"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"This notebook demonstrates how to use `StellarGraph`'s implementation of *Cluster-GCN*, [1], for node classification on a homogeneous graph.\n",
"\n",
"*Cluster-GCN* is an extension of the Graph Convolutional Network (GCN) algorithm, [2], for scalable training of deeper Graph Neural Networks using Stochastic Gradient Descent (SGD).\n",
"\n",
"As a first step, *Cluster-GCN* splits a given graph into `k` non-overlapping subgraphs, i.e., no two subgraphs share a node. In [1], it is suggested that for best classification performance, the *METIS* graph clustering algorithm, [3], should be utilised; *METIS* groups together nodes that form a well connected neighborhood with few connections to other subgraphs. The default clustering algorithm `StellarGraph` uses is the random assignment of nodes to clusters. The user is free to use any suitable clustering algorithm to determine the clusters before training the *Cluster-GCN* model. \n",
"\n",
"This notebook demonstrates how to use either random clustering or METIS. For the latter, it is necessary that 3rd party software has correctly been installed; later, we provide links to websites that host the software and provide detailed installation instructions. \n",
"\n",
"During model training, each subgraph or combination of subgraphs is treated as a mini-batch for estimating the parameters of a *GCN* model. A pass over all subgraphs is considered a training epoch.\n",
"\n",
"*Cluster-GCN* further extends *GCN* from the transductive to the inductive setting but this is not demonstrated in this notebook.\n",
"\n",
"This notebook demonstrates *Cluster-GCN* for node classification using 2 citation network datasets, `Cora` and `PubMed-Diabetes`.\n",
"\n",
"**References**\n",
"\n",
"[1] Cluster-GCN: An Efficient Algorithm for Training Deep and Large Graph Convolutional Networks. W. Chiang, X. Liu, S. Si, Y. Li, S. Bengio, and C. Hsiej, KDD, 2019, arXiv:1905.07953 ([download link](https://arxiv.org/abs/1905.07953))\n",
"\n",
"[2] Semi-Supervised Classification with Graph Convolutional Networks. T. Kipf, M. Welling. ICLR 2017. arXiv:1609.02907 ([download link](https://arxiv.org/abs/1609.02907))\n",
"\n",
"[3] METIS: Serial Graph Partitioning and Fill-reducing Matrix Ordering. ([download link](http://glaros.dtc.umn.edu/gkhome/views/metis))"
]
},
{
"cell_type": "code",
"execution_count": 1,
"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]==1.0.0"
]
},
{
"cell_type": "code",
"execution_count": 2,
"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(\"1.0.0\")\n",
"except AttributeError:\n",
" raise ValueError(\n",
" f\"This notebook requires StellarGraph version 1.0.0, but a different version {sg.__version__} is installed. Please see .\"\n",
" ) from None"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {
"scrolled": true
},
"outputs": [],
"source": [
"import networkx as nx\n",
"import pandas as pd\n",
"import itertools\n",
"import json\n",
"import os\n",
"\n",
"import numpy as np\n",
"\n",
"from networkx.readwrite import json_graph\n",
"\n",
"from sklearn.preprocessing import StandardScaler\n",
"\n",
"import stellargraph as sg\n",
"from stellargraph.mapper import ClusterNodeGenerator\n",
"from stellargraph.layer import ClusterGCN\n",
"from stellargraph import globalvar\n",
"\n",
"from tensorflow.keras import backend as K\n",
"\n",
"from tensorflow.keras import layers, optimizers, losses, metrics, Model\n",
"from sklearn import preprocessing, feature_extraction, model_selection\n",
"from stellargraph import datasets\n",
"from IPython.display import display, HTML\n",
"from IPython.display import display, HTML\n",
"import matplotlib.pyplot as plt\n",
"%matplotlib inline"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Loading the dataset\n",
"\n",
"This notebook demonstrates node classification using the *Cluster-GCN* algorithm using one of two citation networks, `Cora` and `Pubmed`."
]
},
{
"cell_type": "code",
"execution_count": 4,
"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": [
""
]
},
"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": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"display(HTML(datasets.Cora().description))\n",
"display(HTML(datasets.PubMedDiabetes().description))"
]
},
{
"cell_type": "markdown",
"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": 5,
"metadata": {
"tags": [
"DataLoading"
]
},
"outputs": [],
"source": [
"dataset = \"cora\" # can also select 'pubmed'\n",
"\n",
"if dataset == \"cora\":\n",
" G, labels = datasets.Cora().load()\n",
"elif dataset == \"pubmed\":\n",
" G, labels = datasets.PubMedDiabetes().load()"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"StellarGraph: Undirected multigraph\n",
" Nodes: 2708, Edges: 5429\n",
"\n",
" Node types:\n",
" paper: [2708]\n",
" Features: float32 vector, length 1433\n",
" Edge types: paper-cites->paper\n",
"\n",
" Edge types:\n",
" paper-cites->paper: [5429]\n"
]
}
],
"source": [
"print(G.info())"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"{'Case_Based',\n",
" 'Genetic_Algorithms',\n",
" 'Neural_Networks',\n",
" 'Probabilistic_Methods',\n",
" 'Reinforcement_Learning',\n",
" 'Rule_Learning',\n",
" 'Theory'}"
]
},
"execution_count": 6,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"set(labels)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Splitting the data"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"We aim to train a graph-ML model that will predict the **subject** or **label** (depending on the dataset) attribute on the nodes."
]
},
{
"cell_type": "markdown",
"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.\n",
"\n",
"The number of labeled nodes we use for training depends on the dataset. We use 140 labeled nodes for `Cora` and 60 for `Pubmed` training. The validation and test sets have the same sizes for both datasets. We use 500 nodes for validation and the rest for testing."
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {},
"outputs": [],
"source": [
"if dataset == \"cora\":\n",
" train_size = 140\n",
"elif dataset == \"pubmed\":\n",
" train_size = 60\n",
"\n",
"train_labels, test_labels = model_selection.train_test_split(\n",
" labels, train_size=train_size, test_size=None, stratify=labels\n",
")\n",
"val_labels, test_labels = model_selection.train_test_split(\n",
" test_labels, train_size=500, test_size=None, stratify=test_labels\n",
")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Note using stratified sampling gives the following counts:"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"Counter({'Neural_Networks': 42,\n",
" 'Probabilistic_Methods': 22,\n",
" 'Reinforcement_Learning': 11,\n",
" 'Genetic_Algorithms': 22,\n",
" 'Case_Based': 16,\n",
" 'Theory': 18,\n",
" 'Rule_Learning': 9})"
]
},
"execution_count": 8,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"from collections import Counter\n",
"\n",
"Counter(train_labels)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"The training set has class imbalance that might need to be compensated, e.g., via using a weighted cross-entropy loss in model training, with class weights inversely proportional to class support. However, we will ignore the class imbalance in this example, for simplicity."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Converting to numeric arrays"
]
},
{
"cell_type": "markdown",
"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": 10,
"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",
"metadata": {},
"source": [
"Next, we prepare a Pandas DataFrame holding the node attributes we want to use to predict the subject. These are the feature vectors that the Keras model will use as input. `Cora` contains attributes 'w_x' that correspond to words found in that publication. If a word occurs more than once in a publication the relevant attribute will be set to one, otherwise it will be zero. `Pubmed` has similar feature vectors associated with each node but the values are [tf-idf.](https://en.wikipedia.org/wiki/Tf%E2%80%93idf)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Train using cluster GCN"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Graph Clustering \n",
"\n",
"*Cluster-GCN* requires that a graph is clustered into `k` non-overlapping subgraphs. These subgraphs are used as batches to train a *GCN* model. \n",
"\n",
"Any graph clustering method can be used, including random clustering that is the default clustering method in `StellarGraph`. \n",
"\n",
"However, the choice of clustering algorithm can have a large impact on performance. In the *Cluster-GCN* paper, [1], it is suggested that the *METIS* algorithm is used as it produces subgraphs that are well connected with few intra-graph edges. \n",
"\n",
"This demo uses random clustering by default. \n",
"\n",
"#### METIS\n",
"\n",
"In order to use *METIS*, you must download and install the official implemention from [here](http://glaros.dtc.umn.edu/gkhome/views/metis). Also, you must install the Python `metis` library by following the instructions [here.](https://metis.readthedocs.io/en/latest/)"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {},
"outputs": [],
"source": [
"number_of_clusters = 10 # the number of clusters/subgraphs\n",
"clusters_per_batch = 2 # combine two cluster per batch\n",
"random_clusters = True # Set to False if you want to use METIS for clustering"
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {},
"outputs": [],
"source": [
"node_ids = np.array(G.nodes())"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {},
"outputs": [],
"source": [
"if random_clusters:\n",
" # We don't have to specify the cluster because the CluserNodeGenerator will take\n",
" # care of the random clustering for us.\n",
" clusters = number_of_clusters\n",
"else:\n",
" # We are going to use the METIS clustering algorith,\n",
" print(\"Graph clustering using the METIS algorithm.\")\n",
"\n",
" import metis\n",
"\n",
" lil_adj = G.to_adjacency_matrix().tolil()\n",
" adjlist = [tuple(neighbours) for neighbours in lil_adj.rows]\n",
"\n",
" edgecuts, parts = metis.part_graph(adjlist, number_of_clusters)\n",
" parts = np.array(parts)\n",
" clusters = []\n",
" cluster_ids = np.unique(parts)\n",
" for cluster_id in cluster_ids:\n",
" mask = np.where(parts == cluster_id)\n",
" clusters.append(node_ids[mask])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Next we create the `ClusterNodeGenerator` object that will give us access to a generator suitable for model training, evaluation, and prediction via the Keras API. \n",
"\n",
"We specify the number of clusters and the number of clusters to combine per batch, **q**."
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Number of clusters 10\n",
"0 cluster has size 270\n",
"1 cluster has size 270\n",
"2 cluster has size 270\n",
"3 cluster has size 270\n",
"4 cluster has size 270\n",
"5 cluster has size 270\n",
"6 cluster has size 270\n",
"7 cluster has size 270\n",
"8 cluster has size 270\n",
"9 cluster has size 278\n"
]
}
],
"source": [
"generator = ClusterNodeGenerator(G, clusters=clusters, q=clusters_per_batch, lam=0.1)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Now we can specify our machine learning model, we need a few more parameters for this:\n",
"\n",
" * the `layer_sizes` is a list of hidden feature sizes of each layer in the model. In this example we use two GCN layers with 32-dimensional hidden node features at each layer.\n",
" * `activations` is a list of activations applied to each layer's output\n",
" * `dropout=0.5` specifies a 50% dropout at each layer. "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"We create the *Cluster-GCN* model as follows:"
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {
"scrolled": true
},
"outputs": [],
"source": [
"cluster_gcn = ClusterGCN(\n",
" layer_sizes=[32, 32], activations=[\"relu\", \"relu\"], generator=generator, dropout=0.5\n",
")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"To create a Keras model we now expose the input and output tensors of the *Cluster-GCN* model for node prediction, via the `ClusterGCN.in_out_tensors` method:"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {},
"outputs": [],
"source": [
"x_inp, x_out = cluster_gcn.in_out_tensors()"
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"[,\n",
" ,\n",
" ]"
]
},
"execution_count": 16,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"x_inp"
]
},
{
"cell_type": "code",
"execution_count": 18,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
""
]
},
"execution_count": 17,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"x_out"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"We are also going to add a final layer dense layer with softmax output activation. This layers performs classification so we set the number of units to equal the number of classes."
]
},
{
"cell_type": "code",
"execution_count": 19,
"metadata": {},
"outputs": [],
"source": [
"predictions = layers.Dense(units=train_targets.shape[1], activation=\"softmax\")(x_out)"
]
},
{
"cell_type": "code",
"execution_count": 20,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
""
]
},
"execution_count": 19,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"predictions"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Finally, we build the Tensorflow model and compile it specifying the loss function, optimiser, and metrics to monitor."
]
},
{
"cell_type": "code",
"execution_count": 21,
"metadata": {},
"outputs": [],
"source": [
"model = Model(inputs=x_inp, outputs=predictions)\n",
"model.compile(\n",
" optimizer=optimizers.Adam(lr=0.01),\n",
" loss=losses.categorical_crossentropy,\n",
" metrics=[\"acc\"],\n",
")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Train the model"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"We are now ready to train the `ClusterGCN` model, keeping track of its loss and accuracy on the training set, and its generalisation performance on a validation set."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"We need two generators, one for training and one for validation data. We can create such generators by calling the `flow` method of the `ClusterNodeGenerator` object we created earlier and specifying the node IDs and corresponding ground truth target values for each of the two datasets. "
]
},
{
"cell_type": "code",
"execution_count": 22,
"metadata": {},
"outputs": [],
"source": [
"train_gen = generator.flow(train_labels.index, train_targets, name=\"train\")\n",
"val_gen = generator.flow(val_labels.index, val_targets, name=\"val\")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Finally, we are ready to train our `ClusterGCN` model by calling the `fit` method of our Tensorflow Keras model."
]
},
{
"cell_type": "code",
"execution_count": 23,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
" ['...']\n",
" ['...']\n",
"Train for 5 steps, validate for 5 steps\n",
"Epoch 1/20\n",
"5/5 [==============================] - 1s 147ms/step - loss: 1.8824 - acc: 0.2214 - val_loss: 1.7955 - val_acc: 0.3040\n",
"Epoch 2/20\n",
"5/5 [==============================] - 0s 22ms/step - loss: 1.6687 - acc: 0.2929 - val_loss: 1.6810 - val_acc: 0.3020\n",
"Epoch 3/20\n",
"5/5 [==============================] - 0s 21ms/step - loss: 1.5002 - acc: 0.3429 - val_loss: 1.5643 - val_acc: 0.3900\n",
"Epoch 4/20\n",
"5/5 [==============================] - 0s 21ms/step - loss: 1.2585 - acc: 0.5357 - val_loss: 1.4264 - val_acc: 0.5500\n",
"Epoch 5/20\n",
"5/5 [==============================] - 0s 21ms/step - loss: 0.9910 - acc: 0.7071 - val_loss: 1.2739 - val_acc: 0.5880\n",
"Epoch 6/20\n",
"5/5 [==============================] - 0s 22ms/step - loss: 0.8021 - acc: 0.6857 - val_loss: 1.1933 - val_acc: 0.6080\n",
"Epoch 7/20\n",
"5/5 [==============================] - 0s 21ms/step - loss: 0.6358 - acc: 0.7786 - val_loss: 1.0825 - val_acc: 0.6420\n",
"Epoch 8/20\n",
"5/5 [==============================] - 0s 21ms/step - loss: 0.5681 - acc: 0.8357 - val_loss: 1.0657 - val_acc: 0.6380\n",
"Epoch 9/20\n",
"5/5 [==============================] - 0s 22ms/step - loss: 0.4167 - acc: 0.8571 - val_loss: 1.1342 - val_acc: 0.6360\n",
"Epoch 10/20\n",
"5/5 [==============================] - 0s 21ms/step - loss: 0.3251 - acc: 0.9071 - val_loss: 1.2399 - val_acc: 0.6320\n",
"Epoch 11/20\n",
"5/5 [==============================] - 0s 21ms/step - loss: 0.2713 - acc: 0.9071 - val_loss: 1.1463 - val_acc: 0.6500\n",
"Epoch 12/20\n",
"5/5 [==============================] - 0s 21ms/step - loss: 0.3365 - acc: 0.8857 - val_loss: 1.1205 - val_acc: 0.6500\n",
"Epoch 13/20\n",
"5/5 [==============================] - 0s 22ms/step - loss: 0.2272 - acc: 0.9071 - val_loss: 1.1753 - val_acc: 0.6560\n",
"Epoch 14/20\n",
"5/5 [==============================] - 0s 21ms/step - loss: 0.2948 - acc: 0.9000 - val_loss: 1.2997 - val_acc: 0.6340\n",
"Epoch 15/20\n",
"5/5 [==============================] - 0s 21ms/step - loss: 0.2840 - acc: 0.9000 - val_loss: 1.3871 - val_acc: 0.6200\n",
"Epoch 16/20\n",
"5/5 [==============================] - 0s 22ms/step - loss: 0.1464 - acc: 0.9357 - val_loss: 1.4344 - val_acc: 0.6220\n",
"Epoch 17/20\n",
"5/5 [==============================] - 0s 22ms/step - loss: 0.2943 - acc: 0.9214 - val_loss: 1.3791 - val_acc: 0.6200\n",
"Epoch 18/20\n",
"5/5 [==============================] - 0s 21ms/step - loss: 0.1368 - acc: 0.9714 - val_loss: 1.3297 - val_acc: 0.6260\n",
"Epoch 19/20\n",
"5/5 [==============================] - 0s 21ms/step - loss: 0.1462 - acc: 0.9500 - val_loss: 1.3450 - val_acc: 0.6360\n",
"Epoch 20/20\n",
"5/5 [==============================] - 0s 23ms/step - loss: 0.1845 - acc: 0.9429 - val_loss: 1.3614 - val_acc: 0.6380\n"
]
}
],
"source": [
"history = model.fit(\n",
" train_gen, validation_data=val_gen, epochs=20, verbose=1, shuffle=False\n",
")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Plot the training history:"
]
},
{
"cell_type": "code",
"execution_count": 24,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
"
"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"sg.utils.plot_history(history)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Evaluate the best model on the test set.\n",
"\n",
"Note that *Cluster-GCN* performance can be very poor if using random graph clustering. Using *METIS* instead of random graph clustering produces considerably better results."
]
},
{
"cell_type": "code",
"execution_count": 25,
"metadata": {},
"outputs": [],
"source": [
"test_gen = generator.flow(test_labels.index, test_targets)"
]
},
{
"cell_type": "code",
"execution_count": 26,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
" ['...']\n",
"5/5 [==============================] - 0s 7ms/step - loss: 1.4908 - acc: 0.6291\n",
"\n",
"Test Set Metrics:\n",
"\tloss: 1.4908\n",
"\tacc: 0.6291\n"
]
}
],
"source": [
"test_metrics = model.evaluate(test_gen)\n",
"print(\"\\nTest Set Metrics:\")\n",
"for name, val in zip(model.metrics_names, test_metrics):\n",
" print(\"\\t{}: {:0.4f}\".format(name, val))"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Making predictions with the model"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"For predictions to work correctly, we need to remove the extra batch dimensions necessary for the implementation of *Cluster-GCN* to work. We can easily achieve this by adding a layer after the dense predictions layer to remove this extra dimension."
]
},
{
"cell_type": "code",
"execution_count": 27,
"metadata": {},
"outputs": [],
"source": [
"predictions_flat = layers.Lambda(lambda x: K.squeeze(x, 0))(predictions)"
]
},
{
"cell_type": "code",
"execution_count": 28,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"(,\n",
" )"
]
},
"execution_count": 27,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Notice that we have removed the first dimension\n",
"predictions, predictions_flat"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Now let's get the predictions for all nodes.\n",
"\n",
"We need to create a new model using the same as before input Tensor and our new **predictions_flat** Tensor as the output. We are going to re-use the trained model weights."
]
},
{
"cell_type": "code",
"execution_count": 29,
"metadata": {},
"outputs": [],
"source": [
"model_predict = Model(inputs=x_inp, outputs=predictions_flat)"
]
},
{
"cell_type": "code",
"execution_count": 30,
"metadata": {},
"outputs": [],
"source": [
"all_nodes = list(G.nodes())\n",
"all_gen = generator.flow(all_nodes, name=\"all_gen\")\n",
"all_predictions = model_predict.predict(all_gen)"
]
},
{
"cell_type": "code",
"execution_count": 31,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"(2708, 7)"
]
},
"execution_count": 30,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"all_predictions.shape"
]
},
{
"cell_type": "markdown",
"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 specifcation to turn these values back to the original categories."
]
},
{
"cell_type": "code",
"execution_count": 32,
"metadata": {},
"outputs": [],
"source": [
"node_predictions = target_encoding.inverse_transform(all_predictions)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Let's have a look at a few predictions after training the model:"
]
},
{
"cell_type": "code",
"execution_count": 33,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"2708"
]
},
"execution_count": 32,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"len(all_gen.node_order)"
]
},
{
"cell_type": "code",
"execution_count": 34,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
Predicted
\n",
"
True
\n",
"
\n",
" \n",
" \n",
"
\n",
"
35
\n",
"
Genetic_Algorithms
\n",
"
Genetic_Algorithms
\n",
"
\n",
"
\n",
"
40
\n",
"
Genetic_Algorithms
\n",
"
Genetic_Algorithms
\n",
"
\n",
"
\n",
"
114
\n",
"
Reinforcement_Learning
\n",
"
Reinforcement_Learning
\n",
"
\n",
"
\n",
"
117
\n",
"
Reinforcement_Learning
\n",
"
Reinforcement_Learning
\n",
"
\n",
"
\n",
"
128
\n",
"
Reinforcement_Learning
\n",
"
Reinforcement_Learning
\n",
"
\n",
"
\n",
"
130
\n",
"
Probabilistic_Methods
\n",
"
Reinforcement_Learning
\n",
"
\n",
"
\n",
"
164
\n",
"
Theory
\n",
"
Theory
\n",
"
\n",
"
\n",
"
288
\n",
"
Genetic_Algorithms
\n",
"
Reinforcement_Learning
\n",
"
\n",
"
\n",
"
424
\n",
"
Case_Based
\n",
"
Rule_Learning
\n",
"
\n",
"
\n",
"
434
\n",
"
Case_Based
\n",
"
Reinforcement_Learning
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" Predicted True\n",
"35 Genetic_Algorithms Genetic_Algorithms\n",
"40 Genetic_Algorithms Genetic_Algorithms\n",
"114 Reinforcement_Learning Reinforcement_Learning\n",
"117 Reinforcement_Learning Reinforcement_Learning\n",
"128 Reinforcement_Learning Reinforcement_Learning\n",
"130 Probabilistic_Methods Reinforcement_Learning\n",
"164 Theory Theory\n",
"288 Genetic_Algorithms Reinforcement_Learning\n",
"424 Case_Based Rule_Learning\n",
"434 Case_Based Reinforcement_Learning"
]
},
"execution_count": 33,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"results = pd.Series(node_predictions, index=all_gen.node_order)\n",
"df = pd.DataFrame({\"Predicted\": results, \"True\": labels})\n",
"df.head(10)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Node embeddings\n",
"\n",
"Evaluate node embeddings as activations of the output of the last graph convolution layer in the `ClusterGCN` layer stack and visualise them, coloring nodes by their true subject label. We expect to see nice clusters of papers in the node embedding space, with papers of the same subject belonging to the same cluster.\n",
"\n",
"To calculate the node embeddings rather than the class predictions, we create a new model with the same inputs as we used previously `x_inp` but now the output is the embeddings `x_out` rather than the predicted class. Additionally note that the weights trained previously are kept in the new model.\n",
"\n",
"Note that the embeddings from the `ClusterGCN` model have a batch dimension of 1 so we `squeeze` this to get a matrix of $N_{nodes} \\times N_{emb}$."
]
},
{
"cell_type": "code",
"execution_count": 35,
"metadata": {},
"outputs": [],
"source": [
"x_out_flat = layers.Lambda(lambda x: K.squeeze(x, 0))(x_out)\n",
"embedding_model = Model(inputs=x_inp, outputs=x_out_flat)"
]
},
{
"cell_type": "code",
"execution_count": 36,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"5/5 [==============================] - 0s 14ms/step\n"
]
},
{
"data": {
"text/plain": [
"(2708, 32)"
]
},
"execution_count": 35,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"emb = embedding_model.predict(all_gen, verbose=1)\n",
"emb.shape"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Project the embeddings to 2d using either TSNE or PCA transform, and visualise, coloring nodes by their true subject label"
]
},
{
"cell_type": "code",
"execution_count": 37,
"metadata": {},
"outputs": [],
"source": [
"from sklearn.decomposition import PCA\n",
"from sklearn.manifold import TSNE\n",
"import pandas as pd\n",
"import numpy as np"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**Prediction Node Order**\n",
"\n",
"The predictions are not returned in the same order as the input nodes given. The generator object internally maintains the order of predictions. These are stored in the object's member variable `node_order`. We use `node_order` to re-index the `node_data` DataFrame such that the prediction order in `y` corresponds to that of node embeddings in `X`."
]
},
{
"cell_type": "code",
"execution_count": 38,
"metadata": {},
"outputs": [],
"source": [
"X = emb\n",
"y = np.argmax(\n",
" target_encoding.transform(labels.reindex(index=all_gen.node_order)), axis=1,\n",
")"
]
},
{
"cell_type": "code",
"execution_count": 39,
"metadata": {},
"outputs": [],
"source": [
"if X.shape[1] > 2:\n",
" transform = TSNE # or use PCA for speed\n",
"\n",
" trans = transform(n_components=2)\n",
" emb_transformed = pd.DataFrame(trans.fit_transform(X), index=all_gen.node_order)\n",
" emb_transformed[\"label\"] = y\n",
"else:\n",
" emb_transformed = pd.DataFrame(X, index=list(G.nodes()))\n",
" emb_transformed = emb_transformed.rename(columns={\"0\": 0, \"1\": 1})\n",
" emb_transformed[\"label\"] = y"
]
},
{
"cell_type": "code",
"execution_count": 40,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
"