16Dec-GAE&VGAE-Nb.ipynb

{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"name":"16Dec-GAE&VGAE-Nb.ipynb","provenance":[],"authorship_tag":"ABX9TyMrFjTL2E2PBXzsPbBg06sb"},"kernelspec":{"name":"python3","display_name":"Python 3"},"language_info":{"name":"python"}},"cells":[{"cell_type":"markdown","source":["# Installation"],"metadata":{"id":"WBPh-9af7ysR"}},{"cell_type":"code","execution_count":1,"metadata":{"id":"xBQ8wuT3DRaL","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1639753725885,"user_tz":-300,"elapsed":22123,"user":{"displayName":"Muhammad Ali","photoUrl":"https://lh3.googleusercontent.com/a/default-user=s64","userId":"08853600868018567439"}},"outputId":"2e3cd790-7d97-457c-ecfb-5d8cdd52f7d5"},"outputs":[{"output_type":"stream","name":"stdout","text":["\u001b[K     |████████████████████████████████| 7.9 MB 2.6 MB/s \n","\u001b[K     |████████████████████████████████| 3.5 MB 2.6 MB/s \n","\u001b[K     |████████████████████████████████| 407 kB 5.4 MB/s \n","\u001b[K     |████████████████████████████████| 41 kB 374 kB/s \n","\u001b[?25h  Building wheel for torch-geometric (setup.py) ... \u001b[?25l\u001b[?25hdone\n"]}],"source":["# Install required packages.\n","!pip install -q torch-scatter -f https://data.pyg.org/whl/torch-1.10.0+cu113.html\n","!pip install -q torch-sparse -f https://data.pyg.org/whl/torch-1.10.0+cu113.html\n","!pip install -q git+https://github.com/pyg-team/pytorch_geometric.git"]},{"cell_type":"markdown","source":["# Imports"],"metadata":{"id":"LSf9l0zs7-mE"}},{"cell_type":"code","source":["import os.path as osp\n","import torch\n","import torch.nn.functional as F\n","from sklearn.cluster import KMeans\n","from sklearn.metrics.cluster import (v_measure_score, homogeneity_score, completeness_score)\n","from sklearn.manifold import TSNE\n","import matplotlib.pyplot as plt\n","import torch_geometric.transforms as T\n","from torch_geometric.datasets import Planetoid\n","from torch_geometric.nn import GCNConv\n","from torch_geometric.nn.models.autoencoder import ARGVA\n","from torch_geometric.utils import train_test_split_edges"],"metadata":{"id":"MNe6x53-7_Q4","executionInfo":{"status":"ok","timestamp":1639757347376,"user_tz":-300,"elapsed":8393,"user":{"displayName":"Muhammad Ali","photoUrl":"https://lh3.googleusercontent.com/a/default-user=s64","userId":"08853600868018567439"}}},"execution_count":2,"outputs":[]},{"cell_type":"markdown","source":["# Data Part"],"metadata":{"id":"Dj-eWWnAJ6Rt"}},{"cell_type":"code","source":["dataset = 'Cora'\n","path = osp.join('.', 'data', dataset)\n","dataset = Planetoid(path, dataset, transform=T.NormalizeFeatures())\n","data = dataset.get(0)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"U3HvVph6J1Sn","executionInfo":{"status":"ok","timestamp":1639757368675,"user_tz":-300,"elapsed":2088,"user":{"displayName":"Muhammad Ali","photoUrl":"https://lh3.googleusercontent.com/a/default-user=s64","userId":"08853600868018567439"}},"outputId":"439c9ed7-a40f-43ed-d0e5-d08bd5462cb3"},"execution_count":3,"outputs":[{"output_type":"stream","name":"stderr","text":["Downloading https://github.com/kimiyoung/planetoid/raw/master/data/ind.cora.x\n","Downloading https://github.com/kimiyoung/planetoid/raw/master/data/ind.cora.tx\n","Downloading https://github.com/kimiyoung/planetoid/raw/master/data/ind.cora.allx\n","Downloading https://github.com/kimiyoung/planetoid/raw/master/data/ind.cora.y\n","Downloading https://github.com/kimiyoung/planetoid/raw/master/data/ind.cora.ty\n","Downloading https://github.com/kimiyoung/planetoid/raw/master/data/ind.cora.ally\n","Downloading https://github.com/kimiyoung/planetoid/raw/master/data/ind.cora.graph\n","Downloading https://github.com/kimiyoung/planetoid/raw/master/data/ind.cora.test.index\n","Processing...\n","Done!\n"]}]},{"cell_type":"code","source":["num_nodes = data.x.shape[0]"],"metadata":{"id":"Y8yhTsmaJ8BH","executionInfo":{"status":"ok","timestamp":1639757377593,"user_tz":-300,"elapsed":2,"user":{"displayName":"Muhammad Ali","photoUrl":"https://lh3.googleusercontent.com/a/default-user=s64","userId":"08853600868018567439"}}},"execution_count":4,"outputs":[]},{"cell_type":"code","source":["data.train_mask = data.val_mask = data.test_mask = None\n","data = train_test_split_edges(data)\n","data"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"ppkr78IfJ-x3","executionInfo":{"status":"ok","timestamp":1639757400452,"user_tz":-300,"elapsed":566,"user":{"displayName":"Muhammad Ali","photoUrl":"https://lh3.googleusercontent.com/a/default-user=s64","userId":"08853600868018567439"}},"outputId":"68fcb3ea-4a44-4653-c140-47f2f6b37294"},"execution_count":5,"outputs":[{"output_type":"stream","name":"stderr","text":["/usr/local/lib/python3.7/dist-packages/torch_geometric/deprecation.py:13: UserWarning: 'train_test_split_edges' is deprecated, use 'transforms.RandomLinkSplit' instead\n","  warnings.warn(out)\n"]},{"output_type":"execute_result","data":{"text/plain":["Data(x=[2708, 1433], y=[2708], val_pos_edge_index=[2, 263], test_pos_edge_index=[2, 527], train_pos_edge_index=[2, 8976], train_neg_adj_mask=[2708, 2708], val_neg_edge_index=[2, 263], test_neg_edge_index=[2, 527])"]},"metadata":{},"execution_count":5}]},{"cell_type":"markdown","source":["# Model Part"],"metadata":{"id":"GczUicx-KMa5"}},{"cell_type":"code","source":["class VEncoder(torch.nn.Module):\n","  def __init__(self, in_channels, out_channels):\n","    super(VEncoder, self).__init__()\n","    self.conv1=GCNConv(in_channels, 2 * out_channels, cached=True)\n","    self.conv_mu=GCNConv(2 * out_channels, out_channels, cached=True)\n","    self.conv_logstd=GCNConv(2*out_channels, out_channels, cached = True)\n","\n","  def forward(self, x, edge_index):\n","    x=F.relu(self.conv1(x, edge_index))\n","\n","    return self.conv_mu(x, edge_index), self.conv_logstd(x, edge_index)"],"metadata":{"id":"ZEVc0ZDDKEP0","executionInfo":{"status":"ok","timestamp":1639757682717,"user_tz":-300,"elapsed":6,"user":{"displayName":"Muhammad Ali","photoUrl":"https://lh3.googleusercontent.com/a/default-user=s64","userId":"08853600868018567439"}}},"execution_count":6,"outputs":[]},{"cell_type":"code","source":["class Discriminator(torch.nn.Module):\n","  def __init__(self, in_channels, hidden_channels, out_channels):\n","    super(Discriminator, self).__init__()\n","    self.lin1=torch.nn.Linear(in_channels, hidden_channels)\n","    self.lin2=torch.nn.Linear(hidden_channels, hidden_channels)\n","    self.lin3=torch.nn.Linear(hidden_channels, out_channels)\n","\n","  def forward(self, x):\n","\n","    x=F.relu(self.lin1(x))\n","    x=F.relu(self.lin2(x))\n","    x=self.lin3(x)\n","\n","    return x"],"metadata":{"id":"-X35n2eoLJNY","executionInfo":{"status":"ok","timestamp":1639757968244,"user_tz":-300,"elapsed":5,"user":{"displayName":"Muhammad Ali","photoUrl":"https://lh3.googleusercontent.com/a/default-user=s64","userId":"08853600868018567439"}}},"execution_count":7,"outputs":[]},{"cell_type":"code","source":["latent_size = 32\n","encoder = VEncoder(data.num_features, out_channels=latent_size)\n","\n","discriminator = Discriminator(in_channels=latent_size, hidden_channels=64, \n","                              out_channels=1)"],"metadata":{"id":"kZtY4NazMO_A","executionInfo":{"status":"ok","timestamp":1639758116818,"user_tz":-300,"elapsed":4,"user":{"displayName":"Muhammad Ali","photoUrl":"https://lh3.googleusercontent.com/a/default-user=s64","userId":"08853600868018567439"}}},"execution_count":8,"outputs":[]},{"cell_type":"code","source":["model = ARGVA(encoder, discriminator)\n","\n","device = torch.device('cuda' if torch.cuda.is_available() and use_cuda else 'cpu')\n","model, data = model.to(device), data.to(device)"],"metadata":{"id":"Dx9KhCQZMzS1","executionInfo":{"status":"ok","timestamp":1639758208612,"user_tz":-300,"elapsed":375,"user":{"displayName":"Muhammad Ali","photoUrl":"https://lh3.googleusercontent.com/a/default-user=s64","userId":"08853600868018567439"}}},"execution_count":9,"outputs":[]},{"cell_type":"markdown","source":["> optimizers"],"metadata":{"id":"yAiDk-ieNMQq"}},{"cell_type":"code","source":["discriminator_optimizer = torch.optim.Adam(discriminator.parameters(), lr=0.001)\n","encoder_optimizer = torch.optim.Adam(encoder.parameters(), lr=0.005)"],"metadata":{"id":"gIoMyV9ENJqb","executionInfo":{"status":"ok","timestamp":1639758227724,"user_tz":-300,"elapsed":2,"user":{"displayName":"Muhammad Ali","photoUrl":"https://lh3.googleusercontent.com/a/default-user=s64","userId":"08853600868018567439"}}},"execution_count":10,"outputs":[]},{"cell_type":"code","source":["ARGVA??"],"metadata":{"id":"c5cF-KabNOOx","executionInfo":{"status":"ok","timestamp":1639758409526,"user_tz":-300,"elapsed":593,"user":{"displayName":"Muhammad Ali","photoUrl":"https://lh3.googleusercontent.com/a/default-user=s64","userId":"08853600868018567439"}}},"execution_count":14,"outputs":[]},{"cell_type":"code","source":["model.VGAE??"],"metadata":{"id":"nJLm81Z3Nq-v","executionInfo":{"status":"ok","timestamp":1639758468222,"user_tz":-300,"elapsed":4,"user":{"displayName":"Muhammad Ali","photoUrl":"https://lh3.googleusercontent.com/a/default-user=s64","userId":"08853600868018567439"}}},"execution_count":15,"outputs":[]},{"cell_type":"markdown","source":["> So we can see that there are a no of useful attributes which are already implemented for us in PyG"],"metadata":{"id":"x6l3jApbOXsn"}},{"cell_type":"code","source":["model.discriminator_loss??"],"metadata":{"id":"VGZTHtAwPUGb","executionInfo":{"status":"ok","timestamp":1639758781990,"user_tz":-300,"elapsed":4,"user":{"displayName":"Muhammad Ali","photoUrl":"https://lh3.googleusercontent.com/a/default-user=s64","userId":"08853600868018567439"}}},"execution_count":16,"outputs":[]},{"cell_type":"code","source":["def train(model, discriminator):\n","  model.train()\n","  encoder_optimizer.zero_grad()\n","\n","  z= model.encode(data.x, data.train_pos_edge_index)\n","\n","  for i in range(5):\n","    idx=range(num_nodes)\n","    discriminator.train()\n","    discriminator_optimizer.zero_grad()\n","    discriminator_loss=model.discriminator_loss(z[idx])\n","    discriminator_loss.backward(retain_graph=True)\n","    discriminator_optimizer.step()\n","\n","  loss=0\n","  loss = loss + model.reg_loss(z)\n","  loss = loss + model.recon_loss(z, data.train_pos_edge_index)\n","  loss = loss + (1 / data.num_nodes) * model.kl_loss()\n","  loss.backward()\n","\n","  encoder_optimizer.step()\n","  \n","  return loss\n","  "],"metadata":{"id":"jlI5wQjZOJEa","executionInfo":{"status":"ok","timestamp":1639758952979,"user_tz":-300,"elapsed":415,"user":{"displayName":"Muhammad Ali","photoUrl":"https://lh3.googleusercontent.com/a/default-user=s64","userId":"08853600868018567439"}}},"execution_count":21,"outputs":[]},{"cell_type":"code","source":["@torch.no_grad()\n","def test():\n","    model.eval()\n","    z = model.encode(data.x, data.train_pos_edge_index)\n","\n","    # Cluster embedded values using k-means.\n","    kmeans_input = z.cpu().numpy()\n","    kmeans = KMeans(n_clusters=7, random_state=0).fit(kmeans_input)\n","    pred = kmeans.predict(kmeans_input)\n","\n","    labels = data.y.cpu().numpy()\n","    completeness = completeness_score(labels, pred)\n","    hm = homogeneity_score(labels, pred)\n","    nmi = v_measure_score(labels, pred)\n","\n","    auc, ap = model.test(z, data.test_pos_edge_index, data.test_neg_edge_index)\n","\n","    return auc, ap, completeness, hm, nmi"],"metadata":{"id":"TBYKHKoRPvn-","executionInfo":{"status":"ok","timestamp":1639758911369,"user_tz":-300,"elapsed":3,"user":{"displayName":"Muhammad Ali","photoUrl":"https://lh3.googleusercontent.com/a/default-user=s64","userId":"08853600868018567439"}}},"execution_count":18,"outputs":[]},{"cell_type":"code","source":["for epoch in range(1, 101):\n","    loss = train(model, discriminator)\n","    auc, ap, completeness, hm, nmi = test()\n","    print((f'Epoch: {epoch:03d}, Loss: {loss:.3f}, AUC: {auc:.3f}, '\n","           f'AP: {ap:.3f}, Completeness: {completeness:.3f}, '\n","           f'Homogeneity: {hm:.3f}, NMI: {nmi:.3f}'))"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"WRAIhsx5P1J8","executionInfo":{"status":"ok","timestamp":1639759012927,"user_tz":-300,"elapsed":57936,"user":{"displayName":"Muhammad Ali","photoUrl":"https://lh3.googleusercontent.com/a/default-user=s64","userId":"08853600868018567439"}},"outputId":"bdbe8029-1bfe-4ea6-fdae-a477d3107e73"},"execution_count":22,"outputs":[{"output_type":"stream","name":"stdout","text":["Epoch: 001, Loss: 5.650, AUC: 0.724, AP: 0.735, Completeness: 0.095, Homogeneity: 0.096, NMI: 0.096\n","Epoch: 002, Loss: 5.006, AUC: 0.695, AP: 0.715, Completeness: 0.095, Homogeneity: 0.094, NMI: 0.094\n","Epoch: 003, Loss: 4.402, AUC: 0.684, AP: 0.707, Completeness: 0.060, Homogeneity: 0.058, NMI: 0.059\n","Epoch: 004, Loss: 3.949, AUC: 0.685, AP: 0.706, Completeness: 0.109, Homogeneity: 0.106, NMI: 0.107\n","Epoch: 005, Loss: 3.588, AUC: 0.693, AP: 0.712, Completeness: 0.116, Homogeneity: 0.112, NMI: 0.114\n","Epoch: 006, Loss: 3.200, AUC: 0.704, AP: 0.724, Completeness: 0.131, Homogeneity: 0.126, NMI: 0.128\n","Epoch: 007, Loss: 2.920, AUC: 0.708, AP: 0.732, Completeness: 0.167, Homogeneity: 0.160, NMI: 0.163\n","Epoch: 008, Loss: 2.613, AUC: 0.712, AP: 0.738, Completeness: 0.209, Homogeneity: 0.208, NMI: 0.209\n","Epoch: 009, Loss: 2.409, AUC: 0.722, AP: 0.744, Completeness: 0.236, Homogeneity: 0.235, NMI: 0.235\n","Epoch: 010, Loss: 2.222, AUC: 0.732, AP: 0.754, Completeness: 0.242, Homogeneity: 0.246, NMI: 0.244\n","Epoch: 011, Loss: 2.168, AUC: 0.735, AP: 0.758, Completeness: 0.248, Homogeneity: 0.249, NMI: 0.248\n","Epoch: 012, Loss: 2.181, AUC: 0.727, AP: 0.751, Completeness: 0.250, Homogeneity: 0.253, NMI: 0.251\n","Epoch: 013, Loss: 2.295, AUC: 0.730, AP: 0.753, Completeness: 0.273, Homogeneity: 0.277, NMI: 0.275\n","Epoch: 014, Loss: 2.361, AUC: 0.746, AP: 0.770, Completeness: 0.288, Homogeneity: 0.293, NMI: 0.290\n","Epoch: 015, Loss: 2.324, AUC: 0.768, AP: 0.793, Completeness: 0.318, Homogeneity: 0.318, NMI: 0.318\n","Epoch: 016, Loss: 2.271, AUC: 0.782, AP: 0.805, Completeness: 0.334, Homogeneity: 0.331, NMI: 0.333\n","Epoch: 017, Loss: 2.263, AUC: 0.784, AP: 0.805, Completeness: 0.350, Homogeneity: 0.349, NMI: 0.350\n","Epoch: 018, Loss: 2.332, AUC: 0.787, AP: 0.806, Completeness: 0.354, Homogeneity: 0.342, NMI: 0.348\n","Epoch: 019, Loss: 2.442, AUC: 0.792, AP: 0.812, Completeness: 0.358, Homogeneity: 0.345, NMI: 0.352\n","Epoch: 020, Loss: 2.569, AUC: 0.798, AP: 0.817, Completeness: 0.363, Homogeneity: 0.348, NMI: 0.356\n","Epoch: 021, Loss: 2.681, AUC: 0.801, AP: 0.820, Completeness: 0.368, Homogeneity: 0.359, NMI: 0.364\n","Epoch: 022, Loss: 2.809, AUC: 0.805, AP: 0.821, Completeness: 0.378, Homogeneity: 0.373, NMI: 0.375\n","Epoch: 023, Loss: 2.945, AUC: 0.810, AP: 0.825, Completeness: 0.385, Homogeneity: 0.379, NMI: 0.382\n","Epoch: 024, Loss: 3.061, AUC: 0.819, AP: 0.832, Completeness: 0.395, Homogeneity: 0.387, NMI: 0.391\n","Epoch: 025, Loss: 3.147, AUC: 0.825, AP: 0.837, Completeness: 0.403, Homogeneity: 0.396, NMI: 0.400\n","Epoch: 026, Loss: 3.210, AUC: 0.824, AP: 0.834, Completeness: 0.409, Homogeneity: 0.403, NMI: 0.406\n","Epoch: 027, Loss: 3.235, AUC: 0.818, AP: 0.826, Completeness: 0.411, Homogeneity: 0.404, NMI: 0.407\n","Epoch: 028, Loss: 3.381, AUC: 0.812, AP: 0.818, Completeness: 0.414, Homogeneity: 0.406, NMI: 0.410\n","Epoch: 029, Loss: 3.456, AUC: 0.809, AP: 0.816, Completeness: 0.415, Homogeneity: 0.406, NMI: 0.410\n","Epoch: 030, Loss: 3.562, AUC: 0.808, AP: 0.816, Completeness: 0.419, Homogeneity: 0.408, NMI: 0.413\n","Epoch: 031, Loss: 3.657, AUC: 0.807, AP: 0.816, Completeness: 0.419, Homogeneity: 0.408, NMI: 0.414\n","Epoch: 032, Loss: 3.772, AUC: 0.807, AP: 0.817, Completeness: 0.415, Homogeneity: 0.402, NMI: 0.408\n","Epoch: 033, Loss: 3.849, AUC: 0.811, AP: 0.821, Completeness: 0.416, Homogeneity: 0.400, NMI: 0.408\n","Epoch: 034, Loss: 3.846, AUC: 0.818, AP: 0.828, Completeness: 0.410, Homogeneity: 0.391, NMI: 0.400\n","Epoch: 035, Loss: 3.858, AUC: 0.825, AP: 0.836, Completeness: 0.403, Homogeneity: 0.380, NMI: 0.391\n","Epoch: 036, Loss: 3.834, AUC: 0.829, AP: 0.843, Completeness: 0.402, Homogeneity: 0.378, NMI: 0.390\n","Epoch: 037, Loss: 3.891, AUC: 0.826, AP: 0.842, Completeness: 0.388, Homogeneity: 0.363, NMI: 0.375\n","Epoch: 038, Loss: 4.031, AUC: 0.817, AP: 0.836, Completeness: 0.407, Homogeneity: 0.385, NMI: 0.396\n","Epoch: 039, Loss: 4.126, AUC: 0.807, AP: 0.828, Completeness: 0.385, Homogeneity: 0.366, NMI: 0.376\n","Epoch: 040, Loss: 4.100, AUC: 0.801, AP: 0.822, Completeness: 0.378, Homogeneity: 0.359, NMI: 0.369\n","Epoch: 041, Loss: 4.217, AUC: 0.797, AP: 0.819, Completeness: 0.374, Homogeneity: 0.356, NMI: 0.365\n","Epoch: 042, Loss: 4.326, AUC: 0.797, AP: 0.818, Completeness: 0.369, Homogeneity: 0.352, NMI: 0.360\n","Epoch: 043, Loss: 4.416, AUC: 0.801, AP: 0.821, Completeness: 0.366, Homogeneity: 0.350, NMI: 0.358\n","Epoch: 044, Loss: 4.274, AUC: 0.808, AP: 0.825, Completeness: 0.360, Homogeneity: 0.340, NMI: 0.350\n","Epoch: 045, Loss: 4.383, AUC: 0.816, AP: 0.829, Completeness: 0.364, Homogeneity: 0.345, NMI: 0.354\n","Epoch: 046, Loss: 4.300, AUC: 0.824, AP: 0.834, Completeness: 0.364, Homogeneity: 0.346, NMI: 0.355\n","Epoch: 047, Loss: 4.205, AUC: 0.831, AP: 0.838, Completeness: 0.371, Homogeneity: 0.354, NMI: 0.363\n","Epoch: 048, Loss: 4.248, AUC: 0.835, AP: 0.841, Completeness: 0.378, Homogeneity: 0.362, NMI: 0.370\n","Epoch: 049, Loss: 4.226, AUC: 0.838, AP: 0.842, Completeness: 0.389, Homogeneity: 0.372, NMI: 0.380\n","Epoch: 050, Loss: 4.198, AUC: 0.840, AP: 0.844, Completeness: 0.392, Homogeneity: 0.377, NMI: 0.384\n","Epoch: 051, Loss: 4.122, AUC: 0.843, AP: 0.846, Completeness: 0.395, Homogeneity: 0.380, NMI: 0.387\n","Epoch: 052, Loss: 4.023, AUC: 0.847, AP: 0.849, Completeness: 0.395, Homogeneity: 0.382, NMI: 0.389\n","Epoch: 053, Loss: 4.027, AUC: 0.851, AP: 0.852, Completeness: 0.396, Homogeneity: 0.384, NMI: 0.390\n","Epoch: 054, Loss: 3.963, AUC: 0.855, AP: 0.856, Completeness: 0.406, Homogeneity: 0.394, NMI: 0.400\n","Epoch: 055, Loss: 3.963, AUC: 0.860, AP: 0.860, Completeness: 0.413, Homogeneity: 0.404, NMI: 0.408\n","Epoch: 056, Loss: 3.907, AUC: 0.864, AP: 0.863, Completeness: 0.417, Homogeneity: 0.408, NMI: 0.412\n","Epoch: 057, Loss: 3.892, AUC: 0.866, AP: 0.865, Completeness: 0.424, Homogeneity: 0.415, NMI: 0.419\n","Epoch: 058, Loss: 3.792, AUC: 0.868, AP: 0.867, Completeness: 0.429, Homogeneity: 0.422, NMI: 0.425\n","Epoch: 059, Loss: 3.851, AUC: 0.869, AP: 0.868, Completeness: 0.427, Homogeneity: 0.420, NMI: 0.423\n","Epoch: 060, Loss: 3.796, AUC: 0.870, AP: 0.868, Completeness: 0.427, Homogeneity: 0.422, NMI: 0.425\n","Epoch: 061, Loss: 3.768, AUC: 0.871, AP: 0.869, Completeness: 0.446, Homogeneity: 0.412, NMI: 0.428\n","Epoch: 062, Loss: 3.798, AUC: 0.871, AP: 0.869, Completeness: 0.431, Homogeneity: 0.427, NMI: 0.429\n","Epoch: 063, Loss: 3.877, AUC: 0.872, AP: 0.869, Completeness: 0.434, Homogeneity: 0.431, NMI: 0.433\n","Epoch: 064, Loss: 3.879, AUC: 0.872, AP: 0.868, Completeness: 0.428, Homogeneity: 0.426, NMI: 0.427\n","Epoch: 065, Loss: 3.909, AUC: 0.872, AP: 0.868, Completeness: 0.433, Homogeneity: 0.432, NMI: 0.432\n","Epoch: 066, Loss: 3.949, AUC: 0.872, AP: 0.868, Completeness: 0.432, Homogeneity: 0.432, NMI: 0.432\n","Epoch: 067, Loss: 4.031, AUC: 0.872, AP: 0.868, Completeness: 0.442, Homogeneity: 0.414, NMI: 0.427\n","Epoch: 068, Loss: 4.043, AUC: 0.872, AP: 0.868, Completeness: 0.440, Homogeneity: 0.414, NMI: 0.426\n","Epoch: 069, Loss: 4.097, AUC: 0.873, AP: 0.868, Completeness: 0.436, Homogeneity: 0.411, NMI: 0.423\n","Epoch: 070, Loss: 4.129, AUC: 0.873, AP: 0.868, Completeness: 0.434, Homogeneity: 0.410, NMI: 0.422\n","Epoch: 071, Loss: 4.108, AUC: 0.874, AP: 0.869, Completeness: 0.433, Homogeneity: 0.409, NMI: 0.421\n","Epoch: 072, Loss: 4.124, AUC: 0.876, AP: 0.870, Completeness: 0.423, Homogeneity: 0.407, NMI: 0.415\n","Epoch: 073, Loss: 4.134, AUC: 0.877, AP: 0.871, Completeness: 0.421, Homogeneity: 0.407, NMI: 0.414\n","Epoch: 074, Loss: 4.144, AUC: 0.878, AP: 0.872, Completeness: 0.421, Homogeneity: 0.410, NMI: 0.416\n","Epoch: 075, Loss: 4.064, AUC: 0.879, AP: 0.874, Completeness: 0.429, Homogeneity: 0.431, NMI: 0.430\n","Epoch: 076, Loss: 4.031, AUC: 0.880, AP: 0.875, Completeness: 0.422, Homogeneity: 0.417, NMI: 0.420\n","Epoch: 077, Loss: 4.052, AUC: 0.882, AP: 0.877, Completeness: 0.433, Homogeneity: 0.436, NMI: 0.435\n","Epoch: 078, Loss: 4.135, AUC: 0.883, AP: 0.878, Completeness: 0.433, Homogeneity: 0.437, NMI: 0.435\n","Epoch: 079, Loss: 4.068, AUC: 0.884, AP: 0.880, Completeness: 0.433, Homogeneity: 0.440, NMI: 0.436\n","Epoch: 080, Loss: 4.095, AUC: 0.885, AP: 0.882, Completeness: 0.433, Homogeneity: 0.440, NMI: 0.436\n","Epoch: 081, Loss: 4.070, AUC: 0.886, AP: 0.884, Completeness: 0.435, Homogeneity: 0.442, NMI: 0.438\n","Epoch: 082, Loss: 3.953, AUC: 0.887, AP: 0.887, Completeness: 0.434, Homogeneity: 0.441, NMI: 0.437\n","Epoch: 083, Loss: 4.085, AUC: 0.888, AP: 0.888, Completeness: 0.434, Homogeneity: 0.442, NMI: 0.438\n","Epoch: 084, Loss: 3.945, AUC: 0.889, AP: 0.889, Completeness: 0.436, Homogeneity: 0.444, NMI: 0.440\n","Epoch: 085, Loss: 4.025, AUC: 0.890, AP: 0.890, Completeness: 0.446, Homogeneity: 0.455, NMI: 0.450\n","Epoch: 086, Loss: 3.932, AUC: 0.890, AP: 0.891, Completeness: 0.445, Homogeneity: 0.454, NMI: 0.449\n","Epoch: 087, Loss: 4.016, AUC: 0.891, AP: 0.891, Completeness: 0.445, Homogeneity: 0.455, NMI: 0.450\n","Epoch: 088, Loss: 3.984, AUC: 0.891, AP: 0.892, Completeness: 0.448, Homogeneity: 0.457, NMI: 0.452\n","Epoch: 089, Loss: 3.946, AUC: 0.892, AP: 0.892, Completeness: 0.451, Homogeneity: 0.459, NMI: 0.455\n","Epoch: 090, Loss: 3.971, AUC: 0.893, AP: 0.893, Completeness: 0.451, Homogeneity: 0.459, NMI: 0.455\n","Epoch: 091, Loss: 3.950, AUC: 0.893, AP: 0.893, Completeness: 0.452, Homogeneity: 0.459, NMI: 0.455\n","Epoch: 092, Loss: 3.911, AUC: 0.894, AP: 0.893, Completeness: 0.445, Homogeneity: 0.450, NMI: 0.447\n","Epoch: 093, Loss: 3.967, AUC: 0.895, AP: 0.893, Completeness: 0.443, Homogeneity: 0.448, NMI: 0.445\n","Epoch: 094, Loss: 3.905, AUC: 0.895, AP: 0.893, Completeness: 0.442, Homogeneity: 0.447, NMI: 0.445\n","Epoch: 095, Loss: 3.838, AUC: 0.896, AP: 0.894, Completeness: 0.440, Homogeneity: 0.444, NMI: 0.442\n","Epoch: 096, Loss: 3.884, AUC: 0.897, AP: 0.894, Completeness: 0.437, Homogeneity: 0.443, NMI: 0.440\n","Epoch: 097, Loss: 3.845, AUC: 0.897, AP: 0.894, Completeness: 0.438, Homogeneity: 0.444, NMI: 0.441\n","Epoch: 098, Loss: 3.856, AUC: 0.898, AP: 0.894, Completeness: 0.436, Homogeneity: 0.443, NMI: 0.439\n","Epoch: 099, Loss: 3.876, AUC: 0.899, AP: 0.894, Completeness: 0.435, Homogeneity: 0.443, NMI: 0.439\n","Epoch: 100, Loss: 3.768, AUC: 0.899, AP: 0.894, Completeness: 0.435, Homogeneity: 0.444, NMI: 0.439\n"]}]},{"cell_type":"code","source":["@torch.no_grad()\n","def plot_points(colors):\n","    model.eval()\n","    z = model.encode(data.x, data.train_pos_edge_index)\n","    z = TSNE(n_components=2).fit_transform(z.cpu().numpy())\n","    y = data.y.cpu().numpy()\n","\n","    fig = plt.figure(1, figsize=(8, 8))\n","    fig.clf()\n","    for i in range(dataset.num_classes):\n","        plt.scatter(z[y == i, 0], z[y == i, 1], s=20, color=colors[i])\n","    plt.axis('off')\n","    plt.show()"],"metadata":{"id":"U7bnIZiHP320","executionInfo":{"status":"ok","timestamp":1639759013934,"user_tz":-300,"elapsed":4,"user":{"displayName":"Muhammad Ali","photoUrl":"https://lh3.googleusercontent.com/a/default-user=s64","userId":"08853600868018567439"}}},"execution_count":23,"outputs":[]},{"cell_type":"code","source":["#%%\n","colors = [\n","    '#ffc0cb', '#bada55', '#008080', '#420420', '#7fe5f0', '#065535', '#ffd700'\n","]\n","plot_points(colors)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":555},"id":"B3E7rZmMQPjb","executionInfo":{"status":"ok","timestamp":1639759052658,"user_tz":-300,"elapsed":32618,"user":{"displayName":"Muhammad Ali","photoUrl":"https://lh3.googleusercontent.com/a/default-user=s64","userId":"08853600868018567439"}},"outputId":"589f5dde-331f-4797-bb8e-b556d515e7b6"},"execution_count":24,"outputs":[{"output_type":"stream","name":"stderr","text":["/usr/local/lib/python3.7/dist-packages/sklearn/manifold/_t_sne.py:783: FutureWarning: The default initialization in TSNE will change from 'random' to 'pca' in 1.2.\n","  FutureWarning,\n","/usr/local/lib/python3.7/dist-packages/sklearn/manifold/_t_sne.py:793: FutureWarning: The default learning rate in TSNE will change from 200.0 to 'auto' in 1.2.\n","  FutureWarning,\n"]},{"output_type":"display_data","data":{"image/png":"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\n","text/plain":["<Figure size 576x576 with 1 Axes>"]},"metadata":{"needs_background":"light"}}]},{"cell_type":"code","source":[""],"metadata":{"id":"pWKtskRBQP3O"},"execution_count":null,"outputs":[]}]}