ixxi-dante/nw2vec

View on GitHub
projects/correctness/gae-reproduction-citeseer-nw2vec-multitask_arch-shallow_adj-ov=16.ipynb

Summary

Maintainability
Test Coverage
{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Train nw2vec with the original VGAE architecture for Cora embeddings + a kernel for the scalar product decoding + an intermediate layer in decoding"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "\n",
    "# Train on CPU (hide GPU) due to memory constraints\n",
    "os.environ['CUDA_VISIBLE_DEVICES'] = \"\"\n",
    "\n",
    "import contextlib\n",
    "\n",
    "import numpy as np\n",
    "import keras\n",
    "from keras_tqdm import TQDMNotebookCallback as TQDMCallback\n",
    "\n",
    "from nw2vec import layers\n",
    "from nw2vec import ae\n",
    "from nw2vec import utils\n",
    "from nw2vec import batching\n",
    "import settings\n",
    "\n",
    "from gae.input_data import load_data\n",
    "from gae.preprocessing import sparse_to_tuple, mask_test_edges"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "@contextlib.contextmanager\n",
    "def gae_directory():\n",
    "    working_directory = os.path.abspath(os.curdir)\n",
    "    try:\n",
    "        # Move to the GAE directory\n",
    "        os.chdir('../../gae')\n",
    "        yield\n",
    "    finally:\n",
    "        # Move back\n",
    "        os.chdir(working_directory)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load data\n",
    "with gae_directory():\n",
    "    adj, features = load_data('citeseer')\n",
    "    features = features.toarray()\n",
    "\n",
    "#adj_train = mask_test_edges(adj)[0]\n",
    "adj_train, train_edges, val_edges, val_edges_false, test_edges, test_edges_false = mask_test_edges(adj)\n",
    "assert adj_train.diagonal().sum() == 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def build_p_builder(dims, feature_codec='SigmoidBernoulli', adj_kernel=None, use_bias=False,\n",
    "                    embedding_slices=None, with_l1=True):\n",
    "\n",
    "    # Validate arguments and set default values\n",
    "    assert feature_codec in ['SigmoidBernoulli', 'OrthogonalGaussian', 'SoftmaxMultinomial']\n",
    "    if embedding_slices is None:\n",
    "        adj_embedding_slice, v_embedding_slice = [slice(None), slice(None)]\n",
    "    else:\n",
    "        adj_embedding_slice, v_embedding_slice = embedding_slices\n",
    "\n",
    "    # Extract the dimensions we use.\n",
    "    dim_data, dim_l1, _, _ = dims\n",
    "\n",
    "    def p_builder(p_input):\n",
    "        # Get slices of the embeddings for each prediction\n",
    "        p_input_adj = layers.InnerSlice(adj_embedding_slice)(p_input)\n",
    "        p_input_v = layers.InnerSlice(v_embedding_slice)(p_input)\n",
    "        \n",
    "        p_penultimate_adj = p_input_adj\n",
    "        \n",
    "        if with_l1:\n",
    "            p_penultimate_v = keras.layers.Dense(\n",
    "                dim_l1, use_bias=use_bias, activation='relu',\n",
    "                kernel_regularizer='l2', bias_regularizer='l2',\n",
    "                name='p_layer1_v'\n",
    "            )(p_input_v)\n",
    "        else:\n",
    "            p_penultimate_v = p_input_v\n",
    "\n",
    "        # Prepare kwargs for the Bilinear adj decoder, then build it.\n",
    "        adj_kwargs = {}\n",
    "        if adj_kernel is not None:\n",
    "            adj_kwargs['fixed_kernel'] = adj_kernel\n",
    "        else:\n",
    "            adj_kwargs['kernel_regularizer'] = 'l2'\n",
    "        p_adj = layers.Bilinear(0, use_bias=use_bias, name='p_adj',\n",
    "                                bias_regularizer='l2',\n",
    "                                **adj_kwargs)([p_penultimate_adj, p_penultimate_adj])\n",
    "\n",
    "        # Finally build the feature decoder according to the requested codec.\n",
    "        if feature_codec in ['SigmoidBernoulli', 'SoftmaxMultinomial']:\n",
    "            p_v = keras.layers.Dense(dim_data, use_bias=use_bias,\n",
    "                                     kernel_regularizer='l2', bias_regularizer='l2',\n",
    "                                     name='p_v')(p_penultimate_v)\n",
    "        else:\n",
    "            assert feature_codec == 'OrthogonalGaussian'\n",
    "            p_v_μ_flat = keras.layers.Dense(dim_data, use_bias=use_bias,\n",
    "                                            kernel_regularizer='l2', bias_regularizer='l2',\n",
    "                                            name='p_v_mu_flat')(p_penultimate_v)\n",
    "            p_v_logS_flat = keras.layers.Dense(dim_data, use_bias=use_bias,\n",
    "                                               kernel_regularizer='l2', bias_regularizer='l2',\n",
    "                                               name='p_v_logS_flat')(p_penultimate_v)\n",
    "            p_v = keras.layers.Concatenate(name='p_v_mulogS_flat')([p_v_μ_flat, p_v_logS_flat])\n",
    "\n",
    "        return ([p_adj, p_v], ('SigmoidBernoulliScaledAdjacency', feature_codec))\n",
    "\n",
    "    return p_builder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "n_ξ_samples = 1\n",
    "n_nodes = adj_train.shape[0]\n",
    "dim_data, dim_l1, dim_ξ_adj, dim_ξ_v = features.shape[1], 32, 16, 16\n",
    "overlap = 16\n",
    "dims = (dim_data, dim_l1, dim_ξ_adj, dim_ξ_v)\n",
    "loss_weights = {\n",
    "    'q_mulogS_flat': 1e-2 * 1.0 / (dim_ξ_adj - overlap + dim_ξ_v),\n",
    "    'p_adj': 1.0 / (n_nodes * np.log(2)),\n",
    "    'p_v': 1.0 / (features.shape[1] * np.log(2)),\n",
    "}\n",
    "\n",
    "# Actual VAE\n",
    "q_model, q_codecs = ae.build_q(dims, overlap=overlap,\n",
    "                               fullbatcher=batching.fullbatches, minibatcher=batching.pq_batches)\n",
    "p_builder = build_p_builder(dims,\n",
    "                               feature_codec='SigmoidBernoulli',\n",
    "                               adj_kernel=np.eye(16),\n",
    "                               embedding_slices=[slice(dim_ξ_adj),\n",
    "                                                 slice(dim_ξ_adj - overlap, dim_ξ_adj - overlap + dim_ξ_v)],\n",
    "                               with_l1=True)\n",
    "vae, vae_codecs = ae.build_vae(\n",
    "    (q_model, q_codecs), p_builder,\n",
    "    n_ξ_samples,\n",
    "    loss_weights=loss_weights\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "def target_func(batch_adj, required_nodes, final_nodes):\n",
    "    return [\n",
    "        np.zeros(1), # ignored\n",
    "        utils.expand_dims_tile(utils.expand_dims_tile(batch_adj + np.eye(batch_adj.shape[0]), 0, n_ξ_samples), 0, 1),\n",
    "        utils.expand_dims_tile(features[final_nodes], 1, n_ξ_samples),\n",
    "    ]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/sl/.conda/envs/base36/lib/python3.6/site-packages/tensorflow/python/ops/gradients_impl.py:100: UserWarning: Converting sparse IndexedSlices to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"Converting sparse IndexedSlices to a dense Tensor of unknown shape. \"\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "aaad3781d32c4bd19ec93054f7bc1d5d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Training', max=200), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "n_epochs = 200\n",
    "\n",
    "history = vae.fit_fullbatches(batcher_kws={'adj': adj_train, 'features': features, 'target_func': target_func},\n",
    "                              epochs=n_epochs,\n",
    "                              verbose=0, callbacks=[TQDMCallback(show_inner=False)])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "history = {'history': history.history}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x288 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(1, len(history['history']), figsize=(len(history['history']) * 5, 4))\n",
    "for i, (title, values) in enumerate(history['history'].items()):\n",
    "    axes[i].plot(np.array(values))\n",
    "    axes[i].set_title(title)\n",
    "fig.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Precision"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "import scipy\n",
    "from sklearn.metrics import roc_auc_score\n",
    "from sklearn.metrics import average_precision_score\n",
    "from keras import backend as K"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_roc_score(edges_pos, edges_neg, adj_pred):\n",
    "    # Predict on test set of edges\n",
    "    preds = []\n",
    "    pos = []\n",
    "    for e in edges_pos:\n",
    "        preds.append(adj_pred[e[0], e[1]])\n",
    "        pos.append(adj[e[0], e[1]])\n",
    "\n",
    "    preds_neg = []\n",
    "    neg = []\n",
    "    for e in edges_neg:\n",
    "        preds_neg.append(adj_pred[e[0], e[1]])\n",
    "        neg.append(adj[e[0], e[1]])\n",
    "\n",
    "    preds_all = np.hstack([preds, preds_neg])\n",
    "    labels_all = np.hstack([np.ones(len(preds)), np.zeros(len(preds))])\n",
    "    roc_score = roc_auc_score(labels_all, preds_all)\n",
    "    ap_score = average_precision_score(labels_all, preds_all)\n",
    "\n",
    "    return roc_score, ap_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "q_preds, adj_preds, v_preds = zip(*[vae.predict_fullbatch(adj=adj_train, features=features) for _ in range(10)])\n",
    "\n",
    "q_preds = np.array(q_preds)\n",
    "adj_preds = np.array(adj_preds)\n",
    "\n",
    "q_pred = q_preds.mean(0)\n",
    "adj_pred = scipy.special.expit(adj_preds[:, 0, :, :, :]).mean(1).mean(0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.9055283178360101, 0.9133216536345697)"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "get_roc_score(test_edges, test_edges_false, adj_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "for layer in vae.layers:\n",
    "    if hasattr(layer, 'kernel'):\n",
    "        fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(6, 4))#, sharey=True)\n",
    "        im1 = ax1.imshow(K.eval(layer.kernel).T)\n",
    "        ax1.set_title('{} kernel'.format(layer.name))\n",
    "        plt.colorbar(im1, ax=ax1)\n",
    "        if hasattr(layer, 'bias') and layer.bias is not None:\n",
    "            im2 = ax2.imshow(K.eval(K.expand_dims(layer.bias, -1)))\n",
    "            ax2.set_title('{} bias'.format(layer.name))\n",
    "            plt.colorbar(im2, ax=ax2)\n",
    "        else:\n",
    "            ax2.set_visible(False)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}