stellargraph/stellargraph

View on GitHub
demos/ensembles/ensemble-node-classification-example.ipynb

Summary

Maintainability
Test Coverage
{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "0",
   "metadata": {},
   "source": [
    "# Ensemble models for node classification\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1",
   "metadata": {
    "nbsphinx": "hidden",
    "tags": [
     "CloudRunner"
    ]
   },
   "source": [
    "<table><tr><td>Run the latest release of this notebook:</td><td><a href=\"https://mybinder.org/v2/gh/stellargraph/stellargraph/master?urlpath=lab/tree/demos/ensembles/ensemble-node-classification-example.ipynb\" alt=\"Open In Binder\" target=\"_parent\"><img src=\"https://mybinder.org/badge_logo.svg\"/></a></td><td><a href=\"https://colab.research.google.com/github/stellargraph/stellargraph/blob/master/demos/ensembles/ensemble-node-classification-example.ipynb\" alt=\"Open In Colab\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\"/></a></td></tr></table>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2",
   "metadata": {},
   "source": [
    "This notebook demonstrates the use of `stellargraph`'s `Ensemble` class for node attribute inference using the Cora and Pubmed-Diabetes citation datasets.\n",
    "\n",
    "The `Ensemble` class brings ensemble learning to `stellargraph`'s graph neural network models, e.g., `GraphSAGE` and `GCN`, quantifying prediction variance and potentially improving prediction accuracy. \n",
    "\n",
    "**References**\n",
    "\n",
    "1. Inductive Representation Learning on Large Graphs. W.L. Hamilton, R. Ying, and J. Leskovec arXiv:1706.02216 \n",
    "[cs.SI], 2017.\n",
    "\n",
    "\n",
    "2. Semi-Supervised Classification with Graph Convolutional Networks. T. Kipf, M. Welling. ICLR 2017. arXiv:1609.02907 \n",
    "\n",
    "\n",
    "3. Graph Attention Networks. P. Veličković et al. ICLR 2018"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "3",
   "metadata": {
    "nbsphinx": "hidden",
    "tags": [
     "CloudRunner"
    ]
   },
   "outputs": [],
   "source": [
    "# install StellarGraph if running on Google Colab\n",
    "import sys\n",
    "if 'google.colab' in sys.modules:\n",
    "  %pip install -q stellargraph[demos]==1.3.0b"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "4",
   "metadata": {
    "nbsphinx": "hidden",
    "tags": [
     "VersionCheck"
    ]
   },
   "outputs": [],
   "source": [
    "# verify that we're using the correct version of StellarGraph for this notebook\n",
    "import stellargraph as sg\n",
    "\n",
    "try:\n",
    "    sg.utils.validate_notebook_version(\"1.3.0b\")\n",
    "except AttributeError:\n",
    "    raise ValueError(\n",
    "        f\"This notebook requires StellarGraph version 1.3.0b, but a different version {sg.__version__} is installed.  Please see <https://github.com/stellargraph/stellargraph/issues/1172>.\"\n",
    "    ) from None"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "5",
   "metadata": {},
   "outputs": [],
   "source": [
    "import networkx as nx\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import itertools\n",
    "import os\n",
    "\n",
    "from sklearn.decomposition import PCA\n",
    "from sklearn.manifold import TSNE\n",
    "\n",
    "import stellargraph as sg\n",
    "from stellargraph.mapper import GraphSAGENodeGenerator, FullBatchNodeGenerator\n",
    "from stellargraph.layer import GraphSAGE, GCN, GAT\n",
    "from stellargraph import globalvar\n",
    "\n",
    "from stellargraph.ensemble import Ensemble, BaggingEnsemble\n",
    "\n",
    "from tensorflow.keras import layers, optimizers, losses, metrics, Model, models\n",
    "from sklearn import preprocessing, feature_extraction, model_selection\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from stellargraph import datasets\n",
    "from IPython.display import display, HTML\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6",
   "metadata": {},
   "source": [
    "## Loading the network data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "The Cora dataset consists of 2708 scientific publications classified into one of seven classes. The citation network consists of 5429 links. Each publication in the dataset is described by a 0/1-valued word vector indicating the absence/presence of the corresponding word from the dictionary. The dictionary consists of 1433 unique words."
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "The PubMed Diabetes dataset consists of 19717 scientific publications from PubMed database pertaining to diabetes classified into one of three classes. The citation network consists of 44338 links. Each publication in the dataset is described by a TF/IDF weighted word vector from a dictionary which consists of 500 unique words."
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "display(HTML(datasets.Cora().description))\n",
    "display(HTML(datasets.PubMedDiabetes().description))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8",
   "metadata": {},
   "source": [
    "First, we select the dataset to use, either Cora or Pubmed-Diabetes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "9",
   "metadata": {},
   "outputs": [],
   "source": [
    "use_cora = True  # Select the dataset; if False, then Pubmed-Diabetes dataset is used.\n",
    "if use_cora:\n",
    "    dataset = datasets.Cora()\n",
    "else:\n",
    "    dataset = datasets.PubMedDiabetes()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "10",
   "metadata": {},
   "source": [
    "Load the graph data."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "11",
   "metadata": {
    "tags": [
     "DataLoadingLinks"
    ]
   },
   "source": [
    "(See [the \"Loading from Pandas\" demo](../basics/loading-pandas.ipynb) for details on how data can be loaded.)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "12",
   "metadata": {
    "tags": [
     "DataLoading"
    ]
   },
   "outputs": [],
   "source": [
    "G, labels = dataset.load()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "13",
   "metadata": {},
   "source": [
    "We aim to train a graph-ML model that will predict the \"subject\" or \"label\" attribute on the nodes depending on the selected dataset. These subjects are one of 7 or 3 categories for Cora and PubMed-Diabetes respectively:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "14",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'Genetic_Algorithms', 'Neural_Networks', 'Case_Based', 'Rule_Learning', 'Probabilistic_Methods', 'Reinforcement_Learning', 'Theory'}\n"
     ]
    }
   ],
   "source": [
    "# Print the class names for the selected dataset\n",
    "print(set(labels))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "15",
   "metadata": {},
   "source": [
    "### Splitting the data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "16",
   "metadata": {},
   "source": [
    "For machine learning we want to take a subset of the nodes for training, and use the rest for validation and testing. We'll use scikit-learn again to do this"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "17",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_labels, test_labels = model_selection.train_test_split(\n",
    "    labels, train_size=0.2, test_size=None, stratify=labels, random_state=42,  # 140\n",
    ")\n",
    "val_labels, test_labels = model_selection.train_test_split(\n",
    "    test_labels,\n",
    "    train_size=0.2,  # 500,\n",
    "    test_size=None,\n",
    "    stratify=test_labels,\n",
    "    random_state=100,\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "18",
   "metadata": {},
   "source": [
    "### Converting to numeric arrays"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "19",
   "metadata": {},
   "source": [
    "For our categorical target, we will use one-hot vectors that will be fed into a soft-max Keras layer during training."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "20",
   "metadata": {},
   "outputs": [],
   "source": [
    "target_encoding = preprocessing.LabelBinarizer()\n",
    "\n",
    "train_targets = target_encoding.fit_transform(train_labels)\n",
    "val_targets = target_encoding.transform(val_labels)\n",
    "test_targets = target_encoding.transform(test_labels)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "21",
   "metadata": {},
   "source": [
    "### Specify global parameters\n",
    "\n",
    "Here we specify some parameters that control the type of model we are going to use. For example, we specify the base model type, e.g., GCN, GraphSAGE, etc, and the number of estimators in the ensemble as well as model-specific parameters."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "22",
   "metadata": {
    "tags": [
     "parameters"
    ]
   },
   "outputs": [],
   "source": [
    "model_type = \"graphsage\"  # Can be either gcn, gat, or graphsage\n",
    "use_bagging = (\n",
    "    True  # If True, each model in the ensemble is trained on a bootstrapped sample\n",
    ")\n",
    "# of the given training data; otherwise, the same training data are used\n",
    "# for training each model.\n",
    "\n",
    "if model_type == \"graphsage\":\n",
    "    # For GraphSAGE model\n",
    "    batch_size = 50\n",
    "    num_samples = [10, 10]\n",
    "    n_estimators = 5  # The number of estimators in the ensemble\n",
    "    n_predictions = 10  # The number of predictions per estimator per query point\n",
    "    epochs = 50  # The number of training epochs\n",
    "elif model_type == \"gcn\":\n",
    "    # For GCN model\n",
    "    n_estimators = 5  # The number of estimators in the ensemble\n",
    "    n_predictions = 10  # The number of predictions per estimator per query point\n",
    "    epochs = 50  # The number of training epochs\n",
    "elif model_type == \"gat\":\n",
    "    # For GAT model\n",
    "    layer_sizes = [8, train_targets.shape[1]]\n",
    "    attention_heads = 8\n",
    "    n_estimators = 5  # The number of estimators in the ensemble\n",
    "    n_predictions = 10  # The number of predictions per estimator per query point\n",
    "    epochs = 200  # The number of training epochs"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "23",
   "metadata": {},
   "source": [
    "## Creating the base graph machine learning model in Keras"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "24",
   "metadata": {},
   "source": [
    "To feed data from the graph to the Keras model we need a generator that feeds data from the graph into the model. The generators are specialized to the model and the learning task. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "25",
   "metadata": {},
   "source": [
    "For training we use only the training nodes returned from our splitter and the target values. The `shuffle=True` argument is given to the `flow` method to improve training for those generators that support shuffling."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "26",
   "metadata": {},
   "outputs": [],
   "source": [
    "if model_type == \"graphsage\":\n",
    "    generator = GraphSAGENodeGenerator(G, batch_size, num_samples)\n",
    "    train_gen = generator.flow(train_labels.index, train_targets, shuffle=True)\n",
    "elif model_type == \"gcn\":\n",
    "    generator = FullBatchNodeGenerator(G, method=\"gcn\")\n",
    "    train_gen = generator.flow(\n",
    "        train_labels.index, train_targets\n",
    "    )  # does not support shuffle\n",
    "elif model_type == \"gat\":\n",
    "    generator = FullBatchNodeGenerator(G, method=\"gat\")\n",
    "    train_gen = generator.flow(\n",
    "        train_labels.index, train_targets\n",
    "    )  # does not support shuffle"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "27",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "541"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(train_labels.index)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "28",
   "metadata": {},
   "source": [
    "Now we can specify our machine learning model, we need a few more parameters for this but the parameters are model-specific."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "29",
   "metadata": {},
   "outputs": [],
   "source": [
    "if model_type == \"graphsage\":\n",
    "    base_model = GraphSAGE(\n",
    "        layer_sizes=[16, 16], generator=generator, bias=True, dropout=0.5, normalize=\"l2\"\n",
    "    )\n",
    "    x_inp, x_out = base_model.in_out_tensors()\n",
    "    prediction = layers.Dense(units=train_targets.shape[1], activation=\"softmax\")(x_out)\n",
    "elif model_type == \"gcn\":\n",
    "    base_model = GCN(\n",
    "        layer_sizes=[32, train_targets.shape[1]],\n",
    "        generator=generator,\n",
    "        bias=True,\n",
    "        dropout=0.5,\n",
    "        activations=[\"elu\", \"softmax\"],\n",
    "    )\n",
    "    x_inp, x_out = base_model.in_out_tensors()\n",
    "    prediction = x_out\n",
    "elif model_type == \"gat\":\n",
    "    base_model = GAT(\n",
    "        layer_sizes=layer_sizes,\n",
    "        attn_heads=attention_heads,\n",
    "        generator=generator,\n",
    "        bias=True,\n",
    "        in_dropout=0.5,\n",
    "        attn_dropout=0.5,\n",
    "        activations=[\"elu\", \"softmax\"],\n",
    "    )\n",
    "    x_inp, prediction = base_model.in_out_tensors()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "30",
   "metadata": {},
   "source": [
    "Let's have a look at the shape of the output tensor."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "31",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "TensorShape([None, 7])"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prediction.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "32",
   "metadata": {},
   "source": [
    "### Create a Keras model and then an Ensemble"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "33",
   "metadata": {},
   "source": [
    "Now let's create the actual Keras model with the graph inputs `x_inp` provided by the `base_model` and outputs being the predictions from the softmax layer."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "34",
   "metadata": {},
   "outputs": [],
   "source": [
    "model = Model(inputs=x_inp, outputs=prediction)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "35",
   "metadata": {},
   "source": [
    "Next, we create the ensemble model consisting of `n_estimators` models.\n",
    "\n",
    "We are also going to specify that we want to make `n_predictions` per query point per model. These predictions will differ because of the application of `dropout` and, in the case of ensembling GraphSAGE models, the sampling of node neighbourhoods."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "36",
   "metadata": {},
   "outputs": [],
   "source": [
    "if use_bagging:\n",
    "    model = BaggingEnsemble(model, n_estimators=n_estimators, n_predictions=n_predictions)\n",
    "else:\n",
    "    model = Ensemble(model, n_estimators=n_estimators, n_predictions=n_predictions)\n",
    "\n",
    "model.compile(\n",
    "    optimizer=optimizers.Adam(lr=0.005),\n",
    "    loss=losses.categorical_crossentropy,\n",
    "    metrics=[\"acc\"],\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "37",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<stellargraph.ensemble.BaggingEnsemble at 0x1419399d0>"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# The model is of type stellargraph.ensemble.Ensemble but has\n",
    "# a very similar interface to a Keras model\n",
    "model"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "38",
   "metadata": {},
   "source": [
    "The ensemble has `n_estimators` models. Let's have a look at the first model's layers."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "39",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<tensorflow.python.keras.engine.input_layer.InputLayer at 0x141725490>,\n",
       " <tensorflow.python.keras.engine.input_layer.InputLayer at 0x141725590>,\n",
       " <tensorflow.python.keras.engine.input_layer.InputLayer at 0x1416fbf10>,\n",
       " <tensorflow.python.keras.layers.core.Reshape at 0x141725a50>,\n",
       " <tensorflow.python.keras.layers.core.Reshape at 0x141928090>,\n",
       " <tensorflow.python.keras.layers.core.Dropout at 0x1427b83d0>,\n",
       " <tensorflow.python.keras.layers.core.Dropout at 0x141725a90>,\n",
       " <tensorflow.python.keras.layers.core.Dropout at 0x1419c8910>,\n",
       " <tensorflow.python.keras.layers.core.Dropout at 0x142528c50>,\n",
       " <stellargraph.layer.graphsage.MeanAggregator at 0x1093d8f90>,\n",
       " <tensorflow.python.keras.layers.core.Reshape at 0x141928ad0>,\n",
       " <tensorflow.python.keras.layers.core.Dropout at 0x141976610>,\n",
       " <tensorflow.python.keras.layers.core.Dropout at 0x1418e4910>,\n",
       " <stellargraph.layer.graphsage.MeanAggregator at 0x10945b690>,\n",
       " <tensorflow.python.keras.layers.core.Reshape at 0x141970dd0>,\n",
       " <tensorflow.python.keras.layers.core.Lambda at 0x14172ac90>,\n",
       " <tensorflow.python.keras.layers.core.Dense at 0x14172add0>]"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.layers(0)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "40",
   "metadata": {},
   "source": [
    "Train the model, keeping track of its loss and accuracy on the training set, and its performance on the validation set during the training (e.g., for early stopping), and generalization performance of the final model on a held-out test set (we need to create another generator over the test data for this)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "41",
   "metadata": {},
   "outputs": [],
   "source": [
    "val_gen = generator.flow(val_labels.index, val_targets)\n",
    "test_gen = generator.flow(test_labels.index, test_targets)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "42",
   "metadata": {},
   "source": [
    "Note that the amount of time to train the ensemble is linear to `n_estimators`.\n",
    "\n",
    "Also, we are going to use early stopping by monitoring the accuracy on the validation data and stopping if the accuracy does not increase after 10 training epochs (this is the default grace value specified by the `Ensemble` class but we can set it to a different value by using `model.early_stopping_patience=20` for example.)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "43",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n"
     ]
    }
   ],
   "source": [
    "if use_bagging:\n",
    "    # When using bootstrap samples to train each model in the ensemble, we must specify\n",
    "    # the IDs of the training nodes (train_data) and their corresponding target values\n",
    "    # (train_targets)\n",
    "    history = model.fit(\n",
    "        generator,\n",
    "        train_data=train_labels.index,\n",
    "        train_targets=train_targets,\n",
    "        epochs=epochs,\n",
    "        validation_data=val_gen,\n",
    "        verbose=0,\n",
    "        shuffle=False,\n",
    "        bag_size=None,\n",
    "        use_early_stopping=True,  # Enable early stopping\n",
    "        early_stopping_monitor=\"val_acc\",\n",
    "    )\n",
    "else:\n",
    "    history = model.fit(\n",
    "        train_gen,\n",
    "        epochs=epochs,\n",
    "        validation_data=val_gen,\n",
    "        verbose=0,\n",
    "        shuffle=False,\n",
    "        use_early_stopping=True,  # Enable early stopping\n",
    "        early_stopping_monitor=\"val_acc\",\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "44",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sg.utils.plot_history(history)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "45",
   "metadata": {},
   "source": [
    "Now we have trained the model, let's evaluate it on the test set. Note that the `.evaluate()` method of the `Ensemble` class returns mean and standard deviation of each evaluation metric."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "46",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "  ['...']\n",
      "\n",
      "Test Set Metrics of the trained models:\n",
      "\tloss: 0.6872±0.0270\n",
      "\tacc: 0.8085±0.0109\n"
     ]
    }
   ],
   "source": [
    "test_metrics_mean, test_metrics_std = model.evaluate(test_gen)\n",
    "\n",
    "print(\"\\nTest Set Metrics of the trained models:\")\n",
    "for name, m, s in zip(model.metrics_names, test_metrics_mean, test_metrics_std):\n",
    "    print(\"\\t{}: {:0.4f}±{:0.4f}\".format(name, m, s))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "47",
   "metadata": {},
   "source": [
    "### Making predictions with the model"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "48",
   "metadata": {},
   "source": [
    "Now let's get the predictions for all nodes, using a new generator for all nodes:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "49",
   "metadata": {},
   "outputs": [],
   "source": [
    "all_nodes = labels.index\n",
    "all_gen = generator.flow(all_nodes)\n",
    "all_predictions = model.predict(generator=all_gen)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "50",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(5, 10, 2708, 7)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_predictions.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "51",
   "metadata": {},
   "source": [
    "For full-batch methods, the batch dimension is 1 so we will remove any singleton dimensions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "52",
   "metadata": {},
   "outputs": [],
   "source": [
    "all_predictions = np.squeeze(all_predictions)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "53",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(5, 10, 2708, 7)"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_predictions.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "54",
   "metadata": {},
   "source": [
    "These predictions will be the output of the softmax layer, so to get final categories we'll use the `inverse_transform` method of our target attribute specification to turn these values back to the original categories"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "55",
   "metadata": {},
   "source": [
    "For demonstration, we are going to select one of the nodes in the graph, and plot the ensemble's predictions for that node."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "56",
   "metadata": {},
   "outputs": [],
   "source": [
    "selected_query_point = -1"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "57",
   "metadata": {},
   "source": [
    "The array `all_predictions` has dimensionality $MxKxNxF$ where $M$ is the number of estimators in the ensemble (`n_estimators`); $K$ is the number of predictions per query point per estimator (`n_predictions`); $N$ is the number of query points (`len(all_predictions)`); and $F$ is the output dimensionality of the specified layer determined by the shape of the output layer.\n",
    "\n",
    "Since we are only interested in the predictions for a single query node, e.g., `selected_query_point`, we are going to slice the array to extract them."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "58",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(5, 10, 7)"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Select the predictions for the point specified by selected_query_point\n",
    "qp_predictions = all_predictions[:, :, selected_query_point, :]\n",
    "# The shape should be n_estimators x n_predictions x size_output_layer\n",
    "qp_predictions.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "59",
   "metadata": {},
   "source": [
    "Next, to facilitate plotting the predictions using either a density plot or a box plot, we are going to reshape `qp_predictions` to $R\\times F$ where $R$ is equal to $M\\times K$ as above and $F$ is the output dimensionality of the output layer."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "60",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(50, 7)"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "qp_predictions = qp_predictions.reshape(\n",
    "    np.product(qp_predictions.shape[0:-1]), qp_predictions.shape[-1]\n",
    ")\n",
    "qp_predictions.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "61",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{0: 'Case_Based',\n",
       " 1: 'Genetic_Algorithms',\n",
       " 2: 'Neural_Networks',\n",
       " 3: 'Probabilistic_Methods',\n",
       " 4: 'Reinforcement_Learning',\n",
       " 5: 'Rule_Learning',\n",
       " 6: 'Theory'}"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "inv_subject_mapper = {k: v for k, v in enumerate(target_encoding.classes_)}\n",
    "inv_subject_mapper"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "62",
   "metadata": {},
   "source": [
    "We'd like to assess the ensemble's confidence in its predictions in order to decide if we can trust them or not. Utilising density plots, we can visually inspect the ensemble's distribution of prediction probabilities for a node's label.\n",
    "\n",
    "This is better demonstrated if the ensemble's base mode is `GraphSAGE` because the predictions of the base model vary most (when compared to GCN and GAT) due to the random sampling of node neighbours during prediction in addition to the inherent stocasticity of the ensemble itself.\n",
    "\n",
    "If the density plot for the predicted node label is well separated from those of the other labels with little overlap then we can be confident trusting the model's prediction."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "63",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "if model_type not in [\"gcn\", \"gat\"]:\n",
    "    fig, ax = plt.subplots(figsize=(12, 6))\n",
    "    for i in range(qp_predictions.shape[1]):\n",
    "        sns.kdeplot(data=qp_predictions[:, i].reshape((-1,)), label=inv_subject_mapper[i])\n",
    "    plt.xlabel(\"Predicted Probability\")\n",
    "    plt.title(\"Density plots of predicted probabilities for each subject\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "64",
   "metadata": {},
   "source": [
    "An alternative and possibly more informative view of the distribution of node predictions is a box plot."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "65",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'Subject')"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(12, 6))\n",
    "ax.boxplot(x=qp_predictions)\n",
    "ax.set_xticklabels(target_encoding.classes_)\n",
    "ax.tick_params(axis=\"x\", rotation=45)\n",
    "if model_type == \"graphsage\":\n",
    "    y = np.argmax(target_encoding.transform(labels), axis=1)\n",
    "elif model_type == \"gcn\" or model_type == \"gat\":\n",
    "    y = np.argmax(target_encoding.transform(labels.reindex(G.nodes())), axis=1)\n",
    "plt.title(f\"Correct {target_encoding.classes_[y[selected_query_point]]}\")\n",
    "plt.ylabel(\"Predicted Probability\")\n",
    "plt.xlabel(\"Subject\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "66",
   "metadata": {},
   "source": [
    "The above example shows that the ensemble predicts the correct node label with high confidence so we can trust its prediction.\n",
    "\n",
    "(Note that due to the stochastic nature of training neural network algorithms, the above conclusion may not be valid if you re-run the notebook; however, the general conclusion that the use of ensemble learning can be used to quantify the model's uncertainty about its predictions still holds.)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "67",
   "metadata": {},
   "source": [
    "## Node embeddings\n",
    "\n",
    "Evaluate node embeddings as activations of the output of one of the graph convolutional or aggregation layers in the ensemble model, and visualise them, coloring nodes by their subject label.\n",
    "\n",
    "You can find the index of the layer of interest by calling the `Ensemble` class's method `layers`, e.g., `model.layers()`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "68",
   "metadata": {},
   "outputs": [],
   "source": [
    "if model_type == \"graphsage\":\n",
    "    # For GraphSAGE, we are going to use the output activations of the second GraphSAGE layer\n",
    "    # as the node embeddings\n",
    "    emb = model.predict(\n",
    "        generator=generator, predict_data=labels.index, output_layer=-4\n",
    "    )  # this selects the output layer\n",
    "elif model_type == \"gcn\" or model_type == \"gat\":\n",
    "    # For GCN and GAT, we are going to use the output activations of the first GCN or Graph\n",
    "    # Attention layer as the node embeddings\n",
    "    emb = model.predict(\n",
    "        generator=generator, predict_data=labels.index, output_layer=6\n",
    "    )  # this selects the output layer"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "69",
   "metadata": {},
   "source": [
    "The array `emb` has dimensionality $MxKxNxF$ (or $MxKx1xNxF$for full batch methods) where $M$ is the number of estimators in the ensemble (`n_estimators`); $K$ is the number of predictions per query point per estimator (`n_predictions`); $N$ is the number of query points (`len(node_data.index)`); and $F$ is the output dimensionality of the specified layer determined by the shape of the readout layer as specified above."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "70",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(5, 10, 2708, 1, 16)"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "emb.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "71",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(5, 10, 2708, 16)"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "emb = np.squeeze(emb)\n",
    "emb.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "72",
   "metadata": {},
   "source": [
    "Next we are going to average the predictions over the number of models and the number of predictions per query point. \n",
    "\n",
    "The dimensionality of the array will then be **NxF** where N is the number of points to predict (equal to the number of nodes in the graph for this example) and F is the dimensionality of the embeddings that depends on the output shape of the readout layer as specified above.\n",
    "\n",
    "Note that we could have achieved the same by specifying `summarise=True` in the call to the method `predict` above."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "73",
   "metadata": {},
   "outputs": [],
   "source": [
    "emb = np.mean(emb, axis=(0, 1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "74",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2708, 16)"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "emb.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "75",
   "metadata": {},
   "source": [
    "Project the embeddings to 2d using either TSNE or PCA transform, and visualise, coloring nodes by their subject label"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "76",
   "metadata": {},
   "outputs": [],
   "source": [
    "X = emb\n",
    "y = np.argmax(target_encoding.transform(labels), axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "77",
   "metadata": {},
   "outputs": [],
   "source": [
    "if X.shape[1] > 2:\n",
    "    transform = TSNE  # PCA\n",
    "\n",
    "    trans = transform(n_components=2)\n",
    "    emb_transformed = pd.DataFrame(trans.fit_transform(X), index=labels.index)\n",
    "    emb_transformed[\"label\"] = y\n",
    "else:\n",
    "    emb_transformed = pd.DataFrame(X, index=labels.index)\n",
    "    emb_transformed = emb_transformed.rename(columns={\"0\": 0, \"1\": 1})\n",
    "    emb_transformed[\"label\"] = y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "78",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "alpha = 0.7\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(7, 7))\n",
    "ax.scatter(\n",
    "    emb_transformed[0],\n",
    "    emb_transformed[1],\n",
    "    c=emb_transformed[\"label\"].astype(\"category\"),\n",
    "    cmap=\"jet\",\n",
    "    alpha=alpha,\n",
    ")\n",
    "ax.set(aspect=\"equal\", xlabel=\"$X_1$\", ylabel=\"$X_2$\")\n",
    "plt.title(\n",
    "    \"{} visualization of {} embeddings for cora dataset\".format(\n",
    "        model_type, transform.__name__\n",
    "    )\n",
    ")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "79",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "80",
   "metadata": {
    "nbsphinx": "hidden",
    "tags": [
     "CloudRunner"
    ]
   },
   "source": [
    "<table><tr><td>Run the latest release of this notebook:</td><td><a href=\"https://mybinder.org/v2/gh/stellargraph/stellargraph/master?urlpath=lab/tree/demos/ensembles/ensemble-node-classification-example.ipynb\" alt=\"Open In Binder\" target=\"_parent\"><img src=\"https://mybinder.org/badge_logo.svg\"/></a></td><td><a href=\"https://colab.research.google.com/github/stellargraph/stellargraph/blob/master/demos/ensembles/ensemble-node-classification-example.ipynb\" alt=\"Open In Colab\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\"/></a></td></tr></table>"
   ]
  }
 ],
 "metadata": {
  "file_extension": ".py",
  "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": "3.7.5"
  },
  "mimetype": "text/x-python",
  "name": "python",
  "npconvert_exporter": "python",
  "pygments_lexer": "ipython3",
  "version": 3
 },
 "nbformat": 4,
 "nbformat_minor": 4
}