Performance_prediction .ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "authorship_tag": "ABX9TyMUHijncds5mATxfyedgqRo",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/udayameister/sc-DGNN/blob/main/Performance_prediction%20.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Kv1IR8kVxcLm"
      },
      "outputs": [],
      "source": [
        "import torch\n",
        "import torch.nn as nn\n",
        "import torch.optim as optim\n",
        "from torch_geometric.data import DataLoader\n",
        "from torch_geometric.nn import GCNConv, GATConv, EdgeConv\n",
        "from sklearn.metrics import accuracy_score, f1_score, roc_auc_score\n",
        "\n",
        "class GraphNeuralNetwork(nn.Module):\n",
        "    def __init__(self, input_dim, output_dim):\n",
        "        super(GraphNeuralNetwork, self).__init__()\n",
        "\n",
        "        self.edge_processing = nn.Linear(input_dim, input_dim)\n",
        "        self.node_processing = nn.Linear(input_dim, input_dim)\n",
        "        self.enhanced_graph_processing = nn.Linear(input_dim, input_dim)\n",
        "        self.pooling_strategy = torch_geometric.nn.MeanPool()\n",
        "        self.readout = nn.Linear(input_dim, output_dim)\n",
        "\n",
        "    def forward(self, graph_features):\n",
        "        # Edge processing\n",
        "        edge_features = self.edge_processing(graph_features)\n",
        "\n",
        "        # Node processing\n",
        "        node_features = self.node_processing(graph_features)\n",
        "\n",
        "        # Enhanced graph processing\n",
        "        graph_features = self.enhanced_graph_processing(graph_features)\n",
        "\n",
        "        # Pooling strategy\n",
        "        graph_features = self.pooling_strategy(graph_features)\n",
        "\n",
        "        # Readout\n",
        "        output = self.readout(graph_features)\n",
        "\n",
        "        return output\n",
        "\n",
        "# Other models\n",
        "\n",
        "class DeepNeuralNetwork(nn.Module):\n",
        "    def __init__(self, input_dim, output_dim):\n",
        "        super(DeepNeuralNetwork, self).__init__()\n",
        "\n",
        "        self.linear1 = nn.Linear(input_dim, 64)\n",
        "        self.linear2 = nn.Linear(64, 128)\n",
        "        self.linear3 = nn.Linear(128, output_dim)\n",
        "\n",
        "    def forward(self, graph_features):\n",
        "        output = self.linear1(graph_features)\n",
        "        output = nn.functional.relu(output)\n",
        "        output = self.linear2(output)\n",
        "        output = nn.functional.relu(output)\n",
        "        output = self.linear3(output)\n",
        "\n",
        "        return output\n",
        "\n",
        "class GraphConvolutionalNeuralNetwork(nn.Module):\n",
        "    def __init__(self, input_dim, output_dim):\n",
        "        super(GraphConvolutionalNeuralNetwork, self).__init__()\n",
        "\n",
        "        self.conv1 = GCNConv(input_dim, 128)\n",
        "        self.conv2 = GCNConv(128, output_dim)\n",
        "\n",
        "    def forward(self, graph):\n",
        "        output = self.conv1(graph.x, graph.edge_index)\n",
        "        output = nn.functional.relu(output)\n",
        "        output = self.conv2(output, graph.edge_index)\n",
        "\n",
        "        return output\n",
        "\n",
        "class GraphNeuralNetworkModel(nn.Module):\n",
        "    def __init__(self, input_dim, output_dim):\n",
        "        super(GraphNeuralNetworkModel, self).__init__()\n",
        "\n",
        "        self.gnn = GraphNeuralNetwork(input_dim, output_dim)\n",
        "\n",
        "    def forward(self, graph):\n",
        "        output = self.gnn(graph.x)\n",
        "\n",
        "        return output\n",
        "\n",
        "class GraphConvolutionalNetwork(nn.Module):\n",
        "    def __init__(self, input_dim, output_dim):\n",
        "        super(GraphConvolutionalNetwork, self).__init__()\n",
        "\n",
        "        self.conv1 = nn.Conv1d(input_dim, 64, 3)\n",
        "        self.conv2 = nn.Conv1d(64, output_dim, 3)\n",
        "\n",
        "    def forward(self, graph_features):\n",
        "        output = self.conv1(graph_features)\n",
        "        output = nn.functional.relu(output)\n",
        "        output = self.conv2(output)\n",
        "\n",
        "        return output\n",
        "\n",
        "class GraphAttentionNetwork(nn.Module):\n",
        "    def __init__(self, input_dim, output_dim):\n",
        "        super(GraphAttentionNetwork, self).__init__()\n",
        "\n",
        "        self.gat1 = GATConv(input_dim, 64)"
      ]
    }
  ]
}