe5JUkm4afvylpRtJJXQ0GABhNJ+fQbY9LOkvShi5eDwAwutZBt320pLslXZvkjb18fdL2tO3p\n2dnZtssBAPahVdBtH6adMb89yT172ybJVJKJJBODwaDNcgCA/WjzLhdLukXSTJKbuxsJANBEmyP0\ncyVdIek825uHHxd1NBcAYESN37aY5DeS3OEsAIAWuFIUAIog6ABQBEEHgCIIOgAUQdABoAiCDgBF\nEHQAKIKgA0ARBB0AiiDoAFAEQQeAItr8CjoAaGR89bq+Rzjgtq5ZtehrcIQOAEUQdAAogqADQBEE\nHQCKIOgAUARBB4AiCDoAFEHQAaAIgg4ARRB0ACiCoANAEa2Cbnul7WdtP297dVdDAQBG1zjotpdJ\n+qGkT0s6TdJltk/rajAAwGjaHKF/TNLzSV5I8o6kOyVd0s1YAIBRtQn6SZJe2uXxtuFzAIAeLPr9\n0G1PSpocPnzL9rOLvWaPjpf0Wt9DHGTYJ7tjf+xpSewTf3ukzefukw8t5JvaBP1lSSfv8njF8Lnd\nJJmSNNVinUOG7ekkE33PcTBhn+yO/bEn9smemu6TNqdcfi/pw7ZPsX24pEsl3d/i9QAALTQ+Qk+y\nw/bXJD0kaZmktUme6mwyAMBIWp1DT/KgpAc7mqWCJXFqaUTsk92xP/bEPtlTo33iJF0PAgDoAZf+\nA0ARBL1jtr9o+ynb79pesj+557YQu7O91vZ221v6nuVgYftk24/afnr4/8w1fc/UJ9tH2P6d7T8O\n98e3Rn0Ngt69LZI+L+nxvgfpC7eF2KtbJa3se4iDzA5J1yU5TdLZkr66xP87+bek85KcIelMSStt\nnz3KCxD0jiWZSVL54qmF4LYQcyR5XNLrfc9xMEnySpJNw8/flDSjJXy1eXZ6a/jwsOHHSD/kJOhY\nDNwWAiOxPS7pLEkb+p2kX7aX2d4sabuk9UlG2h+Lful/RbZ/JekDe/nSjUnuO9DzAIcy20dLulvS\ntUne6HuePiX5r6QzbR8r6V7bpydZ8M9dCHoDSS7oe4aD3IJuCwHYPkw7Y357knv6nudgkeQfth/V\nzp+7LDjonHLBYuC2EJiXbUu6RdJMkpv7nqdvtgfDI3PZPlLShZKeGeU1CHrHbH/O9jZJ50haZ/uh\nvmc60JLskPT/20LMSPr5Ur8thO07JD0h6VTb22xf3fdMB4FzJV0h6Tzbm4cfF/U9VI9OlPSo7Se1\n86BofZIHRnkBrhQFgCI4QgeAIgg6ABRB0AGgCIIOAEUQdAAogqADQBEEHQCKIOgAUMT/AKgKWDn0\nXthgAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fb3272f10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 1.03681803 -1.21801603  0.26552609]\n",
      "Training epoch 1000\n",
      "Iter: 1000\n",
      "D loss: 0.004949\n",
      "G_loss: 0.003944\n",
      "()\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nIOgAUMT/Afab2sjhHj1DAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fb802a190>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAC8NJREFUeJzt3W2IZQUdx/HfL7cnLbLYW5brNBKyICI9XHoSelqrLcWt\nKFCwtIShFz0SyNpSvgjBMHqgghjUDFossKRQKzcrJFBp1jZbXZ+wXVuzdkzKwhe2+OvF3mqdZufO\nPefM3Ot/vh+QnXPumTl/DvLlcOacM04iAMDT3zPGPQAAoBsEHQCKIOgAUARBB4AiCDoAFEHQAaAI\ngg4ARRB0ACiCoANAEetWc2fr16/P9PT0au4SAJ72du7c+UiS3rDtVjXo09PTmpubW81dAsDTnu19\ny9mOSy4AUARBB4AiCDoAFEHQAaAIgg4ARQwNuu0rbR+wvXuRzz5jO7bXr8x4AIDlWs4Z+lWSNi9c\nafsESe+Q9GDHMwEAGhga9CQ3S3p0kY++IulCSfwNOwCYAI2uodveIumhJL/reB4AQEMjPylq+2hJ\nn9Whyy3L2X5G0owkTU1Njbo7AAVNb71+LPvde+kZY9nvamlyhv4KSSdK+p3tvZI2SLrd9nGLbZxk\nNkk/Sb/XG/oqAgBAQyOfoSf5vaQX/2d5EPV+kkc6nAsAMKLl3LZ4taRbJG20vd/2BSs/FgBgVEPP\n0JOcM+Tz6c6mAQA0xpOiAFAEQQeAIgg6ABRB0AGgCIIOAEUQdAAogqADQBEEHQCKIOgAUARBB4Ai\nCDoAFEHQAaAIgg4ARRB0ACiCoANAEQQdAIog6ABQBEEHgCIIOgAUQdABoAiCDgBFDA267SttH7C9\n+7B1l9m+2/Ydtq+1fezKjgkAGGY5Z+hXSdq8YN0OSackOVXSvZIu6nguAMCIhgY9yc2SHl2w7sYk\nBweLt0rasAKzAQBG0MU19I9I+smRPrQ9Y3vO9tz8/HwHuwMALKZV0G1vk3RQ0vYjbZNkNkk/Sb/X\n67XZHQBgCeuafqPt8yWdKWlTknQ2EQCgkUZBt71Z0oWS3pzk8W5HAgA0sZzbFq+WdIukjbb3275A\n0jckPV/SDtu7bH9rhecEAAwx9Aw9yTmLrL5iBWYBALTAk6IAUARBB4AiCDoAFEHQAaAIgg4ARRB0\nACiCoANAEQQdAIog6ABQBEEHgCIIOgAUQdABoAiCDgBFEHQAKIKgA0ARBB0AiiDoAFAEQQeAIgg6\nABRB0AGgiKFBt32l7QO2dx+27kW2d9i+b/DvC1d2TADAMMs5Q79K0uYF67ZKuinJSZJuGiwDAMZo\naNCT3Czp0QWrt0j6zuDr70h6T8dzAQBG1PQa+kuSPDz4+s+SXtLRPACAhlr/UjRJJOVIn9uesT1n\ne25+fr7t7gAAR9A06H+x/VJJGvx74EgbJplN0k/S7/V6DXcHABimadB/LOm8wdfnSfpRN+MAAJpa\nzm2LV0u6RdJG2/ttXyDpUklvt32fpNMHywCAMVo3bIMk5xzho00dzwIAaIEnRQGgCIIOAEUQdAAo\ngqADQBEEHQCKIOgAUARBB4AiCDoAFEHQAaAIgg4ARQx99B9YTdNbrx/LfvdeesZY9gt0iTN0ACiC\noANAEQQdAIog6ABQBEEHgCIIOgAUQdABoAiCDgBFEHQAKIKgA0ARrYJu+9O277S92/bVtp/T1WAA\ngNE0Drrt4yV9QlI/ySmSjpJ0dleDAQBG0/aSyzpJz7W9TtLRkv7UfiQAQBONg57kIUlfkvSgpIcl\n/T3JjQu3sz1je8723Pz8fPNJAQBLanPJ5YWStkg6UdLLJB1j+9yF2yWZTdJP0u/1es0nBQAsqc0l\nl9Ml/SHJfJJ/SfqhpDd2MxYAYFRtgv6gpNfbPtq2JW2StKebsQAAo2pzDf02SddIul3S7wc/a7aj\nuQAAI2r1J+iSXCzp4o5mAQC0wJOiAFAEQQeAIgg6ABRB0AGgCIIOAEUQdAAogqADQBEEHQCKIOgA\nUESrJ0WBKqa3Xj/uEVbd3kvPGPcI6Bhn6ABQBEEHgCIIOgAUQdABoAiCDgBFEHQAKIKgA0ARBB0A\niiDoAFAEQQeAIloF3faxtq+xfbftPbbf0NVgAIDRtH2Xy9ck/TTJ+20/S9LRHcwEAGigcdBtv0DS\nmySdL0lJnpD0RDdjAQBG1eaSy4mS5iV92/ZvbV9u+5iO5gIAjMhJmn2j3Zd0q6TTktxm+2uSHkvy\nuQXbzUiakaSpqanX7Nu3r+XIqGwtvsYWa0Ob1xXb3pmkP2y7Nmfo+yXtT3LbYPkaSa9euFGS2ST9\nJP1er9didwCApTQOepI/S/qj7Y2DVZsk3dXJVACAkbW9y+XjkrYP7nB5QNKH248EAGiiVdCT7JI0\n9LoOAGDl8aQoABRB0AGgCIIOAEUQdAAogqADQBEEHQCKIOgAUARBB4AiCDoAFEHQAaCItu9yWRPG\n9UrXNq/bBLD2cIYOAEUQdAAogqADQBEEHQCKIOgAUARBB4AiCDoAFEHQAaAIgg4ARRB0ACiiddBt\nH2X7t7av62IgAEAzXZyhf1LSng5+DgCghVZBt71B0hmSLu9mHABAU23P0L8q6UJJT3YwCwCghcav\nz7V9pqQDSXbafssS281ImpGkqampprtbk8b12l4AT09tztBPk3SW7b2Svifpbba/u3CjJLNJ+kn6\nvV6vxe4AAEtpHPQkFyXZkGRa0tmSfpHk3M4mAwCMhPvQAaCITv4EXZJfSfpVFz8LANAMZ+gAUARB\nB4AiCDoAFEHQAaAIgg4ARRB0ACiCoANAEQQdAIog6ABQBEEHgCIIOgAUQdABoAiCDgBFEHQAKIKg\nA0ARBB0AiiDoAFAEQQeAIgg6ABRB0AGgCIIOAEU0DrrtE2z/0vZdtu+0/ckuBwMAjGZdi+89KOkz\nSW63/XxJO23vSHJXR7MBAEbQ+Aw9ycNJbh98/Q9JeyQd39VgAIDRtDlD/y/b05JeJem2RT6bkTQj\nSVNTU433Mb31+sbfCwBrQetfitp+nqQfSPpUkscWfp5kNkk/Sb/X67XdHQDgCFoF3fYzdSjm25P8\nsJuRAABNtLnLxZKukLQnyZe7GwkA0ESbM/TTJH1Q0tts7xr89+6O5gIAjKjxL0WT/FqSO5wFANAC\nT4oCQBEEHQCKIOgAUARBB4AiCDoAFEHQAaAIgg4ARRB0ACiCoANAEQQdAIog6ABQBEEHgCIIOgAU\nQdABoAiCDgBFEHQAKIKgA0ARBB0AiiDoAFAEQQeAIloF3fZm2/fYvt/21q6GAgCMrnHQbR8l6ZuS\n3iXpZEnn2D65q8EAAKNpc4b+Wkn3J3kgyROSvidpSzdjAQBG1Sbox0v642HL+wfrAABjsG6ld2B7\nRtLMYPGftu9Z6X1OqPWSHhn3EBOM47M0js/SJv74+Iutvv3ly9moTdAfknTCYcsbBuueIsmspNkW\n+ynB9lyS/rjnmFQcn6VxfJbG8TmkzSWX30g6yfaJtp8l6WxJP+5mLADAqBqfoSc5aPtjkn4m6ShJ\nVya5s7PJAAAjaXUNPckNkm7oaJbq1vxlpyE4Pkvj+CyN4yPJScY9AwCgAzz6DwBFEPRVZPsLtu+w\nvcv2jbZfNu6ZJonty2zfPThG19o+dtwzTRLbH7B9p+0nba/5OzokXj+yEEFfXZclOTXJKyVdJ+nz\n4x5owuyQdEqSUyXdK+miMc8zaXZLep+km8c9yCTg9SP/j6CvoiSPHbZ4jCR+gXGYJDcmOThYvFWH\nnm3AQJI9Sdbqg3mL4fUjC6z4k6J4KtuXSPqQpL9LeuuYx5lkH5H0/XEPgYm22OtHXjemWSYCQe+Y\n7Z9LOm6Rj7Yl+VGSbZK22b5I0sckXbyqA47ZsOMz2GabpIOStq/mbJNgOccHOBKC3rEkpy9z0+06\ndA//mgr6sONj+3xJZ0ralDV4T+0I//9gma8fWUu4hr6KbJ902OIWSXePa5ZJZHuzpAslnZXk8XHP\ng4nH60cW4MGiVWT7B5I2SnpS0j5JH02yps8oDmf7fknPlvTXwapbk3x0jCNNFNvvlfR1ST1Jf5O0\nK8k7xzvVeNl+t6Sv6n+vH7lkzCONFUEHgCK45AIARRB0ACiCoANAEQQdAIog6ABQBEEHgCIIOgAU\nQdABoIh/A2NAg77FFB93AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fb32d6d90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXAAAAD8CAYAAABuHP8oAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nqPkAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fe3502f90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 1.20803821 -1.14556003  0.11391169]\n",
      "Training epoch 1100\n",
      "Iter: 1100\n",
      "D loss: 0.00333\n",
      "G_loss: 0.002158\n",
      "()\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nXwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fe00c1610>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nUARBB4AifgXVgo0qMNHeXAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9d9c3050>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nCDoAJPE/go4QtnoYVJAAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fb3231550>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 1.21659768 -1.25505292  0.04524595]\n",
      "Training epoch 1200\n",
      "Iter: 1200\n",
      "D loss: 0.003083\n",
      "G_loss: 0.00151\n",
      "()\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nCYMuSU38Hwa2eJDItMo9AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fe34ce650>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n8b8Oh1k2b4FY5wAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fb16dced0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9dac91d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 1.28216577 -1.07958174 -0.02440737]\n",
      "Training epoch 1300\n",
      "Iter: 1300\n",
      "D loss: 0.003541\n",
      "G_loss: 0.002314\n",
      "()\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9dc8a210>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n/C+TOVp+dyb1mAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fb80d13d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9dd692d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 1.44679928 -0.90137827  0.19066785]\n",
      "Training epoch 1400\n",
      "Iter: 1400\n",
      "D loss: 0.002023\n",
      "G_loss: 0.0003641\n",
      "()\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9dd9e3d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9dded950>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYcugAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9dc34f10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 1.54654789 -0.99883056  0.38628727]\n",
      "Training epoch 1500\n",
      "Iter: 1500\n",
      "D loss: 0.00106\n",
      "G_loss: -0.0009896\n",
      "()\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fb164aa90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n4r/1UYz/LJlKNQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fb80aaf50>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAADFNJREFUeJzt3X+sXoVdx/H3RzqiTCIsvTIGq8WEEHHRjNwsuJm5DFSE\nxc5kGohTNkkaTVQ0JkuXJfIvU2P87dJsOIzIYnA4MmDSTQkx2Ygt1g0oE5xsK8LaSbK5HwmSff2j\nD6Z23Pv8OKf3uc+X9ytp7vPj8JzvySnvnJ4+5zRVhSRp9X3HsgeQJI3DoEtSEwZdkpow6JLUhEGX\npCYMuiQ1YdAlqQmDLklNGHRJamLHVq5s586dtXv37q1cpSStvEOHDn25qtamLbelQd+9ezcHDx7c\nylVK0spL8vlZlvOUiyQ1YdAlqQmDLklNGHRJasKgS1ITBl2SmjDoktSEQZekJgy6JDWxpVeKStPs\n3nf3Utb75M3XLGW90pg8QpekJgy6JDVh0CWpCYMuSU0YdElqwqBLUhMGXZKaMOiS1IRBl6QmDLok\nNWHQJamJqUFPckuSY0kePum1303yWJJPJ7kzyTmnd0xJ0jSzHKF/ELjqlNcOAK+pqh8C/g1498hz\nSZLmNDXoVfUA8Owpr91XVc9Pnn4KuPA0zCZJmsMY59B/Cbh3hM+RJA0wKOhJ3gM8D9y2yTJ7kxxM\ncvD48eNDVidJ2sTCQU/yDuAtwM9XVW20XFXtr6r1qlpfW1tbdHWSpCkW+heLklwFvAv4sar6xrgj\nSZIWMcvXFm8HPglckuRokhuAPwHOBg4kOZzkfad5TknSFFOP0Kvquhd5+QOnYRZJ0gBeKSpJTRh0\nSWrCoEtSEwZdkpow6JLUhEGXpCYMuiQ1YdAlqQmDLklNGHRJasKgS1ITBl2SmjDoktSEQZekJgy6\nJDVh0CWpCYMuSU0YdElqwqBLUhMGXZKaMOiS1IRBl6QmDLokNTE16EluSXIsycMnvfaKJAeSPD75\nee7pHVOSNM0sR+gfBK465bV9wCeq6mLgE5PnkqQlmhr0qnoAePaUl/cAt04e3wq8deS5JElzWvQc\n+nlV9fTk8TPAeSPNI0la0OC/FK2qAmqj95PsTXIwycHjx48PXZ0kaQOLBv1LSc4HmPw8ttGCVbW/\nqtaran1tbW3B1UmSplk06HcB108eXw98ZJxxJEmLmuVri7cDnwQuSXI0yQ3AzcCPJ3kcuHLyXJK0\nRDumLVBV123w1hUjzyJJGsArRSWpCYMuSU0YdElqwqBLUhMGXZKaMOiS1IRBl6QmDLokNWHQJakJ\ngy5JTUy99F9ST7v33b20dT958zVLW3dnHqFLUhMGXZKaMOiS1IRBl6QmDLokNWHQJakJgy5JTRh0\nSWrCoEtSEwZdkpow6JLUhEGXpCYGBT3JbyZ5JMnDSW5P8p1jDSZJms/CQU9yAfDrwHpVvQY4A7h2\nrMEkSfMZesplB/BdSXYAZwH/OXwkSdIiFr4felU9leT3gC8A3wTuq6r7Tl0uyV5gL8CuXbsWXZ3U\n1jLvS/5S0/0e8ENOuZwL7AEuAl4FvDzJ209drqr2V9V6Va2vra0tPqkkaVNDTrlcCfxHVR2vqv8B\nPgy8fpyxJEnzGhL0LwCXJzkrSYArgCPjjCVJmtfCQa+qB4E7gIeAz0w+a/9Ic0mS5jToH4muqpuA\nm0aaRZI0gFeKSlITBl2SmjDoktSEQZekJgy6JDVh0CWpCYMuSU0YdElqwqBLUhMGXZKaGHTpv06v\nZd27eSvu2yxpfB6hS1ITBl2SmjDoktSEQZekJgy6JDVh0CWpCYMuSU0YdElqwqBLUhMGXZKaMOiS\n1IRBl6QmBgU9yTlJ7kjyWJIjSX5krMEkSfMZerfFPwQ+VlVvS3ImcNYIM0mSFrBw0JN8D/BG4B0A\nVfUc8Nw4Y0mS5jXkCP0i4DjwF0l+GDgE3FhVXz95oSR7gb0Au3btGrA6bZVl3Yd9mV6K26x+hpxD\n3wFcBvx5Vb0W+Dqw79SFqmp/Va1X1fra2tqA1UmSNjMk6EeBo1X14OT5HZwIvCRpCRYOelU9A3wx\nySWTl64AHh1lKknS3IZ+y+XXgNsm33D5HPDO4SNJkhYxKOhVdRhYH2kWSdIAXikqSU0YdElqwqBL\nUhMGXZKaMOiS1IRBl6QmDLokNWHQJakJgy5JTRh0SWrCoEtSEwZdkpow6JLUhEGXpCYMuiQ1YdAl\nqQmDLklNGHRJasKgS1ITBl2SmjDoktSEQZekJgYHPckZSf4lyUfHGEiStJgxjtBvBI6M8DmSpAEG\nBT3JhcA1wPvHGUeStKihR+h/ALwL+NYIs0iSBlg46EneAhyrqkNTltub5GCSg8ePH190dZKkKYYc\nob8B+OkkTwIfAt6c5K9OXaiq9lfVelWtr62tDVidJGkzCwe9qt5dVRdW1W7gWuAfqurto00mSZqL\n30OXpCZ2jPEhVXU/cP8YnyVJWoxH6JLUhEGXpCYMuiQ1YdAlqQmDLklNGHRJasKgS1ITBl2SmjDo\nktSEQZekJgy6JDUxyr1cJGkeu/fdvewRWvIIXZKaMOiS1IRBl6QmDLokNWHQJakJgy5JTRh0SWrC\noEtSEwZdkpow6JLUhEGXpCYWDnqSVyf5xySPJnkkyY1jDiZJms+Qm3M9D/xWVT2U5GzgUJIDVfXo\nSLNJkuaw8BF6VT1dVQ9NHv83cAS4YKzBJEnzGeUcepLdwGuBB8f4PEnS/AbfDz3JdwN/C/xGVX31\nRd7fC+wF2LVr18Lr8f7JkrS5QUfoSV7GiZjfVlUffrFlqmp/Va1X1fra2tqQ1UmSNjHkWy4BPgAc\nqarfH28kSdIihhyhvwH4BeDNSQ5Pfl090lySpDktfA69qv4JyIizSJIG8EpRSWrCoEtSEwZdkpow\n6JLUhEGXpCYMuiQ1YdAlqQmDLklNGHRJasKgS1ITBl2SmjDoktSEQZekJgy6JDVh0CWpCYMuSU0Y\ndElqwqBLUhMGXZKaMOiS1IRBl6QmDLokNWHQJamJQUFPclWSzyZ5Ism+sYaSJM1v4aAnOQP4U+Cn\ngEuB65JcOtZgkqT5DDlCfx3wRFV9rqqeAz4E7BlnLEnSvIYE/QLgiyc9Pzp5TZK0BDtO9wqS7AX2\nTp5+LclnR/jYncCXR/ic7ajztkHv7eu8bdB7+077tuW9g/7z75tloSFBfwp49UnPL5y89v9U1X5g\n/4D1fJskB6tqfczP3C46bxv03r7O2wa9t6/Ltg055fLPwMVJLkpyJnAtcNc4Y0mS5rXwEXpVPZ/k\nV4G/B84AbqmqR0abTJI0l0Hn0KvqHuCekWaZx6incLaZztsGvbev87ZB7+1rsW2pqmXPIEkagZf+\nS1ITKxH0JD+b5JEk30qy4d9Er+KtCJK8IsmBJI9Pfp67wXJPJvlMksNJDm71nPOYth9ywh9N3v90\nksuWMeeiZti+NyX5ymRfHU7y28uYcxFJbklyLMnDG7y/svtuhm1b2f32f6pq2/8CfgC4BLgfWN9g\nmTOAfwe+HzgT+Ffg0mXPPsO2/Q6wb/J4H/DeDZZ7Eti57Hln2J6p+wG4GrgXCHA58OCy5x55+94E\nfHTZsy64fW8ELgMe3uD9Vd5307ZtZffbC79W4gi9qo5U1bQLklb1VgR7gFsnj28F3rrEWcYwy37Y\nA/xlnfAp4Jwk52/1oAta1d9nM6mqB4BnN1lkZffdDNu28lYi6DNa1VsRnFdVT08ePwOct8FyBXw8\nyaHJ1bfb1Sz7YVX3Fcw+++snpyTuTfKDWzPalljlfTeLld5vp/3S/1kl+Tjwyhd56z1V9ZGtnmdM\nm23byU+qqpJs9LWjH62qp5J8L3AgyWOTIw5tPw8Bu6rqa0muBv4OuHjJM2m6ld9v2yboVXXlwI+Y\n6VYEy7DZtiX5UpLzq+rpyR9dj23wGU9Nfh5Lcicn/ui/HYM+y37YtvtqBlNnr6qvnvT4niR/lmRn\nVXW4D8oq77tNddhvnU65rOqtCO4Crp88vh74tj+NJHl5krNfeAz8BPCif1O/DcyyH+4CfnHyjYnL\nga+cdNppu5u6fUlemSSTx6/jxP9n/7Xlk54eq7zvNtVhv22bI/TNJPkZ4I+BNeDuJIer6ieTvAp4\nf1VdXat7K4Kbgb9JcgPweeDnAE7eNk6cV79z8nttB/DXVfWxJc27qY32Q5Jfnrz/Pk5cXXw18ATw\nDeCdy5p3XjNu39uAX0nyPPBN4NqafI1iu0tyOye+7bEzyVHgJuBlsPr7boZtW9n99gKvFJWkJjqd\ncpGklzSDLklNGHRJasKgS1ITBl2SmjDoktSEQZekJgy6JDXxvz5F5WNhdW0GAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fb167bbd0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 1.79292989 -0.97180986  0.51098967]\n",
      "Training epoch 1600\n",
      "Iter: 1600\n",
      "D loss: 0.002953\n",
      "G_loss: -0.002461\n",
      "()\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fb17f3090>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fb17954d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAD8CAYAAAB5Pm/hAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nDwDJUfQAkBxFDwDJUfQAkNz/ACj9DPdARYjTAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fd87e7fd0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 1.82164252 -0.94051939  0.50860089]\n",
      "Training epoch 1700\n",
      "Iter: 1700\n",
      "D loss: 0.008033\n",
      "G_loss: 0.004372\n",
      "()\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAADTRJREFUeJzt3X+MZeVdx/H3p7uQlkKEZkdKgXEwISRIVMikwWIqKdQg\nNNI/GgOxlVaSiSYqNSbN1iYS/6NqGjUam03BYkSahoIQKLXbSkNMWnQXt+XH0oJ12y5CdysRSmuC\nq1//mEOyjjNzf5wz9+48vF/JZM699+Q+nzwMn33mzJxnUlVIkra/1807gCRpGBa6JDXCQpekRljo\nktQIC12SGmGhS1IjLHRJaoSFLkmNsNAlqRE7ZznYrl27amlpaZZDStK2t3///u9V1cKo82Za6EtL\nS+zbt2+WQ0rStpfkW+Oc5yUXSWqEhS5JjbDQJakRFrokNcJCl6RGWOiS1AgLXZIaYaFLUiMsdElq\nxEzvFNX2sLT7gbmNfeiWa+Y2trTduUKXpEZY6JLUCAtdkhphoUtSIyx0SWqEhS5JjbDQJakRFrok\nNcJCl6RGWOiS1AgLXZIaMbLQk9yW5EiSx4977g+TPJXka0nuSXL61saUJI0yzgr9k8BVa57bC1xU\nVT8JfAP48MC5JEkTGlnoVfUw8MKa5z5fVce6h18BztmCbJKkCQxxDf1XgQcHeB9JUg+9Cj3JR4Bj\nwB2bnLOSZF+SfUePHu0znCRpE1MXepL3A+8CfrmqaqPzqmpPVS1X1fLCwsK0w0mSRpjqLxYluQr4\nEPBzVfXDYSNJkqYxzq8t3gl8GbggyeEkNwJ/BpwG7E1yIMnHtzinJGmEkSv0qrp+nadv3YIskqQe\nvFNUkhphoUtSIyx0SWqEhS5JjbDQJakRFrokNcJCl6RGWOiS1AgLXZIaYaFLUiMsdElqhIUuSY2w\n0CWpERa6JDXCQpekRljoktQIC12SGmGhS1IjLHRJaoSFLkmNsNAlqREWuiQ1wkKXpEaMLPQktyU5\nkuTx4557U5K9SZ7uPp+xtTElSaOMs0L/JHDVmud2A1+sqvOBL3aPJUlzNLLQq+ph4IU1T18L3N4d\n3w68e+BckqQJTXsN/cyqeq47fh44c6A8kqQp7ez7BlVVSWqj15OsACsAi4uLfYeT1ICl3Q/MZdxD\nt1wzl3FnZdoV+neTnAXQfT6y0YlVtaeqlqtqeWFhYcrhJEmjTFvo9wE3dMc3APcOE0eSNK1xfm3x\nTuDLwAVJDie5EbgFeGeSp4Eru8eSpDkaeQ29qq7f4KUrBs4iSerBO0UlqREWuiQ1wkKXpEZY6JLU\nCAtdkhphoUtSIyx0SWqEhS5JjbDQJakRFrokNaL39rnSkNxWVZqeK3RJaoSFLkmNsNAlqREWuiQ1\nwkKXpEZY6JLUCAtdkhphoUtSIyx0SWqEhS5JjbDQJakRFrokNaJXoSf57SRPJHk8yZ1JXj9UMEnS\nZKYu9CRnA78FLFfVRcAO4LqhgkmSJtP3kstO4A1JdgKnAP/WP5IkaRpTF3pVPQv8EfBt4Dngxar6\n/Nrzkqwk2Zdk39GjR6dPKknaVJ9LLmcA1wLnAW8B3pjkvWvPq6o9VbVcVcsLCwvTJ5UkbarPJZcr\ngX+tqqNV9V/A3cDbhoklSZpUn0L/NnBpklOSBLgCODhMLEnSpPpcQ38EuAt4FHise689A+WSJE2o\n1x+JrqqbgZsHyiJJ6sE7RSWpERa6JDXCQpekRljoktQIC12SGmGhS1IjLHRJaoSFLkmNsNAlqREW\nuiQ1otet/5K0nSztfmBuYx+65ZotH8MVuiQ1wkKXpEZY6JLUCAtdkhphoUtSIyx0SWqEhS5JjbDQ\nJakRFrokNcJCl6RGWOiS1AgLXZIa0avQk5ye5K4kTyU5mORnhgomSZpM390W/wT4XFW9J8nJwCkD\nZJIkTWHqQk/yI8DbgfcDVNUrwCvDxJIkTarPCv084Cjwl0l+CtgP3FRVPzj+pCQrwArA4uJij+Gk\nNs1rj+5Z7M+t2epzDX0ncAnwF1V1MfADYPfak6pqT1UtV9XywsJCj+EkSZvpU+iHgcNV9Uj3+C5W\nC16SNAdTF3pVPQ98J8kF3VNXAE8OkkqSNLG+v+Xym8Ad3W+4fBP4QP9IkqRp9Cr0qjoALA+URZLU\ng3eKSlIjLHRJaoSFLkmNsNAlqREWuiQ1wkKXpEZY6JLUCAtdkhphoUtSIyx0SWpE371ctIXmtU/2\na5FzrRa4QpekRljoktQIC12SGmGhS1IjLHRJaoSFLkmNsNAlqREWuiQ1wkKXpEZY6JLUCAtdkhrR\nu9CT7Ejyz0nuHyKQJGk6Q6zQbwIODvA+kqQeehV6knOAa4BPDBNHkjStviv0PwY+BPzPAFkkST1M\nvR96kncBR6pqf5LLNzlvBVgBWFxcnHa4uXKvbEnbQZ8V+mXALyY5BHwKeEeSv157UlXtqarlqlpe\nWFjoMZwkaTNTF3pVfbiqzqmqJeA64O+r6r2DJZMkTcTfQ5ekRgzyN0Wr6kvAl4Z4L0nSdFyhS1Ij\nLHRJaoSFLkmNsNAlqREWuiQ1wkKXpEZY6JLUCAtdkhphoUtSIyx0SWqEhS5JjRhkLxdJ24/7/LfH\nFbokNcJCl6RGWOiS1AgLXZIaYaFLUiMsdElqhIUuSY2w0CWpERa6JDXCQpekRljoktSIqQs9yblJ\nHkryZJInktw0ZDBJ0mT6bM51DPidqno0yWnA/iR7q+rJgbJJkiYw9Qq9qp6rqke74+8DB4Gzhwom\nSZrMINfQkywBFwOPDPF+kqTJ9d4PPcmpwGeAD1bVS+u8vgKsACwuLk49jns3S9Lmeq3Qk5zEapnf\nUVV3r3dOVe2pquWqWl5YWOgznCRpE31+yyXArcDBqvrYcJEkSdPos0K/DHgf8I4kB7qPqwfKJUma\n0NTX0KvqH4AMmEWS1IN3ikpSIyx0SWqEhS5JjbDQJakRFrokNcJCl6RGWOiS1AgLXZIaYaFLUiMs\ndElqhIUuSY2w0CWpERa6JDXCQpekRljoktQIC12SGmGhS1IjLHRJaoSFLkmNsNAlqREWuiQ1wkKX\npEZY6JLUiF6FnuSqJF9P8kyS3UOFkiRNbupCT7ID+HPgF4ALgeuTXDhUMEnSZPqs0N8KPFNV36yq\nV4BPAdcOE0uSNKk+hX428J3jHh/unpMkzcHOrR4gyQqw0j18OcnXBx5iF/C9gd9zCOaajLkmY67J\nzD1XPrru0+Pm+rFxxuhT6M8C5x73+Jzuuf+jqvYAe3qMs6kk+6pqeavef1rmmoy5JmOuybxWcvW5\n5PJPwPlJzktyMnAdcN8wsSRJk5p6hV5Vx5L8BvB3wA7gtqp6YrBkkqSJ9LqGXlWfBT47UJZpbdnl\nnJ7MNRlzTcZck3lN5EpVDfl+kqQ58dZ/SWrEtiv0JG9KsjfJ093nMzY471CSx5IcSLJvC/Nsuv1B\nVv1p9/rXklyyVVkmzHV5khe7+TmQ5PdmkOm2JEeSPL7B6/Oaq1G5Zj5X3bjnJnkoyZNJnkhy0zrn\nzHzOxsw1j6+v1yf5xyRf7XL9/jrnzGO+xsk1zHxV1bb6AP4A2N0d7wY+usF5h4BdW5xlB/AvwI8D\nJwNfBS5cc87VwINAgEuBR2YwR+Pkuhy4f8b/7d4OXAI8vsHrM5+rMXPNfK66cc8CLumOTwO+cYJ8\nfY2Tax5fXwFO7Y5PAh4BLj0B5mucXIPM17ZbobO6vcDt3fHtwLvnmGWc7Q+uBf6qVn0FOD3JWSdA\nrpmrqoeBFzY5ZR5zNU6uuaiq56rq0e74+8BB/v/d2DOfszFzzVw3By93D0/qPtb+kHAe8zVOrkFs\nx0I/s6qe646fB87c4LwCvpBkf3e36lYYZ/uDeWyRMO6Yb+u+7XwwyU9scaZxnMjbScx1rpIsARez\nuro73lznbJNcMIc5S7IjyQHgCLC3qk6I+RojFwwwX1t+6/80knwBePM6L33k+AdVVUk2+pfuZ6vq\n2SQ/CuxN8lS3EtOqR4HFqno5ydXA3wLnzznTiWquc5XkVOAzwAer6qVZjTvKiFxzmbOq+m/gp5Oc\nDtyT5KKqWvdnI7M0Rq5B5uuEXKFX1ZVVddE6H/cC3331W6Tu85EN3uPZ7vMR4B5WL0MMbZztD8ba\nImHWuarqpVe/DazV+wlOSrJri3ONMo+5Gmmec5XkJFZL846qunudU+YyZ6Nyzfvrq6r+A3gIuGrN\nS3P9Gtso11DzdUIW+gj3ATd0xzcA9649Ickbk5z26jHw88BW/Cs9zvYH9wG/0v10/VLgxeMuGW2V\nkbmSvDlJuuO3svq18O9bnGuUeczVSPOaq27MW4GDVfWxDU6b+ZyNk2sec5ZkoVsBk+QNwDuBp9ac\nNo/5GplrqPk6IS+5jHAL8OkkNwLfAn4JIMlbgE9U1dWsXle/p5ufncDfVNXnhg5SG2x/kOTXutc/\nzuqdtFcDzwA/BD4wdI4pc70H+PUkx4D/BK6r7sftWyXJnaz+NH9XksPAzaz+gGhuczVmrpnPVecy\n4H3AY931V4DfBRaPyzaPORsn1zzm7Czg9qz+8Z3XAZ+uqvvn/f/jmLkGmS/vFJWkRmzHSy6SpHVY\n6JLUCAtdkhphoUtSIyx0SWqEhS5JjbDQJakRFrokNeJ/AYeFnRcIbDGeAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fb17b8190>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9dea71d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nqQmDLklNGHRJasKgS1ITBl2SmvgfcLHeSmXYw28AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9dfde190>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 1.58536446 -1.04804301  0.51924175]\n",
      "Training epoch 1800\n",
      "Iter: 1800\n",
      "D loss: 0.006754\n",
      "G_loss: 0.002748\n",
      "()\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fb328f3d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fb332efd0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nUiMsdElqhIUuSY2w0CWpEf8Pns8MIz6sIDgAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fb804ad90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 1.26493788 -1.24722457  0.54286826]\n",
      "Training epoch 1900\n",
      "Iter: 1900\n",
      "D loss: 0.005539\n",
      "G_loss: 0.0007035\n",
      "()\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nxMO+AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9d9f96d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nskT3AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9dfea250>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nUhH/B1XlFHeh8Xs5AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fe00c9350>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 1.01008201 -1.35460746  0.66635871]\n",
      "Training epoch 2000\n",
      "Iter: 2000\n",
      "D loss: 0.004782\n",
      "G_loss: -0.0006933\n",
      "()\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAW4AAAD8CAYAAABXe05zAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nsbglqTH/CyPGavQIjMtaAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9da77050>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX4AAAD8CAYAAABw1c+bAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fe005d090>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n0CWpERa6JDXCQpekRljoktSI/wOQG0Y3864ivgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fd871cbd0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 0.78025752 -1.23265254  0.58919692]\n",
      "Training epoch 2100\n",
      "Iter: 2100\n",
      "D loss: 0.004036\n",
      "G_loss: -0.000306\n",
      "()\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fe005b790>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nJakJgy5JTRh0SWrCoEtSEwZdkpr4P1Zxc4S5ipR0AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9db50e10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nM0sya6g6RQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fb3363150>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 0.49555561 -1.10714269  0.51629198]\n",
      "Training epoch 2200\n",
      "Iter: 2200\n",
      "D loss: 0.002441\n",
      "G_loss: -0.002929\n",
      "()\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\niUXVPioAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9db72cd0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9ddf8d90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9dcb6350>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 0.20300867 -1.16329014  0.53777868]\n",
      "Training epoch 2300\n",
      "Iter: 2300\n",
      "D loss: 0.001937\n",
      "G_loss: -0.004058\n",
      "()\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fb17c0f50>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9dcb6490>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQpekRljoktSI/wN4CdoUaVGhMwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9dc54590>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 0.02757948 -1.07353973  0.5281412 ]\n",
      "Training epoch 2400\n",
      "Iter: 2400\n",
      "D loss: 0.003768\n",
      "G_loss: -0.0003566\n",
      "()\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nrcGSAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fb80d9c10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX4AAAD8CAYAAABw1c+bAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fb3244350>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXkAAAD8CAYAAACSCdTiAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fe005e850>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[-0.1604237  -1.03399849  0.51664531]\n",
      "Training epoch 2500\n",
      "Iter: 2500\n",
      "D loss: 0.003976\n",
      "G_loss: -0.00029\n",
      "()\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fd87299d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRljoktQIC12SGmGhS1Ij/h9rIlewMfJlWwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fb80b4350>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAD8CAYAAACMwORRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nVwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fe005db90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 0.06920671 -0.89499158  0.44140276]\n",
      "Training epoch 2600\n",
      "Iter: 2600\n",
      "D loss: 0.002914\n",
      "G_loss: -0.001571\n",
      "()\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fb337cf10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9dfc2dd0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXYAAAD8CAYAAABjAo9vAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nsdglqTEWuyQ15v8Akp3han6oReQAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9dacc290>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 0.34926867 -1.07029819  0.56313717]\n",
      "Training epoch 2700\n",
      "Iter: 2700\n",
      "D loss: 0.002428\n",
      "G_loss: -0.001861\n",
      "()\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9db11cd0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fb16aa0d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRvw/86hJdAhO028AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9daf2fd0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 0.57017493 -1.18510818  0.30485779]\n",
      "Training epoch 2800\n",
      "Iter: 2800\n",
      "D loss: 0.002618\n",
      "G_loss: -0.00168\n",
      "()\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQZekJgy6JDVh0CWpif8HrzMwu0aEbjMAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9dfa4050>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9d9c4350>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9de4c790>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 0.58154434 -0.96689051  0.44317442]\n",
      "Training epoch 2900\n",
      "Iter: 2900\n",
      "D loss: 0.001627\n",
      "G_loss: -0.002505\n",
      "()\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nFEHQAaAIgg4ARfwfsxgr+t6FctAAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9de75fd0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nIyx0SWqEhS5JjbDQJakR/wdIA9mcrx6kwgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fe34c9f10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nuiQ1wkKXpEZY6JLUiP8FElB27IE9G+0AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9df75f10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 0.63106841 -0.90578157  0.52533048]\n",
      "Training epoch 3000\n",
      "Iter: 3000\n",
      "D loss: 0.001797\n",
      "G_loss: -0.003126\n",
      "()\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nA7ueqTzTjDYEAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9dda8350>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAD8CAYAAAB5Pm/hAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9dc77a10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nswAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9deb4f90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 0.63167834 -0.96474922  0.62452638]\n",
      "Training epoch 3100\n",
      "Iter: 3100\n",
      "D loss: 0.001525\n",
      "G_loss: -0.002937\n",
      "()\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7f9dd74410>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fb333efd0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSY2w0CWpERa6JDXCQpekRvwv6Vx2LKyH2eIAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7fb80aef50>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 0.63048744 -0.94344193  0.61500883]\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-15-4832b1bf6b2c>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m     62\u001b[0m             _, D_loss_curr, _ = sess.run(\n\u001b[1;32m     63\u001b[0m                 \u001b[0;34m[\u001b[0m\u001b[0mD_solver\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mD_loss\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mclip_D\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 64\u001b[0;31m                 feed_dict=feed_dict)\n\u001b[0m\u001b[1;32m     65\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     66\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/home/evan/.local/lib/python2.7/site-packages/tensorflow/python/client/session.pyc\u001b[0m in \u001b[0;36mrun\u001b[0;34m(self, fetches, feed_dict, options, run_metadata)\u001b[0m\n\u001b[1;32m    765\u001b[0m     \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    766\u001b[0m       result = self._run(None, fetches, feed_dict, options_ptr,\n\u001b[0;32m--> 767\u001b[0;31m                          run_metadata_ptr)\n\u001b[0m\u001b[1;32m    768\u001b[0m       \u001b[0;32mif\u001b[0m \u001b[0mrun_metadata\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    769\u001b[0m         \u001b[0mproto_data\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtf_session\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mTF_GetBuffer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mrun_metadata_ptr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/home/evan/.local/lib/python2.7/site-packages/tensorflow/python/client/session.pyc\u001b[0m in \u001b[0;36m_run\u001b[0;34m(self, handle, fetches, feed_dict, options, run_metadata)\u001b[0m\n\u001b[1;32m    936\u001b[0m                 ' to a larger type (e.g. int64).')\n\u001b[1;32m    937\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 938\u001b[0;31m           \u001b[0mnp_val\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0masarray\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msubfeed_val\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdtype\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0msubfeed_dtype\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    939\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    940\u001b[0m           \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0msubfeed_t\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget_shape\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mis_compatible_with\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnp_val\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshape\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/home/evan/.local/lib/python2.7/site-packages/numpy/core/numeric.pyc\u001b[0m in \u001b[0;36masarray\u001b[0;34m(a, dtype, order)\u001b[0m\n\u001b[1;32m    529\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    530\u001b[0m     \"\"\"\n\u001b[0;32m--> 531\u001b[0;31m     \u001b[0;32mreturn\u001b[0m \u001b[0marray\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0ma\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdtype\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcopy\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mFalse\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0morder\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0morder\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    532\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    533\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "G_pred, G_pred_output_i = generator(z, is_training=False)\n",
    "G_sample, G_sample_output_i = generator(z, is_training=True)\n",
    "\n",
    "D_real  = discriminator(label_placeholder)\n",
    "D_fake  = discriminator(G_sample)\n",
    "\n",
    "# D_loss = -tf.reduce_mean(tf.log(D_real) + tf.log(1. - D_fake))\n",
    "# G_loss = -tf.reduce_mean(tf.log(D_fake))\n",
    "\n",
    "# Alternative losses:\n",
    "# -------------------\n",
    "\n",
    "#D_loss_real = tf.reduce_mean(tf.nn.sigmoid_cross_entropy_with_logits(logits=D_logit_real, labels=ones_like))\n",
    "#D_loss_fake = tf.reduce_mean(tf.nn.sigmoid_cross_entropy_with_logits(logits=D_logit_fake, labels=zeros_like))\n",
    "#D_loss = D_loss_real + D_loss_fake\n",
    "#G_loss = tf.reduce_mean(tf.nn.sigmoid_cross_entropy_with_logits(logits=D_logit_fake, labels=ones_like))\n",
    "\n",
    "#D_loss_real = tf.reduce_mean(tf.nn.sigmoid_cross_entropy_with_logits(logits=D_logit_real, labels=tf.ones_like(D_logit_real)))\n",
    "#D_loss_fake = tf.reduce_mean(tf.nn.sigmoid_cross_entropy_with_logits(logits=D_logit_fake, labels=tf.zeros_like(D_logit_fake)))\n",
    "#D_loss = D_loss_real + D_loss_fake\n",
    "#G_loss = tf.reduce_mean(tf.nn.sigmoid_cross_entropy_with_logits(logits=D_logit_fake, labels=tf.ones_like(D_logit_fake)))\n",
    "\n",
    "#WGAN:\n",
    "D_loss = tf.reduce_mean(D_real) - tf.reduce_mean(D_fake)\n",
    "G_loss = -tf.reduce_mean(D_fake)\n",
    "clip_D = [p.assign(tf.clip_by_value(p, -0.01, 0.01)) for p in theta_D]\n",
    "\n",
    "#D_solver = tf.train.AdamOptimizer(learning_rate=1e-4).minimize(D_loss, var_list=theta_D)\n",
    "#G_solver = tf.train.AdamOptimizer(learning_rate=1e-4).minimize(G_loss, var_list=theta_G)\n",
    "\n",
    "D_solver = (tf.train.RMSPropOptimizer(learning_rate=5e-4)\n",
    "            .minimize(-D_loss, var_list=theta_D))\n",
    "G_solver = (tf.train.RMSPropOptimizer(learning_rate=5e-4)\n",
    "            .minimize(G_loss, var_list=theta_G))\n",
    "\n",
    "mb_size = S\n",
    "\n",
    "sess = tf.Session()\n",
    "sess.run(tf.global_variables_initializer())\n",
    "\n",
    "train_dataset = [features[0] for i in range(0,64)]\n",
    "n_train = len(train_dataset)\n",
    "preds = []\n",
    "\n",
    "d_losses = []\n",
    "g_losses = []\n",
    "\n",
    "for it in range(100000):\n",
    "    if it % 100 == 0:\n",
    "        print(\"Training epoch %d\" %it)\n",
    "    batch_sched = list(range(0, n_train+1,S))\n",
    "    for j in range(0, len(batch_sched)-1):\n",
    "        start = batch_sched[j]\n",
    "        stop = batch_sched[j+1]\n",
    "        feed_dict = construct_feed_dict(train_dataset, start, stop)\n",
    "\n",
    "        _, D_loss_curr = sess.run([D_solver, D_loss], feed_dict=feed_dict)\n",
    "        \n",
    "        for _ in range(5):\n",
    "            feed_dict = construct_feed_dict(train_dataset, start, stop)\n",
    "\n",
    "            _, D_loss_curr, _ = sess.run(\n",
    "                [D_solver, D_loss, clip_D],\n",
    "                feed_dict=feed_dict)\n",
    "            \n",
    "        \n",
    "        _, G_loss_curr = sess.run([G_solver, G_loss], feed_dict=feed_dict)\n",
    "        \n",
    "        d_losses.append(D_loss_curr)\n",
    "        g_losses.append(G_loss_curr)\n",
    "\n",
    "        \n",
    "    if it % 100 == 0:\n",
    "        print('Iter: {}'.format(it))\n",
    "        print('D loss: {:.4}'. format(D_loss_curr))\n",
    "        print('G_loss: {:.4}'.format(G_loss_curr))\n",
    "        print()\n",
    "\n",
    "        #print(\"output_i\")\n",
    "        #print(sess.run(G_sample_output_i, feed_dict=feed_dict))\n",
    "        #print(sess.run(G_sample, feed_dict=feed_dict))\n",
    "\n",
    "    if it % 100 == 0:\n",
    "        samples = sess.run(G_sample, feed_dict=feed_dict)\n",
    "        plt.hist(samples[:,0], bins=10)\n",
    "        plt.show()\n",
    "        plt.hist(samples[:,1], bins=10)\n",
    "        plt.show()\n",
    "        plt.hist(samples[:,2], bins=10)\n",
    "        plt.show()\n",
    "        print(np.mean(samples, axis=0))\n",
    "        #print(samples[:10])\n",
    "        #plt.hist(samples, bins=50)\n",
    "        #plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "array([ 0.61143659, -0.99900387,  0.48252666])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'np' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-7-aa94017d45a3>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mwhere\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfeed_dict\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mdihed_indices\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m!=\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m: name 'np' is not defined"
     ]
    }
   ],
   "source": [
    "np.where(feed_dict[dihed_indices][0] != 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'feed_dict' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-8-c9e9ad47ecf2>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mfeed_dict\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mlabel_placeholder\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m: name 'feed_dict' is not defined"
     ]
    }
   ],
   "source": [
    "feed_dict[label_placeholder]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "feed_dict[non_zero_inds]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    " [-0.91292151,  3.09695411,  1.06725919,  2.02969492,  1.9801807 , 2.27330438],"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 0.61143659, -0.99900387,  0.48252666],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ],\n",
       "       [ 0.        ,  0.        ,  0.        ]])"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "features[0][1][1].toarray()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "sess.run(G_sample, feed_dict=feed_dict)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "array([[ 0.91292151],\n",
    "       [ 3.09695411],\n",
    "       [ 1.06725919],"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "plt.hist(sess.run(G_sample, feed_dict=feed_dict))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "batch_sched"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "for it in range(20001):\n",
    "    if it % 100 == 0:\n",
    "        print(\"Training epoch %d\" %it)\n",
    "    batch_sched = list(range(0, n_train+1,S))\n",
    "    for j in range(0, len(batch_sched)-1):\n",
    "        start = batch_sched[j]\n",
    "        stop = batch_sched[j+1]\n",
    "        feed_dict = construct_feed_dict(train_dataset, start, stop)\n",
    "\n",
    "        _, D_loss_curr = sess.run([D_solver, D_loss], feed_dict=feed_dict)\n",
    "        _, G_loss_curr = sess.run([G_solver, G_loss], feed_dict=feed_dict)\n",
    "\n",
    "    if it % 1000 == 0:\n",
    "        print('Iter: {}'.format(it))\n",
    "        print('D loss: {:.4}'. format(D_loss_curr))\n",
    "        print('G_loss: {:.4}'.format(G_loss_curr))\n",
    "        print()\n",
    "\n",
    "    if it % 1000 == 0:\n",
    "        samples = sess.run(G_sample, feed_dict=feed_dict)\n",
    "        plt.hist(samples, bins=50)\n",
    "        plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "preds = []\n",
    "for k in range(0,100):\n",
    "    feed_dict = construct_feed_dict(train_dataset, 0, 1)\n",
    "    samples = sess.run(G_sample, feed_dict=feed_dict)[2]\n",
    "    preds.append(samples)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "plt.hist(np.concatenate(preds))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "preds[4]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "train_dataset[0][1][1].toarray()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "res = gan.predict()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "res"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "plt.hist(res, bins=50)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "features[0][1][1].toarray()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "features[0][1][1].toarray()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "preds = []\n",
    "for i in range(0,100):\n",
    "    preds.append(gan.predict(features)[0][2])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "np.mean(preds)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "features[0][2].atom_features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}