Elliptic_dataset_GCN.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Elliptic_dataset_GCN.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "authorship_tag": "ABX9TyNsGRIrHK4wb7FxrMp2vHe7",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/JungWoo-Chae/GCN_Elliptic_dataset/blob/main/Elliptic_dataset_GCN.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5XF8TSMqilz2"
      },
      "source": [
        "# Bitcoin Fraud Detection System with GCN"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Ksc_Ogwxwwl2"
      },
      "source": [
        "## Pytorch Geometric Environment Setting"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LCidzPVYwxN9",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "51b2e2f5-88e4-44a1-e378-b024f038d8fb"
      },
      "source": [
        "# Install required packages.\n",
        "!pip install -q torch-scatter==latest+cu101 -f https://pytorch-geometric.com/whl/torch-1.7.0.html\n",
        "!pip install -q torch-sparse==latest+cu101 -f https://pytorch-geometric.com/whl/torch-1.7.0.html\n",
        "!pip install -q git+https://github.com/rusty1s/pytorch_geometric.git"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "\u001b[K     |████████████████████████████████| 11.9MB 45.7MB/s \n",
            "\u001b[K     |████████████████████████████████| 24.3MB 1.2MB/s \n",
            "\u001b[K     |████████████████████████████████| 235kB 14.2MB/s \n",
            "\u001b[K     |████████████████████████████████| 2.2MB 42.0MB/s \n",
            "\u001b[K     |████████████████████████████████| 51kB 9.6MB/s \n",
            "\u001b[?25h  Building wheel for torch-geometric (setup.py) ... \u001b[?25l\u001b[?25hdone\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "yvMENKuLwVvB"
      },
      "source": [
        "## Library Import"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8fFna02HwQxB"
      },
      "source": [
        "import numpy as np\n",
        "import networkx as nx\n",
        "import os\n",
        "import pandas as pd\n",
        "from sklearn.metrics import f1_score, precision_score, recall_score\n",
        "from sklearn.manifold import TSNE\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "import torch\n",
        "import torch.nn as nn\n",
        "import torch.nn.functional as F\n",
        "from torch.nn import Embedding\n",
        "from torch.nn import Parameter\n",
        "from torch_geometric.data import Data,DataLoader\n",
        "from torch_geometric.nn import GCNConv\n",
        "from torch_geometric.utils.convert import to_networkx\n",
        "from torch_geometric.utils import to_undirected"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BLSPDBtLy-aT"
      },
      "source": [
        "device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "sdBcVJbxjAsb"
      },
      "source": [
        "# **Please insert Kaggle username and kaggle key**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TZLcvgEfiinM",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "41c10030-cb40-4beb-a369-a31d4c49d0f9"
      },
      "source": [
        "os.environ['KAGGLE_USERNAME'] = \"@@@@@@@@@\" # username from the json file\n",
        "os.environ['KAGGLE_KEY'] = \"####################\" # key from the json file\n",
        "!kaggle datasets download -d ellipticco/elliptic-data-set\n",
        "!unzip elliptic-data-set.zip\n",
        "!mkdir elliptic_bitcoin_dataset_cont"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Downloading elliptic-data-set.zip to /content\n",
            " 94% 137M/146M [00:01<00:00, 96.8MB/s]\n",
            "100% 146M/146M [00:01<00:00, 104MB/s] \n",
            "Archive:  elliptic-data-set.zip\n",
            "  inflating: elliptic_bitcoin_dataset/elliptic_txs_classes.csv  \n",
            "  inflating: elliptic_bitcoin_dataset/elliptic_txs_edgelist.csv  \n",
            "  inflating: elliptic_bitcoin_dataset/elliptic_txs_features.csv  \n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "wfVZ-ihYit3K"
      },
      "source": [
        "## Data Preparation"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4q6mAv_0jNwD",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "a55d0b5a-66d4-47ec-f016-da45c788d746"
      },
      "source": [
        "# Load Dataframe\n",
        "df_edge = pd.read_csv('elliptic_bitcoin_dataset/elliptic_txs_edgelist.csv')\n",
        "df_class = pd.read_csv('elliptic_bitcoin_dataset/elliptic_txs_classes.csv')\n",
        "df_features = pd.read_csv('elliptic_bitcoin_dataset/elliptic_txs_features.csv',header=None)\n",
        "\n",
        "# Setting Column name\n",
        "df_features.columns = ['id', 'time step'] + [f'trans_feat_{i}' for i in range(93)] + [f'agg_feat_{i}' for i in range(72)]\n",
        "\n",
        "print('Number of edges: {}'.format(len(df_edge)))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Number of edges: 234355\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "kqkMUFOmmJ9y"
      },
      "source": [
        "## Get Node Index"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bWoUAV16mLHS",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "8a1a490f-0766-40e9-eb1a-bea999b215d8"
      },
      "source": [
        "all_nodes = list(set(df_edge['txId1']).union(set(df_edge['txId2'])).union(set(df_class['txId'])).union(set(df_features['id'])))\n",
        "nodes_df = pd.DataFrame(all_nodes,columns=['id']).reset_index()\n",
        "\n",
        "print('Number of nodes: {}'.format(len(nodes_df)))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Number of nodes: 203769\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "03RoQLdPorg9"
      },
      "source": [
        "## Fix id index"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4dot3YJdotoU",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        },
        "outputId": "64cb69b5-2f69-45ae-ac09-e5c99352cf42"
      },
      "source": [
        "df_edge = df_edge.join(nodes_df.rename(columns={'id':'txId1'}).set_index('txId1'),on='txId1',how='inner') \\\n",
        "       .join(nodes_df.rename(columns={'id':'txId2'}).set_index('txId2'),on='txId2',how='inner',rsuffix='2') \\\n",
        "       .drop(columns=['txId1','txId2']) \\\n",
        "       .rename(columns={'index':'txId1','index2':'txId2'})\n",
        "df_edge.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>txId1</th>\n",
              "      <th>txId2</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>153796</td>\n",
              "      <td>56598</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>56486</td>\n",
              "      <td>137561</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3575</th>\n",
              "      <td>153260</td>\n",
              "      <td>137561</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1387</th>\n",
              "      <td>153676</td>\n",
              "      <td>137561</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6890</th>\n",
              "      <td>170223</td>\n",
              "      <td>137561</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "       txId1   txId2\n",
              "0     153796   56598\n",
              "1      56486  137561\n",
              "3575  153260  137561\n",
              "1387  153676  137561\n",
              "6890  170223  137561"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "HGivrPVeo5Uy",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        },
        "outputId": "de565cc6-a3ad-4113-e94b-d9b7a26af663"
      },
      "source": [
        "df_class = df_class.join(nodes_df.rename(columns={'id':'txId'}).set_index('txId'),on='txId',how='inner') \\\n",
        "        .drop(columns=['txId']).rename(columns={'index':'txId'})[['txId','class']]\n",
        "df_class.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>txId</th>\n",
              "      <th>class</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>153796</td>\n",
              "      <td>unknown</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>56598</td>\n",
              "      <td>unknown</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>56486</td>\n",
              "      <td>unknown</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>137561</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>161325</td>\n",
              "      <td>unknown</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "     txId    class\n",
              "0  153796  unknown\n",
              "1   56598  unknown\n",
              "2   56486  unknown\n",
              "3  137561        2\n",
              "4  161325  unknown"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lpAjxw4VqprC",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 270
        },
        "outputId": "45579c58-d6b6-4330-94a4-ad8dcadcce39"
      },
      "source": [
        "df_features = df_features.join(nodes_df.set_index('id'),on='id',how='inner') \\\n",
        "        .drop(columns=['id']).rename(columns={'index':'id'})\n",
        "df_features = df_features [ ['id']+list(df_features.drop(columns=['id']).columns) ]\n",
        "df_features.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>id</th>\n",
              "      <th>time step</th>\n",
              "      <th>trans_feat_0</th>\n",
              "      <th>trans_feat_1</th>\n",
              "      <th>trans_feat_2</th>\n",
              "      <th>trans_feat_3</th>\n",
              "      <th>trans_feat_4</th>\n",
              "      <th>trans_feat_5</th>\n",
              "      <th>trans_feat_6</th>\n",
              "      <th>trans_feat_7</th>\n",
              "      <th>trans_feat_8</th>\n",
              "      <th>trans_feat_9</th>\n",
              "      <th>trans_feat_10</th>\n",
              "      <th>trans_feat_11</th>\n",
              "      <th>trans_feat_12</th>\n",
              "      <th>trans_feat_13</th>\n",
              "      <th>trans_feat_14</th>\n",
              "      <th>trans_feat_15</th>\n",
              "      <th>trans_feat_16</th>\n",
              "      <th>trans_feat_17</th>\n",
              "      <th>trans_feat_18</th>\n",
              "      <th>trans_feat_19</th>\n",
              "      <th>trans_feat_20</th>\n",
              "      <th>trans_feat_21</th>\n",
              "      <th>trans_feat_22</th>\n",
              "      <th>trans_feat_23</th>\n",
              "      <th>trans_feat_24</th>\n",
              "      <th>trans_feat_25</th>\n",
              "      <th>trans_feat_26</th>\n",
              "      <th>trans_feat_27</th>\n",
              "      <th>trans_feat_28</th>\n",
              "      <th>trans_feat_29</th>\n",
              "      <th>trans_feat_30</th>\n",
              "      <th>trans_feat_31</th>\n",
              "      <th>trans_feat_32</th>\n",
              "      <th>trans_feat_33</th>\n",
              "      <th>trans_feat_34</th>\n",
              "      <th>trans_feat_35</th>\n",
              "      <th>trans_feat_36</th>\n",
              "      <th>trans_feat_37</th>\n",
              "      <th>...</th>\n",
              "      <th>agg_feat_32</th>\n",
              "      <th>agg_feat_33</th>\n",
              "      <th>agg_feat_34</th>\n",
              "      <th>agg_feat_35</th>\n",
              "      <th>agg_feat_36</th>\n",
              "      <th>agg_feat_37</th>\n",
              "      <th>agg_feat_38</th>\n",
              "      <th>agg_feat_39</th>\n",
              "      <th>agg_feat_40</th>\n",
              "      <th>agg_feat_41</th>\n",
              "      <th>agg_feat_42</th>\n",
              "      <th>agg_feat_43</th>\n",
              "      <th>agg_feat_44</th>\n",
              "      <th>agg_feat_45</th>\n",
              "      <th>agg_feat_46</th>\n",
              "      <th>agg_feat_47</th>\n",
              "      <th>agg_feat_48</th>\n",
              "      <th>agg_feat_49</th>\n",
              "      <th>agg_feat_50</th>\n",
              "      <th>agg_feat_51</th>\n",
              "      <th>agg_feat_52</th>\n",
              "      <th>agg_feat_53</th>\n",
              "      <th>agg_feat_54</th>\n",
              "      <th>agg_feat_55</th>\n",
              "      <th>agg_feat_56</th>\n",
              "      <th>agg_feat_57</th>\n",
              "      <th>agg_feat_58</th>\n",
              "      <th>agg_feat_59</th>\n",
              "      <th>agg_feat_60</th>\n",
              "      <th>agg_feat_61</th>\n",
              "      <th>agg_feat_62</th>\n",
              "      <th>agg_feat_63</th>\n",
              "      <th>agg_feat_64</th>\n",
              "      <th>agg_feat_65</th>\n",
              "      <th>agg_feat_66</th>\n",
              "      <th>agg_feat_67</th>\n",
              "      <th>agg_feat_68</th>\n",
              "      <th>agg_feat_69</th>\n",
              "      <th>agg_feat_70</th>\n",
              "      <th>agg_feat_71</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>153796</td>\n",
              "      <td>1</td>\n",
              "      <td>-0.171469</td>\n",
              "      <td>-0.184668</td>\n",
              "      <td>-1.201369</td>\n",
              "      <td>-0.121970</td>\n",
              "      <td>-0.043875</td>\n",
              "      <td>-0.113002</td>\n",
              "      <td>-0.061584</td>\n",
              "      <td>-0.162097</td>\n",
              "      <td>-0.167933</td>\n",
              "      <td>-0.049707</td>\n",
              "      <td>-0.164402</td>\n",
              "      <td>-0.028741</td>\n",
              "      <td>-0.035391</td>\n",
              "      <td>-0.042955</td>\n",
              "      <td>-0.013282</td>\n",
              "      <td>-0.057195</td>\n",
              "      <td>-0.169609</td>\n",
              "      <td>-0.171154</td>\n",
              "      <td>-0.174473</td>\n",
              "      <td>-1.373657</td>\n",
              "      <td>-1.371460</td>\n",
              "      <td>-0.139731</td>\n",
              "      <td>-0.148912</td>\n",
              "      <td>-0.080147</td>\n",
              "      <td>-0.155661</td>\n",
              "      <td>-0.010763</td>\n",
              "      <td>-0.012107</td>\n",
              "      <td>-0.139733</td>\n",
              "      <td>-0.148907</td>\n",
              "      <td>-0.080147</td>\n",
              "      <td>-0.155661</td>\n",
              "      <td>-0.010669</td>\n",
              "      <td>-0.012005</td>\n",
              "      <td>-0.024669</td>\n",
              "      <td>-0.031272</td>\n",
              "      <td>-0.023045</td>\n",
              "      <td>-0.026215</td>\n",
              "      <td>0.001428</td>\n",
              "      <td>...</td>\n",
              "      <td>-0.097719</td>\n",
              "      <td>-0.127462</td>\n",
              "      <td>0.003143</td>\n",
              "      <td>0.002426</td>\n",
              "      <td>-0.120950</td>\n",
              "      <td>-0.199145</td>\n",
              "      <td>-0.187993</td>\n",
              "      <td>-0.212948</td>\n",
              "      <td>1.064205</td>\n",
              "      <td>1.063787</td>\n",
              "      <td>-1.373782</td>\n",
              "      <td>-1.354735</td>\n",
              "      <td>-0.297975</td>\n",
              "      <td>-1.403698</td>\n",
              "      <td>1.342003</td>\n",
              "      <td>1.340733</td>\n",
              "      <td>-0.171601</td>\n",
              "      <td>-0.458162</td>\n",
              "      <td>-0.423588</td>\n",
              "      <td>-0.440883</td>\n",
              "      <td>-1.015963</td>\n",
              "      <td>-1.016230</td>\n",
              "      <td>-0.968903</td>\n",
              "      <td>-0.375715</td>\n",
              "      <td>0.759748</td>\n",
              "      <td>-0.768329</td>\n",
              "      <td>1.488113</td>\n",
              "      <td>1.487932</td>\n",
              "      <td>-0.216814</td>\n",
              "      <td>-0.605631</td>\n",
              "      <td>-0.562153</td>\n",
              "      <td>-0.600999</td>\n",
              "      <td>1.461330</td>\n",
              "      <td>1.461369</td>\n",
              "      <td>0.018279</td>\n",
              "      <td>-0.087490</td>\n",
              "      <td>-0.131155</td>\n",
              "      <td>-0.097524</td>\n",
              "      <td>-0.120613</td>\n",
              "      <td>-0.119792</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>56598</td>\n",
              "      <td>1</td>\n",
              "      <td>-0.171484</td>\n",
              "      <td>-0.184668</td>\n",
              "      <td>-1.201369</td>\n",
              "      <td>-0.121970</td>\n",
              "      <td>-0.043875</td>\n",
              "      <td>-0.113002</td>\n",
              "      <td>-0.061584</td>\n",
              "      <td>-0.162112</td>\n",
              "      <td>-0.167948</td>\n",
              "      <td>-0.049707</td>\n",
              "      <td>-0.164417</td>\n",
              "      <td>-0.028741</td>\n",
              "      <td>-0.035391</td>\n",
              "      <td>-0.042955</td>\n",
              "      <td>-0.013282</td>\n",
              "      <td>-0.055327</td>\n",
              "      <td>-0.169757</td>\n",
              "      <td>-0.171477</td>\n",
              "      <td>-0.174490</td>\n",
              "      <td>0.887058</td>\n",
              "      <td>0.884557</td>\n",
              "      <td>-0.139731</td>\n",
              "      <td>-0.148912</td>\n",
              "      <td>-0.080147</td>\n",
              "      <td>-0.155661</td>\n",
              "      <td>-0.010763</td>\n",
              "      <td>-0.012107</td>\n",
              "      <td>-0.139733</td>\n",
              "      <td>-0.148907</td>\n",
              "      <td>-0.080147</td>\n",
              "      <td>-0.155661</td>\n",
              "      <td>-0.010669</td>\n",
              "      <td>-0.012005</td>\n",
              "      <td>-0.024669</td>\n",
              "      <td>-0.031272</td>\n",
              "      <td>-0.023045</td>\n",
              "      <td>-0.026215</td>\n",
              "      <td>0.001428</td>\n",
              "      <td>...</td>\n",
              "      <td>-0.097719</td>\n",
              "      <td>-0.127462</td>\n",
              "      <td>0.003143</td>\n",
              "      <td>0.002426</td>\n",
              "      <td>-0.121330</td>\n",
              "      <td>-0.110933</td>\n",
              "      <td>-0.075909</td>\n",
              "      <td>-0.111641</td>\n",
              "      <td>-1.159649</td>\n",
              "      <td>-1.160129</td>\n",
              "      <td>-1.373723</td>\n",
              "      <td>-1.353918</td>\n",
              "      <td>-0.295982</td>\n",
              "      <td>-1.403215</td>\n",
              "      <td>-0.975738</td>\n",
              "      <td>-0.975237</td>\n",
              "      <td>-0.168742</td>\n",
              "      <td>-0.263290</td>\n",
              "      <td>-0.186389</td>\n",
              "      <td>-0.250875</td>\n",
              "      <td>-1.015963</td>\n",
              "      <td>-1.016230</td>\n",
              "      <td>-0.968903</td>\n",
              "      <td>0.146997</td>\n",
              "      <td>1.366287</td>\n",
              "      <td>-0.464773</td>\n",
              "      <td>-1.116918</td>\n",
              "      <td>-1.116948</td>\n",
              "      <td>-0.216814</td>\n",
              "      <td>0.634272</td>\n",
              "      <td>0.947382</td>\n",
              "      <td>0.673103</td>\n",
              "      <td>-0.979074</td>\n",
              "      <td>-0.978556</td>\n",
              "      <td>0.018279</td>\n",
              "      <td>-0.087490</td>\n",
              "      <td>-0.131155</td>\n",
              "      <td>-0.097524</td>\n",
              "      <td>-0.120613</td>\n",
              "      <td>-0.119792</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>56486</td>\n",
              "      <td>1</td>\n",
              "      <td>-0.172107</td>\n",
              "      <td>-0.184668</td>\n",
              "      <td>-1.201369</td>\n",
              "      <td>-0.121970</td>\n",
              "      <td>-0.043875</td>\n",
              "      <td>-0.113002</td>\n",
              "      <td>-0.061584</td>\n",
              "      <td>-0.162749</td>\n",
              "      <td>-0.168576</td>\n",
              "      <td>-0.049707</td>\n",
              "      <td>-0.165054</td>\n",
              "      <td>-0.028741</td>\n",
              "      <td>-0.035391</td>\n",
              "      <td>-0.042955</td>\n",
              "      <td>-0.013282</td>\n",
              "      <td>-0.055298</td>\n",
              "      <td>-0.170400</td>\n",
              "      <td>-0.172217</td>\n",
              "      <td>-0.175227</td>\n",
              "      <td>0.887058</td>\n",
              "      <td>0.884557</td>\n",
              "      <td>-0.139729</td>\n",
              "      <td>-0.148911</td>\n",
              "      <td>-0.080147</td>\n",
              "      <td>-0.155660</td>\n",
              "      <td>-0.010763</td>\n",
              "      <td>-0.012107</td>\n",
              "      <td>-0.139731</td>\n",
              "      <td>-0.148906</td>\n",
              "      <td>-0.080147</td>\n",
              "      <td>-0.155660</td>\n",
              "      <td>-0.010669</td>\n",
              "      <td>-0.012005</td>\n",
              "      <td>-0.024669</td>\n",
              "      <td>-0.031272</td>\n",
              "      <td>-0.023045</td>\n",
              "      <td>-0.026215</td>\n",
              "      <td>0.001428</td>\n",
              "      <td>...</td>\n",
              "      <td>-0.097719</td>\n",
              "      <td>-0.129496</td>\n",
              "      <td>0.003143</td>\n",
              "      <td>0.002426</td>\n",
              "      <td>-0.122974</td>\n",
              "      <td>-0.041556</td>\n",
              "      <td>0.012549</td>\n",
              "      <td>-0.032244</td>\n",
              "      <td>-1.159649</td>\n",
              "      <td>-1.160129</td>\n",
              "      <td>-1.373902</td>\n",
              "      <td>-1.356210</td>\n",
              "      <td>-0.301548</td>\n",
              "      <td>-1.404577</td>\n",
              "      <td>-0.975738</td>\n",
              "      <td>-0.975237</td>\n",
              "      <td>-0.168742</td>\n",
              "      <td>-0.192468</td>\n",
              "      <td>-0.099790</td>\n",
              "      <td>-0.182133</td>\n",
              "      <td>-1.015963</td>\n",
              "      <td>-1.016230</td>\n",
              "      <td>-0.968903</td>\n",
              "      <td>-1.421138</td>\n",
              "      <td>-0.453330</td>\n",
              "      <td>-1.375441</td>\n",
              "      <td>-1.116918</td>\n",
              "      <td>-1.116948</td>\n",
              "      <td>-0.216814</td>\n",
              "      <td>0.407161</td>\n",
              "      <td>0.670883</td>\n",
              "      <td>0.439728</td>\n",
              "      <td>-0.979074</td>\n",
              "      <td>-0.978556</td>\n",
              "      <td>-0.098889</td>\n",
              "      <td>-0.106715</td>\n",
              "      <td>-0.131155</td>\n",
              "      <td>-0.183671</td>\n",
              "      <td>-0.120613</td>\n",
              "      <td>-0.119792</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>137561</td>\n",
              "      <td>1</td>\n",
              "      <td>0.163054</td>\n",
              "      <td>1.963790</td>\n",
              "      <td>-0.646376</td>\n",
              "      <td>12.409294</td>\n",
              "      <td>-0.063725</td>\n",
              "      <td>9.782742</td>\n",
              "      <td>12.414558</td>\n",
              "      <td>-0.163645</td>\n",
              "      <td>-0.115831</td>\n",
              "      <td>0.043598</td>\n",
              "      <td>-0.163905</td>\n",
              "      <td>1.059993</td>\n",
              "      <td>2.397457</td>\n",
              "      <td>-0.063200</td>\n",
              "      <td>-0.013282</td>\n",
              "      <td>4.800240</td>\n",
              "      <td>0.174329</td>\n",
              "      <td>-0.172908</td>\n",
              "      <td>0.619399</td>\n",
              "      <td>-0.243300</td>\n",
              "      <td>-0.243452</td>\n",
              "      <td>-0.139732</td>\n",
              "      <td>-0.146895</td>\n",
              "      <td>-0.079667</td>\n",
              "      <td>-0.155554</td>\n",
              "      <td>0.773847</td>\n",
              "      <td>0.378516</td>\n",
              "      <td>-0.139734</td>\n",
              "      <td>-0.146891</td>\n",
              "      <td>-0.079667</td>\n",
              "      <td>-0.155553</td>\n",
              "      <td>0.773932</td>\n",
              "      <td>0.378593</td>\n",
              "      <td>-0.024669</td>\n",
              "      <td>-0.031271</td>\n",
              "      <td>-0.023044</td>\n",
              "      <td>-0.026215</td>\n",
              "      <td>0.101008</td>\n",
              "      <td>...</td>\n",
              "      <td>0.019771</td>\n",
              "      <td>-0.119195</td>\n",
              "      <td>0.147881</td>\n",
              "      <td>0.956170</td>\n",
              "      <td>0.759756</td>\n",
              "      <td>-0.041556</td>\n",
              "      <td>-0.188716</td>\n",
              "      <td>0.149949</td>\n",
              "      <td>-0.047722</td>\n",
              "      <td>-0.048171</td>\n",
              "      <td>-1.373723</td>\n",
              "      <td>-1.356131</td>\n",
              "      <td>-0.301754</td>\n",
              "      <td>-1.404445</td>\n",
              "      <td>0.183133</td>\n",
              "      <td>0.182748</td>\n",
              "      <td>-0.163577</td>\n",
              "      <td>-0.456561</td>\n",
              "      <td>-0.424672</td>\n",
              "      <td>-0.436915</td>\n",
              "      <td>0.143001</td>\n",
              "      <td>0.142525</td>\n",
              "      <td>-0.425896</td>\n",
              "      <td>-1.421138</td>\n",
              "      <td>-1.059868</td>\n",
              "      <td>-1.071885</td>\n",
              "      <td>0.185597</td>\n",
              "      <td>0.185492</td>\n",
              "      <td>-0.216814</td>\n",
              "      <td>-0.617907</td>\n",
              "      <td>-0.577099</td>\n",
              "      <td>-0.613614</td>\n",
              "      <td>0.241128</td>\n",
              "      <td>0.241406</td>\n",
              "      <td>1.072793</td>\n",
              "      <td>0.085530</td>\n",
              "      <td>-0.131155</td>\n",
              "      <td>0.677799</td>\n",
              "      <td>-0.120613</td>\n",
              "      <td>-0.119792</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>161325</td>\n",
              "      <td>1</td>\n",
              "      <td>1.011523</td>\n",
              "      <td>-0.081127</td>\n",
              "      <td>-1.201369</td>\n",
              "      <td>1.153668</td>\n",
              "      <td>0.333276</td>\n",
              "      <td>1.312656</td>\n",
              "      <td>-0.061584</td>\n",
              "      <td>-0.163523</td>\n",
              "      <td>0.041399</td>\n",
              "      <td>0.935886</td>\n",
              "      <td>-0.098695</td>\n",
              "      <td>1.203038</td>\n",
              "      <td>0.988902</td>\n",
              "      <td>0.341693</td>\n",
              "      <td>-0.013282</td>\n",
              "      <td>-0.057392</td>\n",
              "      <td>0.058329</td>\n",
              "      <td>-0.032930</td>\n",
              "      <td>-0.042709</td>\n",
              "      <td>0.065455</td>\n",
              "      <td>0.324947</td>\n",
              "      <td>-0.139732</td>\n",
              "      <td>-0.148854</td>\n",
              "      <td>-0.080105</td>\n",
              "      <td>-0.155638</td>\n",
              "      <td>1.313051</td>\n",
              "      <td>1.106965</td>\n",
              "      <td>-0.139734</td>\n",
              "      <td>-0.148850</td>\n",
              "      <td>-0.080104</td>\n",
              "      <td>-0.155638</td>\n",
              "      <td>1.313130</td>\n",
              "      <td>1.106995</td>\n",
              "      <td>-0.024669</td>\n",
              "      <td>-0.031272</td>\n",
              "      <td>-0.023045</td>\n",
              "      <td>-0.026215</td>\n",
              "      <td>0.001428</td>\n",
              "      <td>...</td>\n",
              "      <td>-0.077909</td>\n",
              "      <td>-0.093120</td>\n",
              "      <td>1.587690</td>\n",
              "      <td>1.586539</td>\n",
              "      <td>-0.101820</td>\n",
              "      <td>0.704558</td>\n",
              "      <td>0.448898</td>\n",
              "      <td>0.659662</td>\n",
              "      <td>0.311091</td>\n",
              "      <td>0.302198</td>\n",
              "      <td>-1.373962</td>\n",
              "      <td>-1.355393</td>\n",
              "      <td>-0.300941</td>\n",
              "      <td>-1.404583</td>\n",
              "      <td>0.064846</td>\n",
              "      <td>-0.035232</td>\n",
              "      <td>-0.168742</td>\n",
              "      <td>-0.442156</td>\n",
              "      <td>-0.416540</td>\n",
              "      <td>-0.418273</td>\n",
              "      <td>0.367605</td>\n",
              "      <td>0.392035</td>\n",
              "      <td>-0.968903</td>\n",
              "      <td>-1.943850</td>\n",
              "      <td>-1.059868</td>\n",
              "      <td>-1.678997</td>\n",
              "      <td>0.185597</td>\n",
              "      <td>0.185492</td>\n",
              "      <td>-0.169472</td>\n",
              "      <td>-0.470592</td>\n",
              "      <td>-0.511871</td>\n",
              "      <td>-0.400422</td>\n",
              "      <td>0.517257</td>\n",
              "      <td>0.579382</td>\n",
              "      <td>0.018279</td>\n",
              "      <td>0.277775</td>\n",
              "      <td>0.326394</td>\n",
              "      <td>1.293750</td>\n",
              "      <td>0.178136</td>\n",
              "      <td>0.179117</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5 rows × 167 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "       id  time step  trans_feat_0  ...  agg_feat_69  agg_feat_70  agg_feat_71\n",
              "0  153796          1     -0.171469  ...    -0.097524    -0.120613    -0.119792\n",
              "1   56598          1     -0.171484  ...    -0.097524    -0.120613    -0.119792\n",
              "2   56486          1     -0.172107  ...    -0.183671    -0.120613    -0.119792\n",
              "3  137561          1      0.163054  ...     0.677799    -0.120613    -0.119792\n",
              "4  161325          1      1.011523  ...     1.293750     0.178136     0.179117\n",
              "\n",
              "[5 rows x 167 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qFFg89HyraP6"
      },
      "source": [
        "df_edge_time = df_edge.join(df_features[['id','time step']].rename(columns={'id':'txId1'}).set_index('txId1'),on='txId1',how='left',rsuffix='1') \\\n",
        ".join(df_features[['id','time step']].rename(columns={'id':'txId2'}).set_index('txId2'),on='txId2',how='left',rsuffix='2')\n",
        "df_edge_time['is_time_same'] = df_edge_time['time step'] == df_edge_time['time step2']\n",
        "df_edge_time_fin = df_edge_time[['txId1','txId2','time step']].rename(columns={'txId1':'source','txId2':'target','time step':'time'})"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3hLkqgWQtRoj"
      },
      "source": [
        "## Create csv from Dataframe"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0oRwz171sdF7"
      },
      "source": [
        "df_features.drop(columns=['time step']).to_csv('elliptic_bitcoin_dataset_cont/elliptic_txs_features.csv',index=False,header=None)\n",
        "df_class.rename(columns={'txId':'nid','class':'label'})[['nid','label']].sort_values(by='nid').to_csv('elliptic_bitcoin_dataset_cont/elliptic_txs_classes.csv',index=False,header=None)\n",
        "df_features[['id','time step']].rename(columns={'id':'nid','time step':'time'})[['nid','time']].sort_values(by='nid').to_csv('elliptic_bitcoin_dataset_cont/elliptic_txs_nodetime.csv',index=False,header=None)\n",
        "df_edge_time_fin[['source','target','time']].to_csv('elliptic_bitcoin_dataset_cont/elliptic_txs_edgelist_timed.csv',index=False,header=None)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "glU24wM7uuO6"
      },
      "source": [
        "## Graph Preprocessing"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ebolR4V10TaZ",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        },
        "outputId": "e93ff300-adab-49e9-87ba-a869a60ca223"
      },
      "source": [
        "node_label = df_class.rename(columns={'txId':'nid','class':'label'})[['nid','label']].sort_values(by='nid').merge(df_features[['id','time step']].rename(columns={'id':'nid','time step':'time'}),on='nid',how='left')\n",
        "node_label['label'] =  node_label['label'].apply(lambda x: '3'  if x =='unknown' else x).astype(int)-1\n",
        "node_label.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>nid</th>\n",
              "      <th>label</th>\n",
              "      <th>time</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>17</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>13</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>10</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>27</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>4</td>\n",
              "      <td>2</td>\n",
              "      <td>31</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   nid  label  time\n",
              "0    0      1    17\n",
              "1    1      2    13\n",
              "2    2      1    10\n",
              "3    3      2    27\n",
              "4    4      2    31"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_f0dLVSrvjUD",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 253
        },
        "outputId": "787ab6e5-914d-48fd-ff2c-d3931d46b8e9"
      },
      "source": [
        "merged_nodes_df = node_label.merge(df_features.rename(columns={'id':'nid','time step':'time'}).drop(columns=['time']),on='nid',how='left')\n",
        "merged_nodes_df.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>nid</th>\n",
              "      <th>label</th>\n",
              "      <th>time</th>\n",
              "      <th>trans_feat_0</th>\n",
              "      <th>trans_feat_1</th>\n",
              "      <th>trans_feat_2</th>\n",
              "      <th>trans_feat_3</th>\n",
              "      <th>trans_feat_4</th>\n",
              "      <th>trans_feat_5</th>\n",
              "      <th>trans_feat_6</th>\n",
              "      <th>trans_feat_7</th>\n",
              "      <th>trans_feat_8</th>\n",
              "      <th>trans_feat_9</th>\n",
              "      <th>trans_feat_10</th>\n",
              "      <th>trans_feat_11</th>\n",
              "      <th>trans_feat_12</th>\n",
              "      <th>trans_feat_13</th>\n",
              "      <th>trans_feat_14</th>\n",
              "      <th>trans_feat_15</th>\n",
              "      <th>trans_feat_16</th>\n",
              "      <th>trans_feat_17</th>\n",
              "      <th>trans_feat_18</th>\n",
              "      <th>trans_feat_19</th>\n",
              "      <th>trans_feat_20</th>\n",
              "      <th>trans_feat_21</th>\n",
              "      <th>trans_feat_22</th>\n",
              "      <th>trans_feat_23</th>\n",
              "      <th>trans_feat_24</th>\n",
              "      <th>trans_feat_25</th>\n",
              "      <th>trans_feat_26</th>\n",
              "      <th>trans_feat_27</th>\n",
              "      <th>trans_feat_28</th>\n",
              "      <th>trans_feat_29</th>\n",
              "      <th>trans_feat_30</th>\n",
              "      <th>trans_feat_31</th>\n",
              "      <th>trans_feat_32</th>\n",
              "      <th>trans_feat_33</th>\n",
              "      <th>trans_feat_34</th>\n",
              "      <th>trans_feat_35</th>\n",
              "      <th>trans_feat_36</th>\n",
              "      <th>...</th>\n",
              "      <th>agg_feat_32</th>\n",
              "      <th>agg_feat_33</th>\n",
              "      <th>agg_feat_34</th>\n",
              "      <th>agg_feat_35</th>\n",
              "      <th>agg_feat_36</th>\n",
              "      <th>agg_feat_37</th>\n",
              "      <th>agg_feat_38</th>\n",
              "      <th>agg_feat_39</th>\n",
              "      <th>agg_feat_40</th>\n",
              "      <th>agg_feat_41</th>\n",
              "      <th>agg_feat_42</th>\n",
              "      <th>agg_feat_43</th>\n",
              "      <th>agg_feat_44</th>\n",
              "      <th>agg_feat_45</th>\n",
              "      <th>agg_feat_46</th>\n",
              "      <th>agg_feat_47</th>\n",
              "      <th>agg_feat_48</th>\n",
              "      <th>agg_feat_49</th>\n",
              "      <th>agg_feat_50</th>\n",
              "      <th>agg_feat_51</th>\n",
              "      <th>agg_feat_52</th>\n",
              "      <th>agg_feat_53</th>\n",
              "      <th>agg_feat_54</th>\n",
              "      <th>agg_feat_55</th>\n",
              "      <th>agg_feat_56</th>\n",
              "      <th>agg_feat_57</th>\n",
              "      <th>agg_feat_58</th>\n",
              "      <th>agg_feat_59</th>\n",
              "      <th>agg_feat_60</th>\n",
              "      <th>agg_feat_61</th>\n",
              "      <th>agg_feat_62</th>\n",
              "      <th>agg_feat_63</th>\n",
              "      <th>agg_feat_64</th>\n",
              "      <th>agg_feat_65</th>\n",
              "      <th>agg_feat_66</th>\n",
              "      <th>agg_feat_67</th>\n",
              "      <th>agg_feat_68</th>\n",
              "      <th>agg_feat_69</th>\n",
              "      <th>agg_feat_70</th>\n",
              "      <th>agg_feat_71</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>17</td>\n",
              "      <td>0.218933</td>\n",
              "      <td>0.048298</td>\n",
              "      <td>-1.201369</td>\n",
              "      <td>2.054118</td>\n",
              "      <td>-0.063725</td>\n",
              "      <td>1.480381</td>\n",
              "      <td>2.981378</td>\n",
              "      <td>-0.157193</td>\n",
              "      <td>-0.101173</td>\n",
              "      <td>0.127603</td>\n",
              "      <td>-0.152598</td>\n",
              "      <td>-0.579930</td>\n",
              "      <td>-0.234867</td>\n",
              "      <td>-0.063200</td>\n",
              "      <td>-0.013282</td>\n",
              "      <td>5.608007</td>\n",
              "      <td>0.231778</td>\n",
              "      <td>-0.172908</td>\n",
              "      <td>0.751708</td>\n",
              "      <td>-0.243300</td>\n",
              "      <td>-0.243452</td>\n",
              "      <td>-0.139728</td>\n",
              "      <td>-0.146095</td>\n",
              "      <td>-0.078754</td>\n",
              "      <td>-0.154979</td>\n",
              "      <td>0.108984</td>\n",
              "      <td>-0.106066</td>\n",
              "      <td>-0.139730</td>\n",
              "      <td>-0.146090</td>\n",
              "      <td>-0.078753</td>\n",
              "      <td>-0.154979</td>\n",
              "      <td>0.109077</td>\n",
              "      <td>-0.105958</td>\n",
              "      <td>-0.024669</td>\n",
              "      <td>-0.031272</td>\n",
              "      <td>-0.023045</td>\n",
              "      <td>-0.026215</td>\n",
              "      <td>...</td>\n",
              "      <td>-0.068457</td>\n",
              "      <td>-0.115668</td>\n",
              "      <td>-1.627558</td>\n",
              "      <td>-0.897054</td>\n",
              "      <td>0.906864</td>\n",
              "      <td>-0.015139</td>\n",
              "      <td>-0.188716</td>\n",
              "      <td>0.210674</td>\n",
              "      <td>-0.047722</td>\n",
              "      <td>-0.048171</td>\n",
              "      <td>-0.407359</td>\n",
              "      <td>-0.506459</td>\n",
              "      <td>-0.301754</td>\n",
              "      <td>-0.459161</td>\n",
              "      <td>0.183133</td>\n",
              "      <td>0.182748</td>\n",
              "      <td>-0.168742</td>\n",
              "      <td>-0.458162</td>\n",
              "      <td>-0.424672</td>\n",
              "      <td>-0.440022</td>\n",
              "      <td>0.143001</td>\n",
              "      <td>0.142525</td>\n",
              "      <td>-0.968903</td>\n",
              "      <td>-1.943850</td>\n",
              "      <td>-1.059868</td>\n",
              "      <td>-1.678997</td>\n",
              "      <td>0.185597</td>\n",
              "      <td>0.185492</td>\n",
              "      <td>-0.216814</td>\n",
              "      <td>-0.617907</td>\n",
              "      <td>-0.577099</td>\n",
              "      <td>-0.613614</td>\n",
              "      <td>0.241128</td>\n",
              "      <td>0.241406</td>\n",
              "      <td>0.135448</td>\n",
              "      <td>-0.068266</td>\n",
              "      <td>-0.131155</td>\n",
              "      <td>-0.011377</td>\n",
              "      <td>-0.120613</td>\n",
              "      <td>-0.119792</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>13</td>\n",
              "      <td>-0.167004</td>\n",
              "      <td>-0.178843</td>\n",
              "      <td>0.463609</td>\n",
              "      <td>-0.121970</td>\n",
              "      <td>-0.024025</td>\n",
              "      <td>-0.113002</td>\n",
              "      <td>-0.061584</td>\n",
              "      <td>-0.157528</td>\n",
              "      <td>-0.163428</td>\n",
              "      <td>-0.049707</td>\n",
              "      <td>-0.159838</td>\n",
              "      <td>-0.028741</td>\n",
              "      <td>-0.035391</td>\n",
              "      <td>-0.022711</td>\n",
              "      <td>-0.013282</td>\n",
              "      <td>-0.057378</td>\n",
              "      <td>-0.165010</td>\n",
              "      <td>-0.167148</td>\n",
              "      <td>-0.171545</td>\n",
              "      <td>-1.222435</td>\n",
              "      <td>-1.371460</td>\n",
              "      <td>-0.139729</td>\n",
              "      <td>-0.148911</td>\n",
              "      <td>-0.080147</td>\n",
              "      <td>-0.155660</td>\n",
              "      <td>-0.010763</td>\n",
              "      <td>-0.012107</td>\n",
              "      <td>-0.139731</td>\n",
              "      <td>-0.148906</td>\n",
              "      <td>-0.080147</td>\n",
              "      <td>-0.155660</td>\n",
              "      <td>-0.010669</td>\n",
              "      <td>-0.012005</td>\n",
              "      <td>-0.024669</td>\n",
              "      <td>-0.031272</td>\n",
              "      <td>-0.023045</td>\n",
              "      <td>-0.026215</td>\n",
              "      <td>...</td>\n",
              "      <td>-0.097719</td>\n",
              "      <td>-0.127462</td>\n",
              "      <td>0.003143</td>\n",
              "      <td>0.002426</td>\n",
              "      <td>-0.122609</td>\n",
              "      <td>-0.197599</td>\n",
              "      <td>-0.186370</td>\n",
              "      <td>-0.212179</td>\n",
              "      <td>-1.074554</td>\n",
              "      <td>-1.160129</td>\n",
              "      <td>-0.649055</td>\n",
              "      <td>-0.706879</td>\n",
              "      <td>-0.278174</td>\n",
              "      <td>-0.690193</td>\n",
              "      <td>0.405208</td>\n",
              "      <td>0.761740</td>\n",
              "      <td>-0.165007</td>\n",
              "      <td>-0.391245</td>\n",
              "      <td>-0.363145</td>\n",
              "      <td>-0.358710</td>\n",
              "      <td>1.214012</td>\n",
              "      <td>1.301281</td>\n",
              "      <td>1.203124</td>\n",
              "      <td>0.669709</td>\n",
              "      <td>-0.564631</td>\n",
              "      <td>1.154192</td>\n",
              "      <td>1.313609</td>\n",
              "      <td>1.313439</td>\n",
              "      <td>-0.216814</td>\n",
              "      <td>3.519195</td>\n",
              "      <td>3.447003</td>\n",
              "      <td>2.346677</td>\n",
              "      <td>0.289197</td>\n",
              "      <td>0.851388</td>\n",
              "      <td>-0.098889</td>\n",
              "      <td>-0.049041</td>\n",
              "      <td>-0.030745</td>\n",
              "      <td>-0.068808</td>\n",
              "      <td>-1.194450</td>\n",
              "      <td>-0.940388</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>10</td>\n",
              "      <td>-0.172886</td>\n",
              "      <td>-0.184668</td>\n",
              "      <td>-1.201369</td>\n",
              "      <td>0.178180</td>\n",
              "      <td>-0.043875</td>\n",
              "      <td>0.222447</td>\n",
              "      <td>-0.061584</td>\n",
              "      <td>-0.163645</td>\n",
              "      <td>-0.169369</td>\n",
              "      <td>-0.049186</td>\n",
              "      <td>-0.165930</td>\n",
              "      <td>0.886679</td>\n",
              "      <td>1.189967</td>\n",
              "      <td>-0.042955</td>\n",
              "      <td>-0.013282</td>\n",
              "      <td>-0.056917</td>\n",
              "      <td>-0.171086</td>\n",
              "      <td>-0.172873</td>\n",
              "      <td>-0.176150</td>\n",
              "      <td>0.887058</td>\n",
              "      <td>0.884557</td>\n",
              "      <td>-0.139732</td>\n",
              "      <td>-0.148905</td>\n",
              "      <td>-0.080141</td>\n",
              "      <td>-0.155660</td>\n",
              "      <td>-0.185016</td>\n",
              "      <td>-0.815353</td>\n",
              "      <td>-0.139734</td>\n",
              "      <td>-0.148900</td>\n",
              "      <td>-0.080140</td>\n",
              "      <td>-0.155660</td>\n",
              "      <td>-0.184919</td>\n",
              "      <td>-0.815199</td>\n",
              "      <td>-0.024669</td>\n",
              "      <td>-0.031272</td>\n",
              "      <td>-0.023045</td>\n",
              "      <td>-0.026215</td>\n",
              "      <td>...</td>\n",
              "      <td>0.587033</td>\n",
              "      <td>0.127938</td>\n",
              "      <td>2.738218</td>\n",
              "      <td>1.164709</td>\n",
              "      <td>-0.124476</td>\n",
              "      <td>-0.200151</td>\n",
              "      <td>-0.188466</td>\n",
              "      <td>-0.214831</td>\n",
              "      <td>1.064205</td>\n",
              "      <td>1.063787</td>\n",
              "      <td>-0.830446</td>\n",
              "      <td>-0.877983</td>\n",
              "      <td>-0.300517</td>\n",
              "      <td>-0.872756</td>\n",
              "      <td>-0.975738</td>\n",
              "      <td>-0.975237</td>\n",
              "      <td>-0.168742</td>\n",
              "      <td>-0.458162</td>\n",
              "      <td>-0.424672</td>\n",
              "      <td>-0.440022</td>\n",
              "      <td>0.143001</td>\n",
              "      <td>0.142525</td>\n",
              "      <td>-0.968903</td>\n",
              "      <td>-1.943850</td>\n",
              "      <td>-1.059868</td>\n",
              "      <td>-1.678997</td>\n",
              "      <td>0.185597</td>\n",
              "      <td>0.185492</td>\n",
              "      <td>-0.169472</td>\n",
              "      <td>-0.605631</td>\n",
              "      <td>-0.577099</td>\n",
              "      <td>-0.588384</td>\n",
              "      <td>0.241128</td>\n",
              "      <td>0.241406</td>\n",
              "      <td>0.018279</td>\n",
              "      <td>-0.087490</td>\n",
              "      <td>-0.131155</td>\n",
              "      <td>-0.097524</td>\n",
              "      <td>-0.120613</td>\n",
              "      <td>-0.119792</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>27</td>\n",
              "      <td>-0.068066</td>\n",
              "      <td>-0.172527</td>\n",
              "      <td>0.463609</td>\n",
              "      <td>-0.121970</td>\n",
              "      <td>-0.043875</td>\n",
              "      <td>-0.113002</td>\n",
              "      <td>-0.061584</td>\n",
              "      <td>-0.056299</td>\n",
              "      <td>-0.063618</td>\n",
              "      <td>-0.049707</td>\n",
              "      <td>-0.058723</td>\n",
              "      <td>-0.028741</td>\n",
              "      <td>-0.035391</td>\n",
              "      <td>-0.042955</td>\n",
              "      <td>-0.013282</td>\n",
              "      <td>-0.046132</td>\n",
              "      <td>-0.064087</td>\n",
              "      <td>-0.050781</td>\n",
              "      <td>-0.052054</td>\n",
              "      <td>0.887058</td>\n",
              "      <td>0.884557</td>\n",
              "      <td>-0.139686</td>\n",
              "      <td>-0.148883</td>\n",
              "      <td>-0.080147</td>\n",
              "      <td>-0.155623</td>\n",
              "      <td>-0.010763</td>\n",
              "      <td>-0.012107</td>\n",
              "      <td>-0.139688</td>\n",
              "      <td>-0.148879</td>\n",
              "      <td>-0.080147</td>\n",
              "      <td>-0.155623</td>\n",
              "      <td>-0.010669</td>\n",
              "      <td>-0.012005</td>\n",
              "      <td>-0.024669</td>\n",
              "      <td>-0.031272</td>\n",
              "      <td>-0.023045</td>\n",
              "      <td>-0.026215</td>\n",
              "      <td>...</td>\n",
              "      <td>-0.097719</td>\n",
              "      <td>-0.127462</td>\n",
              "      <td>0.003143</td>\n",
              "      <td>0.002426</td>\n",
              "      <td>-0.122845</td>\n",
              "      <td>-0.151191</td>\n",
              "      <td>-0.126676</td>\n",
              "      <td>-0.158223</td>\n",
              "      <td>1.064205</td>\n",
              "      <td>1.063787</td>\n",
              "      <td>0.196191</td>\n",
              "      <td>0.102624</td>\n",
              "      <td>-0.097194</td>\n",
              "      <td>0.174843</td>\n",
              "      <td>-0.975738</td>\n",
              "      <td>-0.975237</td>\n",
              "      <td>-0.163898</td>\n",
              "      <td>-0.456661</td>\n",
              "      <td>-0.424672</td>\n",
              "      <td>-0.437108</td>\n",
              "      <td>0.143001</td>\n",
              "      <td>0.142525</td>\n",
              "      <td>0.660117</td>\n",
              "      <td>-0.375715</td>\n",
              "      <td>-1.059868</td>\n",
              "      <td>0.142339</td>\n",
              "      <td>0.185597</td>\n",
              "      <td>0.185492</td>\n",
              "      <td>-0.216814</td>\n",
              "      <td>-0.617907</td>\n",
              "      <td>-0.577099</td>\n",
              "      <td>-0.613614</td>\n",
              "      <td>0.241128</td>\n",
              "      <td>0.241406</td>\n",
              "      <td>0.018279</td>\n",
              "      <td>-0.087490</td>\n",
              "      <td>-0.131155</td>\n",
              "      <td>-0.097524</td>\n",
              "      <td>-0.120613</td>\n",
              "      <td>-0.119792</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>4</td>\n",
              "      <td>2</td>\n",
              "      <td>31</td>\n",
              "      <td>-0.172141</td>\n",
              "      <td>-0.128652</td>\n",
              "      <td>0.463609</td>\n",
              "      <td>-0.121970</td>\n",
              "      <td>-0.043875</td>\n",
              "      <td>-0.113002</td>\n",
              "      <td>-0.061584</td>\n",
              "      <td>-0.162783</td>\n",
              "      <td>-0.168609</td>\n",
              "      <td>-0.049707</td>\n",
              "      <td>-0.165087</td>\n",
              "      <td>-0.028741</td>\n",
              "      <td>-0.035391</td>\n",
              "      <td>-0.042955</td>\n",
              "      <td>-0.013282</td>\n",
              "      <td>-0.052842</td>\n",
              "      <td>-0.170609</td>\n",
              "      <td>-0.172659</td>\n",
              "      <td>-0.175268</td>\n",
              "      <td>0.887058</td>\n",
              "      <td>0.884557</td>\n",
              "      <td>-0.139731</td>\n",
              "      <td>-0.148912</td>\n",
              "      <td>-0.080147</td>\n",
              "      <td>-0.155662</td>\n",
              "      <td>-0.010763</td>\n",
              "      <td>-0.012107</td>\n",
              "      <td>-0.139733</td>\n",
              "      <td>-0.148907</td>\n",
              "      <td>-0.080147</td>\n",
              "      <td>-0.155661</td>\n",
              "      <td>-0.010669</td>\n",
              "      <td>-0.012005</td>\n",
              "      <td>-0.024669</td>\n",
              "      <td>-0.031272</td>\n",
              "      <td>-0.023045</td>\n",
              "      <td>-0.026215</td>\n",
              "      <td>...</td>\n",
              "      <td>-0.097719</td>\n",
              "      <td>-0.127462</td>\n",
              "      <td>0.003143</td>\n",
              "      <td>0.002426</td>\n",
              "      <td>-0.123515</td>\n",
              "      <td>-0.198132</td>\n",
              "      <td>-0.186122</td>\n",
              "      <td>-0.212313</td>\n",
              "      <td>-1.159649</td>\n",
              "      <td>-1.160129</td>\n",
              "      <td>0.437857</td>\n",
              "      <td>0.237062</td>\n",
              "      <td>-0.300792</td>\n",
              "      <td>0.367823</td>\n",
              "      <td>-0.975738</td>\n",
              "      <td>-0.975237</td>\n",
              "      <td>-0.146391</td>\n",
              "      <td>-0.376825</td>\n",
              "      <td>-0.333687</td>\n",
              "      <td>-0.354351</td>\n",
              "      <td>-1.015963</td>\n",
              "      <td>-1.016230</td>\n",
              "      <td>0.660117</td>\n",
              "      <td>0.669709</td>\n",
              "      <td>0.153209</td>\n",
              "      <td>0.749451</td>\n",
              "      <td>-1.116918</td>\n",
              "      <td>-1.116948</td>\n",
              "      <td>-0.216814</td>\n",
              "      <td>-0.538112</td>\n",
              "      <td>-0.479951</td>\n",
              "      <td>-0.531617</td>\n",
              "      <td>-0.979074</td>\n",
              "      <td>-0.978556</td>\n",
              "      <td>0.018279</td>\n",
              "      <td>-0.087490</td>\n",
              "      <td>-0.131155</td>\n",
              "      <td>-0.097524</td>\n",
              "      <td>-0.120613</td>\n",
              "      <td>-0.119792</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5 rows × 168 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "   nid  label  time  ...  agg_feat_69  agg_feat_70  agg_feat_71\n",
              "0    0      1    17  ...    -0.011377    -0.120613    -0.119792\n",
              "1    1      2    13  ...    -0.068808    -1.194450    -0.940388\n",
              "2    2      1    10  ...    -0.097524    -0.120613    -0.119792\n",
              "3    3      2    27  ...    -0.097524    -0.120613    -0.119792\n",
              "4    4      2    31  ...    -0.097524    -0.120613    -0.119792\n",
              "\n",
              "[5 rows x 168 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zquQgCXPsMIN"
      },
      "source": [
        "train_dataset=[]\n",
        "test_dataset=[]\n",
        "for i in range(49):\n",
        "    nodes_df_tmp=merged_nodes_df[merged_nodes_df['time']==i+1].reset_index()\n",
        "    nodes_df_tmp['index']=nodes_df_tmp.index\n",
        "    df_edge_tmp = df_edge_time_fin.join(nodes_df_tmp.rename(columns={'nid':'source'})[['source','index']].set_index('source'),on='source',how='inner')\\\n",
        "        .join(nodes_df_tmp.rename(columns={'nid':'target'})[['target','index']].set_index('target'),on='target',how='inner',rsuffix='2') \\\n",
        "        .drop(columns=['source','target']) \\\n",
        "        .rename(columns={'index':'source','index2':'target'})\n",
        "    x = torch.tensor(np.array(nodes_df_tmp.sort_values(by='index').drop(columns=['index','nid','label'])), dtype=torch.float)\n",
        "    edge_index = torch.tensor(np.array(df_edge_tmp[['source','target']]).T, dtype=torch.long)\n",
        "    edge_index = to_undirected(edge_index)\n",
        "    mask = nodes_df_tmp['label']!=2\n",
        "    y=torch.tensor(np.array(nodes_df_tmp['label']))\n",
        "\n",
        "    if i+1<35:\n",
        "        data = Data(x=x,edge_index=edge_index, train_mask=mask, y=y)\n",
        "        train_dataset.append(data)\n",
        "    else:\n",
        "        data = Data(x=x,edge_index=edge_index, test_mask=mask, y=y)\n",
        "        test_dataset.append(data)\n",
        "\n",
        "train_loader = DataLoader(train_dataset, batch_size=1, shuffle=True)\n",
        "test_loader = DataLoader(test_dataset, batch_size=1, shuffle=False)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xHLjqRlXInHD"
      },
      "source": [
        "## Model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "h7tKytc-C4PT"
      },
      "source": [
        "class GCN(torch.nn.Module):\n",
        "    def __init__(self, num_node_features, hidden_channels, use_skip=False):\n",
        "        super(GCN, self).__init__()\n",
        "        self.conv1 = GCNConv(num_node_features, hidden_channels[0])\n",
        "        self.conv2 = GCNConv(hidden_channels[0], 2)\n",
        "        self.use_skip = use_skip\n",
        "        if self.use_skip:\n",
        "            self.weight = nn.init.xavier_normal_(Parameter(torch.Tensor(num_node_features, 2)))\n",
        "\n",
        "\n",
        "    def forward(self, data):\n",
        "        x = self.conv1(data.x, data.edge_index)\n",
        "        x = x.relu()\n",
        "        x = F.dropout(x, p=0.5, training=self.training)\n",
        "        x = self.conv2(x, data.edge_index)\n",
        "        if self.use_skip:\n",
        "            x = F.softmax(x+torch.matmul(data.x, self.weight), dim=-1)\n",
        "        else:\n",
        "            x = F.softmax(x, dim=-1)\n",
        "        return x\n",
        "\n",
        "    def embed(self, data):\n",
        "        x = self.conv1(data.x, data.edge_index)\n",
        "        return x"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZBQLzM2zKv55",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "ce6e2d8c-1019-4dca-91d3-a5dc87d37aec"
      },
      "source": [
        "model = GCN(num_node_features=data.num_node_features ,hidden_channels=[100])\n",
        "model.to(device)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "GCN(\n",
              "  (conv1): GCNConv(166, 100)\n",
              "  (conv2): GCNConv(100, 2)\n",
              ")"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 20
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "GM-0T8IJKA-m"
      },
      "source": [
        "## Train"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "i96wwQ1RQ4vw"
      },
      "source": [
        "#### Hyperparameter"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "y3JFJ6inQ4gK"
      },
      "source": [
        "patience = 50\n",
        "lr = 0.001\n",
        "epoches = 1000"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "V5dxOv0KKCap",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "14d270e6-09b7-42cd-d3d0-0b9f135cf03f"
      },
      "source": [
        "optimizer = torch.optim.Adam(model.parameters(), lr=lr)\n",
        "criterion = torch.nn.CrossEntropyLoss(weight=torch.tensor([0.7,0.3]).to(device))\n",
        "\n",
        "\n",
        "train_losses=[]\n",
        "val_losses =[]\n",
        "accuracies=[]\n",
        "if1=[]\n",
        "precisions=[]\n",
        "recalls=[]\n",
        "iterations=[]\n",
        "\n",
        "for epoch in range(epoches):\n",
        "    \n",
        "    model.train()\n",
        "    train_loss = 0\n",
        "    for data in train_loader:\n",
        "        data = data.to(device)\n",
        "        optimizer.zero_grad()\n",
        "        out = model(data)\n",
        "        loss = criterion(out[data.train_mask], data.y[data.train_mask])\n",
        "        _, pred = out[data.train_mask].max(dim=1)\n",
        "        loss.backward()\n",
        "        train_loss += loss.item() * data.num_graphs\n",
        "        optimizer.step()\n",
        "    train_loss /= len(train_loader.dataset)\n",
        "    \n",
        "\n",
        "    if (epoch+1)%50==0:\n",
        "        model.eval()\n",
        "        ys, preds = [], []\n",
        "        val_loss = 0\n",
        "        for data in test_loader:\n",
        "            data = data.to(device)\n",
        "            out = model(data)\n",
        "            loss = criterion(out[data.test_mask], data.y[data.test_mask])\n",
        "            val_loss += loss.item() * data.num_graphs\n",
        "            _, pred = out[data.test_mask].max(dim=1)\n",
        "            ys.append(data.y[data.test_mask].cpu())\n",
        "            preds.append(pred.cpu())\n",
        "\n",
        "        y, pred = torch.cat(ys, dim=0).numpy(), torch.cat(preds, dim=0).numpy()\n",
        "        val_loss /= len(test_loader.dataset)\n",
        "        f1 = f1_score(y, pred, average=None)\n",
        "        mf1 = f1_score(y, pred, average='micro')\n",
        "        precision = precision_score(y, pred, average=None)\n",
        "        recall = recall_score(y, pred, average=None)\n",
        "\n",
        "        iterations.append(epoch+1)\n",
        "        train_losses.append(train_loss)\n",
        "        val_losses.append(val_loss)\n",
        "        if1.append(f1[0])\n",
        "        accuracies.append(mf1)\n",
        "        precisions.append(precision[0])\n",
        "        recalls.append(recall[0])\n",
        "\n",
        "        print('Epoch: {:02d}, Train_Loss: {:.4f}, Val_Loss: {:.4f}, Precision: {:.4f}, Recall: {:.4f}, Illicit f1: {:.4f}, F1: {:.4f}'.format(epoch+1, train_loss, val_loss, precision[0], recall[0], f1[0], mf1))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch: 50, Train_Loss: 0.3916, Val_Loss: 0.4068, Precision: 0.5137, Recall: 0.6233, Illicit f1: 0.5632, F1: 0.9372\n",
            "Epoch: 100, Train_Loss: 0.3762, Val_Loss: 0.3989, Precision: 0.8003, Recall: 0.5069, Illicit f1: 0.6207, F1: 0.9597\n",
            "Epoch: 150, Train_Loss: 0.3679, Val_Loss: 0.4023, Precision: 0.8539, Recall: 0.4534, Illicit f1: 0.5923, F1: 0.9594\n",
            "Epoch: 200, Train_Loss: 0.3623, Val_Loss: 0.4004, Precision: 0.8409, Recall: 0.4783, Illicit f1: 0.6098, F1: 0.9602\n",
            "Epoch: 250, Train_Loss: 0.3587, Val_Loss: 0.3971, Precision: 0.7650, Recall: 0.5291, Illicit f1: 0.6255, F1: 0.9588\n",
            "Epoch: 300, Train_Loss: 0.3561, Val_Loss: 0.3985, Precision: 0.7890, Recall: 0.5143, Illicit f1: 0.6227, F1: 0.9595\n",
            "Epoch: 350, Train_Loss: 0.3553, Val_Loss: 0.4025, Precision: 0.8665, Recall: 0.4497, Illicit f1: 0.5921, F1: 0.9597\n",
            "Epoch: 400, Train_Loss: 0.3533, Val_Loss: 0.3995, Precision: 0.8347, Recall: 0.4848, Illicit f1: 0.6133, F1: 0.9603\n",
            "Epoch: 450, Train_Loss: 0.3523, Val_Loss: 0.3945, Precision: 0.7227, Recall: 0.5799, Illicit f1: 0.6434, F1: 0.9582\n",
            "Epoch: 500, Train_Loss: 0.3514, Val_Loss: 0.3976, Precision: 0.8174, Recall: 0.5125, Illicit f1: 0.6300, F1: 0.9609\n",
            "Epoch: 550, Train_Loss: 0.3514, Val_Loss: 0.3950, Precision: 0.7807, Recall: 0.5457, Illicit f1: 0.6424, F1: 0.9605\n",
            "Epoch: 600, Train_Loss: 0.3503, Val_Loss: 0.3942, Precision: 0.8183, Recall: 0.5365, Illicit f1: 0.6481, F1: 0.9621\n",
            "Epoch: 650, Train_Loss: 0.3495, Val_Loss: 0.3994, Precision: 0.8739, Recall: 0.4801, Illicit f1: 0.6198, F1: 0.9617\n",
            "Epoch: 700, Train_Loss: 0.3495, Val_Loss: 0.3988, Precision: 0.8876, Recall: 0.4811, Illicit f1: 0.6240, F1: 0.9623\n",
            "Epoch: 750, Train_Loss: 0.3492, Val_Loss: 0.4012, Precision: 0.8930, Recall: 0.4626, Illicit f1: 0.6095, F1: 0.9615\n",
            "Epoch: 800, Train_Loss: 0.3485, Val_Loss: 0.3962, Precision: 0.8124, Recall: 0.5319, Illicit f1: 0.6429, F1: 0.9616\n",
            "Epoch: 850, Train_Loss: 0.3480, Val_Loss: 0.3962, Precision: 0.8351, Recall: 0.5235, Illicit f1: 0.6436, F1: 0.9623\n",
            "Epoch: 900, Train_Loss: 0.3471, Val_Loss: 0.3958, Precision: 0.8419, Recall: 0.5162, Illicit f1: 0.6400, F1: 0.9623\n",
            "Epoch: 950, Train_Loss: 0.3473, Val_Loss: 0.3946, Precision: 0.7971, Recall: 0.5512, Illicit f1: 0.6517, F1: 0.9617\n",
            "Epoch: 1000, Train_Loss: 0.3462, Val_Loss: 0.3926, Precision: 0.8193, Recall: 0.5568, Illicit f1: 0.6630, F1: 0.9632\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Adu5SekkqlTs",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 269
        },
        "outputId": "8ee04673-3670-432a-934f-9b6edb886094"
      },
      "source": [
        "a,b,c,d = train_losses, val_losses, if1, accuracies\n",
        "plt.plot(np.array(a), 'r', label='Train loss')\n",
        "plt.plot(np.array(b), 'g', label='Valid loss')\n",
        "plt.plot(np.array(c), 'black', label='Illicit F1')\n",
        "plt.plot(np.array(d), 'orange', label='F1')\n",
        "plt.ylim([0,1.0])\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "eu7zu7NcMdwU"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}