(Sider) Drug_based_strategy.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true,
    "id": "dgBTU57OWYI0"
   },
   "source": [
    "# Install Library\n",
    "\n",
    "[RDKit ](https://github.com/rdkit/rdkit)\n",
    "\n",
    "[DGL](https://github.com/dmlc/dgl/)\n",
    "\n",
    "[DGL-LifeSci](https://github.com/awslabs/dgl-lifesci)\n",
    "\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "hidden": true,
    "id": "40Q4MY7Ch_QZ"
   },
   "outputs": [],
   "source": [
    "%%capture\n",
    "!git clone https://github.com/joerg84/Graph_Powered_ML_Workshop.git\n",
    "!rsync -av Graph_Powered_ML_Workshop/ ./ --exclude=.git"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "hidden": true,
    "id": "Eb-EbaQziDd5"
   },
   "outputs": [],
   "source": [
    "!pip3 install dgl"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "hidden": true,
    "id": "271tuyJSQ-m3"
   },
   "outputs": [],
   "source": [
    "%%capture\n",
    "!pip install rdkit-pypi\n",
    "!pip install dgl-cu111 \n",
    "!pip install dgllife"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true,
    "id": "xtojkovzWYI2"
   },
   "source": [
    "# Import Library"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "hidden": true,
    "id": "MoyaI9ZVQ6pH"
   },
   "outputs": [],
   "source": [
    "import dgl\n",
    "import sys\n",
    "import torch\n",
    "import random\n",
    "import cv2\n",
    "import torchvision\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "import torch.nn as nn\n",
    "import torch.nn.functional as F\n",
    "import matplotlib.pyplot as plt\n",
    "import torch.optim as optim\n",
    "\n",
    "from numpy import array\n",
    "from numpy import argmax\n",
    "from tensorflow.keras.utils import to_categorical\n",
    "from dgllife.model import MLPPredictor\n",
    "from dgllife.utils import smiles_to_bigraph, CanonicalAtomFeaturizer, AttentiveFPAtomFeaturizer\n",
    "from tqdm.notebook import tqdm, trange\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "from utils.general import DATASET, get_dataset, separate_active_and_inactive_data, get_embedding_vector_class, count_lablel,data_generator\n",
    "from utils.gcn_pre_trained import get_sider_model\n",
    "from model.heterogeneous_siamese_sider import siamese_model_attentiveFp_sider, siamese_model_Canonical_sider\n",
    "\n",
    "device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true,
    "id": "1RVgRpTmQ5rp",
    "jp-MarkdownHeadingCollapsed": true,
    "tags": []
   },
   "source": [
    "# Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "hidden": true,
    "id": "NdqTu7HrWYI4",
    "outputId": "8e48e2ee-a215-410d-f978-7c81c7c684e7"
   },
   "outputs": [],
   "source": [
    "cache_path='./sider_dglgraph.bin'\n",
    "\n",
    "df = get_dataset(\"sider\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "hidden": true,
    "id": "vwQ9fKAaWYI5"
   },
   "outputs": [],
   "source": [
    "sider_tasks = df.columns.values[1:].tolist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "hidden": true,
    "id": "uZFIo92cWYI6",
    "outputId": "0d173830-f770-4ddd-a1b0-ef780c3f33a7"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Hepatobiliary disorders : Counter({1: 743, 0: 684})\n",
      "Metabolism and nutrition disorders : Counter({1: 996, 0: 431})\n",
      "Product issues : Counter({0: 1405, 1: 22})\n",
      "Eye disorders : Counter({1: 876, 0: 551})\n",
      "Investigations : Counter({1: 1151, 0: 276})\n",
      "Musculoskeletal and connective tissue disorders : Counter({1: 997, 0: 430})\n",
      "Gastrointestinal disorders : Counter({1: 1298, 0: 129})\n",
      "Social circumstances : Counter({0: 1176, 1: 251})\n",
      "Immune system disorders : Counter({1: 1024, 0: 403})\n",
      "Reproductive system and breast disorders : Counter({1: 727, 0: 700})\n",
      "Neoplasms benign, malignant and unspecified (incl cysts and polyps) : Counter({0: 1051, 1: 376})\n",
      "General disorders and administration site conditions : Counter({1: 1292, 0: 135})\n",
      "Endocrine disorders : Counter({0: 1104, 1: 323})\n",
      "Surgical and medical procedures : Counter({0: 1214, 1: 213})\n",
      "Vascular disorders : Counter({1: 1108, 0: 319})\n",
      "Blood and lymphatic system disorders : Counter({1: 885, 0: 542})\n",
      "Skin and subcutaneous tissue disorders : Counter({1: 1318, 0: 109})\n",
      "Congenital, familial and genetic disorders : Counter({0: 1174, 1: 253})\n",
      "Infections and infestations : Counter({1: 1006, 0: 421})\n",
      "Respiratory, thoracic and mediastinal disorders : Counter({1: 1060, 0: 367})\n",
      "Psychiatric disorders : Counter({1: 1016, 0: 411})\n",
      "Renal and urinary disorders : Counter({1: 911, 0: 516})\n",
      "Pregnancy, puerperium and perinatal conditions : Counter({0: 1302, 1: 125})\n",
      "Ear and labyrinth disorders : Counter({0: 768, 1: 659})\n",
      "Cardiac disorders : Counter({1: 988, 0: 439})\n",
      "Nervous system disorders : Counter({1: 1304, 0: 123})\n",
      "Injury, poisoning and procedural complications : Counter({1: 946, 0: 481})\n"
     ]
    }
   ],
   "source": [
    "from collections import Counter\n",
    "\n",
    "one = []\n",
    "zero = []\n",
    "\n",
    "for task in sider_tasks:\n",
    "  data = df[task]\n",
    "  print(task ,\":\" ,Counter(data))\n",
    "  zero.append(Counter(data)[0])\n",
    "  one.append(Counter(data)[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 894
    },
    "hidden": true,
    "id": "e0RRCiSZWYI7",
    "outputId": "9f45a788-984a-4b13-8fd3-f77a64cb5c02"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 2000x1500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Importing the matplotlib library\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "# Declaring the figure or the plot (y, x) or (width, height)\n",
    "plt.figure(figsize=[20, 15])\n",
    "# Data to be plotted\n",
    "X = np.arange(1,len(sider_tasks)+1)\n",
    "\n",
    "plt.bar(X + 0.20, zero, color = 'g', width = 0.25)\n",
    "plt.bar(X + 0.4, one, color = 'b', width = 0.25)\n",
    "# Creating the legend of the bars in the plot\n",
    "plt.legend(['Active' , 'inactive' ])\n",
    "# Overiding the x axis with the country names\n",
    "plt.xticks([i + 0.25 for i in range(1,28)], X)\n",
    "# Giving the tilte for the plot\n",
    "plt.title(\"Sider dataset diagram\")\n",
    "# Namimg the x and y axis\n",
    "plt.xlabel('sider_tasks')\n",
    "plt.ylabel('Cases')\n",
    "# Saving the plot as a 'png'\n",
    "plt.savefig('4BarPlot.png')\n",
    "# Displaying the bar plot\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true,
    "id": "6grIE_JeqkUZ"
   },
   "source": [
    "# Required functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "hidden": true,
    "id": "m9K0ArLh6BC1"
   },
   "outputs": [],
   "source": [
    "def create_dataset_with_gcn(dataset, class_embed_vector, GCN, tasks ):\n",
    "    \n",
    "    created_data = []\n",
    "    data = np.arange(27)\n",
    "    onehot_encoded = to_categorical(data)\n",
    "    \n",
    "    for i, data in enumerate(dataset):\n",
    "        smiles, g, labels, mask = data\n",
    "        g = g.to(device)\n",
    "        g = dgl.add_self_loop(g)\n",
    "        graph_feats = g.ndata.pop('h')\n",
    "        embbed = GCN(g, graph_feats)\n",
    "        embbed = embbed.to('cpu')\n",
    "        embbed = embbed.detach().numpy()\n",
    "        for j, label in enumerate(labels):\n",
    "            a = (embbed, onehot_encoded[j], class_embed_vector[j], label, tasks[j])\n",
    "            created_data.append(a)\n",
    "    print('Data created!!')\n",
    "    return created_data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "OXvAkNGUWYI-"
   },
   "source": [
    "# drug-based strategy with BioAct-Het"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true,
    "id": "l3p5SEIwvM_1"
   },
   "source": [
    "## Classification with BioAct-Het and AttentiveFp GCN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "hidden": true,
    "id": "lDS5UguKr_x_",
    "outputId": "95c39725-8a28-4f88-d9ec-4ab30d76cf2e"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading GCN_attentivefp_SIDER_pre_trained.pth from https://data.dgl.ai/dgllife/pre_trained/gcn_attentivefp_sider.pth...\n",
      "Pretrained model loaded\n"
     ]
    }
   ],
   "source": [
    "model_name = 'GCN_attentivefp_SIDER'\n",
    "gcn_model = get_sider_model(model_name)\n",
    "gcn_model.eval()\n",
    "gcn_model = gcn_model.to(device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "hidden": true,
    "id": "MAx_oKuzvrGE",
    "outputId": "c12b82ed-20a6-44eb-9d8d-9199641fc5de",
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1033\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1176\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1165\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1189\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1180\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1266\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1046\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1089\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1052\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1170\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "class vector created!!\n",
      "class vector created!!\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1284\n",
      "Processing dgl graphs from scratch...\n",
      "Data created!!\n",
      "Data created!!\n",
      "train positive label: 19825 - train negative label: 14843\n",
      "Test positive label: 2043 - Test negative label: 1818\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 3s 12ms/step - loss: 0.5508 - accuracy: 0.7325 - mae: 0.3605 - mse: 0.1824 - auc: 0.7885\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 4s 13ms/step - loss: 0.4694 - accuracy: 0.7884 - mae: 0.3043 - mse: 0.1517 - auc: 0.8555\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 4s 13ms/step - loss: 0.4379 - accuracy: 0.8002 - mae: 0.2821 - mse: 0.1404 - auc: 0.8766\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 4s 13ms/step - loss: 0.4197 - accuracy: 0.8093 - mae: 0.2709 - mse: 0.1344 - auc: 0.8871\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 4s 13ms/step - loss: 0.4122 - accuracy: 0.8114 - mae: 0.2662 - mse: 0.1321 - auc: 0.8911\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 3s 13ms/step - loss: 0.4047 - accuracy: 0.8157 - mae: 0.2606 - mse: 0.1296 - auc: 0.8951\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 3s 13ms/step - loss: 0.4001 - accuracy: 0.8169 - mae: 0.2579 - mse: 0.1278 - auc: 0.8981\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 3s 12ms/step - loss: 0.3955 - accuracy: 0.8201 - mae: 0.2546 - mse: 0.1261 - auc: 0.9008\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 3s 12ms/step - loss: 0.3888 - accuracy: 0.8236 - mae: 0.2500 - mse: 0.1241 - auc: 0.9038\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 3s 12ms/step - loss: 0.3879 - accuracy: 0.8236 - mae: 0.2496 - mse: 0.1235 - auc: 0.9047\n",
      "82\n",
      "82\n",
      "81\n",
      "81\n",
      "81\n",
      "80\n",
      "80\n",
      "78\n",
      "73\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 3s 12ms/step - loss: 0.3859 - accuracy: 0.8238 - mae: 0.2478 - mse: 0.1230 - auc: 0.9054\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 3s 12ms/step - loss: 0.3837 - accuracy: 0.8245 - mae: 0.2460 - mse: 0.1220 - auc: 0.9068\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 3s 12ms/step - loss: 0.3780 - accuracy: 0.8276 - mae: 0.2433 - mse: 0.1206 - auc: 0.9091\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 3s 12ms/step - loss: 0.3773 - accuracy: 0.8298 - mae: 0.2413 - mse: 0.1197 - auc: 0.9100\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 3s 12ms/step - loss: 0.3742 - accuracy: 0.8292 - mae: 0.2402 - mse: 0.1195 - auc: 0.9107\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 3s 12ms/step - loss: 0.3736 - accuracy: 0.8298 - mae: 0.2394 - mse: 0.1187 - auc: 0.9116\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 3s 12ms/step - loss: 0.3725 - accuracy: 0.8289 - mae: 0.2393 - mse: 0.1188 - auc: 0.9116\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 3s 12ms/step - loss: 0.3698 - accuracy: 0.8302 - mae: 0.2370 - mse: 0.1176 - auc: 0.9134\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 3s 12ms/step - loss: 0.3662 - accuracy: 0.8337 - mae: 0.2346 - mse: 0.1165 - auc: 0.9149\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 3s 12ms/step - loss: 0.3668 - accuracy: 0.8334 - mae: 0.2350 - mse: 0.1167 - auc: 0.9145\n",
      "83\n",
      "83\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 3s 12ms/step - loss: 0.3625 - accuracy: 0.8349 - mae: 0.2325 - mse: 0.1154 - auc: 0.9163\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 3s 12ms/step - loss: 0.3613 - accuracy: 0.8366 - mae: 0.2306 - mse: 0.1147 - auc: 0.9171\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 3s 12ms/step - loss: 0.3610 - accuracy: 0.8348 - mae: 0.2314 - mse: 0.1150 - auc: 0.9173\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 3s 11ms/step - loss: 0.3597 - accuracy: 0.8351 - mae: 0.2303 - mse: 0.1143 - auc: 0.9180\n",
      "Epoch 5/10\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "271/271 [==============================] - 3s 11ms/step - loss: 0.3576 - accuracy: 0.8360 - mae: 0.2291 - mse: 0.1137 - auc: 0.9189\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 3s 13ms/step - loss: 0.3563 - accuracy: 0.8383 - mae: 0.2265 - mse: 0.1130 - auc: 0.9198\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 3s 13ms/step - loss: 0.3533 - accuracy: 0.8397 - mae: 0.2258 - mse: 0.1124 - auc: 0.9208\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 3s 13ms/step - loss: 0.3529 - accuracy: 0.8396 - mae: 0.2253 - mse: 0.1120 - auc: 0.9213\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 3s 12ms/step - loss: 0.3509 - accuracy: 0.8395 - mae: 0.2233 - mse: 0.1110 - auc: 0.9224\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 3s 12ms/step - loss: 0.3486 - accuracy: 0.8406 - mae: 0.2224 - mse: 0.1107 - auc: 0.9228\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 3s 12ms/step - loss: 0.3464 - accuracy: 0.8441 - mae: 0.2211 - mse: 0.1097 - auc: 0.9242\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 3s 12ms/step - loss: 0.3426 - accuracy: 0.8433 - mae: 0.2193 - mse: 0.1089 - auc: 0.9256\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 3s 12ms/step - loss: 0.3422 - accuracy: 0.8439 - mae: 0.2177 - mse: 0.1083 - auc: 0.9260\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 3s 12ms/step - loss: 0.3384 - accuracy: 0.8464 - mae: 0.2167 - mse: 0.1075 - auc: 0.9272\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 3s 13ms/step - loss: 0.3385 - accuracy: 0.8469 - mae: 0.2154 - mse: 0.1073 - auc: 0.9276\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 3s 12ms/step - loss: 0.3406 - accuracy: 0.8444 - mae: 0.2171 - mse: 0.1079 - auc: 0.9267\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 3s 12ms/step - loss: 0.3344 - accuracy: 0.8471 - mae: 0.2128 - mse: 0.1059 - auc: 0.9292\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 3s 13ms/step - loss: 0.3337 - accuracy: 0.8487 - mae: 0.2124 - mse: 0.1057 - auc: 0.9296\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 3s 12ms/step - loss: 0.3338 - accuracy: 0.8487 - mae: 0.2127 - mse: 0.1057 - auc: 0.9296\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 3s 13ms/step - loss: 0.3360 - accuracy: 0.8473 - mae: 0.2144 - mse: 0.1063 - auc: 0.9288\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "4\n",
      "121/121 [==============================] - 0s 2ms/step - loss: 1.0264 - accuracy: 0.7477 - mae: 0.2582 - mse: 0.2051 - auc: 0.8300\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1035\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1168\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1173\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1007\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1187\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1179\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1265\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1055\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1091\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1048\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1169\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "class vector created!!\n",
      "class vector created!!\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1284\n",
      "Processing dgl graphs from scratch...\n",
      "Data created!!\n",
      "Data created!!\n",
      "train positive label: 19821 - train negative label: 14847\n",
      "Test positive label: 2047 - Test negative label: 1814\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.5517 - accuracy: 0.7196 - mae: 0.3669 - mse: 0.1840 - auc_1: 0.7853\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.4635 - accuracy: 0.7857 - mae: 0.3010 - mse: 0.1498 - auc_1: 0.8599\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.4398 - accuracy: 0.8000 - mae: 0.2832 - mse: 0.1408 - auc_1: 0.8761\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.4205 - accuracy: 0.8077 - mae: 0.2715 - mse: 0.1349 - auc_1: 0.8868\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.4133 - accuracy: 0.8109 - mae: 0.2658 - mse: 0.1319 - auc_1: 0.8914\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.4034 - accuracy: 0.8143 - mae: 0.2599 - mse: 0.1291 - auc_1: 0.8962\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.3981 - accuracy: 0.8193 - mae: 0.2561 - mse: 0.1269 - auc_1: 0.8996\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.3950 - accuracy: 0.8201 - mae: 0.2547 - mse: 0.1263 - auc_1: 0.9007\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.3876 - accuracy: 0.8252 - mae: 0.2489 - mse: 0.1237 - auc_1: 0.9044\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.3851 - accuracy: 0.8262 - mae: 0.2475 - mse: 0.1228 - auc_1: 0.9057\n",
      "82\n",
      "82\n",
      "81\n",
      "81\n",
      "81\n",
      "80\n",
      "80\n",
      "78\n",
      "71\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.3835 - accuracy: 0.8257 - mae: 0.2464 - mse: 0.1224 - auc_1: 0.9064\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 4s 13ms/step - loss: 0.3820 - accuracy: 0.8269 - mae: 0.2454 - mse: 0.1216 - auc_1: 0.9075\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.3767 - accuracy: 0.8275 - mae: 0.2425 - mse: 0.1202 - auc_1: 0.9098\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3788 - accuracy: 0.8271 - mae: 0.2438 - mse: 0.1211 - auc_1: 0.9087\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3762 - accuracy: 0.8287 - mae: 0.2416 - mse: 0.1196 - auc_1: 0.9107\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3740 - accuracy: 0.8292 - mae: 0.2400 - mse: 0.1191 - auc_1: 0.9115\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3694 - accuracy: 0.8324 - mae: 0.2378 - mse: 0.1176 - auc_1: 0.9137\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3654 - accuracy: 0.8324 - mae: 0.2347 - mse: 0.1166 - auc_1: 0.9151\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3675 - accuracy: 0.8335 - mae: 0.2355 - mse: 0.1168 - auc_1: 0.9147\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3623 - accuracy: 0.8348 - mae: 0.2325 - mse: 0.1153 - auc_1: 0.9169\n",
      "83\n",
      "83\n",
      "83\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3602 - accuracy: 0.8343 - mae: 0.2308 - mse: 0.1144 - auc_1: 0.9181\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3588 - accuracy: 0.8362 - mae: 0.2303 - mse: 0.1141 - auc_1: 0.9187\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3581 - accuracy: 0.8382 - mae: 0.2295 - mse: 0.1135 - auc_1: 0.9193\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3543 - accuracy: 0.8354 - mae: 0.2272 - mse: 0.1132 - auc_1: 0.9201\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3520 - accuracy: 0.8382 - mae: 0.2257 - mse: 0.1119 - auc_1: 0.9217\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3485 - accuracy: 0.8429 - mae: 0.2232 - mse: 0.1110 - auc_1: 0.9230\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3467 - accuracy: 0.8419 - mae: 0.2218 - mse: 0.1103 - auc_1: 0.9237\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3464 - accuracy: 0.8422 - mae: 0.2218 - mse: 0.1100 - auc_1: 0.9242\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3450 - accuracy: 0.8413 - mae: 0.2212 - mse: 0.1101 - auc_1: 0.9244\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3410 - accuracy: 0.8435 - mae: 0.2186 - mse: 0.1086 - auc_1: 0.9264\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3407 - accuracy: 0.8428 - mae: 0.2185 - mse: 0.1083 - auc_1: 0.9267\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3415 - accuracy: 0.8435 - mae: 0.2181 - mse: 0.1084 - auc_1: 0.9263\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3376 - accuracy: 0.8439 - mae: 0.2157 - mse: 0.1073 - auc_1: 0.9280\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3326 - accuracy: 0.8469 - mae: 0.2136 - mse: 0.1061 - auc_1: 0.9298\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3335 - accuracy: 0.8472 - mae: 0.2125 - mse: 0.1057 - auc_1: 0.9299\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3336 - accuracy: 0.8473 - mae: 0.2136 - mse: 0.1059 - auc_1: 0.9297\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3303 - accuracy: 0.8474 - mae: 0.2115 - mse: 0.1050 - auc_1: 0.9310\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3303 - accuracy: 0.8486 - mae: 0.2115 - mse: 0.1051 - auc_1: 0.9310\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3255 - accuracy: 0.8490 - mae: 0.2091 - mse: 0.1039 - auc_1: 0.9326\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3257 - accuracy: 0.8512 - mae: 0.2078 - mse: 0.1035 - auc_1: 0.9329\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3247 - accuracy: 0.8520 - mae: 0.2079 - mse: 0.1029 - auc_1: 0.9337\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3268 - accuracy: 0.8512 - mae: 0.2085 - mse: 0.1036 - auc_1: 0.9326\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3203 - accuracy: 0.8536 - mae: 0.2050 - mse: 0.1019 - auc_1: 0.9349\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3181 - accuracy: 0.8547 - mae: 0.2032 - mse: 0.1011 - auc_1: 0.9360\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3180 - accuracy: 0.8534 - mae: 0.2033 - mse: 0.1012 - auc_1: 0.9359\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3173 - accuracy: 0.8526 - mae: 0.2035 - mse: 0.1011 - auc_1: 0.9362\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3146 - accuracy: 0.8551 - mae: 0.2016 - mse: 0.1002 - auc_1: 0.9372\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3121 - accuracy: 0.8552 - mae: 0.1999 - mse: 0.0995 - auc_1: 0.9382\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3113 - accuracy: 0.8561 - mae: 0.2000 - mse: 0.0993 - auc_1: 0.9383\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3111 - accuracy: 0.8560 - mae: 0.1991 - mse: 0.0992 - auc_1: 0.9386\n",
      "85\n",
      "85\n",
      "85\n",
      "85\n",
      "85\n",
      "85\n",
      "85\n",
      "85\n",
      "85\n",
      "5\n",
      "121/121 [==============================] - 0s 3ms/step - loss: 0.8692 - accuracy: 0.7534 - mae: 0.2599 - mse: 0.1925 - auc_1: 0.8297\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1047\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1168\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1162\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1184\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1171\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1264\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1061\n",
      "Processing dgl graphs from scratch...\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1091\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1057\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1170\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "class vector created!!\n",
      "class vector created!!\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1284\n",
      "Processing dgl graphs from scratch...\n",
      "Data created!!\n",
      "Data created!!\n",
      "train positive label: 19739 - train negative label: 14929\n",
      "Test positive label: 2129 - Test negative label: 1732\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.5527 - accuracy: 0.7223 - mae: 0.3673 - mse: 0.1843 - auc_2: 0.7888\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.4675 - accuracy: 0.7829 - mae: 0.3037 - mse: 0.1512 - auc_2: 0.8579\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.4379 - accuracy: 0.8015 - mae: 0.2835 - mse: 0.1407 - auc_2: 0.8769\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.4234 - accuracy: 0.8065 - mae: 0.2732 - mse: 0.1357 - auc_2: 0.8856\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.4122 - accuracy: 0.8101 - mae: 0.2665 - mse: 0.1321 - auc_2: 0.8918\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.4044 - accuracy: 0.8123 - mae: 0.2603 - mse: 0.1296 - auc_2: 0.8956\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.3999 - accuracy: 0.8153 - mae: 0.2575 - mse: 0.1278 - auc_2: 0.8985\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3936 - accuracy: 0.8189 - mae: 0.2534 - mse: 0.1259 - auc_2: 0.9015\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3919 - accuracy: 0.8194 - mae: 0.2519 - mse: 0.1252 - auc_2: 0.9026\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3909 - accuracy: 0.8221 - mae: 0.2506 - mse: 0.1246 - auc_2: 0.9031\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "80\n",
      "80\n",
      "78\n",
      "72\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3840 - accuracy: 0.8223 - mae: 0.2478 - mse: 0.1228 - auc_2: 0.9064\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3794 - accuracy: 0.8236 - mae: 0.2443 - mse: 0.1215 - auc_2: 0.9083\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3808 - accuracy: 0.8239 - mae: 0.2442 - mse: 0.1216 - auc_2: 0.9080\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3796 - accuracy: 0.8236 - mae: 0.2449 - mse: 0.1215 - auc_2: 0.9084\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3765 - accuracy: 0.8247 - mae: 0.2423 - mse: 0.1204 - auc_2: 0.9099\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3752 - accuracy: 0.8248 - mae: 0.2416 - mse: 0.1201 - auc_2: 0.9105\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3736 - accuracy: 0.8274 - mae: 0.2402 - mse: 0.1193 - auc_2: 0.9114\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3737 - accuracy: 0.8249 - mae: 0.2408 - mse: 0.1195 - auc_2: 0.9113\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3706 - accuracy: 0.8285 - mae: 0.2385 - mse: 0.1184 - auc_2: 0.9129\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 4s 15ms/step - loss: 0.3690 - accuracy: 0.8299 - mae: 0.2382 - mse: 0.1182 - auc_2: 0.9132\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "2\n",
      "121/121 [==============================] - 0s 3ms/step - loss: 0.5797 - accuracy: 0.7713 - mae: 0.2757 - mse: 0.1661 - auc_2: 0.8428\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1032\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1169\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1159\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1184\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1167\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1264\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1065\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1096\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1064\n",
      "Processing dgl graphs from scratch...\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1176\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "class vector created!!\n",
      "class vector created!!\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1284\n",
      "Processing dgl graphs from scratch...\n",
      "Data created!!\n",
      "Data created!!\n",
      "train positive label: 19520 - train negative label: 15148\n",
      "Test positive label: 2348 - Test negative label: 1513\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 4s 16ms/step - loss: 0.5579 - accuracy: 0.7196 - mae: 0.3687 - mse: 0.1861 - auc_3: 0.7841\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 4s 16ms/step - loss: 0.4737 - accuracy: 0.7834 - mae: 0.3081 - mse: 0.1537 - auc_3: 0.8543\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 4s 16ms/step - loss: 0.4430 - accuracy: 0.8009 - mae: 0.2857 - mse: 0.1421 - auc_3: 0.8743\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 4s 16ms/step - loss: 0.4267 - accuracy: 0.8058 - mae: 0.2744 - mse: 0.1364 - auc_3: 0.8841\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 4s 16ms/step - loss: 0.4161 - accuracy: 0.8103 - mae: 0.2676 - mse: 0.1326 - auc_3: 0.8907\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 4s 16ms/step - loss: 0.4097 - accuracy: 0.8142 - mae: 0.2629 - mse: 0.1307 - auc_3: 0.8937\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 4s 16ms/step - loss: 0.4027 - accuracy: 0.8169 - mae: 0.2589 - mse: 0.1287 - auc_3: 0.8969\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 4s 16ms/step - loss: 0.3998 - accuracy: 0.8167 - mae: 0.2569 - mse: 0.1275 - auc_3: 0.8989\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 4s 16ms/step - loss: 0.3933 - accuracy: 0.8196 - mae: 0.2529 - mse: 0.1257 - auc_3: 0.9020\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 4s 17ms/step - loss: 0.3917 - accuracy: 0.8232 - mae: 0.2512 - mse: 0.1244 - auc_3: 0.9033\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "80\n",
      "80\n",
      "78\n",
      "71\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 4s 16ms/step - loss: 0.3874 - accuracy: 0.8209 - mae: 0.2483 - mse: 0.1235 - auc_3: 0.9052\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 4s 16ms/step - loss: 0.3856 - accuracy: 0.8255 - mae: 0.2472 - mse: 0.1230 - auc_3: 0.9057\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.3821 - accuracy: 0.8242 - mae: 0.2455 - mse: 0.1221 - auc_3: 0.9075\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 4s 16ms/step - loss: 0.3793 - accuracy: 0.8279 - mae: 0.2434 - mse: 0.1208 - auc_3: 0.9092\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 4s 17ms/step - loss: 0.3769 - accuracy: 0.8273 - mae: 0.2416 - mse: 0.1199 - auc_3: 0.9104\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 4s 17ms/step - loss: 0.3724 - accuracy: 0.8318 - mae: 0.2389 - mse: 0.1187 - auc_3: 0.9123\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 4s 16ms/step - loss: 0.3727 - accuracy: 0.8284 - mae: 0.2392 - mse: 0.1187 - auc_3: 0.9123\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 4s 17ms/step - loss: 0.3708 - accuracy: 0.8305 - mae: 0.2377 - mse: 0.1179 - auc_3: 0.9134\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 4s 17ms/step - loss: 0.3690 - accuracy: 0.8298 - mae: 0.2370 - mse: 0.1176 - auc_3: 0.9137\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 4s 17ms/step - loss: 0.3680 - accuracy: 0.8330 - mae: 0.2354 - mse: 0.1171 - auc_3: 0.9145\n",
      "82\n",
      "83\n",
      "82\n",
      "83\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 4s 17ms/step - loss: 0.3679 - accuracy: 0.8319 - mae: 0.2353 - mse: 0.1174 - auc_3: 0.9143\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 4s 17ms/step - loss: 0.3657 - accuracy: 0.8329 - mae: 0.2348 - mse: 0.1162 - auc_3: 0.9157\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 4s 16ms/step - loss: 0.3621 - accuracy: 0.8369 - mae: 0.2307 - mse: 0.1148 - auc_3: 0.9176\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 4s 16ms/step - loss: 0.3608 - accuracy: 0.8357 - mae: 0.2306 - mse: 0.1146 - auc_3: 0.9179\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 4s 17ms/step - loss: 0.3589 - accuracy: 0.8373 - mae: 0.2293 - mse: 0.1141 - auc_3: 0.9187\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 4s 16ms/step - loss: 0.3584 - accuracy: 0.8354 - mae: 0.2291 - mse: 0.1140 - auc_3: 0.9187\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 4s 16ms/step - loss: 0.3563 - accuracy: 0.8398 - mae: 0.2267 - mse: 0.1130 - auc_3: 0.9202\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 4s 16ms/step - loss: 0.3536 - accuracy: 0.8378 - mae: 0.2256 - mse: 0.1126 - auc_3: 0.9209\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 4s 16ms/step - loss: 0.3515 - accuracy: 0.8391 - mae: 0.2247 - mse: 0.1117 - auc_3: 0.9220\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 4s 16ms/step - loss: 0.3495 - accuracy: 0.8402 - mae: 0.2235 - mse: 0.1110 - auc_3: 0.9230\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 4s 16ms/step - loss: 0.3478 - accuracy: 0.8406 - mae: 0.2221 - mse: 0.1103 - auc_3: 0.9239\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 4s 17ms/step - loss: 0.3472 - accuracy: 0.8414 - mae: 0.2207 - mse: 0.1102 - auc_3: 0.9241\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 4s 16ms/step - loss: 0.3493 - accuracy: 0.8414 - mae: 0.2226 - mse: 0.1107 - auc_3: 0.9233\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 4s 17ms/step - loss: 0.3417 - accuracy: 0.8457 - mae: 0.2175 - mse: 0.1082 - auc_3: 0.9265\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 4s 16ms/step - loss: 0.3420 - accuracy: 0.8440 - mae: 0.2183 - mse: 0.1088 - auc_3: 0.9261\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 4s 17ms/step - loss: 0.3417 - accuracy: 0.8442 - mae: 0.2173 - mse: 0.1083 - auc_3: 0.9265\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 4s 17ms/step - loss: 0.3412 - accuracy: 0.8445 - mae: 0.2173 - mse: 0.1083 - auc_3: 0.9266\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 4s 16ms/step - loss: 0.3439 - accuracy: 0.8444 - mae: 0.2184 - mse: 0.1088 - auc_3: 0.9258\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 4s 17ms/step - loss: 0.3397 - accuracy: 0.8467 - mae: 0.2163 - mse: 0.1074 - auc_3: 0.9276\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 4s 17ms/step - loss: 0.3345 - accuracy: 0.8450 - mae: 0.2145 - mse: 0.1066 - auc_3: 0.9291\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "4\n",
      "121/121 [==============================] - 0s 3ms/step - loss: 0.8787 - accuracy: 0.7827 - mae: 0.2267 - mse: 0.1803 - auc_3: 0.8317\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1035\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1167\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1160\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1181\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1167\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1264\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1055\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1004\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1097\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1073\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1172\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "class vector created!!\n",
      "class vector created!!\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1284\n",
      "Processing dgl graphs from scratch...\n",
      "Data created!!\n",
      "Data created!!\n",
      "train positive label: 19537 - train negative label: 15131\n",
      "Test positive label: 2331 - Test negative label: 1530\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 6s 21ms/step - loss: 0.5649 - accuracy: 0.7181 - mae: 0.3706 - mse: 0.1879 - auc_4: 0.7778\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 6s 20ms/step - loss: 0.4743 - accuracy: 0.7846 - mae: 0.3079 - mse: 0.1537 - auc_4: 0.8534\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 6s 20ms/step - loss: 0.4447 - accuracy: 0.7989 - mae: 0.2871 - mse: 0.1426 - auc_4: 0.8735\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 6s 20ms/step - loss: 0.4303 - accuracy: 0.8041 - mae: 0.2774 - mse: 0.1376 - auc_4: 0.8821\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 6s 21ms/step - loss: 0.4173 - accuracy: 0.8067 - mae: 0.2692 - mse: 0.1338 - auc_4: 0.8889\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 6s 20ms/step - loss: 0.4129 - accuracy: 0.8098 - mae: 0.2662 - mse: 0.1321 - auc_4: 0.8917\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 6s 21ms/step - loss: 0.4045 - accuracy: 0.8140 - mae: 0.2606 - mse: 0.1292 - auc_4: 0.8964\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 6s 20ms/step - loss: 0.4023 - accuracy: 0.8151 - mae: 0.2591 - mse: 0.1285 - auc_4: 0.8974\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 6s 20ms/step - loss: 0.3980 - accuracy: 0.8181 - mae: 0.2555 - mse: 0.1268 - auc_4: 0.9001\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 6s 20ms/step - loss: 0.3934 - accuracy: 0.8188 - mae: 0.2532 - mse: 0.1259 - auc_4: 0.9016\n",
      "81\n",
      "81\n",
      "81\n",
      "80\n",
      "80\n",
      "80\n",
      "79\n",
      "78\n",
      "71\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 6s 20ms/step - loss: 0.3914 - accuracy: 0.8222 - mae: 0.2515 - mse: 0.1245 - auc_4: 0.9034\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 6s 20ms/step - loss: 0.3874 - accuracy: 0.8202 - mae: 0.2490 - mse: 0.1238 - auc_4: 0.9048\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 6s 20ms/step - loss: 0.3882 - accuracy: 0.8225 - mae: 0.2491 - mse: 0.1237 - auc_4: 0.9046\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 6s 21ms/step - loss: 0.3852 - accuracy: 0.8242 - mae: 0.2467 - mse: 0.1226 - auc_4: 0.9064\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 6s 21ms/step - loss: 0.3805 - accuracy: 0.8229 - mae: 0.2450 - mse: 0.1217 - auc_4: 0.9080\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 6s 21ms/step - loss: 0.3774 - accuracy: 0.8255 - mae: 0.2424 - mse: 0.1206 - auc_4: 0.9096\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 6s 21ms/step - loss: 0.3733 - accuracy: 0.8283 - mae: 0.2395 - mse: 0.1190 - auc_4: 0.9118\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 6s 21ms/step - loss: 0.3742 - accuracy: 0.8273 - mae: 0.2400 - mse: 0.1192 - auc_4: 0.9117\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 6s 21ms/step - loss: 0.3736 - accuracy: 0.8293 - mae: 0.2387 - mse: 0.1188 - auc_4: 0.9120\n",
      "Epoch 10/10\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "271/271 [==============================] - 6s 21ms/step - loss: 0.3708 - accuracy: 0.8292 - mae: 0.2379 - mse: 0.1183 - auc_4: 0.9130\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "2\n",
      "121/121 [==============================] - 0s 3ms/step - loss: 0.5871 - accuracy: 0.7949 - mae: 0.2350 - mse: 0.1532 - auc_4: 0.8643\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1036\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1164\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1161\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1188\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1175\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1263\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1061\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1090\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1053\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1174\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "class vector created!!\n",
      "class vector created!!\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1284\n",
      "Processing dgl graphs from scratch...\n",
      "Data created!!\n",
      "Data created!!\n",
      "train positive label: 19640 - train negative label: 15028\n",
      "Test positive label: 2228 - Test negative label: 1633\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.5520 - accuracy: 0.7265 - mae: 0.3641 - mse: 0.1833 - auc_5: 0.7890\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.4749 - accuracy: 0.7842 - mae: 0.3085 - mse: 0.1535 - auc_5: 0.8532\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.4432 - accuracy: 0.7997 - mae: 0.2862 - mse: 0.1421 - auc_5: 0.8740\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.4257 - accuracy: 0.8085 - mae: 0.2741 - mse: 0.1363 - auc_5: 0.8844\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.4166 - accuracy: 0.8122 - mae: 0.2682 - mse: 0.1331 - auc_5: 0.8894\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.4098 - accuracy: 0.8146 - mae: 0.2634 - mse: 0.1308 - auc_5: 0.8931\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.4030 - accuracy: 0.8168 - mae: 0.2592 - mse: 0.1287 - auc_5: 0.8968\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.3996 - accuracy: 0.8155 - mae: 0.2574 - mse: 0.1277 - auc_5: 0.8987\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.3924 - accuracy: 0.8204 - mae: 0.2526 - mse: 0.1253 - auc_5: 0.9023\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.3921 - accuracy: 0.8198 - mae: 0.2522 - mse: 0.1251 - auc_5: 0.9026\n",
      "82\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "80\n",
      "79\n",
      "78\n",
      "72\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.3906 - accuracy: 0.8226 - mae: 0.2509 - mse: 0.1243 - auc_5: 0.9039\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.3843 - accuracy: 0.8253 - mae: 0.2468 - mse: 0.1226 - auc_5: 0.9064\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.3832 - accuracy: 0.8265 - mae: 0.2463 - mse: 0.1219 - auc_5: 0.9074\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.3789 - accuracy: 0.8287 - mae: 0.2430 - mse: 0.1204 - auc_5: 0.9094\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.3781 - accuracy: 0.8264 - mae: 0.2420 - mse: 0.1202 - auc_5: 0.9099\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.3753 - accuracy: 0.8293 - mae: 0.2407 - mse: 0.1195 - auc_5: 0.9110\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.3698 - accuracy: 0.8304 - mae: 0.2369 - mse: 0.1176 - auc_5: 0.9137\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.3704 - accuracy: 0.8332 - mae: 0.2367 - mse: 0.1174 - auc_5: 0.9137\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.3684 - accuracy: 0.8312 - mae: 0.2360 - mse: 0.1172 - auc_5: 0.9142\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.3654 - accuracy: 0.8319 - mae: 0.2341 - mse: 0.1163 - auc_5: 0.9157\n",
      "83\n",
      "83\n",
      "83\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.3605 - accuracy: 0.8352 - mae: 0.2306 - mse: 0.1146 - auc_5: 0.9180\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.3609 - accuracy: 0.8339 - mae: 0.2315 - mse: 0.1149 - auc_5: 0.9176\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.3565 - accuracy: 0.8373 - mae: 0.2278 - mse: 0.1130 - auc_5: 0.9201\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 4s 13ms/step - loss: 0.3574 - accuracy: 0.8366 - mae: 0.2278 - mse: 0.1134 - auc_5: 0.9195\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.3558 - accuracy: 0.8379 - mae: 0.2271 - mse: 0.1128 - auc_5: 0.9202\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.3531 - accuracy: 0.8379 - mae: 0.2260 - mse: 0.1124 - auc_5: 0.9210\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 4s 13ms/step - loss: 0.3481 - accuracy: 0.8408 - mae: 0.2222 - mse: 0.1106 - auc_5: 0.9235\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 4s 13ms/step - loss: 0.3496 - accuracy: 0.8399 - mae: 0.2232 - mse: 0.1110 - auc_5: 0.9229\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 4s 13ms/step - loss: 0.3491 - accuracy: 0.8433 - mae: 0.2228 - mse: 0.1104 - auc_5: 0.9234\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.3465 - accuracy: 0.8404 - mae: 0.2204 - mse: 0.1100 - auc_5: 0.9242\n",
      "84\n",
      "83\n",
      "84\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.3428 - accuracy: 0.8427 - mae: 0.2187 - mse: 0.1090 - auc_5: 0.9257\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.3425 - accuracy: 0.8410 - mae: 0.2189 - mse: 0.1089 - auc_5: 0.9260\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 4s 13ms/step - loss: 0.3397 - accuracy: 0.8466 - mae: 0.2171 - mse: 0.1080 - auc_5: 0.9270\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.3397 - accuracy: 0.8454 - mae: 0.2161 - mse: 0.1076 - auc_5: 0.9274\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.3370 - accuracy: 0.8468 - mae: 0.2149 - mse: 0.1068 - auc_5: 0.9285\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.3369 - accuracy: 0.8476 - mae: 0.2145 - mse: 0.1067 - auc_5: 0.9284\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.3355 - accuracy: 0.8477 - mae: 0.2130 - mse: 0.1059 - auc_5: 0.9294\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.3328 - accuracy: 0.8484 - mae: 0.2115 - mse: 0.1053 - auc_5: 0.9303\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.3314 - accuracy: 0.8487 - mae: 0.2115 - mse: 0.1051 - auc_5: 0.9309\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 4s 14ms/step - loss: 0.3332 - accuracy: 0.8476 - mae: 0.2120 - mse: 0.1059 - auc_5: 0.9297\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "4\n",
      "121/121 [==============================] - 0s 2ms/step - loss: 0.5704 - accuracy: 0.7773 - mae: 0.2700 - mse: 0.1697 - auc_5: 0.8303\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1034\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1162\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1161\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1182\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1174\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1265\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1057\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1093\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1055\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1171\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "class vector created!!\n",
      "class vector created!!\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1284\n",
      "Processing dgl graphs from scratch...\n",
      "Data created!!\n",
      "Data created!!\n",
      "train positive label: 19638 - train negative label: 15030\n",
      "Test positive label: 2230 - Test negative label: 1631\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.5478 - accuracy: 0.7349 - mae: 0.3620 - mse: 0.1821 - auc_6: 0.7915\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.4701 - accuracy: 0.7886 - mae: 0.3050 - mse: 0.1521 - auc_6: 0.8551\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.4430 - accuracy: 0.7986 - mae: 0.2863 - mse: 0.1425 - auc_6: 0.8734\n",
      "Epoch 4/10\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "271/271 [==============================] - 5s 17ms/step - loss: 0.4264 - accuracy: 0.8064 - mae: 0.2747 - mse: 0.1364 - auc_6: 0.8840\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.4145 - accuracy: 0.8109 - mae: 0.2669 - mse: 0.1327 - auc_6: 0.8902\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.4063 - accuracy: 0.8150 - mae: 0.2611 - mse: 0.1298 - auc_6: 0.8949\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.4019 - accuracy: 0.8178 - mae: 0.2581 - mse: 0.1278 - auc_6: 0.8981\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.3960 - accuracy: 0.8192 - mae: 0.2546 - mse: 0.1264 - auc_6: 0.9005\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.3936 - accuracy: 0.8204 - mae: 0.2521 - mse: 0.1254 - auc_6: 0.9020\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.3882 - accuracy: 0.8240 - mae: 0.2490 - mse: 0.1234 - auc_6: 0.9049\n",
      "82\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "80\n",
      "79\n",
      "78\n",
      "73\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.3878 - accuracy: 0.8252 - mae: 0.2491 - mse: 0.1232 - auc_6: 0.9052\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.3815 - accuracy: 0.8259 - mae: 0.2446 - mse: 0.1216 - auc_6: 0.9078\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.3809 - accuracy: 0.8269 - mae: 0.2447 - mse: 0.1215 - auc_6: 0.9080\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.3766 - accuracy: 0.8264 - mae: 0.2422 - mse: 0.1203 - auc_6: 0.9099\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.3761 - accuracy: 0.8274 - mae: 0.2415 - mse: 0.1199 - auc_6: 0.9103\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.3739 - accuracy: 0.8298 - mae: 0.2400 - mse: 0.1192 - auc_6: 0.9113\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.3706 - accuracy: 0.8320 - mae: 0.2376 - mse: 0.1179 - auc_6: 0.9131\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.3706 - accuracy: 0.8304 - mae: 0.2376 - mse: 0.1179 - auc_6: 0.9131\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.3673 - accuracy: 0.8313 - mae: 0.2357 - mse: 0.1170 - auc_6: 0.9146\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.3658 - accuracy: 0.8311 - mae: 0.2351 - mse: 0.1167 - auc_6: 0.9151\n",
      "83\n",
      "83\n",
      "83\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.3624 - accuracy: 0.8347 - mae: 0.2325 - mse: 0.1154 - auc_6: 0.9169\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.3626 - accuracy: 0.8330 - mae: 0.2328 - mse: 0.1154 - auc_6: 0.9169\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.3581 - accuracy: 0.8369 - mae: 0.2296 - mse: 0.1140 - auc_6: 0.9189\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.3591 - accuracy: 0.8375 - mae: 0.2292 - mse: 0.1144 - auc_6: 0.9182\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.3559 - accuracy: 0.8381 - mae: 0.2276 - mse: 0.1130 - auc_6: 0.9203\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.3510 - accuracy: 0.8389 - mae: 0.2253 - mse: 0.1120 - auc_6: 0.9218\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.3541 - accuracy: 0.8375 - mae: 0.2271 - mse: 0.1126 - auc_6: 0.9209\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.3487 - accuracy: 0.8394 - mae: 0.2235 - mse: 0.1108 - auc_6: 0.9233\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.3433 - accuracy: 0.8423 - mae: 0.2196 - mse: 0.1093 - auc_6: 0.9255\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.3469 - accuracy: 0.8412 - mae: 0.2222 - mse: 0.1102 - auc_6: 0.9242\n",
      "84\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.3439 - accuracy: 0.8423 - mae: 0.2203 - mse: 0.1092 - auc_6: 0.9254\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.3403 - accuracy: 0.8432 - mae: 0.2176 - mse: 0.1083 - auc_6: 0.9268\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.3401 - accuracy: 0.8454 - mae: 0.2172 - mse: 0.1077 - auc_6: 0.9273\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.3377 - accuracy: 0.8449 - mae: 0.2158 - mse: 0.1074 - auc_6: 0.9279\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.3373 - accuracy: 0.8455 - mae: 0.2159 - mse: 0.1072 - auc_6: 0.9282\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.3347 - accuracy: 0.8452 - mae: 0.2144 - mse: 0.1066 - auc_6: 0.9292\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.3325 - accuracy: 0.8469 - mae: 0.2124 - mse: 0.1054 - auc_6: 0.9304\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.3343 - accuracy: 0.8476 - mae: 0.2138 - mse: 0.1061 - auc_6: 0.9296\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.3336 - accuracy: 0.8456 - mae: 0.2131 - mse: 0.1061 - auc_6: 0.9298\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 5s 17ms/step - loss: 0.3297 - accuracy: 0.8460 - mae: 0.2114 - mse: 0.1053 - auc_6: 0.9310\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "4\n",
      "121/121 [==============================] - 0s 2ms/step - loss: 0.5094 - accuracy: 0.7920 - mae: 0.2413 - mse: 0.1511 - auc_6: 0.8731\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1040\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1167\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1163\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1003\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1183\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1174\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1265\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1074\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1012\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1098\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1062\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1176\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "class vector created!!\n",
      "class vector created!!\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1285\n",
      "Processing dgl graphs from scratch...\n",
      "Data created!!\n",
      "Data created!!\n",
      "train positive label: 19567 - train negative label: 15128\n",
      "Test positive label: 2301 - Test negative label: 1533\n",
      "Epoch 1/10\n",
      "272/272 [==============================] - 5s 18ms/step - loss: 0.5340 - accuracy: 0.7430 - mae: 0.3506 - mse: 0.1761 - auc_7: 0.8068\n",
      "Epoch 2/10\n",
      "272/272 [==============================] - 5s 18ms/step - loss: 0.4656 - accuracy: 0.7902 - mae: 0.3011 - mse: 0.1500 - auc_7: 0.8598\n",
      "Epoch 3/10\n",
      "272/272 [==============================] - 5s 18ms/step - loss: 0.4404 - accuracy: 0.8005 - mae: 0.2837 - mse: 0.1413 - auc_7: 0.8754\n",
      "Epoch 4/10\n",
      "272/272 [==============================] - 5s 18ms/step - loss: 0.4241 - accuracy: 0.8084 - mae: 0.2731 - mse: 0.1355 - auc_7: 0.8855\n",
      "Epoch 5/10\n",
      "272/272 [==============================] - 5s 18ms/step - loss: 0.4170 - accuracy: 0.8118 - mae: 0.2680 - mse: 0.1332 - auc_7: 0.8895\n",
      "Epoch 6/10\n",
      "272/272 [==============================] - 5s 18ms/step - loss: 0.4079 - accuracy: 0.8157 - mae: 0.2624 - mse: 0.1302 - auc_7: 0.8945\n",
      "Epoch 7/10\n",
      "272/272 [==============================] - 5s 18ms/step - loss: 0.4038 - accuracy: 0.8163 - mae: 0.2597 - mse: 0.1287 - auc_7: 0.8970\n",
      "Epoch 8/10\n",
      "272/272 [==============================] - 5s 18ms/step - loss: 0.4028 - accuracy: 0.8194 - mae: 0.2586 - mse: 0.1281 - auc_7: 0.8975\n",
      "Epoch 9/10\n",
      "272/272 [==============================] - 5s 18ms/step - loss: 0.3946 - accuracy: 0.8211 - mae: 0.2530 - mse: 0.1257 - auc_7: 0.9016\n",
      "Epoch 10/10\n",
      "272/272 [==============================] - 5s 18ms/step - loss: 0.3945 - accuracy: 0.8213 - mae: 0.2533 - mse: 0.1255 - auc_7: 0.9020\n",
      "82\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "80\n",
      "80\n",
      "79\n",
      "74\n",
      "Epoch 1/10\n",
      "272/272 [==============================] - 5s 18ms/step - loss: 0.3862 - accuracy: 0.8242 - mae: 0.2488 - mse: 0.1236 - auc_7: 0.9049\n",
      "Epoch 2/10\n",
      "272/272 [==============================] - 5s 18ms/step - loss: 0.3867 - accuracy: 0.8254 - mae: 0.2480 - mse: 0.1231 - auc_7: 0.9055\n",
      "Epoch 3/10\n",
      "272/272 [==============================] - 5s 18ms/step - loss: 0.3865 - accuracy: 0.8226 - mae: 0.2484 - mse: 0.1231 - auc_7: 0.9057\n",
      "Epoch 4/10\n",
      "272/272 [==============================] - 5s 18ms/step - loss: 0.3825 - accuracy: 0.8253 - mae: 0.2460 - mse: 0.1216 - auc_7: 0.9080\n",
      "Epoch 5/10\n",
      "272/272 [==============================] - 5s 18ms/step - loss: 0.3791 - accuracy: 0.8255 - mae: 0.2429 - mse: 0.1209 - auc_7: 0.9092\n",
      "Epoch 6/10\n",
      "272/272 [==============================] - 5s 18ms/step - loss: 0.3796 - accuracy: 0.8271 - mae: 0.2436 - mse: 0.1208 - auc_7: 0.9094\n",
      "Epoch 7/10\n",
      "272/272 [==============================] - 5s 18ms/step - loss: 0.3744 - accuracy: 0.8296 - mae: 0.2412 - mse: 0.1196 - auc_7: 0.9111\n",
      "Epoch 8/10\n",
      "272/272 [==============================] - 5s 18ms/step - loss: 0.3751 - accuracy: 0.8301 - mae: 0.2402 - mse: 0.1192 - auc_7: 0.9114\n",
      "Epoch 9/10\n",
      "272/272 [==============================] - 5s 18ms/step - loss: 0.3719 - accuracy: 0.8307 - mae: 0.2393 - mse: 0.1188 - auc_7: 0.9123\n",
      "Epoch 10/10\n",
      "272/272 [==============================] - 5s 18ms/step - loss: 0.3712 - accuracy: 0.8325 - mae: 0.2381 - mse: 0.1181 - auc_7: 0.9131\n",
      "83\n",
      "83\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "Epoch 1/10\n",
      "272/272 [==============================] - 5s 18ms/step - loss: 0.3711 - accuracy: 0.8301 - mae: 0.2379 - mse: 0.1181 - auc_7: 0.9133\n",
      "Epoch 2/10\n",
      "272/272 [==============================] - 5s 18ms/step - loss: 0.3664 - accuracy: 0.8316 - mae: 0.2354 - mse: 0.1168 - auc_7: 0.9152\n",
      "Epoch 3/10\n",
      "272/272 [==============================] - 5s 18ms/step - loss: 0.3661 - accuracy: 0.8335 - mae: 0.2338 - mse: 0.1162 - auc_7: 0.9157\n",
      "Epoch 4/10\n",
      "272/272 [==============================] - 5s 18ms/step - loss: 0.3624 - accuracy: 0.8346 - mae: 0.2322 - mse: 0.1153 - auc_7: 0.9172\n",
      "Epoch 5/10\n",
      "272/272 [==============================] - 5s 18ms/step - loss: 0.3630 - accuracy: 0.8347 - mae: 0.2325 - mse: 0.1157 - auc_7: 0.9166\n",
      "Epoch 6/10\n",
      "272/272 [==============================] - 5s 18ms/step - loss: 0.3587 - accuracy: 0.8360 - mae: 0.2300 - mse: 0.1141 - auc_7: 0.9189\n",
      "Epoch 7/10\n",
      "272/272 [==============================] - 5s 18ms/step - loss: 0.3577 - accuracy: 0.8372 - mae: 0.2289 - mse: 0.1138 - auc_7: 0.9191\n",
      "Epoch 8/10\n",
      "272/272 [==============================] - 5s 18ms/step - loss: 0.3568 - accuracy: 0.8386 - mae: 0.2284 - mse: 0.1130 - auc_7: 0.9202\n",
      "Epoch 9/10\n",
      "272/272 [==============================] - 5s 18ms/step - loss: 0.3565 - accuracy: 0.8371 - mae: 0.2280 - mse: 0.1134 - auc_7: 0.9197\n",
      "Epoch 10/10\n",
      "272/272 [==============================] - 5s 18ms/step - loss: 0.3549 - accuracy: 0.8374 - mae: 0.2267 - mse: 0.1125 - auc_7: 0.9210\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "3\n",
      "120/120 [==============================] - 0s 2ms/step - loss: 0.5688 - accuracy: 0.7705 - mae: 0.2502 - mse: 0.1615 - auc_7: 0.8571\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1029\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1169\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1159\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1191\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1172\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1264\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Processing molecule 1000/1059\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1093\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1053\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1172\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "class vector created!!\n",
      "class vector created!!\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1285\n",
      "Processing dgl graphs from scratch...\n",
      "Data created!!\n",
      "Data created!!\n",
      "train positive label: 19691 - train negative label: 15004\n",
      "Test positive label: 2177 - Test negative label: 1657\n",
      "Epoch 1/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.5393 - accuracy: 0.7359 - mae: 0.3557 - mse: 0.1787 - auc_8: 0.8004\n",
      "Epoch 2/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.4633 - accuracy: 0.7901 - mae: 0.2997 - mse: 0.1495 - auc_8: 0.8613\n",
      "Epoch 3/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.4379 - accuracy: 0.8002 - mae: 0.2831 - mse: 0.1408 - auc_8: 0.8766\n",
      "Epoch 4/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.4203 - accuracy: 0.8056 - mae: 0.2715 - mse: 0.1349 - auc_8: 0.8868\n",
      "Epoch 5/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.4133 - accuracy: 0.8120 - mae: 0.2658 - mse: 0.1320 - auc_8: 0.8913\n",
      "Epoch 6/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.4052 - accuracy: 0.8165 - mae: 0.2609 - mse: 0.1294 - auc_8: 0.8958\n",
      "Epoch 7/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.4015 - accuracy: 0.8178 - mae: 0.2580 - mse: 0.1281 - auc_8: 0.8978\n",
      "Epoch 8/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3965 - accuracy: 0.8188 - mae: 0.2553 - mse: 0.1264 - auc_8: 0.9005\n",
      "Epoch 9/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3909 - accuracy: 0.8214 - mae: 0.2516 - mse: 0.1250 - auc_8: 0.9029\n",
      "Epoch 10/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3864 - accuracy: 0.8232 - mae: 0.2486 - mse: 0.1236 - auc_8: 0.9048\n",
      "82\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "80\n",
      "80\n",
      "79\n",
      "73\n",
      "Epoch 1/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3828 - accuracy: 0.8251 - mae: 0.2463 - mse: 0.1223 - auc_8: 0.9070\n",
      "Epoch 2/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3846 - accuracy: 0.8250 - mae: 0.2475 - mse: 0.1224 - auc_8: 0.9067\n",
      "Epoch 3/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3794 - accuracy: 0.8259 - mae: 0.2444 - mse: 0.1214 - auc_8: 0.9083\n",
      "Epoch 4/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3782 - accuracy: 0.8292 - mae: 0.2418 - mse: 0.1200 - auc_8: 0.9100\n",
      "Epoch 5/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3753 - accuracy: 0.8309 - mae: 0.2409 - mse: 0.1193 - auc_8: 0.9111\n",
      "Epoch 6/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3700 - accuracy: 0.8319 - mae: 0.2373 - mse: 0.1177 - auc_8: 0.9135\n",
      "Epoch 7/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3699 - accuracy: 0.8318 - mae: 0.2371 - mse: 0.1176 - auc_8: 0.9138\n",
      "Epoch 8/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3642 - accuracy: 0.8342 - mae: 0.2331 - mse: 0.1158 - auc_8: 0.9162\n",
      "Epoch 9/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3607 - accuracy: 0.8366 - mae: 0.2311 - mse: 0.1146 - auc_8: 0.9178\n",
      "Epoch 10/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3651 - accuracy: 0.8339 - mae: 0.2339 - mse: 0.1161 - auc_8: 0.9158\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "Epoch 1/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3580 - accuracy: 0.8394 - mae: 0.2288 - mse: 0.1137 - auc_8: 0.9190\n",
      "Epoch 2/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3555 - accuracy: 0.8385 - mae: 0.2268 - mse: 0.1130 - auc_8: 0.9201\n",
      "Epoch 3/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3549 - accuracy: 0.8390 - mae: 0.2275 - mse: 0.1129 - auc_8: 0.9204\n",
      "Epoch 4/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3503 - accuracy: 0.8440 - mae: 0.2235 - mse: 0.1109 - auc_8: 0.9228\n",
      "Epoch 5/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3488 - accuracy: 0.8432 - mae: 0.2230 - mse: 0.1105 - auc_8: 0.9235\n",
      "Epoch 6/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3481 - accuracy: 0.8433 - mae: 0.2218 - mse: 0.1102 - auc_8: 0.9238\n",
      "Epoch 7/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3468 - accuracy: 0.8441 - mae: 0.2219 - mse: 0.1101 - auc_8: 0.9241\n",
      "Epoch 8/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3441 - accuracy: 0.8419 - mae: 0.2191 - mse: 0.1092 - auc_8: 0.9255\n",
      "Epoch 9/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3411 - accuracy: 0.8433 - mae: 0.2185 - mse: 0.1086 - auc_8: 0.9264\n",
      "Epoch 10/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3381 - accuracy: 0.8475 - mae: 0.2164 - mse: 0.1077 - auc_8: 0.9276\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "83\n",
      "83\n",
      "83\n",
      "Epoch 1/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3397 - accuracy: 0.8465 - mae: 0.2171 - mse: 0.1076 - auc_8: 0.9275\n",
      "Epoch 2/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3362 - accuracy: 0.8478 - mae: 0.2145 - mse: 0.1065 - auc_8: 0.9288\n",
      "Epoch 3/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3347 - accuracy: 0.8487 - mae: 0.2138 - mse: 0.1058 - auc_8: 0.9297\n",
      "Epoch 4/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3322 - accuracy: 0.8484 - mae: 0.2119 - mse: 0.1056 - auc_8: 0.9302\n",
      "Epoch 5/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3319 - accuracy: 0.8501 - mae: 0.2122 - mse: 0.1052 - auc_8: 0.9306\n",
      "Epoch 6/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3321 - accuracy: 0.8495 - mae: 0.2111 - mse: 0.1050 - auc_8: 0.9308\n",
      "Epoch 7/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3281 - accuracy: 0.8499 - mae: 0.2094 - mse: 0.1042 - auc_8: 0.9321\n",
      "Epoch 8/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3255 - accuracy: 0.8526 - mae: 0.2079 - mse: 0.1032 - auc_8: 0.9332\n",
      "Epoch 9/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3239 - accuracy: 0.8532 - mae: 0.2070 - mse: 0.1028 - auc_8: 0.9339\n",
      "Epoch 10/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3230 - accuracy: 0.8531 - mae: 0.2060 - mse: 0.1028 - auc_8: 0.9340\n",
      "85\n",
      "85\n",
      "84\n",
      "84\n",
      "85\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "Epoch 1/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3212 - accuracy: 0.8539 - mae: 0.2049 - mse: 0.1019 - auc_8: 0.9351\n",
      "Epoch 2/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3200 - accuracy: 0.8543 - mae: 0.2043 - mse: 0.1015 - auc_8: 0.9355\n",
      "Epoch 3/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3192 - accuracy: 0.8547 - mae: 0.2047 - mse: 0.1015 - auc_8: 0.9356\n",
      "Epoch 4/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3189 - accuracy: 0.8552 - mae: 0.2035 - mse: 0.1013 - auc_8: 0.9358\n",
      "Epoch 5/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3185 - accuracy: 0.8545 - mae: 0.2039 - mse: 0.1011 - auc_8: 0.9361\n",
      "Epoch 6/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3184 - accuracy: 0.8547 - mae: 0.2029 - mse: 0.1011 - auc_8: 0.9360\n",
      "Epoch 7/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3114 - accuracy: 0.8556 - mae: 0.2001 - mse: 0.0994 - auc_8: 0.9384\n",
      "Epoch 8/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3113 - accuracy: 0.8568 - mae: 0.1991 - mse: 0.0995 - auc_8: 0.9383\n",
      "Epoch 9/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3105 - accuracy: 0.8572 - mae: 0.1986 - mse: 0.0988 - auc_8: 0.9390\n",
      "Epoch 10/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3123 - accuracy: 0.8555 - mae: 0.2002 - mse: 0.0997 - auc_8: 0.9381\n",
      "85\n",
      "85\n",
      "85\n",
      "85\n",
      "85\n",
      "85\n",
      "85\n",
      "85\n",
      "85\n",
      "5\n",
      "120/120 [==============================] - 0s 2ms/step - loss: 0.5710 - accuracy: 0.7851 - mae: 0.2435 - mse: 0.1603 - auc_8: 0.8642\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1038\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1172\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1165\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1008\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1193\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1177\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1265\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1051\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1088\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1049\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1168\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "class vector created!!\n",
      "class vector created!!\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1285\n",
      "Processing dgl graphs from scratch...\n",
      "Data created!!\n",
      "Data created!!\n",
      "train positive label: 19834 - train negative label: 14861\n",
      "Test positive label: 2034 - Test negative label: 1800\n",
      "Epoch 1/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.5354 - accuracy: 0.7421 - mae: 0.3516 - mse: 0.1766 - auc_9: 0.8042\n",
      "Epoch 2/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.4650 - accuracy: 0.7892 - mae: 0.3012 - mse: 0.1499 - auc_9: 0.8589\n",
      "Epoch 3/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.4420 - accuracy: 0.7976 - mae: 0.2854 - mse: 0.1416 - auc_9: 0.8742\n",
      "Epoch 4/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.4259 - accuracy: 0.8036 - mae: 0.2754 - mse: 0.1365 - auc_9: 0.8839\n",
      "Epoch 5/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.4162 - accuracy: 0.8109 - mae: 0.2681 - mse: 0.1331 - auc_9: 0.8894\n",
      "Epoch 6/10\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "272/272 [==============================] - 4s 14ms/step - loss: 0.4075 - accuracy: 0.8146 - mae: 0.2619 - mse: 0.1303 - auc_9: 0.8938\n",
      "Epoch 7/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.4029 - accuracy: 0.8137 - mae: 0.2598 - mse: 0.1289 - auc_9: 0.8964\n",
      "Epoch 8/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3958 - accuracy: 0.8168 - mae: 0.2546 - mse: 0.1268 - auc_9: 0.8999\n",
      "Epoch 9/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3956 - accuracy: 0.8184 - mae: 0.2544 - mse: 0.1264 - auc_9: 0.9002\n",
      "Epoch 10/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3904 - accuracy: 0.8206 - mae: 0.2515 - mse: 0.1250 - auc_9: 0.9027\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "80\n",
      "79\n",
      "78\n",
      "74\n",
      "Epoch 1/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3881 - accuracy: 0.8234 - mae: 0.2499 - mse: 0.1238 - auc_9: 0.9045\n",
      "Epoch 2/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3820 - accuracy: 0.8235 - mae: 0.2454 - mse: 0.1221 - auc_9: 0.9070\n",
      "Epoch 3/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3849 - accuracy: 0.8220 - mae: 0.2466 - mse: 0.1226 - auc_9: 0.9063\n",
      "Epoch 4/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3815 - accuracy: 0.8233 - mae: 0.2454 - mse: 0.1221 - auc_9: 0.9072\n",
      "Epoch 5/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3781 - accuracy: 0.8259 - mae: 0.2434 - mse: 0.1206 - auc_9: 0.9092\n",
      "Epoch 6/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3729 - accuracy: 0.8283 - mae: 0.2392 - mse: 0.1194 - auc_9: 0.9112\n",
      "Epoch 7/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3729 - accuracy: 0.8289 - mae: 0.2395 - mse: 0.1189 - auc_9: 0.9117\n",
      "Epoch 8/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3703 - accuracy: 0.8306 - mae: 0.2380 - mse: 0.1180 - auc_9: 0.9131\n",
      "Epoch 9/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3723 - accuracy: 0.8298 - mae: 0.2378 - mse: 0.1186 - auc_9: 0.9120\n",
      "Epoch 10/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3707 - accuracy: 0.8280 - mae: 0.2389 - mse: 0.1182 - auc_9: 0.9129\n",
      "82\n",
      "83\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "Epoch 1/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3661 - accuracy: 0.8319 - mae: 0.2351 - mse: 0.1167 - auc_9: 0.9151\n",
      "Epoch 2/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3656 - accuracy: 0.8318 - mae: 0.2340 - mse: 0.1164 - auc_9: 0.9156\n",
      "Epoch 3/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3644 - accuracy: 0.8330 - mae: 0.2342 - mse: 0.1161 - auc_9: 0.9157\n",
      "Epoch 4/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3625 - accuracy: 0.8337 - mae: 0.2320 - mse: 0.1156 - auc_9: 0.9166\n",
      "Epoch 5/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3617 - accuracy: 0.8364 - mae: 0.2327 - mse: 0.1150 - auc_9: 0.9174\n",
      "Epoch 6/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3595 - accuracy: 0.8353 - mae: 0.2304 - mse: 0.1147 - auc_9: 0.9178\n",
      "Epoch 7/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3587 - accuracy: 0.8357 - mae: 0.2315 - mse: 0.1145 - auc_9: 0.9184\n",
      "Epoch 8/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3568 - accuracy: 0.8365 - mae: 0.2287 - mse: 0.1133 - auc_9: 0.9198\n",
      "Epoch 9/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3559 - accuracy: 0.8354 - mae: 0.2285 - mse: 0.1136 - auc_9: 0.9196\n",
      "Epoch 10/10\n",
      "272/272 [==============================] - 4s 14ms/step - loss: 0.3544 - accuracy: 0.8377 - mae: 0.2266 - mse: 0.1130 - auc_9: 0.9202\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "3\n",
      "120/120 [==============================] - 0s 2ms/step - loss: 0.5906 - accuracy: 0.7517 - mae: 0.2581 - mse: 0.1681 - auc_9: 0.8602\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import KFold\n",
    "\n",
    "Epoch_S = 10\n",
    "\n",
    "def evaluate_model(df, k = 10 , shuffle = False, tasks = sider_tasks):\n",
    "    result =[]    \n",
    "\n",
    "    kf = KFold(n_splits=10, shuffle= shuffle, random_state=None)\n",
    "    \n",
    "    for train_index, test_index in kf.split(df):\n",
    "\n",
    "        train = df.iloc[train_index]\n",
    "        test =  df.iloc[test_index]\n",
    "        \n",
    "        #Calculation of embedded vectors for each class\n",
    "        df_train_positive, df_train_negative = Separate_active_and_inactive_data(train, tasks)\n",
    "        df_test_positive, df_test_negative = Separate_active_and_inactive_data(test, tasks)\n",
    "        \n",
    "        dataset_positive_train = [DATASET(d, smiles_to_bigraph, AttentiveFPAtomFeaturizer(), cache_file_path = cache_path) for d in df_train_positive]\n",
    "        dataset_negative_train = [DATASET(d, smiles_to_bigraph, AttentiveFPAtomFeaturizer(), cache_file_path = cache_path) for d in df_train_negative]    \n",
    "        dataset_positive_test = [DATASET(d, smiles_to_bigraph, AttentiveFPAtomFeaturizer(), cache_file_path = cache_path) for d in df_test_positive]\n",
    "        dataset_negative_test = [DATASET(d, smiles_to_bigraph, AttentiveFPAtomFeaturizer(), cache_file_path = cache_path) for d in df_test_negative]\n",
    "        \n",
    "        embed_class_sider_train = get_embedding_vector_class(dataset_positive_train, dataset_negative_train, radius=2, size = 512)\n",
    "        embed_class_sider_test = get_embedding_vector_class(dataset_positive_test, dataset_negative_test, radius=2, size = 512)\n",
    "       \n",
    "        #create_dataset\n",
    "        train_dataset = DATASET(train, smiles_to_bigraph, AttentiveFPAtomFeaturizer(), cache_file_path = cache_path)\n",
    "        test_dataset = DATASET(test, smiles_to_bigraph, AttentiveFPAtomFeaturizer(), cache_file_path = cache_path)\n",
    "                \n",
    "        train_ds = create_dataset_with_gcn(train_dataset, embed_class_sider_train, gcn_model, tasks)\n",
    "        valid_ds = create_dataset_with_gcn(test_dataset, embed_class_sider_test, gcn_model, tasks)\n",
    "        \n",
    "        label_pos , label_neg = count_lablel(train_ds)\n",
    "        print(f'train positive label: {label_pos} - train negative label: {label_neg}')\n",
    "\n",
    "        label_pos , label_neg = count_lablel(valid_ds)\n",
    "        print(f'Test positive label: {label_pos} - Test negative label: {label_neg}')\n",
    "\n",
    "        l_train = []\n",
    "        r_train = []\n",
    "        lbls_train = []\n",
    "        \n",
    "        l_valid = []\n",
    "        r_valid = []\n",
    "        lbls_valid = []\n",
    "\n",
    "        for i , data in enumerate(train_ds):\n",
    "            embbed_drug, onehot_task, embbed_task, lbl, task_name = data\n",
    "            l_train.append(embbed_drug[0])\n",
    "            r_train.append(embbed_task)\n",
    "            lbls_train.append(lbl.tolist())\n",
    "        \n",
    "        for i , data in enumerate(valid_ds):\n",
    "            embbed_drug, onehot_task, embbed_task, lbl, task_name = data\n",
    "            l_valid.append(embbed_drug[0])\n",
    "            r_valid.append(embbed_task)\n",
    "            lbls_valid.append(lbl.tolist())\n",
    "\n",
    "        l_train = np.array(l_train).reshape(-1,1024,1)\n",
    "        r_train = np.array(r_train).reshape(-1,512,1)\n",
    "        lbls_train = np.array(lbls_train)\n",
    "\n",
    "        l_valid = np.array(l_valid).reshape(-1,1024,1)\n",
    "        r_valid = np.array(r_valid).reshape(-1,512,1)\n",
    "        lbls_valid = np.array(lbls_valid)\n",
    "\n",
    "        # create neural network model\n",
    "        siamese_net = siamese_model_attentiveFp_sider()\n",
    "        history = History()\n",
    "        P = siamese_net.fit([l_train, r_train], lbls_train, epochs = Epoch_S, batch_size = 128, callbacks=[history])\n",
    "\n",
    "        for j in range(100):\n",
    "            C=1\n",
    "            Before = int(P.history['accuracy'][-1]*100)\n",
    "            for i in range(2,Epoch_S+1):\n",
    "                if  int(P.history['accuracy'][-i]*100)== Before:\n",
    "                    C=C+1\n",
    "                else:\n",
    "                    C=1\n",
    "                Before=int(P.history['accuracy'][-i]*100)\n",
    "                print(Before)\n",
    "            if C==Epoch_S:\n",
    "                break\n",
    "            P = siamese_net.fit([l_train, r_train], lbls_train, epochs = Epoch_S, batch_size = 128, callbacks=[history])\n",
    "        print(j+1)\n",
    "        \n",
    "        score  = siamese_net.evaluate([l_valid,r_valid],lbls_valid, verbose=1)\n",
    "        a = (score[1],score[4])\n",
    "        result.append(a)\n",
    "    \n",
    "    return result\n",
    " \n",
    " \n",
    "scores = evaluate_model(df, 10, False, sider_tasks)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "hidden": true,
    "id": "lsGn5AonEiDO",
    "outputId": "3a3b33ce-f18c-43de-ebb3-80edae943d4d"
   },
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'scores' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[1;32m~\\AppData\\Local\\Temp\\ipykernel_27092\\345061674.py\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[0mscores\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m: name 'scores' is not defined"
     ]
    }
   ],
   "source": [
    "scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "hidden": true,
    "id": "BavvQrdWMri1",
    "outputId": "23f952b1-4dc0-456a-8aa7-dbff5a071628"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "accuracy= 0.7726572275161743 AUC= 0.8483467936515808\n"
     ]
    }
   ],
   "source": [
    "acc = []\n",
    "auc = []\n",
    "for i in scores:\n",
    "    acc.append(i[0])\n",
    "    auc.append(i[1])\n",
    "print(f'accuracy= {np.mean(acc)} AUC= {np.mean(auc)}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true,
    "id": "9JiD9OHJnX55"
   },
   "source": [
    "## Classification with BioAct-Het and Canonical GCN "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "hidden": true,
    "id": "UePCH8kkoBE5",
    "outputId": "33ff0759-7a62-4997-ec01-ffcf1b04aaec"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading GCN_canonical_SIDER_pre_trained.pth from https://data.dgl.ai/dgllife/pre_trained/gcn_canonical_sider.pth...\n",
      "Pretrained model loaded\n"
     ]
    }
   ],
   "source": [
    "model_name = 'GCN_canonical_SIDER'\n",
    "gcn_model = get_sider_model(model_name)\n",
    "gcn_model.eval()\n",
    "gcn_model = gcn_model.to(device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "hidden": true,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1033\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1176\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1165\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1189\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1180\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1266\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1046\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1089\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1052\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1170\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "class vector created!!\n",
      "class vector created!!\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1284\n",
      "Processing dgl graphs from scratch...\n",
      "Data created!!\n",
      "Data created!!\n",
      "train positive label: 19825 - train negative label: 14843\n",
      "Test positive label: 2043 - Test negative label: 1818\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 2s 6ms/step - loss: 0.5360 - accuracy: 0.7480 - mae: 0.3531 - mse: 0.1761 - auc_13: 0.8041\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 2s 6ms/step - loss: 0.4937 - accuracy: 0.7737 - mae: 0.3236 - mse: 0.1610 - auc_13: 0.8360\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 2s 6ms/step - loss: 0.4812 - accuracy: 0.7794 - mae: 0.3147 - mse: 0.1564 - auc_13: 0.8454\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 2s 6ms/step - loss: 0.4753 - accuracy: 0.7798 - mae: 0.3111 - mse: 0.1548 - auc_13: 0.8483\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 2s 6ms/step - loss: 0.4721 - accuracy: 0.7828 - mae: 0.3082 - mse: 0.1533 - auc_13: 0.8514\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 2s 6ms/step - loss: 0.4662 - accuracy: 0.7878 - mae: 0.3043 - mse: 0.1513 - auc_13: 0.8554\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 2s 6ms/step - loss: 0.4618 - accuracy: 0.7896 - mae: 0.3011 - mse: 0.1497 - auc_13: 0.8583\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 2s 6ms/step - loss: 0.4576 - accuracy: 0.7897 - mae: 0.2977 - mse: 0.1481 - auc_13: 0.8612\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4534 - accuracy: 0.7921 - mae: 0.2949 - mse: 0.1465 - auc_13: 0.8641\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 2s 6ms/step - loss: 0.4498 - accuracy: 0.7939 - mae: 0.2922 - mse: 0.1452 - auc_13: 0.8671\n",
      "79\n",
      "78\n",
      "78\n",
      "78\n",
      "78\n",
      "77\n",
      "77\n",
      "77\n",
      "74\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 2s 6ms/step - loss: 0.4473 - accuracy: 0.7965 - mae: 0.2902 - mse: 0.1442 - auc_13: 0.8685\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 2s 6ms/step - loss: 0.4426 - accuracy: 0.7960 - mae: 0.2874 - mse: 0.1428 - auc_13: 0.8714\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 2s 6ms/step - loss: 0.4367 - accuracy: 0.8025 - mae: 0.2825 - mse: 0.1404 - auc_13: 0.8752\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 2s 6ms/step - loss: 0.4336 - accuracy: 0.8020 - mae: 0.2808 - mse: 0.1394 - auc_13: 0.8774\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 2s 6ms/step - loss: 0.4333 - accuracy: 0.8027 - mae: 0.2799 - mse: 0.1391 - auc_13: 0.8780\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4280 - accuracy: 0.8042 - mae: 0.2769 - mse: 0.1375 - auc_13: 0.8811\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 2s 6ms/step - loss: 0.4239 - accuracy: 0.8079 - mae: 0.2733 - mse: 0.1359 - auc_13: 0.8834\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4203 - accuracy: 0.8102 - mae: 0.2713 - mse: 0.1347 - auc_13: 0.8854\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 2s 6ms/step - loss: 0.4172 - accuracy: 0.8101 - mae: 0.2697 - mse: 0.1337 - auc_13: 0.8875\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 2s 6ms/step - loss: 0.4142 - accuracy: 0.8121 - mae: 0.2658 - mse: 0.1324 - auc_13: 0.8897\n",
      "81\n",
      "81\n",
      "80\n",
      "80\n",
      "80\n",
      "80\n",
      "80\n",
      "79\n",
      "79\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 2s 6ms/step - loss: 0.4119 - accuracy: 0.8141 - mae: 0.2644 - mse: 0.1312 - auc_13: 0.8911\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4108 - accuracy: 0.8164 - mae: 0.2639 - mse: 0.1310 - auc_13: 0.8913\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4027 - accuracy: 0.8182 - mae: 0.2584 - mse: 0.1284 - auc_13: 0.8957\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4018 - accuracy: 0.8174 - mae: 0.2586 - mse: 0.1285 - auc_13: 0.8960\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 2s 6ms/step - loss: 0.3988 - accuracy: 0.8210 - mae: 0.2559 - mse: 0.1269 - auc_13: 0.8982\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3951 - accuracy: 0.8217 - mae: 0.2533 - mse: 0.1261 - auc_13: 0.8997\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3943 - accuracy: 0.8219 - mae: 0.2533 - mse: 0.1260 - auc_13: 0.9002\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3929 - accuracy: 0.8242 - mae: 0.2515 - mse: 0.1249 - auc_13: 0.9014\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3898 - accuracy: 0.8232 - mae: 0.2494 - mse: 0.1240 - auc_13: 0.9028\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 2s 6ms/step - loss: 0.3900 - accuracy: 0.8251 - mae: 0.2493 - mse: 0.1240 - auc_13: 0.9029\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3896 - accuracy: 0.8229 - mae: 0.2502 - mse: 0.1243 - auc_13: 0.9026\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3842 - accuracy: 0.8253 - mae: 0.2462 - mse: 0.1223 - auc_13: 0.9057\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3819 - accuracy: 0.8274 - mae: 0.2441 - mse: 0.1217 - auc_13: 0.9066\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3811 - accuracy: 0.8279 - mae: 0.2443 - mse: 0.1215 - auc_13: 0.9069\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3791 - accuracy: 0.8279 - mae: 0.2423 - mse: 0.1205 - auc_13: 0.9085\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3759 - accuracy: 0.8314 - mae: 0.2404 - mse: 0.1194 - auc_13: 0.9100\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3765 - accuracy: 0.8298 - mae: 0.2406 - mse: 0.1198 - auc_13: 0.9097\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3709 - accuracy: 0.8307 - mae: 0.2373 - mse: 0.1179 - auc_13: 0.9126\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3737 - accuracy: 0.8316 - mae: 0.2393 - mse: 0.1187 - auc_13: 0.9111\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3704 - accuracy: 0.8310 - mae: 0.2365 - mse: 0.1179 - auc_13: 0.9126\n",
      "83\n",
      "83\n",
      "82\n",
      "83\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3689 - accuracy: 0.8345 - mae: 0.2363 - mse: 0.1172 - auc_13: 0.9134\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3708 - accuracy: 0.8334 - mae: 0.2365 - mse: 0.1178 - auc_13: 0.9126\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3677 - accuracy: 0.8306 - mae: 0.2353 - mse: 0.1173 - auc_13: 0.9136\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3680 - accuracy: 0.8322 - mae: 0.2353 - mse: 0.1168 - auc_13: 0.9139\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3655 - accuracy: 0.8330 - mae: 0.2335 - mse: 0.1162 - auc_13: 0.9151\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3623 - accuracy: 0.8369 - mae: 0.2309 - mse: 0.1150 - auc_13: 0.9167\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3635 - accuracy: 0.8355 - mae: 0.2316 - mse: 0.1154 - auc_13: 0.9161\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3607 - accuracy: 0.8360 - mae: 0.2304 - mse: 0.1149 - auc_13: 0.9171\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3605 - accuracy: 0.8372 - mae: 0.2304 - mse: 0.1145 - auc_13: 0.9172\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3601 - accuracy: 0.8381 - mae: 0.2292 - mse: 0.1141 - auc_13: 0.9178\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "5\n",
      "121/121 [==============================] - 0s 835us/step - loss: 0.6200 - accuracy: 0.7418 - mae: 0.2835 - mse: 0.1853 - auc_13: 0.8251\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1035\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1168\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1173\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1007\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1187\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1179\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1265\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1055\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1091\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1048\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1169\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "class vector created!!\n",
      "class vector created!!\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1284\n",
      "Processing dgl graphs from scratch...\n",
      "Data created!!\n",
      "Data created!!\n",
      "train positive label: 19821 - train negative label: 14847\n",
      "Test positive label: 2047 - Test negative label: 1814\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.5446 - accuracy: 0.7365 - mae: 0.3626 - mse: 0.1814 - auc_14: 0.7953\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4911 - accuracy: 0.7753 - mae: 0.3213 - mse: 0.1601 - auc_14: 0.8377\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4811 - accuracy: 0.7802 - mae: 0.3143 - mse: 0.1562 - auc_14: 0.8454\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4759 - accuracy: 0.7826 - mae: 0.3108 - mse: 0.1546 - auc_14: 0.8487\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4684 - accuracy: 0.7859 - mae: 0.3053 - mse: 0.1518 - auc_14: 0.8543\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4632 - accuracy: 0.7885 - mae: 0.3017 - mse: 0.1501 - auc_14: 0.8572\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4595 - accuracy: 0.7896 - mae: 0.2987 - mse: 0.1486 - auc_14: 0.8604\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4544 - accuracy: 0.7915 - mae: 0.2947 - mse: 0.1468 - auc_14: 0.8639\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4514 - accuracy: 0.7933 - mae: 0.2931 - mse: 0.1456 - auc_14: 0.8662\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4497 - accuracy: 0.7959 - mae: 0.2909 - mse: 0.1447 - auc_14: 0.8676\n",
      "79\n",
      "79\n",
      "78\n",
      "78\n",
      "78\n",
      "78\n",
      "78\n",
      "77\n",
      "73\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4436 - accuracy: 0.7974 - mae: 0.2877 - mse: 0.1432 - auc_14: 0.8705\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4386 - accuracy: 0.8013 - mae: 0.2834 - mse: 0.1410 - auc_14: 0.8745\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4389 - accuracy: 0.8006 - mae: 0.2841 - mse: 0.1411 - auc_14: 0.8748\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4352 - accuracy: 0.8027 - mae: 0.2806 - mse: 0.1399 - auc_14: 0.8765\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4323 - accuracy: 0.8030 - mae: 0.2798 - mse: 0.1390 - auc_14: 0.8784\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4276 - accuracy: 0.8039 - mae: 0.2765 - mse: 0.1375 - auc_14: 0.8811\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4270 - accuracy: 0.8042 - mae: 0.2759 - mse: 0.1371 - auc_14: 0.8819\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4237 - accuracy: 0.8062 - mae: 0.2734 - mse: 0.1359 - auc_14: 0.8838\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4185 - accuracy: 0.8078 - mae: 0.2701 - mse: 0.1344 - auc_14: 0.8866\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4169 - accuracy: 0.8096 - mae: 0.2700 - mse: 0.1339 - auc_14: 0.8877\n",
      "80\n",
      "80\n",
      "80\n",
      "80\n",
      "80\n",
      "80\n",
      "80\n",
      "80\n",
      "79\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4145 - accuracy: 0.8104 - mae: 0.2678 - mse: 0.1331 - auc_14: 0.8888\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4121 - accuracy: 0.8127 - mae: 0.2653 - mse: 0.1318 - auc_14: 0.8905\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4109 - accuracy: 0.8131 - mae: 0.2645 - mse: 0.1316 - auc_14: 0.8913\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4062 - accuracy: 0.8150 - mae: 0.2615 - mse: 0.1297 - auc_14: 0.8943\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4051 - accuracy: 0.8152 - mae: 0.2605 - mse: 0.1297 - auc_14: 0.8944\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4026 - accuracy: 0.8165 - mae: 0.2589 - mse: 0.1286 - auc_14: 0.8959\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4024 - accuracy: 0.8173 - mae: 0.2586 - mse: 0.1284 - auc_14: 0.8962\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3988 - accuracy: 0.8178 - mae: 0.2568 - mse: 0.1276 - auc_14: 0.8977\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3965 - accuracy: 0.8198 - mae: 0.2547 - mse: 0.1270 - auc_14: 0.8989\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3959 - accuracy: 0.8212 - mae: 0.2540 - mse: 0.1262 - auc_14: 0.8998\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3946 - accuracy: 0.8212 - mae: 0.2535 - mse: 0.1256 - auc_14: 0.9006\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3929 - accuracy: 0.8213 - mae: 0.2517 - mse: 0.1254 - auc_14: 0.9012\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3895 - accuracy: 0.8241 - mae: 0.2502 - mse: 0.1240 - auc_14: 0.9030\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3887 - accuracy: 0.8242 - mae: 0.2492 - mse: 0.1242 - auc_14: 0.9031\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3889 - accuracy: 0.8239 - mae: 0.2501 - mse: 0.1240 - auc_14: 0.9034\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3859 - accuracy: 0.8242 - mae: 0.2475 - mse: 0.1231 - auc_14: 0.9047\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3859 - accuracy: 0.8261 - mae: 0.2466 - mse: 0.1230 - auc_14: 0.9047\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3836 - accuracy: 0.8257 - mae: 0.2461 - mse: 0.1222 - auc_14: 0.9060\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3840 - accuracy: 0.8253 - mae: 0.2457 - mse: 0.1225 - auc_14: 0.9056\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3804 - accuracy: 0.8273 - mae: 0.2442 - mse: 0.1213 - auc_14: 0.9077\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "4\n",
      "121/121 [==============================] - 0s 802us/step - loss: 0.5962 - accuracy: 0.7529 - mae: 0.2808 - mse: 0.1822 - auc_14: 0.8265\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1047\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1168\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1162\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1184\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1171\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1264\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Processing molecule 1000/1061\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1091\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1057\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1170\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "class vector created!!\n",
      "class vector created!!\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1284\n",
      "Processing dgl graphs from scratch...\n",
      "Data created!!\n",
      "Data created!!\n",
      "train positive label: 19739 - train negative label: 14929\n",
      "Test positive label: 2129 - Test negative label: 1732\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.5395 - accuracy: 0.7482 - mae: 0.3567 - mse: 0.1782 - auc_15: 0.7990\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.4939 - accuracy: 0.7725 - mae: 0.3245 - mse: 0.1612 - auc_15: 0.8362\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.4852 - accuracy: 0.7779 - mae: 0.3172 - mse: 0.1578 - auc_15: 0.8424\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.4757 - accuracy: 0.7804 - mae: 0.3113 - mse: 0.1546 - auc_15: 0.8491\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.4721 - accuracy: 0.7835 - mae: 0.3076 - mse: 0.1531 - auc_15: 0.8518\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.4661 - accuracy: 0.7842 - mae: 0.3042 - mse: 0.1513 - auc_15: 0.8558\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.4605 - accuracy: 0.7880 - mae: 0.3004 - mse: 0.1494 - auc_15: 0.8594\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.4586 - accuracy: 0.7902 - mae: 0.2986 - mse: 0.1485 - auc_15: 0.8608\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.4537 - accuracy: 0.7926 - mae: 0.2949 - mse: 0.1464 - auc_15: 0.8652\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.4496 - accuracy: 0.7935 - mae: 0.2919 - mse: 0.1454 - auc_15: 0.8669\n",
      "79\n",
      "79\n",
      "78\n",
      "78\n",
      "78\n",
      "78\n",
      "77\n",
      "77\n",
      "74\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.4474 - accuracy: 0.7938 - mae: 0.2910 - mse: 0.1446 - auc_15: 0.8685\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.4401 - accuracy: 0.7982 - mae: 0.2858 - mse: 0.1421 - auc_15: 0.8729\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.4399 - accuracy: 0.7973 - mae: 0.2856 - mse: 0.1420 - auc_15: 0.8736\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.4357 - accuracy: 0.7987 - mae: 0.2820 - mse: 0.1403 - auc_15: 0.8764\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.4316 - accuracy: 0.8014 - mae: 0.2799 - mse: 0.1393 - auc_15: 0.8784\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.4316 - accuracy: 0.8014 - mae: 0.2795 - mse: 0.1388 - auc_15: 0.8791\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.4267 - accuracy: 0.8030 - mae: 0.2769 - mse: 0.1376 - auc_15: 0.8814\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.4244 - accuracy: 0.8049 - mae: 0.2741 - mse: 0.1365 - auc_15: 0.8832\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.4233 - accuracy: 0.8052 - mae: 0.2741 - mse: 0.1363 - auc_15: 0.8837\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.4207 - accuracy: 0.8091 - mae: 0.2721 - mse: 0.1351 - auc_15: 0.8855\n",
      "80\n",
      "80\n",
      "80\n",
      "80\n",
      "80\n",
      "79\n",
      "79\n",
      "79\n",
      "79\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 3s 10ms/step - loss: 0.4160 - accuracy: 0.8092 - mae: 0.2686 - mse: 0.1335 - auc_15: 0.8882\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.4148 - accuracy: 0.8100 - mae: 0.2683 - mse: 0.1331 - auc_15: 0.8890\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.4112 - accuracy: 0.8114 - mae: 0.2647 - mse: 0.1319 - auc_15: 0.8909\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.4090 - accuracy: 0.8111 - mae: 0.2642 - mse: 0.1313 - auc_15: 0.8921\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.4102 - accuracy: 0.8111 - mae: 0.2648 - mse: 0.1315 - auc_15: 0.8918\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.4052 - accuracy: 0.8142 - mae: 0.2611 - mse: 0.1301 - auc_15: 0.8940\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.4029 - accuracy: 0.8145 - mae: 0.2603 - mse: 0.1294 - auc_15: 0.8953\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.4006 - accuracy: 0.8169 - mae: 0.2584 - mse: 0.1282 - auc_15: 0.8970\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.3974 - accuracy: 0.8172 - mae: 0.2560 - mse: 0.1272 - auc_15: 0.8988\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.3999 - accuracy: 0.8177 - mae: 0.2576 - mse: 0.1280 - auc_15: 0.8975\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "80\n",
      "80\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.3976 - accuracy: 0.8177 - mae: 0.2557 - mse: 0.1271 - auc_15: 0.8989\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.3933 - accuracy: 0.8200 - mae: 0.2532 - mse: 0.1257 - auc_15: 0.9010\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.3904 - accuracy: 0.8201 - mae: 0.2514 - mse: 0.1250 - auc_15: 0.9023\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.3889 - accuracy: 0.8220 - mae: 0.2493 - mse: 0.1241 - auc_15: 0.9035\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.3910 - accuracy: 0.8225 - mae: 0.2511 - mse: 0.1248 - auc_15: 0.9023\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.3876 - accuracy: 0.8222 - mae: 0.2490 - mse: 0.1241 - auc_15: 0.9037\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.3858 - accuracy: 0.8234 - mae: 0.2480 - mse: 0.1231 - auc_15: 0.9050\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.3832 - accuracy: 0.8249 - mae: 0.2458 - mse: 0.1225 - auc_15: 0.9061\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.3822 - accuracy: 0.8253 - mae: 0.2456 - mse: 0.1222 - auc_15: 0.9066\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.3820 - accuracy: 0.8245 - mae: 0.2449 - mse: 0.1221 - auc_15: 0.9067\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "81\n",
      "81\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.3801 - accuracy: 0.8262 - mae: 0.2444 - mse: 0.1216 - auc_15: 0.9075\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.3819 - accuracy: 0.8261 - mae: 0.2447 - mse: 0.1216 - auc_15: 0.9073\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.3755 - accuracy: 0.8271 - mae: 0.2406 - mse: 0.1199 - auc_15: 0.9099\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.3762 - accuracy: 0.8280 - mae: 0.2421 - mse: 0.1202 - auc_15: 0.9096\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.3751 - accuracy: 0.8296 - mae: 0.2400 - mse: 0.1196 - auc_15: 0.9104\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.3715 - accuracy: 0.8311 - mae: 0.2384 - mse: 0.1186 - auc_15: 0.9120\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.3694 - accuracy: 0.8311 - mae: 0.2366 - mse: 0.1181 - auc_15: 0.9129\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.3733 - accuracy: 0.8326 - mae: 0.2386 - mse: 0.1186 - auc_15: 0.9115\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.3702 - accuracy: 0.8312 - mae: 0.2375 - mse: 0.1181 - auc_15: 0.9126\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.3686 - accuracy: 0.8335 - mae: 0.2352 - mse: 0.1172 - auc_15: 0.9137\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 3s 10ms/step - loss: 0.3696 - accuracy: 0.8330 - mae: 0.2364 - mse: 0.1176 - auc_15: 0.9132\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 3s 10ms/step - loss: 0.3665 - accuracy: 0.8332 - mae: 0.2343 - mse: 0.1167 - auc_15: 0.9146\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 3s 10ms/step - loss: 0.3673 - accuracy: 0.8332 - mae: 0.2354 - mse: 0.1171 - auc_15: 0.9141\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.3642 - accuracy: 0.8363 - mae: 0.2327 - mse: 0.1160 - auc_15: 0.9155\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.3645 - accuracy: 0.8356 - mae: 0.2329 - mse: 0.1158 - auc_15: 0.9157\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.3616 - accuracy: 0.8370 - mae: 0.2314 - mse: 0.1153 - auc_15: 0.9166\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.3622 - accuracy: 0.8364 - mae: 0.2311 - mse: 0.1152 - auc_15: 0.9166\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.3599 - accuracy: 0.8370 - mae: 0.2297 - mse: 0.1146 - auc_15: 0.9176\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.3588 - accuracy: 0.8359 - mae: 0.2297 - mse: 0.1145 - auc_15: 0.9180\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 3s 9ms/step - loss: 0.3599 - accuracy: 0.8359 - mae: 0.2306 - mse: 0.1148 - auc_15: 0.9177\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "6\n",
      "121/121 [==============================] - 0s 1ms/step - loss: 0.5335 - accuracy: 0.7620 - mae: 0.3083 - mse: 0.1705 - auc_15: 0.8201\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1032\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1169\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1159\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1184\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1167\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1264\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1065\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1096\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1064\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1176\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "class vector created!!\n",
      "class vector created!!\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1284\n",
      "Processing dgl graphs from scratch...\n",
      "Data created!!\n",
      "Data created!!\n",
      "train positive label: 19520 - train negative label: 15148\n",
      "Test positive label: 2348 - Test negative label: 1513\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.5305 - accuracy: 0.7486 - mae: 0.3519 - mse: 0.1752 - auc_16: 0.8074\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4932 - accuracy: 0.7739 - mae: 0.3234 - mse: 0.1607 - auc_16: 0.8381\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4792 - accuracy: 0.7803 - mae: 0.3135 - mse: 0.1557 - auc_16: 0.8479\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4772 - accuracy: 0.7823 - mae: 0.3124 - mse: 0.1547 - auc_16: 0.8500\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4663 - accuracy: 0.7840 - mae: 0.3040 - mse: 0.1511 - auc_16: 0.8573\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4627 - accuracy: 0.7873 - mae: 0.3018 - mse: 0.1498 - auc_16: 0.8596\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.4562 - accuracy: 0.7902 - mae: 0.2968 - mse: 0.1475 - auc_16: 0.8635\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.4536 - accuracy: 0.7929 - mae: 0.2940 - mse: 0.1465 - auc_16: 0.8660\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.4490 - accuracy: 0.7935 - mae: 0.2920 - mse: 0.1448 - auc_16: 0.8692\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.4456 - accuracy: 0.7947 - mae: 0.2894 - mse: 0.1436 - auc_16: 0.8713\n",
      "79\n",
      "79\n",
      "79\n",
      "78\n",
      "78\n",
      "78\n",
      "78\n",
      "77\n",
      "74\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.4413 - accuracy: 0.7970 - mae: 0.2866 - mse: 0.1422 - auc_16: 0.8740\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4370 - accuracy: 0.7987 - mae: 0.2835 - mse: 0.1408 - auc_16: 0.8763\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4322 - accuracy: 0.8011 - mae: 0.2805 - mse: 0.1391 - auc_16: 0.8793\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4308 - accuracy: 0.8015 - mae: 0.2791 - mse: 0.1386 - auc_16: 0.8799\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4255 - accuracy: 0.8041 - mae: 0.2752 - mse: 0.1368 - auc_16: 0.8832\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4248 - accuracy: 0.8053 - mae: 0.2749 - mse: 0.1363 - auc_16: 0.8842\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4203 - accuracy: 0.8059 - mae: 0.2718 - mse: 0.1350 - auc_16: 0.8864\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4172 - accuracy: 0.8093 - mae: 0.2697 - mse: 0.1336 - auc_16: 0.8883\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4133 - accuracy: 0.8094 - mae: 0.2672 - mse: 0.1326 - auc_16: 0.8906\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4104 - accuracy: 0.8114 - mae: 0.2641 - mse: 0.1315 - auc_16: 0.8921\n",
      "80\n",
      "80\n",
      "80\n",
      "80\n",
      "80\n",
      "80\n",
      "80\n",
      "79\n",
      "79\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4094 - accuracy: 0.8128 - mae: 0.2651 - mse: 0.1312 - auc_16: 0.8927\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4055 - accuracy: 0.8145 - mae: 0.2610 - mse: 0.1300 - auc_16: 0.8946\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4045 - accuracy: 0.8133 - mae: 0.2610 - mse: 0.1295 - auc_16: 0.8955\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4017 - accuracy: 0.8168 - mae: 0.2586 - mse: 0.1285 - auc_16: 0.8968\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4003 - accuracy: 0.8182 - mae: 0.2575 - mse: 0.1278 - auc_16: 0.8980\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3987 - accuracy: 0.8182 - mae: 0.2566 - mse: 0.1275 - auc_16: 0.8986\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3941 - accuracy: 0.8194 - mae: 0.2533 - mse: 0.1257 - auc_16: 0.9014\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3954 - accuracy: 0.8221 - mae: 0.2538 - mse: 0.1259 - auc_16: 0.9006\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3901 - accuracy: 0.8218 - mae: 0.2509 - mse: 0.1246 - auc_16: 0.9033\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3900 - accuracy: 0.8234 - mae: 0.2509 - mse: 0.1245 - auc_16: 0.9030\n",
      "82\n",
      "82\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3904 - accuracy: 0.8203 - mae: 0.2508 - mse: 0.1247 - auc_16: 0.9032\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3882 - accuracy: 0.8230 - mae: 0.2495 - mse: 0.1240 - auc_16: 0.9041\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3854 - accuracy: 0.8240 - mae: 0.2471 - mse: 0.1230 - auc_16: 0.9055\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3831 - accuracy: 0.8242 - mae: 0.2462 - mse: 0.1223 - auc_16: 0.9068\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3820 - accuracy: 0.8262 - mae: 0.2453 - mse: 0.1221 - auc_16: 0.9069\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3807 - accuracy: 0.8271 - mae: 0.2443 - mse: 0.1212 - auc_16: 0.9081\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3804 - accuracy: 0.8265 - mae: 0.2435 - mse: 0.1213 - auc_16: 0.9081\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3781 - accuracy: 0.8292 - mae: 0.2423 - mse: 0.1205 - auc_16: 0.9093\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3789 - accuracy: 0.8281 - mae: 0.2434 - mse: 0.1208 - auc_16: 0.9090\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3733 - accuracy: 0.8318 - mae: 0.2385 - mse: 0.1185 - auc_16: 0.9120\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3734 - accuracy: 0.8302 - mae: 0.2391 - mse: 0.1189 - auc_16: 0.9117\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3773 - accuracy: 0.8288 - mae: 0.2409 - mse: 0.1202 - auc_16: 0.9098\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3742 - accuracy: 0.8294 - mae: 0.2401 - mse: 0.1190 - auc_16: 0.9116\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3710 - accuracy: 0.8329 - mae: 0.2375 - mse: 0.1177 - auc_16: 0.9132\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3670 - accuracy: 0.8354 - mae: 0.2345 - mse: 0.1163 - auc_16: 0.9151\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3681 - accuracy: 0.8323 - mae: 0.2357 - mse: 0.1172 - auc_16: 0.9143\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3682 - accuracy: 0.8328 - mae: 0.2355 - mse: 0.1170 - auc_16: 0.9144\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3657 - accuracy: 0.8352 - mae: 0.2343 - mse: 0.1160 - auc_16: 0.9156\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3674 - accuracy: 0.8341 - mae: 0.2350 - mse: 0.1169 - auc_16: 0.9146\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3644 - accuracy: 0.8337 - mae: 0.2334 - mse: 0.1161 - auc_16: 0.9158\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "82\n",
      "82\n",
      "83\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3622 - accuracy: 0.8346 - mae: 0.2318 - mse: 0.1156 - auc_16: 0.9167\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3599 - accuracy: 0.8372 - mae: 0.2310 - mse: 0.1146 - auc_16: 0.9180\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3617 - accuracy: 0.8354 - mae: 0.2312 - mse: 0.1153 - auc_16: 0.9170\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3631 - accuracy: 0.8358 - mae: 0.2324 - mse: 0.1151 - auc_16: 0.9169\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3583 - accuracy: 0.8382 - mae: 0.2286 - mse: 0.1138 - auc_16: 0.9189\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3585 - accuracy: 0.8383 - mae: 0.2293 - mse: 0.1138 - auc_16: 0.9188\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3567 - accuracy: 0.8384 - mae: 0.2279 - mse: 0.1135 - auc_16: 0.9195\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3566 - accuracy: 0.8397 - mae: 0.2279 - mse: 0.1131 - auc_16: 0.9198\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3559 - accuracy: 0.8403 - mae: 0.2278 - mse: 0.1132 - auc_16: 0.9198\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3556 - accuracy: 0.8396 - mae: 0.2271 - mse: 0.1131 - auc_16: 0.9200\n",
      "84\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3526 - accuracy: 0.8399 - mae: 0.2253 - mse: 0.1124 - auc_16: 0.9212\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3520 - accuracy: 0.8424 - mae: 0.2248 - mse: 0.1114 - auc_16: 0.9221\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3498 - accuracy: 0.8426 - mae: 0.2228 - mse: 0.1112 - auc_16: 0.9226\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3516 - accuracy: 0.8426 - mae: 0.2241 - mse: 0.1116 - auc_16: 0.9218\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3511 - accuracy: 0.8408 - mae: 0.2246 - mse: 0.1116 - auc_16: 0.9222\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3503 - accuracy: 0.8396 - mae: 0.2240 - mse: 0.1116 - auc_16: 0.9223\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3493 - accuracy: 0.8439 - mae: 0.2236 - mse: 0.1110 - auc_16: 0.9228\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3474 - accuracy: 0.8434 - mae: 0.2215 - mse: 0.1105 - auc_16: 0.9236\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3474 - accuracy: 0.8442 - mae: 0.2216 - mse: 0.1102 - auc_16: 0.9238\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3448 - accuracy: 0.8444 - mae: 0.2200 - mse: 0.1093 - auc_16: 0.9251\n",
      "84\n",
      "84\n",
      "84\n",
      "83\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "83\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3452 - accuracy: 0.8437 - mae: 0.2198 - mse: 0.1096 - auc_16: 0.9247\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3461 - accuracy: 0.8436 - mae: 0.2212 - mse: 0.1100 - auc_16: 0.9242\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3425 - accuracy: 0.8452 - mae: 0.2185 - mse: 0.1089 - auc_16: 0.9258\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3428 - accuracy: 0.8447 - mae: 0.2188 - mse: 0.1088 - auc_16: 0.9258\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3419 - accuracy: 0.8471 - mae: 0.2177 - mse: 0.1083 - auc_16: 0.9264\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3444 - accuracy: 0.8430 - mae: 0.2196 - mse: 0.1094 - auc_16: 0.9253\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3427 - accuracy: 0.8459 - mae: 0.2183 - mse: 0.1088 - auc_16: 0.9259\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3428 - accuracy: 0.8459 - mae: 0.2188 - mse: 0.1088 - auc_16: 0.9258\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3401 - accuracy: 0.8457 - mae: 0.2172 - mse: 0.1082 - auc_16: 0.9269\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3393 - accuracy: 0.8478 - mae: 0.2163 - mse: 0.1075 - auc_16: 0.9274\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "8\n",
      "121/121 [==============================] - 0s 1ms/step - loss: 0.6584 - accuracy: 0.7700 - mae: 0.2605 - mse: 0.1806 - auc_16: 0.8130\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1035\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1167\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1160\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1181\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1167\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1264\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1055\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1004\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1097\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1073\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1172\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "class vector created!!\n",
      "class vector created!!\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1284\n",
      "Processing dgl graphs from scratch...\n",
      "Data created!!\n",
      "Data created!!\n",
      "train positive label: 19537 - train negative label: 15131\n",
      "Test positive label: 2331 - Test negative label: 1530\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.5296 - accuracy: 0.7435 - mae: 0.3528 - mse: 0.1759 - auc_17: 0.8053\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4911 - accuracy: 0.7746 - mae: 0.3218 - mse: 0.1604 - auc_17: 0.8385\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4835 - accuracy: 0.7785 - mae: 0.3159 - mse: 0.1572 - auc_17: 0.8447\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4757 - accuracy: 0.7834 - mae: 0.3112 - mse: 0.1545 - auc_17: 0.8498\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4685 - accuracy: 0.7853 - mae: 0.3050 - mse: 0.1520 - auc_17: 0.8547\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4665 - accuracy: 0.7858 - mae: 0.3046 - mse: 0.1512 - auc_17: 0.8563\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4600 - accuracy: 0.7881 - mae: 0.2992 - mse: 0.1487 - auc_17: 0.8613\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4546 - accuracy: 0.7903 - mae: 0.2964 - mse: 0.1472 - auc_17: 0.8641\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4519 - accuracy: 0.7931 - mae: 0.2938 - mse: 0.1460 - auc_17: 0.8664\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4467 - accuracy: 0.7949 - mae: 0.2904 - mse: 0.1444 - auc_17: 0.8695\n",
      "79\n",
      "79\n",
      "78\n",
      "78\n",
      "78\n",
      "78\n",
      "77\n",
      "77\n",
      "74\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4420 - accuracy: 0.7958 - mae: 0.2869 - mse: 0.1428 - auc_17: 0.8727\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4387 - accuracy: 0.7981 - mae: 0.2848 - mse: 0.1415 - auc_17: 0.8749\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4355 - accuracy: 0.8011 - mae: 0.2827 - mse: 0.1401 - auc_17: 0.8772\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4305 - accuracy: 0.8027 - mae: 0.2787 - mse: 0.1385 - auc_17: 0.8801\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4282 - accuracy: 0.8037 - mae: 0.2772 - mse: 0.1376 - auc_17: 0.8820\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4236 - accuracy: 0.8053 - mae: 0.2739 - mse: 0.1361 - auc_17: 0.8845\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4220 - accuracy: 0.8069 - mae: 0.2728 - mse: 0.1353 - auc_17: 0.8856\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4178 - accuracy: 0.8083 - mae: 0.2705 - mse: 0.1341 - auc_17: 0.8878\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4132 - accuracy: 0.8099 - mae: 0.2672 - mse: 0.1328 - auc_17: 0.8901\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4126 - accuracy: 0.8106 - mae: 0.2657 - mse: 0.1323 - auc_17: 0.8906\n",
      "80\n",
      "80\n",
      "80\n",
      "80\n",
      "80\n",
      "80\n",
      "80\n",
      "79\n",
      "79\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4084 - accuracy: 0.8142 - mae: 0.2634 - mse: 0.1309 - auc_17: 0.8932\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4064 - accuracy: 0.8122 - mae: 0.2623 - mse: 0.1303 - auc_17: 0.8944\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4051 - accuracy: 0.8137 - mae: 0.2614 - mse: 0.1297 - auc_17: 0.8952\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4010 - accuracy: 0.8176 - mae: 0.2585 - mse: 0.1283 - auc_17: 0.8974\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.4004 - accuracy: 0.8161 - mae: 0.2582 - mse: 0.1282 - auc_17: 0.8977\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3999 - accuracy: 0.8162 - mae: 0.2577 - mse: 0.1281 - auc_17: 0.8979\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3955 - accuracy: 0.8188 - mae: 0.2545 - mse: 0.1262 - auc_17: 0.9006\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3933 - accuracy: 0.8195 - mae: 0.2531 - mse: 0.1257 - auc_17: 0.9015\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3914 - accuracy: 0.8201 - mae: 0.2518 - mse: 0.1251 - auc_17: 0.9024\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3895 - accuracy: 0.8227 - mae: 0.2502 - mse: 0.1243 - auc_17: 0.9035\n",
      "82\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3880 - accuracy: 0.8229 - mae: 0.2498 - mse: 0.1238 - auc_17: 0.9043\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3848 - accuracy: 0.8253 - mae: 0.2469 - mse: 0.1226 - auc_17: 0.9061\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3828 - accuracy: 0.8254 - mae: 0.2459 - mse: 0.1221 - auc_17: 0.9069\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3837 - accuracy: 0.8247 - mae: 0.2462 - mse: 0.1224 - auc_17: 0.9065\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3805 - accuracy: 0.8263 - mae: 0.2447 - mse: 0.1216 - auc_17: 0.9079\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3782 - accuracy: 0.8278 - mae: 0.2429 - mse: 0.1209 - auc_17: 0.9089\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3790 - accuracy: 0.8274 - mae: 0.2438 - mse: 0.1210 - auc_17: 0.9086\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3752 - accuracy: 0.8294 - mae: 0.2407 - mse: 0.1199 - auc_17: 0.9102\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3745 - accuracy: 0.8292 - mae: 0.2400 - mse: 0.1194 - auc_17: 0.9110\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3705 - accuracy: 0.8309 - mae: 0.2375 - mse: 0.1179 - auc_17: 0.9131\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3726 - accuracy: 0.8317 - mae: 0.2385 - mse: 0.1184 - auc_17: 0.9122\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3716 - accuracy: 0.8299 - mae: 0.2384 - mse: 0.1185 - auc_17: 0.9124\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3700 - accuracy: 0.8312 - mae: 0.2373 - mse: 0.1180 - auc_17: 0.9130\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3705 - accuracy: 0.8313 - mae: 0.2374 - mse: 0.1182 - auc_17: 0.9128\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3684 - accuracy: 0.8316 - mae: 0.2360 - mse: 0.1176 - auc_17: 0.9137\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3648 - accuracy: 0.8345 - mae: 0.2337 - mse: 0.1161 - auc_17: 0.9157\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3643 - accuracy: 0.8339 - mae: 0.2338 - mse: 0.1163 - auc_17: 0.9158\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3647 - accuracy: 0.8339 - mae: 0.2328 - mse: 0.1159 - auc_17: 0.9160\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3637 - accuracy: 0.8358 - mae: 0.2332 - mse: 0.1157 - auc_17: 0.9163\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3612 - accuracy: 0.8368 - mae: 0.2309 - mse: 0.1150 - auc_17: 0.9172\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "82\n",
      "83\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3604 - accuracy: 0.8350 - mae: 0.2306 - mse: 0.1149 - auc_17: 0.9177\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3604 - accuracy: 0.8367 - mae: 0.2308 - mse: 0.1148 - auc_17: 0.9177\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3566 - accuracy: 0.8385 - mae: 0.2287 - mse: 0.1137 - auc_17: 0.9193\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3580 - accuracy: 0.8368 - mae: 0.2288 - mse: 0.1140 - auc_17: 0.9188\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3587 - accuracy: 0.8396 - mae: 0.2289 - mse: 0.1138 - auc_17: 0.9187\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3564 - accuracy: 0.8377 - mae: 0.2284 - mse: 0.1136 - auc_17: 0.9194\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3551 - accuracy: 0.8391 - mae: 0.2268 - mse: 0.1130 - auc_17: 0.9202\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3547 - accuracy: 0.8384 - mae: 0.2274 - mse: 0.1131 - auc_17: 0.9202\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3522 - accuracy: 0.8394 - mae: 0.2254 - mse: 0.1123 - auc_17: 0.9214\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 2s 7ms/step - loss: 0.3550 - accuracy: 0.8397 - mae: 0.2273 - mse: 0.1132 - auc_17: 0.9200\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "6\n",
      "121/121 [==============================] - 0s 960us/step - loss: 0.5924 - accuracy: 0.7767 - mae: 0.2686 - mse: 0.1713 - auc_17: 0.8194\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1036\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1164\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1161\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1188\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1175\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1263\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1061\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1090\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1053\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1174\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "class vector created!!\n",
      "class vector created!!\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1284\n",
      "Processing dgl graphs from scratch...\n",
      "Data created!!\n",
      "Data created!!\n",
      "train positive label: 19640 - train negative label: 15028\n",
      "Test positive label: 2228 - Test negative label: 1633\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.5355 - accuracy: 0.7438 - mae: 0.3541 - mse: 0.1770 - auc_18: 0.8046\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4943 - accuracy: 0.7714 - mae: 0.3241 - mse: 0.1614 - auc_18: 0.8366\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.4853 - accuracy: 0.7784 - mae: 0.3174 - mse: 0.1577 - auc_18: 0.8438\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4774 - accuracy: 0.7814 - mae: 0.3116 - mse: 0.1551 - auc_18: 0.8486\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4719 - accuracy: 0.7832 - mae: 0.3080 - mse: 0.1531 - auc_18: 0.8523\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4656 - accuracy: 0.7879 - mae: 0.3037 - mse: 0.1507 - auc_18: 0.8569\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4609 - accuracy: 0.7904 - mae: 0.2992 - mse: 0.1489 - auc_18: 0.8601\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.4575 - accuracy: 0.7901 - mae: 0.2975 - mse: 0.1479 - auc_18: 0.8625\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.4520 - accuracy: 0.7938 - mae: 0.2934 - mse: 0.1459 - auc_18: 0.8662\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.4479 - accuracy: 0.7953 - mae: 0.2910 - mse: 0.1444 - auc_18: 0.8690\n",
      "79\n",
      "79\n",
      "79\n",
      "78\n",
      "78\n",
      "78\n",
      "77\n",
      "77\n",
      "74\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.4413 - accuracy: 0.8002 - mae: 0.2861 - mse: 0.1422 - auc_18: 0.8731\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.4389 - accuracy: 0.7973 - mae: 0.2849 - mse: 0.1417 - auc_18: 0.8743\n",
      "Epoch 3/10\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "271/271 [==============================] - 2s 9ms/step - loss: 0.4341 - accuracy: 0.8022 - mae: 0.2811 - mse: 0.1397 - auc_18: 0.8777\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.4301 - accuracy: 0.8034 - mae: 0.2787 - mse: 0.1385 - auc_18: 0.8800\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.4283 - accuracy: 0.8050 - mae: 0.2768 - mse: 0.1376 - auc_18: 0.8814\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.4256 - accuracy: 0.8042 - mae: 0.2750 - mse: 0.1368 - auc_18: 0.8831\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.4203 - accuracy: 0.8092 - mae: 0.2716 - mse: 0.1349 - auc_18: 0.8862\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.4191 - accuracy: 0.8089 - mae: 0.2702 - mse: 0.1342 - auc_18: 0.8872\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.4141 - accuracy: 0.8117 - mae: 0.2668 - mse: 0.1324 - auc_18: 0.8903\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.4129 - accuracy: 0.8118 - mae: 0.2662 - mse: 0.1323 - auc_18: 0.8905\n",
      "81\n",
      "80\n",
      "80\n",
      "80\n",
      "80\n",
      "80\n",
      "80\n",
      "79\n",
      "80\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.4097 - accuracy: 0.8140 - mae: 0.2637 - mse: 0.1313 - auc_18: 0.8918\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4082 - accuracy: 0.8155 - mae: 0.2627 - mse: 0.1302 - auc_18: 0.8937\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4061 - accuracy: 0.8160 - mae: 0.2618 - mse: 0.1299 - auc_18: 0.8943\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4010 - accuracy: 0.8169 - mae: 0.2581 - mse: 0.1284 - auc_18: 0.8969\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.4010 - accuracy: 0.8163 - mae: 0.2576 - mse: 0.1281 - auc_18: 0.8973\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3991 - accuracy: 0.8180 - mae: 0.2572 - mse: 0.1276 - auc_18: 0.8983\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3947 - accuracy: 0.8210 - mae: 0.2532 - mse: 0.1259 - auc_18: 0.9007\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3951 - accuracy: 0.8207 - mae: 0.2535 - mse: 0.1263 - auc_18: 0.9001\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3914 - accuracy: 0.8221 - mae: 0.2508 - mse: 0.1250 - auc_18: 0.9023\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3887 - accuracy: 0.8233 - mae: 0.2493 - mse: 0.1242 - auc_18: 0.9035\n",
      "82\n",
      "82\n",
      "82\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3872 - accuracy: 0.8248 - mae: 0.2481 - mse: 0.1234 - auc_18: 0.9043\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3868 - accuracy: 0.8251 - mae: 0.2476 - mse: 0.1231 - auc_18: 0.9050\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3834 - accuracy: 0.8261 - mae: 0.2458 - mse: 0.1222 - auc_18: 0.9065\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3828 - accuracy: 0.8270 - mae: 0.2453 - mse: 0.1219 - auc_18: 0.9067\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3806 - accuracy: 0.8273 - mae: 0.2440 - mse: 0.1213 - auc_18: 0.9078\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3790 - accuracy: 0.8275 - mae: 0.2428 - mse: 0.1209 - auc_18: 0.9085\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3772 - accuracy: 0.8281 - mae: 0.2417 - mse: 0.1204 - auc_18: 0.9092\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3764 - accuracy: 0.8284 - mae: 0.2415 - mse: 0.1199 - auc_18: 0.9098\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3735 - accuracy: 0.8311 - mae: 0.2391 - mse: 0.1188 - auc_18: 0.9115\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3755 - accuracy: 0.8300 - mae: 0.2404 - mse: 0.1195 - auc_18: 0.9104\n",
      "83\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3732 - accuracy: 0.8294 - mae: 0.2386 - mse: 0.1190 - auc_18: 0.9114\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3716 - accuracy: 0.8321 - mae: 0.2376 - mse: 0.1181 - auc_18: 0.9125\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3709 - accuracy: 0.8318 - mae: 0.2370 - mse: 0.1183 - auc_18: 0.9124\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3666 - accuracy: 0.8323 - mae: 0.2350 - mse: 0.1167 - auc_18: 0.9147\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3688 - accuracy: 0.8333 - mae: 0.2354 - mse: 0.1173 - auc_18: 0.9137\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3689 - accuracy: 0.8341 - mae: 0.2362 - mse: 0.1174 - auc_18: 0.9136\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3670 - accuracy: 0.8341 - mae: 0.2343 - mse: 0.1168 - auc_18: 0.9145\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3646 - accuracy: 0.8347 - mae: 0.2327 - mse: 0.1159 - auc_18: 0.9156\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3640 - accuracy: 0.8366 - mae: 0.2322 - mse: 0.1156 - auc_18: 0.9160\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3605 - accuracy: 0.8376 - mae: 0.2303 - mse: 0.1147 - auc_18: 0.9174\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "82\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3615 - accuracy: 0.8367 - mae: 0.2315 - mse: 0.1149 - auc_18: 0.9173\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3621 - accuracy: 0.8362 - mae: 0.2305 - mse: 0.1152 - auc_18: 0.9167\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3590 - accuracy: 0.8368 - mae: 0.2295 - mse: 0.1144 - auc_18: 0.9182\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3590 - accuracy: 0.8373 - mae: 0.2296 - mse: 0.1144 - auc_18: 0.9181\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3562 - accuracy: 0.8378 - mae: 0.2272 - mse: 0.1137 - auc_18: 0.9192\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3578 - accuracy: 0.8382 - mae: 0.2287 - mse: 0.1138 - auc_18: 0.9188\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3567 - accuracy: 0.8377 - mae: 0.2285 - mse: 0.1136 - auc_18: 0.9192\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3539 - accuracy: 0.8411 - mae: 0.2262 - mse: 0.1123 - auc_18: 0.9208\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3539 - accuracy: 0.8409 - mae: 0.2253 - mse: 0.1120 - auc_18: 0.9209\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3526 - accuracy: 0.8420 - mae: 0.2245 - mse: 0.1119 - auc_18: 0.9213\n",
      "84\n",
      "84\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3507 - accuracy: 0.8407 - mae: 0.2238 - mse: 0.1112 - auc_18: 0.9223\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3480 - accuracy: 0.8423 - mae: 0.2216 - mse: 0.1106 - auc_18: 0.9232\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3519 - accuracy: 0.8403 - mae: 0.2245 - mse: 0.1119 - auc_18: 0.9215\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3489 - accuracy: 0.8448 - mae: 0.2220 - mse: 0.1107 - auc_18: 0.9228\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3494 - accuracy: 0.8415 - mae: 0.2232 - mse: 0.1108 - auc_18: 0.9229\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3491 - accuracy: 0.8394 - mae: 0.2222 - mse: 0.1108 - auc_18: 0.9230\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3476 - accuracy: 0.8428 - mae: 0.2217 - mse: 0.1102 - auc_18: 0.9237\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3477 - accuracy: 0.8422 - mae: 0.2218 - mse: 0.1104 - auc_18: 0.9236\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3452 - accuracy: 0.8456 - mae: 0.2197 - mse: 0.1093 - auc_18: 0.9246\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3452 - accuracy: 0.8455 - mae: 0.2192 - mse: 0.1094 - auc_18: 0.9246\n",
      "84\n",
      "84\n",
      "84\n",
      "83\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3434 - accuracy: 0.8440 - mae: 0.2191 - mse: 0.1092 - auc_18: 0.9253\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3451 - accuracy: 0.8441 - mae: 0.2199 - mse: 0.1096 - auc_18: 0.9246\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3427 - accuracy: 0.8441 - mae: 0.2186 - mse: 0.1092 - auc_18: 0.9253\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3450 - accuracy: 0.8430 - mae: 0.2195 - mse: 0.1096 - auc_18: 0.9246\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3432 - accuracy: 0.8443 - mae: 0.2191 - mse: 0.1087 - auc_18: 0.9256\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3396 - accuracy: 0.8462 - mae: 0.2166 - mse: 0.1079 - auc_18: 0.9271\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3403 - accuracy: 0.8445 - mae: 0.2173 - mse: 0.1084 - auc_18: 0.9266\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3399 - accuracy: 0.8465 - mae: 0.2162 - mse: 0.1079 - auc_18: 0.9269\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.3393 - accuracy: 0.8458 - mae: 0.2153 - mse: 0.1077 - auc_18: 0.9271\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3365 - accuracy: 0.8479 - mae: 0.2146 - mse: 0.1068 - auc_18: 0.9285\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "8\n",
      "121/121 [==============================] - 0s 984us/step - loss: 0.5608 - accuracy: 0.7659 - mae: 0.2841 - mse: 0.1720 - auc_18: 0.8278\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1034\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1162\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1161\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1182\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1174\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1265\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1057\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1093\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1055\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1171\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "class vector created!!\n",
      "class vector created!!\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1284\n",
      "Processing dgl graphs from scratch...\n",
      "Data created!!\n",
      "Data created!!\n",
      "train positive label: 19638 - train negative label: 15030\n",
      "Test positive label: 2230 - Test negative label: 1631\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.5369 - accuracy: 0.7380 - mae: 0.3565 - mse: 0.1783 - auc_19: 0.7997\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.4978 - accuracy: 0.7707 - mae: 0.3263 - mse: 0.1627 - auc_19: 0.8334\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4869 - accuracy: 0.7744 - mae: 0.3191 - mse: 0.1588 - auc_19: 0.8415\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.4809 - accuracy: 0.7781 - mae: 0.3147 - mse: 0.1567 - auc_19: 0.8453\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4734 - accuracy: 0.7814 - mae: 0.3093 - mse: 0.1540 - auc_19: 0.8507\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4705 - accuracy: 0.7819 - mae: 0.3066 - mse: 0.1526 - auc_19: 0.8536\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4660 - accuracy: 0.7843 - mae: 0.3036 - mse: 0.1510 - auc_19: 0.8567\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4615 - accuracy: 0.7866 - mae: 0.3012 - mse: 0.1498 - auc_19: 0.8590\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.4571 - accuracy: 0.7901 - mae: 0.2975 - mse: 0.1480 - auc_19: 0.8623\n",
      "Epoch 10/10\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4534 - accuracy: 0.7921 - mae: 0.2952 - mse: 0.1468 - auc_19: 0.8646\n",
      "79\n",
      "78\n",
      "78\n",
      "78\n",
      "78\n",
      "77\n",
      "77\n",
      "77\n",
      "73\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4471 - accuracy: 0.7959 - mae: 0.2905 - mse: 0.1443 - auc_19: 0.8692\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4474 - accuracy: 0.7972 - mae: 0.2895 - mse: 0.1438 - auc_19: 0.8697\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.4414 - accuracy: 0.7971 - mae: 0.2862 - mse: 0.1426 - auc_19: 0.8729\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.4382 - accuracy: 0.8005 - mae: 0.2838 - mse: 0.1411 - auc_19: 0.8752\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4358 - accuracy: 0.8006 - mae: 0.2825 - mse: 0.1402 - auc_19: 0.8770\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 2s 9ms/step - loss: 0.4313 - accuracy: 0.8045 - mae: 0.2785 - mse: 0.1385 - auc_19: 0.8796\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4294 - accuracy: 0.8049 - mae: 0.2771 - mse: 0.1379 - auc_19: 0.8806\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4275 - accuracy: 0.8063 - mae: 0.2758 - mse: 0.1373 - auc_19: 0.8816\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4242 - accuracy: 0.8066 - mae: 0.2738 - mse: 0.1362 - auc_19: 0.8840\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4188 - accuracy: 0.8114 - mae: 0.2704 - mse: 0.1340 - auc_19: 0.8873\n",
      "80\n",
      "80\n",
      "80\n",
      "80\n",
      "80\n",
      "80\n",
      "79\n",
      "79\n",
      "79\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4186 - accuracy: 0.8092 - mae: 0.2695 - mse: 0.1339 - auc_19: 0.8876\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4164 - accuracy: 0.8106 - mae: 0.2692 - mse: 0.1336 - auc_19: 0.8882\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4127 - accuracy: 0.8130 - mae: 0.2656 - mse: 0.1321 - auc_19: 0.8905\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4096 - accuracy: 0.8148 - mae: 0.2630 - mse: 0.1309 - auc_19: 0.8924\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4072 - accuracy: 0.8144 - mae: 0.2621 - mse: 0.1303 - auc_19: 0.8937\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4037 - accuracy: 0.8176 - mae: 0.2598 - mse: 0.1291 - auc_19: 0.8956\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4047 - accuracy: 0.8181 - mae: 0.2599 - mse: 0.1292 - auc_19: 0.8950\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.4028 - accuracy: 0.8166 - mae: 0.2585 - mse: 0.1288 - auc_19: 0.8961\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3985 - accuracy: 0.8180 - mae: 0.2562 - mse: 0.1275 - auc_19: 0.8983\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3984 - accuracy: 0.8208 - mae: 0.2555 - mse: 0.1271 - auc_19: 0.8985\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "80\n",
      "Epoch 1/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3955 - accuracy: 0.8205 - mae: 0.2540 - mse: 0.1264 - auc_19: 0.8999\n",
      "Epoch 2/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3926 - accuracy: 0.8215 - mae: 0.2521 - mse: 0.1251 - auc_19: 0.9017\n",
      "Epoch 3/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3923 - accuracy: 0.8210 - mae: 0.2511 - mse: 0.1252 - auc_19: 0.9018\n",
      "Epoch 4/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3886 - accuracy: 0.8242 - mae: 0.2493 - mse: 0.1237 - auc_19: 0.9037\n",
      "Epoch 5/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3867 - accuracy: 0.8250 - mae: 0.2476 - mse: 0.1231 - auc_19: 0.9047\n",
      "Epoch 6/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3869 - accuracy: 0.8246 - mae: 0.2475 - mse: 0.1233 - auc_19: 0.9046\n",
      "Epoch 7/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3867 - accuracy: 0.8254 - mae: 0.2483 - mse: 0.1230 - auc_19: 0.9049\n",
      "Epoch 8/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3810 - accuracy: 0.8274 - mae: 0.2442 - mse: 0.1214 - auc_19: 0.9076\n",
      "Epoch 9/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3807 - accuracy: 0.8274 - mae: 0.2440 - mse: 0.1213 - auc_19: 0.9077\n",
      "Epoch 10/10\n",
      "271/271 [==============================] - 2s 8ms/step - loss: 0.3789 - accuracy: 0.8269 - mae: 0.2432 - mse: 0.1207 - auc_19: 0.9087\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "4\n",
      "121/121 [==============================] - 0s 1ms/step - loss: 0.5388 - accuracy: 0.7843 - mae: 0.2664 - mse: 0.1607 - auc_19: 0.8391\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1040\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1167\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1163\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1003\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1183\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1174\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1265\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1074\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1012\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1098\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1062\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1176\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "class vector created!!\n",
      "class vector created!!\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1285\n",
      "Processing dgl graphs from scratch...\n",
      "Data created!!\n",
      "Data created!!\n",
      "train positive label: 19567 - train negative label: 15128\n",
      "Test positive label: 2301 - Test negative label: 1533\n",
      "Epoch 1/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.5312 - accuracy: 0.7515 - mae: 0.3533 - mse: 0.1756 - auc_20: 0.8072\n",
      "Epoch 2/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.4930 - accuracy: 0.7746 - mae: 0.3233 - mse: 0.1607 - auc_20: 0.8381\n",
      "Epoch 3/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.4843 - accuracy: 0.7774 - mae: 0.3168 - mse: 0.1573 - auc_20: 0.8442\n",
      "Epoch 4/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.4754 - accuracy: 0.7808 - mae: 0.3105 - mse: 0.1545 - auc_20: 0.8500\n",
      "Epoch 5/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.4724 - accuracy: 0.7830 - mae: 0.3088 - mse: 0.1532 - auc_20: 0.8526\n",
      "Epoch 6/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.4702 - accuracy: 0.7846 - mae: 0.3074 - mse: 0.1525 - auc_20: 0.8540\n",
      "Epoch 7/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.4647 - accuracy: 0.7864 - mae: 0.3029 - mse: 0.1507 - auc_20: 0.8575\n",
      "Epoch 8/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.4614 - accuracy: 0.7883 - mae: 0.3004 - mse: 0.1495 - auc_20: 0.8593\n",
      "Epoch 9/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.4564 - accuracy: 0.7903 - mae: 0.2966 - mse: 0.1476 - auc_20: 0.8633\n",
      "Epoch 10/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.4535 - accuracy: 0.7917 - mae: 0.2959 - mse: 0.1469 - auc_20: 0.8649\n",
      "79\n",
      "78\n",
      "78\n",
      "78\n",
      "78\n",
      "78\n",
      "77\n",
      "77\n",
      "75\n",
      "Epoch 1/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.4500 - accuracy: 0.7935 - mae: 0.2920 - mse: 0.1454 - auc_20: 0.8675\n",
      "Epoch 2/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.4471 - accuracy: 0.7954 - mae: 0.2909 - mse: 0.1444 - auc_20: 0.8695\n",
      "Epoch 3/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.4422 - accuracy: 0.7979 - mae: 0.2875 - mse: 0.1427 - auc_20: 0.8725\n",
      "Epoch 4/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.4392 - accuracy: 0.7969 - mae: 0.2853 - mse: 0.1417 - auc_20: 0.8747\n",
      "Epoch 5/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.4360 - accuracy: 0.7993 - mae: 0.2837 - mse: 0.1406 - auc_20: 0.8766\n",
      "Epoch 6/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.4321 - accuracy: 0.8004 - mae: 0.2798 - mse: 0.1394 - auc_20: 0.8789\n",
      "Epoch 7/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.4274 - accuracy: 0.8043 - mae: 0.2768 - mse: 0.1374 - auc_20: 0.8820\n",
      "Epoch 8/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.4281 - accuracy: 0.8042 - mae: 0.2767 - mse: 0.1378 - auc_20: 0.8814\n",
      "Epoch 9/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.4208 - accuracy: 0.8096 - mae: 0.2717 - mse: 0.1349 - auc_20: 0.8860\n",
      "Epoch 10/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.4202 - accuracy: 0.8093 - mae: 0.2708 - mse: 0.1346 - auc_20: 0.8867\n",
      "80\n",
      "80\n",
      "80\n",
      "80\n",
      "79\n",
      "79\n",
      "79\n",
      "79\n",
      "79\n",
      "Epoch 1/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.4175 - accuracy: 0.8088 - mae: 0.2690 - mse: 0.1340 - auc_20: 0.8879\n",
      "Epoch 2/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.4161 - accuracy: 0.8103 - mae: 0.2692 - mse: 0.1335 - auc_20: 0.8886\n",
      "Epoch 3/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.4103 - accuracy: 0.8109 - mae: 0.2648 - mse: 0.1315 - auc_20: 0.8921\n",
      "Epoch 4/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.4088 - accuracy: 0.8127 - mae: 0.2634 - mse: 0.1310 - auc_20: 0.8928\n",
      "Epoch 5/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.4066 - accuracy: 0.8133 - mae: 0.2635 - mse: 0.1304 - auc_20: 0.8940\n",
      "Epoch 6/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.4046 - accuracy: 0.8149 - mae: 0.2602 - mse: 0.1295 - auc_20: 0.8953\n",
      "Epoch 7/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.4023 - accuracy: 0.8162 - mae: 0.2589 - mse: 0.1287 - auc_20: 0.8967\n",
      "Epoch 8/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.4005 - accuracy: 0.8175 - mae: 0.2576 - mse: 0.1282 - auc_20: 0.8975\n",
      "Epoch 9/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.3972 - accuracy: 0.8185 - mae: 0.2553 - mse: 0.1269 - auc_20: 0.8993\n",
      "Epoch 10/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.3970 - accuracy: 0.8185 - mae: 0.2555 - mse: 0.1269 - auc_20: 0.8993\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "80\n",
      "Epoch 1/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.3944 - accuracy: 0.8209 - mae: 0.2526 - mse: 0.1260 - auc_20: 0.9006\n",
      "Epoch 2/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.3910 - accuracy: 0.8218 - mae: 0.2510 - mse: 0.1248 - auc_20: 0.9027\n",
      "Epoch 3/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.3903 - accuracy: 0.8218 - mae: 0.2508 - mse: 0.1247 - auc_20: 0.9028\n",
      "Epoch 4/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.3913 - accuracy: 0.8219 - mae: 0.2512 - mse: 0.1249 - auc_20: 0.9026\n",
      "Epoch 5/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.3851 - accuracy: 0.8261 - mae: 0.2465 - mse: 0.1229 - auc_20: 0.9056\n",
      "Epoch 6/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.3851 - accuracy: 0.8241 - mae: 0.2478 - mse: 0.1232 - auc_20: 0.9053\n",
      "Epoch 7/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.3831 - accuracy: 0.8246 - mae: 0.2462 - mse: 0.1222 - auc_20: 0.9068\n",
      "Epoch 8/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.3842 - accuracy: 0.8255 - mae: 0.2463 - mse: 0.1225 - auc_20: 0.9060\n",
      "Epoch 9/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.3835 - accuracy: 0.8259 - mae: 0.2451 - mse: 0.1217 - auc_20: 0.9069\n",
      "Epoch 10/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.3795 - accuracy: 0.8280 - mae: 0.2428 - mse: 0.1209 - auc_20: 0.9086\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "4\n",
      "120/120 [==============================] - 0s 1ms/step - loss: 0.6197 - accuracy: 0.7660 - mae: 0.2670 - mse: 0.1751 - auc_20: 0.8311\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1029\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1169\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1159\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1191\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1172\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1264\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1059\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1093\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1053\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1172\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "class vector created!!\n",
      "class vector created!!\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1285\n",
      "Processing dgl graphs from scratch...\n",
      "Data created!!\n",
      "Data created!!\n",
      "train positive label: 19691 - train negative label: 15004\n",
      "Test positive label: 2177 - Test negative label: 1657\n",
      "Epoch 1/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.5477 - accuracy: 0.7307 - mae: 0.3685 - mse: 0.1834 - auc_21: 0.7858\n",
      "Epoch 2/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.4980 - accuracy: 0.7716 - mae: 0.3267 - mse: 0.1628 - auc_21: 0.8329\n",
      "Epoch 3/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.4866 - accuracy: 0.7782 - mae: 0.3195 - mse: 0.1584 - auc_21: 0.8415\n",
      "Epoch 4/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.4789 - accuracy: 0.7819 - mae: 0.3135 - mse: 0.1557 - auc_21: 0.8467\n",
      "Epoch 5/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.4729 - accuracy: 0.7834 - mae: 0.3103 - mse: 0.1539 - auc_21: 0.8504\n",
      "Epoch 6/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.4662 - accuracy: 0.7857 - mae: 0.3048 - mse: 0.1513 - auc_21: 0.8559\n",
      "Epoch 7/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.4618 - accuracy: 0.7883 - mae: 0.3017 - mse: 0.1498 - auc_21: 0.8587\n",
      "Epoch 8/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.4607 - accuracy: 0.7921 - mae: 0.3002 - mse: 0.1487 - auc_21: 0.8603\n",
      "Epoch 9/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.4537 - accuracy: 0.7925 - mae: 0.2948 - mse: 0.1466 - auc_21: 0.8647\n",
      "Epoch 10/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.4499 - accuracy: 0.7939 - mae: 0.2935 - mse: 0.1453 - auc_21: 0.8680\n",
      "79\n",
      "79\n",
      "78\n",
      "78\n",
      "78\n",
      "78\n",
      "77\n",
      "77\n",
      "73\n",
      "Epoch 1/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.4453 - accuracy: 0.7967 - mae: 0.2893 - mse: 0.1438 - auc_21: 0.8704\n",
      "Epoch 2/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.4410 - accuracy: 0.7959 - mae: 0.2866 - mse: 0.1423 - auc_21: 0.8733\n",
      "Epoch 3/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.4393 - accuracy: 0.7985 - mae: 0.2857 - mse: 0.1416 - auc_21: 0.8746\n",
      "Epoch 4/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.4339 - accuracy: 0.8000 - mae: 0.2821 - mse: 0.1398 - auc_21: 0.8780\n",
      "Epoch 5/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.4299 - accuracy: 0.8035 - mae: 0.2778 - mse: 0.1380 - auc_21: 0.8808\n",
      "Epoch 6/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.4244 - accuracy: 0.8063 - mae: 0.2748 - mse: 0.1367 - auc_21: 0.8829\n",
      "Epoch 7/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.4246 - accuracy: 0.8055 - mae: 0.2751 - mse: 0.1365 - auc_21: 0.8837\n",
      "Epoch 8/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.4214 - accuracy: 0.8080 - mae: 0.2722 - mse: 0.1352 - auc_21: 0.8857\n",
      "Epoch 9/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.4181 - accuracy: 0.8094 - mae: 0.2707 - mse: 0.1343 - auc_21: 0.8870\n",
      "Epoch 10/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.4145 - accuracy: 0.8105 - mae: 0.2672 - mse: 0.1329 - auc_21: 0.8896\n",
      "80\n",
      "80\n",
      "80\n",
      "80\n",
      "80\n",
      "79\n",
      "79\n",
      "79\n",
      "79\n",
      "Epoch 1/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.4137 - accuracy: 0.8093 - mae: 0.2676 - mse: 0.1330 - auc_21: 0.8897\n",
      "Epoch 2/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.4097 - accuracy: 0.8124 - mae: 0.2639 - mse: 0.1311 - auc_21: 0.8926\n",
      "Epoch 3/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.4089 - accuracy: 0.8131 - mae: 0.2645 - mse: 0.1308 - auc_21: 0.8930\n",
      "Epoch 4/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.4048 - accuracy: 0.8147 - mae: 0.2612 - mse: 0.1297 - auc_21: 0.8949\n",
      "Epoch 5/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.4070 - accuracy: 0.8139 - mae: 0.2620 - mse: 0.1303 - auc_21: 0.8940\n",
      "Epoch 6/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3995 - accuracy: 0.8167 - mae: 0.2579 - mse: 0.1281 - auc_21: 0.8977\n",
      "Epoch 7/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.4019 - accuracy: 0.8176 - mae: 0.2585 - mse: 0.1286 - auc_21: 0.8964\n",
      "Epoch 8/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.4005 - accuracy: 0.8188 - mae: 0.2580 - mse: 0.1278 - auc_21: 0.8976\n",
      "Epoch 9/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3979 - accuracy: 0.8186 - mae: 0.2564 - mse: 0.1273 - auc_21: 0.8988\n",
      "Epoch 10/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3935 - accuracy: 0.8201 - mae: 0.2530 - mse: 0.1257 - auc_21: 0.9012\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "80\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3895 - accuracy: 0.8231 - mae: 0.2505 - mse: 0.1244 - auc_21: 0.9033\n",
      "Epoch 2/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3904 - accuracy: 0.8233 - mae: 0.2508 - mse: 0.1245 - auc_21: 0.9030\n",
      "Epoch 3/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3888 - accuracy: 0.8247 - mae: 0.2494 - mse: 0.1239 - auc_21: 0.9037\n",
      "Epoch 4/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3862 - accuracy: 0.8251 - mae: 0.2481 - mse: 0.1233 - auc_21: 0.9047\n",
      "Epoch 5/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3836 - accuracy: 0.8260 - mae: 0.2463 - mse: 0.1223 - auc_21: 0.9061\n",
      "Epoch 6/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3823 - accuracy: 0.8276 - mae: 0.2452 - mse: 0.1218 - auc_21: 0.9069\n",
      "Epoch 7/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3819 - accuracy: 0.8261 - mae: 0.2452 - mse: 0.1219 - auc_21: 0.9069\n",
      "Epoch 8/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3820 - accuracy: 0.8260 - mae: 0.2443 - mse: 0.1215 - auc_21: 0.9074\n",
      "Epoch 9/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3771 - accuracy: 0.8291 - mae: 0.2422 - mse: 0.1200 - auc_21: 0.9096\n",
      "Epoch 10/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3765 - accuracy: 0.8302 - mae: 0.2414 - mse: 0.1199 - auc_21: 0.9097\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "Epoch 1/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3759 - accuracy: 0.8291 - mae: 0.2411 - mse: 0.1196 - auc_21: 0.9103\n",
      "Epoch 2/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3754 - accuracy: 0.8298 - mae: 0.2396 - mse: 0.1193 - auc_21: 0.9106\n",
      "Epoch 3/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3735 - accuracy: 0.8312 - mae: 0.2390 - mse: 0.1189 - auc_21: 0.9113\n",
      "Epoch 4/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3726 - accuracy: 0.8313 - mae: 0.2384 - mse: 0.1186 - auc_21: 0.9118\n",
      "Epoch 5/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3710 - accuracy: 0.8315 - mae: 0.2380 - mse: 0.1183 - auc_21: 0.9123\n",
      "Epoch 6/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3682 - accuracy: 0.8321 - mae: 0.2353 - mse: 0.1174 - auc_21: 0.9137\n",
      "Epoch 7/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3667 - accuracy: 0.8327 - mae: 0.2342 - mse: 0.1168 - auc_21: 0.9145\n",
      "Epoch 8/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3684 - accuracy: 0.8321 - mae: 0.2362 - mse: 0.1173 - auc_21: 0.9137\n",
      "Epoch 9/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3701 - accuracy: 0.8341 - mae: 0.2369 - mse: 0.1176 - auc_21: 0.9130\n",
      "Epoch 10/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3647 - accuracy: 0.8346 - mae: 0.2335 - mse: 0.1161 - auc_21: 0.9156\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "82\n",
      "82\n",
      "Epoch 1/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3673 - accuracy: 0.8341 - mae: 0.2344 - mse: 0.1167 - auc_21: 0.9146\n",
      "Epoch 2/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3642 - accuracy: 0.8345 - mae: 0.2337 - mse: 0.1161 - auc_21: 0.9156\n",
      "Epoch 3/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3630 - accuracy: 0.8359 - mae: 0.2323 - mse: 0.1154 - auc_21: 0.9165\n",
      "Epoch 4/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3621 - accuracy: 0.8362 - mae: 0.2312 - mse: 0.1151 - auc_21: 0.9168\n",
      "Epoch 5/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3605 - accuracy: 0.8370 - mae: 0.2295 - mse: 0.1144 - auc_21: 0.9177\n",
      "Epoch 6/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3590 - accuracy: 0.8359 - mae: 0.2295 - mse: 0.1143 - auc_21: 0.9182\n",
      "Epoch 7/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3555 - accuracy: 0.8412 - mae: 0.2264 - mse: 0.1127 - auc_21: 0.9200\n",
      "Epoch 8/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3546 - accuracy: 0.8412 - mae: 0.2265 - mse: 0.1125 - auc_21: 0.9203\n",
      "Epoch 9/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3551 - accuracy: 0.8419 - mae: 0.2257 - mse: 0.1123 - auc_21: 0.9203\n",
      "Epoch 10/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3565 - accuracy: 0.8386 - mae: 0.2276 - mse: 0.1133 - auc_21: 0.9194\n",
      "84\n",
      "84\n",
      "84\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "Epoch 1/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3529 - accuracy: 0.8388 - mae: 0.2262 - mse: 0.1121 - auc_21: 0.9212\n",
      "Epoch 2/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3522 - accuracy: 0.8402 - mae: 0.2243 - mse: 0.1118 - auc_21: 0.9216\n",
      "Epoch 3/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3541 - accuracy: 0.8392 - mae: 0.2254 - mse: 0.1123 - auc_21: 0.9208\n",
      "Epoch 4/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3550 - accuracy: 0.8370 - mae: 0.2272 - mse: 0.1131 - auc_21: 0.9200\n",
      "Epoch 5/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3508 - accuracy: 0.8372 - mae: 0.2255 - mse: 0.1120 - auc_21: 0.9218\n",
      "Epoch 6/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3486 - accuracy: 0.8432 - mae: 0.2222 - mse: 0.1105 - auc_21: 0.9233\n",
      "Epoch 7/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3480 - accuracy: 0.8422 - mae: 0.2223 - mse: 0.1107 - auc_21: 0.9232\n",
      "Epoch 8/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3470 - accuracy: 0.8454 - mae: 0.2211 - mse: 0.1098 - auc_21: 0.9238\n",
      "Epoch 9/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3464 - accuracy: 0.8452 - mae: 0.2201 - mse: 0.1097 - auc_21: 0.9242\n",
      "Epoch 10/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3458 - accuracy: 0.8445 - mae: 0.2200 - mse: 0.1097 - auc_21: 0.9243\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "83\n",
      "83\n",
      "83\n",
      "84\n",
      "83\n",
      "Epoch 1/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3436 - accuracy: 0.8460 - mae: 0.2195 - mse: 0.1088 - auc_21: 0.9255\n",
      "Epoch 2/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3465 - accuracy: 0.8428 - mae: 0.2211 - mse: 0.1099 - auc_21: 0.9241\n",
      "Epoch 3/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3457 - accuracy: 0.8437 - mae: 0.2200 - mse: 0.1098 - auc_21: 0.9242\n",
      "Epoch 4/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3446 - accuracy: 0.8440 - mae: 0.2199 - mse: 0.1095 - auc_21: 0.9246\n",
      "Epoch 5/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3439 - accuracy: 0.8450 - mae: 0.2184 - mse: 0.1089 - auc_21: 0.9253\n",
      "Epoch 6/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3406 - accuracy: 0.8471 - mae: 0.2170 - mse: 0.1080 - auc_21: 0.9266\n",
      "Epoch 7/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3408 - accuracy: 0.8475 - mae: 0.2158 - mse: 0.1077 - auc_21: 0.9267\n",
      "Epoch 8/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3420 - accuracy: 0.8440 - mae: 0.2185 - mse: 0.1087 - auc_21: 0.9259\n",
      "Epoch 9/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3400 - accuracy: 0.8475 - mae: 0.2167 - mse: 0.1079 - auc_21: 0.9268\n",
      "Epoch 10/10\n",
      "272/272 [==============================] - 2s 7ms/step - loss: 0.3388 - accuracy: 0.8465 - mae: 0.2152 - mse: 0.1075 - auc_21: 0.9273\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "84\n",
      "8\n",
      "120/120 [==============================] - 0s 1ms/step - loss: 0.6649 - accuracy: 0.7230 - mae: 0.3570 - mse: 0.2064 - auc_21: 0.7505\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1038\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1172\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1165\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1008\n",
      "Processing dgl graphs from scratch...\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1193\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1177\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1265\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1051\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1088\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1049\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1168\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "Processing dgl graphs from scratch...\n",
      "class vector created!!\n",
      "class vector created!!\n",
      "Processing dgl graphs from scratch...\n",
      "Processing molecule 1000/1285\n",
      "Processing dgl graphs from scratch...\n",
      "Data created!!\n",
      "Data created!!\n",
      "train positive label: 19834 - train negative label: 14861\n",
      "Test positive label: 2034 - Test negative label: 1800\n",
      "Epoch 1/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.5348 - accuracy: 0.7490 - mae: 0.3535 - mse: 0.1764 - auc_22: 0.8030\n",
      "Epoch 2/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.4963 - accuracy: 0.7711 - mae: 0.3251 - mse: 0.1618 - auc_22: 0.8353\n",
      "Epoch 3/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.4849 - accuracy: 0.7765 - mae: 0.3187 - mse: 0.1582 - auc_22: 0.8419\n",
      "Epoch 4/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.4777 - accuracy: 0.7796 - mae: 0.3126 - mse: 0.1556 - auc_22: 0.8474\n",
      "Epoch 5/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.4731 - accuracy: 0.7816 - mae: 0.3085 - mse: 0.1536 - auc_22: 0.8512\n",
      "Epoch 6/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.4684 - accuracy: 0.7848 - mae: 0.3058 - mse: 0.1518 - auc_22: 0.8545\n",
      "Epoch 7/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.4638 - accuracy: 0.7858 - mae: 0.3024 - mse: 0.1507 - auc_22: 0.8569\n",
      "Epoch 8/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.4599 - accuracy: 0.7885 - mae: 0.2999 - mse: 0.1489 - auc_22: 0.8606\n",
      "Epoch 9/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.4542 - accuracy: 0.7896 - mae: 0.2964 - mse: 0.1471 - auc_22: 0.8638\n",
      "Epoch 10/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.4503 - accuracy: 0.7932 - mae: 0.2921 - mse: 0.1454 - auc_22: 0.8673\n",
      "78\n",
      "78\n",
      "78\n",
      "78\n",
      "78\n",
      "77\n",
      "77\n",
      "77\n",
      "74\n",
      "Epoch 1/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.4474 - accuracy: 0.7960 - mae: 0.2917 - mse: 0.1446 - auc_22: 0.8684\n",
      "Epoch 2/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.4429 - accuracy: 0.7952 - mae: 0.2872 - mse: 0.1430 - auc_22: 0.8717\n",
      "Epoch 3/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.4379 - accuracy: 0.7990 - mae: 0.2844 - mse: 0.1414 - auc_22: 0.8744\n",
      "Epoch 4/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.4344 - accuracy: 0.7994 - mae: 0.2823 - mse: 0.1401 - auc_22: 0.8770\n",
      "Epoch 5/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.4309 - accuracy: 0.8032 - mae: 0.2789 - mse: 0.1386 - auc_22: 0.8791\n",
      "Epoch 6/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.4312 - accuracy: 0.8044 - mae: 0.2795 - mse: 0.1389 - auc_22: 0.8786\n",
      "Epoch 7/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.4230 - accuracy: 0.8070 - mae: 0.2737 - mse: 0.1359 - auc_22: 0.8839\n",
      "Epoch 8/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.4236 - accuracy: 0.8075 - mae: 0.2735 - mse: 0.1360 - auc_22: 0.8837\n",
      "Epoch 9/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.4179 - accuracy: 0.8086 - mae: 0.2703 - mse: 0.1343 - auc_22: 0.8868\n",
      "Epoch 10/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.4143 - accuracy: 0.8103 - mae: 0.2673 - mse: 0.1328 - auc_22: 0.8894\n",
      "80\n",
      "80\n",
      "80\n",
      "80\n",
      "80\n",
      "79\n",
      "79\n",
      "79\n",
      "79\n",
      "Epoch 1/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.4110 - accuracy: 0.8125 - mae: 0.2651 - mse: 0.1315 - auc_22: 0.8913\n",
      "Epoch 2/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.4093 - accuracy: 0.8117 - mae: 0.2635 - mse: 0.1310 - auc_22: 0.8923\n",
      "Epoch 3/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.4074 - accuracy: 0.8109 - mae: 0.2627 - mse: 0.1308 - auc_22: 0.8929\n",
      "Epoch 4/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.4049 - accuracy: 0.8145 - mae: 0.2607 - mse: 0.1295 - auc_22: 0.8946\n",
      "Epoch 5/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.4019 - accuracy: 0.8169 - mae: 0.2586 - mse: 0.1283 - auc_22: 0.8965\n",
      "Epoch 6/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.4001 - accuracy: 0.8163 - mae: 0.2576 - mse: 0.1283 - auc_22: 0.8969\n",
      "Epoch 7/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.3967 - accuracy: 0.8209 - mae: 0.2552 - mse: 0.1268 - auc_22: 0.8989\n",
      "Epoch 8/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.3942 - accuracy: 0.8208 - mae: 0.2537 - mse: 0.1259 - auc_22: 0.9003\n",
      "Epoch 9/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.3916 - accuracy: 0.8212 - mae: 0.2512 - mse: 0.1249 - auc_22: 0.9019\n",
      "Epoch 10/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.3913 - accuracy: 0.8211 - mae: 0.2511 - mse: 0.1252 - auc_22: 0.9017\n",
      "82\n",
      "82\n",
      "82\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "81\n",
      "Epoch 1/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.3896 - accuracy: 0.8218 - mae: 0.2500 - mse: 0.1243 - auc_22: 0.9030\n",
      "Epoch 2/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.3833 - accuracy: 0.8227 - mae: 0.2465 - mse: 0.1227 - auc_22: 0.9057\n",
      "Epoch 3/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.3847 - accuracy: 0.8253 - mae: 0.2469 - mse: 0.1227 - auc_22: 0.9053\n",
      "Epoch 4/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.3864 - accuracy: 0.8237 - mae: 0.2470 - mse: 0.1231 - auc_22: 0.9047\n",
      "Epoch 5/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.3831 - accuracy: 0.8253 - mae: 0.2454 - mse: 0.1219 - auc_22: 0.9064\n",
      "Epoch 6/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.3823 - accuracy: 0.8264 - mae: 0.2453 - mse: 0.1221 - auc_22: 0.9061\n",
      "Epoch 7/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.3831 - accuracy: 0.8245 - mae: 0.2461 - mse: 0.1224 - auc_22: 0.9058\n",
      "Epoch 8/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.3775 - accuracy: 0.8285 - mae: 0.2414 - mse: 0.1203 - auc_22: 0.9089\n",
      "Epoch 9/10\n",
      "272/272 [==============================] - 2s 8ms/step - loss: 0.3753 - accuracy: 0.8301 - mae: 0.2402 - mse: 0.1194 - auc_22: 0.9101\n",
      "Epoch 10/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.3755 - accuracy: 0.8280 - mae: 0.2405 - mse: 0.1199 - auc_22: 0.9096\n",
      "83\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "82\n",
      "Epoch 1/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.3716 - accuracy: 0.8312 - mae: 0.2380 - mse: 0.1182 - auc_22: 0.9120\n",
      "Epoch 2/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.3731 - accuracy: 0.8310 - mae: 0.2386 - mse: 0.1186 - auc_22: 0.9115\n",
      "Epoch 3/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.3731 - accuracy: 0.8300 - mae: 0.2393 - mse: 0.1189 - auc_22: 0.9110\n",
      "Epoch 4/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.3687 - accuracy: 0.8338 - mae: 0.2351 - mse: 0.1170 - auc_22: 0.9135\n",
      "Epoch 5/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.3733 - accuracy: 0.8321 - mae: 0.2386 - mse: 0.1188 - auc_22: 0.9112\n",
      "Epoch 6/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.3689 - accuracy: 0.8331 - mae: 0.2356 - mse: 0.1174 - auc_22: 0.9132\n",
      "Epoch 7/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.3634 - accuracy: 0.8364 - mae: 0.2316 - mse: 0.1153 - auc_22: 0.9160\n",
      "Epoch 8/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.3661 - accuracy: 0.8337 - mae: 0.2335 - mse: 0.1164 - auc_22: 0.9147\n",
      "Epoch 9/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.3665 - accuracy: 0.8342 - mae: 0.2345 - mse: 0.1168 - auc_22: 0.9141\n",
      "Epoch 10/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.3598 - accuracy: 0.8367 - mae: 0.2291 - mse: 0.1146 - auc_22: 0.9174\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "82\n",
      "83\n",
      "83\n",
      "Epoch 1/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.3621 - accuracy: 0.8369 - mae: 0.2313 - mse: 0.1153 - auc_22: 0.9163\n",
      "Epoch 2/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.3598 - accuracy: 0.8376 - mae: 0.2298 - mse: 0.1144 - auc_22: 0.9176\n",
      "Epoch 3/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.3594 - accuracy: 0.8363 - mae: 0.2295 - mse: 0.1141 - auc_22: 0.9181\n",
      "Epoch 4/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.3572 - accuracy: 0.8394 - mae: 0.2279 - mse: 0.1136 - auc_22: 0.9187\n",
      "Epoch 5/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.3584 - accuracy: 0.8395 - mae: 0.2281 - mse: 0.1138 - auc_22: 0.9183\n",
      "Epoch 6/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.3555 - accuracy: 0.8391 - mae: 0.2273 - mse: 0.1129 - auc_22: 0.9199\n",
      "Epoch 7/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.3554 - accuracy: 0.8390 - mae: 0.2266 - mse: 0.1130 - auc_22: 0.9196\n",
      "Epoch 8/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.3578 - accuracy: 0.8393 - mae: 0.2280 - mse: 0.1135 - auc_22: 0.9184\n",
      "Epoch 9/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.3550 - accuracy: 0.8393 - mae: 0.2263 - mse: 0.1127 - auc_22: 0.9201\n",
      "Epoch 10/10\n",
      "272/272 [==============================] - 2s 9ms/step - loss: 0.3540 - accuracy: 0.8380 - mae: 0.2260 - mse: 0.1125 - auc_22: 0.9204\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "83\n",
      "6\n",
      "120/120 [==============================] - 0s 1ms/step - loss: 0.5883 - accuracy: 0.7770 - mae: 0.2869 - mse: 0.1747 - auc_22: 0.8399\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import KFold\n",
    "\n",
    "Epoch_S = 10\n",
    "\n",
    "def evaluate_model(df, k = 10 , shuffle = False , tasks = sider_tasks):\n",
    "    result =[]    \n",
    "\n",
    "    kf = KFold(n_splits=10, shuffle = shuffle, random_state=None)\n",
    "    \n",
    "    for train_index, test_index in kf.split(df):\n",
    "\n",
    "        train = df.iloc[train_index]\n",
    "        test =  df.iloc[test_index]\n",
    "        \n",
    "        #Calculation of embedded vectors for each class\n",
    "        df_train_positive, df_train_negative = Separate_active_and_inactive_data(train, tasks)\n",
    "        df_test_positive, df_test_negative = Separate_active_and_inactive_data(test, tasks)\n",
    "        \n",
    "        dataset_positive_train = [DATASET(d, smiles_to_bigraph, AttentiveFPAtomFeaturizer(), cache_file_path = cache_path) for d in df_train_positive]\n",
    "        dataset_negative_train = [DATASET(d, smiles_to_bigraph, AttentiveFPAtomFeaturizer(), cache_file_path = cache_path) for d in df_train_negative]    \n",
    "        dataset_positive_test = [DATASET(d, smiles_to_bigraph, AttentiveFPAtomFeaturizer(), cache_file_path = cache_path) for d in df_test_positive]\n",
    "        dataset_negative_test = [DATASET(d, smiles_to_bigraph, AttentiveFPAtomFeaturizer(), cache_file_path = cache_path) for d in df_test_negative]\n",
    "        \n",
    "        embed_class_sider_train = get_embedding_vector_class(dataset_positive_train, dataset_negative_train, radius=2, size = 512)\n",
    "        embed_class_sider_test = get_embedding_vector_class(dataset_positive_test, dataset_negative_test, radius=2, size = 512)\n",
    "       \n",
    "        #create_dataset\n",
    "        train_dataset = DATASET(train, smiles_to_bigraph, CanonicalAtomFeaturizer(), cache_file_path = cache_path)\n",
    "        test_dataset = DATASET(test, smiles_to_bigraph, CanonicalAtomFeaturizer(), cache_file_path = cache_path)\n",
    "                \n",
    "        train_ds = create_dataset_with_gcn(train_dataset, embed_class_sider_train, gcn_model, tasks)\n",
    "        valid_ds = create_dataset_with_gcn(test_dataset, embed_class_sider_test, gcn_model, tasks)\n",
    "        \n",
    "        label_pos , label_neg = count_lablel(train_ds)\n",
    "        print(f'train positive label: {label_pos} - train negative label: {label_neg}')\n",
    "\n",
    "        label_pos , label_neg = count_lablel(valid_ds)\n",
    "        print(f'Test positive label: {label_pos} - Test negative label: {label_neg}')\n",
    "\n",
    "        l_train = []\n",
    "        r_train = []\n",
    "        lbls_train = []\n",
    "        l_valid = []\n",
    "        r_valid = []\n",
    "        lbls_valid = []\n",
    "\n",
    "        for i , data in enumerate(train_ds):\n",
    "            embbed_drug, onehot_task, embbed_task, lbl, task_name = data\n",
    "            l_train.append(embbed_drug[0])\n",
    "            r_train.append(embbed_task)\n",
    "            lbls_train.append(lbl.tolist())\n",
    "        \n",
    "        for i , data in enumerate(valid_ds):\n",
    "            embbed_drug, onehot_task, embbed_task, lbl, task_name = data\n",
    "            l_valid.append(embbed_drug[0])\n",
    "            r_valid.append(embbed_task)\n",
    "            lbls_valid.append(lbl.tolist())\n",
    "\n",
    "        l_train = np.array(l_train).reshape(-1,512,1)\n",
    "        r_train = np.array(r_train).reshape(-1,512,1)\n",
    "        lbls_train = np.array(lbls_train)\n",
    "\n",
    "        l_valid = np.array(l_valid).reshape(-1,512,1)\n",
    "        r_valid = np.array(r_valid).reshape(-1,512,1)\n",
    "        lbls_valid = np.array(lbls_valid)\n",
    "\n",
    "        # create neural network model\n",
    "        siamese_net = siamese_model_Canonical_sider()\n",
    "        history = History()\n",
    "        P = siamese_net.fit([l_train, r_train], lbls_train, epochs = Epoch_S, batch_size = 128, callbacks=[history])\n",
    "\n",
    "        for j in range(100):\n",
    "            C=1\n",
    "            Before = int(P.history['accuracy'][-1]*100)\n",
    "            for i in range(2,Epoch_S+1):\n",
    "                if  int(P.history['accuracy'][-i]*100)== Before:\n",
    "                    C=C+1\n",
    "                else:\n",
    "                    C=1\n",
    "                Before=int(P.history['accuracy'][-i]*100)\n",
    "                print(Before)\n",
    "            if C==Epoch_S:\n",
    "                break\n",
    "            P = siamese_net.fit([l_train, r_train], lbls_train, epochs = Epoch_S, batch_size = 128, callbacks=[history])\n",
    "        print(j+1)\n",
    "        \n",
    "        score  = siamese_net.evaluate([l_valid,r_valid],lbls_valid, verbose=1)\n",
    "        a = (score[1],score[4])\n",
    "        result.append(a)\n",
    "    \n",
    "    return result\n",
    " \n",
    " \n",
    "scores = evaluate_model(df, 10, False, sider_tasks)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "hidden": true,
    "id": "FygQzNG5nc-c",
    "outputId": "4e236b5c-a9d0-4a33-ab35-a46ae5953438"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[(0.7417767643928528, 0.8250888586044312),\n",
       " (0.752913773059845, 0.8265397548675537),\n",
       " (0.7619787454605103, 0.8200584650039673),\n",
       " (0.7700077891349792, 0.81299889087677),\n",
       " (0.7767418026924133, 0.8194084763526917),\n",
       " (0.7658637762069702, 0.8277521133422852),\n",
       " (0.7842527627944946, 0.8390901684761047),\n",
       " (0.7660406827926636, 0.8311195969581604),\n",
       " (0.7230046987533569, 0.7504766583442688),\n",
       " (0.7769953012466431, 0.8399366140365601)]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "hidden": true,
    "id": "XgMgDdpznc7s",
    "outputId": "6c375659-416a-4b28-8800-f5b166b55652"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "accuracy= 0.7619576096534729 AUC= 0.8192469596862793\n"
     ]
    }
   ],
   "source": [
    "acc = []\n",
    "auc = []\n",
    "for i in scores:\n",
    "    acc.append(i[0])\n",
    "    auc.append(i[1])\n",
    "print(f'accuracy= {np.mean(acc)} AUC= {np.mean(auc)}')"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [
    "_t1qbQt2DHT5",
    "iUo5FQhSN3rL",
    "oJ3RuoQFju2Y",
    "exyHakmjOvRz",
    "VxhrO0wEl_ms",
    "6grIE_JeqkUZ",
    "XSskFEV1j9Jf",
    "eCQzwMhntoWw",
    "gjTlqZRSq85G",
    "-fWqCCmFzMkY",
    "Uhkoil50zSnm",
    "DWoMkG_w4yG9",
    "Tt9TjAEhujAB",
    "9JiD9OHJnX55",
    "2_8xlDC9nTQB",
    "jPBVB5rSJvpQ",
    "TkHNj5nRS2xW"
   ],
   "provenance": []
  },
  "hide_input": false,
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.16"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}