lung/predictor_LUAD_LUSC.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 124
    },
    "colab_type": "code",
    "id": "OM7BoWVULIgX",
    "outputId": "a9f13766-24fd-4405-fa18-e6e588ea332c"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Go to this URL in a browser: https://accounts.google.com/o/oauth2/auth?client_id=947318989803-6bn6qk8qdgf4n4g3pfee6491hc0brc4i.apps.googleusercontent.com&redirect_uri=urn%3aietf%3awg%3aoauth%3a2.0%3aoob&response_type=code&scope=email%20https%3a%2f%2fwww.googleapis.com%2fauth%2fdocs.test%20https%3a%2f%2fwww.googleapis.com%2fauth%2fdrive%20https%3a%2f%2fwww.googleapis.com%2fauth%2fdrive.photos.readonly%20https%3a%2f%2fwww.googleapis.com%2fauth%2fpeopleapi.readonly\n",
      "\n",
      "Enter your authorization code:\n",
      "··········\n",
      "Mounted at /content/drive\n"
     ]
    }
   ],
   "source": [
    "from google.colab import drive\n",
    "drive.mount('/content/drive')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 72
    },
    "colab_type": "code",
    "id": "cZ0Axhw2Hfop",
    "outputId": "3d0cf708-fb57-4171-8e0c-8a8209c29871"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/statsmodels/tools/_testing.py:19: FutureWarning: pandas.util.testing is deprecated. Use the functions in the public API at pandas.testing instead.\n",
      "  import pandas.util.testing as tm\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import os,sys, gc\n",
    "import matplotlib.pyplot as plt\n",
    "from scipy.stats import entropy\n",
    "sys.path.append(\"/content/drive/My Drive/phd\")\n",
    "sys.path.append(\"/content/drive/My Drive/phd/hsbm-occam\")\n",
    "from hsbmpy import get_max_available_L"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "Yhh35xNYHfo1"
   },
   "outputs": [],
   "source": [
    "algorithm = \"topsbm\"\n",
    "directory='/content/drive/My Drive/phd/datasets/cancers/lung'\n",
    "L = get_max_available_L(directory, algorithm)-1\n",
    "os.chdir(directory)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 344
    },
    "colab_type": "code",
    "id": "XXZfXhOYHfo5",
    "outputId": "374b7337-5be3-4b29-dc98-32d49e2a27a0"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>cases.0.submitter_id</th>\n",
       "      <th>cases.0.diagnoses.0.morphology</th>\n",
       "      <th>cases.0.diagnoses.0.primary_diagnosis</th>\n",
       "      <th>cases.0.diagnoses.0.tissue_or_organ_of_origin</th>\n",
       "      <th>cases.0.diagnoses.0.tumor_grade</th>\n",
       "      <th>cases.0.diagnoses.0.tumor_stage</th>\n",
       "      <th>cases.0.project.disease_type</th>\n",
       "      <th>cases.0.project.primary_site</th>\n",
       "      <th>cases.0.samples.0.longest_dimension</th>\n",
       "      <th>cases.0.samples.0.portions.0.analytes.0.aliquots.0.submitter_id</th>\n",
       "      <th>longest_dimension</th>\n",
       "      <th>cancer.type</th>\n",
       "      <th>Subtype_mRNA</th>\n",
       "      <th>Subtype_DNAmeth</th>\n",
       "      <th>Subtype_Integrative</th>\n",
       "      <th>Subtype_Selected</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>file_name</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2d683c1f-3a48-4bb8-84f9-bad6347c008b.FPKM.txt.gz</th>\n",
       "      <td>TCGA-05-4249</td>\n",
       "      <td>8140/3</td>\n",
       "      <td>Adenocarcinoma, NOS</td>\n",
       "      <td>Lower lobe, lung</td>\n",
       "      <td>not reported</td>\n",
       "      <td>stage ib</td>\n",
       "      <td>Lung Adenocarcinoma</td>\n",
       "      <td>Lung</td>\n",
       "      <td>1.4</td>\n",
       "      <td>TCGA-05-4249-01A-01R-1107-07</td>\n",
       "      <td>1.6666666666666667</td>\n",
       "      <td>LUAD</td>\n",
       "      <td>unknown</td>\n",
       "      <td>unknown</td>\n",
       "      <td>5</td>\n",
       "      <td>LUAD.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>780ab54f-9100-447b-ba9e-75480e7be3d9.FPKM.txt.gz</th>\n",
       "      <td>TCGA-05-4382</td>\n",
       "      <td>8255/3</td>\n",
       "      <td>Adenocarcinoma with mixed subtypes</td>\n",
       "      <td>Upper lobe, lung</td>\n",
       "      <td>not reported</td>\n",
       "      <td>stage ib</td>\n",
       "      <td>Lung Adenocarcinoma</td>\n",
       "      <td>Lung</td>\n",
       "      <td>0.9</td>\n",
       "      <td>TCGA-05-4382-01A-01R-1206-07</td>\n",
       "      <td>1.6666666666666667</td>\n",
       "      <td>LUAD</td>\n",
       "      <td>unknown</td>\n",
       "      <td>unknown</td>\n",
       "      <td>6</td>\n",
       "      <td>LUAD.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>f8332a4b-0365-4732-a6b8-a3998872e24c.FPKM.txt.gz</th>\n",
       "      <td>TCGA-05-4384</td>\n",
       "      <td>8255/3</td>\n",
       "      <td>Adenocarcinoma with mixed subtypes</td>\n",
       "      <td>Lower lobe, lung</td>\n",
       "      <td>not reported</td>\n",
       "      <td>stage iiia</td>\n",
       "      <td>Lung Adenocarcinoma</td>\n",
       "      <td>Lung</td>\n",
       "      <td>0.8</td>\n",
       "      <td>TCGA-05-4384-01A-01R-1755-07</td>\n",
       "      <td>1.6666666666666667</td>\n",
       "      <td>LUAD</td>\n",
       "      <td>unknown</td>\n",
       "      <td>low</td>\n",
       "      <td>1</td>\n",
       "      <td>LUAD.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>755f981f-f957-48ec-9779-cf9887e19ab0.FPKM.txt.gz</th>\n",
       "      <td>TCGA-05-4389</td>\n",
       "      <td>8255/3</td>\n",
       "      <td>Adenocarcinoma with mixed subtypes</td>\n",
       "      <td>Upper lobe, lung</td>\n",
       "      <td>not reported</td>\n",
       "      <td>stage ia</td>\n",
       "      <td>Lung Adenocarcinoma</td>\n",
       "      <td>Lung</td>\n",
       "      <td>1.3</td>\n",
       "      <td>TCGA-05-4389-01A-01R-1206-07</td>\n",
       "      <td>1.6666666666666667</td>\n",
       "      <td>LUAD</td>\n",
       "      <td>unknown</td>\n",
       "      <td>unknown</td>\n",
       "      <td>2</td>\n",
       "      <td>LUAD.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>d0d7fae1-355e-4ada-a469-b69d59c1edfe.FPKM.txt.gz</th>\n",
       "      <td>TCGA-05-4390</td>\n",
       "      <td>8255/3</td>\n",
       "      <td>Adenocarcinoma with mixed subtypes</td>\n",
       "      <td>Upper lobe, lung</td>\n",
       "      <td>not reported</td>\n",
       "      <td>stage ib</td>\n",
       "      <td>Lung Adenocarcinoma</td>\n",
       "      <td>Lung</td>\n",
       "      <td>1.8</td>\n",
       "      <td>TCGA-05-4390-01A-02R-1755-07</td>\n",
       "      <td>3.3333333333333335</td>\n",
       "      <td>LUAD</td>\n",
       "      <td>unknown</td>\n",
       "      <td>intermediate</td>\n",
       "      <td>2</td>\n",
       "      <td>LUAD.2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                 cases.0.submitter_id  ... Subtype_Selected\n",
       "file_name                                                              ...                 \n",
       "2d683c1f-3a48-4bb8-84f9-bad6347c008b.FPKM.txt.gz         TCGA-05-4249  ...           LUAD.5\n",
       "780ab54f-9100-447b-ba9e-75480e7be3d9.FPKM.txt.gz         TCGA-05-4382  ...           LUAD.6\n",
       "f8332a4b-0365-4732-a6b8-a3998872e24c.FPKM.txt.gz         TCGA-05-4384  ...           LUAD.1\n",
       "755f981f-f957-48ec-9779-cf9887e19ab0.FPKM.txt.gz         TCGA-05-4389  ...           LUAD.2\n",
       "d0d7fae1-355e-4ada-a469-b69d59c1edfe.FPKM.txt.gz         TCGA-05-4390  ...           LUAD.2\n",
       "\n",
       "[5 rows x 16 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "label = 'cancer.type'\n",
    "\n",
    "df_topics = pd.read_csv(\"%s/%s_level_%d_topic-dist.csv\"%(algorithm,algorithm,L)).set_index('doc').drop('i_doc', axis=1)\n",
    "df_words = pd.read_csv(\"%s/%s_level_%d_word-dist.csv\"%(algorithm,algorithm,L), index_col=0)\n",
    "df_words.index=[g[:15] for g in df_words.index]\n",
    "df = pd.read_csv(\"mainTable.csv\", index_col=0).reindex(index=df_words.index)\n",
    "df = df.divide(df.sum(0),1).transpose().fillna(0)\n",
    "df_files=pd.read_csv(\"files.dat\", index_col=0)\n",
    "df_topics.insert(0,'tissue', df_files.reindex(index=df_topics.index)[label])\n",
    "df_topic_tissue = df_topics.groupby('tissue').mean()\n",
    "df_files.replace(\"Discordant_LUSC\",\"LUAD\", inplace=True)\n",
    "df_files.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "LdnogDbsHfpP"
   },
   "source": [
    "## NN based predictor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 486
    },
    "colab_type": "code",
    "id": "KB8z21PkHfpR",
    "outputId": "61dcd97f-4df1-4799-8bdf-db2e29c535f4"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[name: \"/device:CPU:0\"\n",
       " device_type: \"CPU\"\n",
       " memory_limit: 268435456\n",
       " locality {\n",
       " }\n",
       " incarnation: 16786936562873749751, name: \"/device:XLA_CPU:0\"\n",
       " device_type: \"XLA_CPU\"\n",
       " memory_limit: 17179869184\n",
       " locality {\n",
       " }\n",
       " incarnation: 17928951809544487343\n",
       " physical_device_desc: \"device: XLA_CPU device\", name: \"/device:XLA_GPU:0\"\n",
       " device_type: \"XLA_GPU\"\n",
       " memory_limit: 17179869184\n",
       " locality {\n",
       " }\n",
       " incarnation: 17877213065920856225\n",
       " physical_device_desc: \"device: XLA_GPU device\", name: \"/device:GPU:0\"\n",
       " device_type: \"GPU\"\n",
       " memory_limit: 11150726272\n",
       " locality {\n",
       "   bus_id: 1\n",
       "   links {\n",
       "   }\n",
       " }\n",
       " incarnation: 9933453681815736776\n",
       " physical_device_desc: \"device: 0, name: Tesla K80, pci bus id: 0000:00:04.0, compute capability: 3.7\"]"
      ]
     },
     "execution_count": 5,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Dense, Convolution1D\n",
    "from tensorflow.keras.losses import binary_crossentropy,mean_squared_error, categorical_crossentropy\n",
    "from tensorflow.keras.optimizers import SGD, Adam\n",
    "from tensorflow.keras.regularizers import l1_l2\n",
    "from tensorflow.keras.datasets import mnist\n",
    "from tensorflow.keras.utils import plot_model,to_categorical\n",
    "from tensorflow.keras.callbacks import Callback, CSVLogger, EarlyStopping\n",
    "from tensorflow.keras.models import load_model\n",
    "from tensorflow.keras import backend as K\n",
    "from tensorflow.keras.wrappers.scikit_learn import KerasClassifier\n",
    "from tensorflow.python.framework.graph_util import convert_variables_to_constants\n",
    "from tensorflow.python.client.device_lib import list_local_devices\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.model_selection import train_test_split, GridSearchCV\n",
    "import numpy as np\n",
    "import os,sys, gc\n",
    "list_local_devices()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "id": "J8SmE2OrHfpV",
    "outputId": "38c66f08-6c22-4baa-ad57-b5b78aa8a39a"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(326, 12) (326,) (82, 12) (82,)\n"
     ]
    }
   ],
   "source": [
    "df_labels=df_files.copy()\n",
    "df_labels=df_labels.reindex(index=df_topics.index)\n",
    "\n",
    "uniq = len(df_labels[label].unique())\n",
    "\n",
    "X_train = df_topics.drop('tissue',1)\n",
    "X_train = X_train.subtract(X_train.mean(0),1).divide(0.5*(X_train.max(0)-X_train.min(0)),1).values.astype(float) #SGD transform\n",
    "Y_train = to_categorical(np.unique(df_labels[label], return_inverse=True)[1])\n",
    "inputs = X_train.shape[1]\n",
    "\n",
    "Y_train = np.argmax(Y_train, axis=-1)\n",
    "uniq = 1\n",
    "activation_func = \"sigmoid\"\n",
    "lr = 0.01\n",
    "momentum = 0.95\n",
    "l1 = 0.01\n",
    "l2 = 0.0001\n",
    "loss=binary_crossentropy\n",
    "\n",
    "classes=np.unique(df_labels[df_labels.index.isin(df.index)][label], return_inverse=True)[0]\n",
    "\n",
    "X_train, X_test, Y_train, Y_test = train_test_split(X_train, Y_train, random_state=42, train_size=0.8)\n",
    "\n",
    "print(X_train.shape, Y_train.shape, X_test.shape, Y_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "colab_type": "code",
    "id": "9A-pjO9lHfpa",
    "outputId": "9176140b-92fa-4556-82d5-12f7735a5d36"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "dense (Dense)                (None, 20)                260       \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 1)                 21        \n",
      "=================================================================\n",
      "Total params: 281\n",
      "Trainable params: 281\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n",
      "None\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 15,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "K.clear_session()\n",
    "\n",
    "os.system(\"rm -rf log.csv\")\n",
    "csv_logger = CSVLogger('log.csv', append=True, separator=',')\n",
    "es = EarlyStopping(monitor='val_loss', min_delta=1e-3, mode='min', patience=25)\n",
    "\n",
    "model=Sequential()\n",
    "model.add(Dense(units=20, input_dim=inputs, use_bias=True, bias_initializer=\"ones\", activation=\"relu\", kernel_regularizer=l1_l2(l1=l1, l2=l2)))\n",
    "model.add(Dense(units=uniq, activation=activation_func))\n",
    "model.compile(loss=loss, optimizer=SGD(lr=lr, momentum=momentum), metrics=['accuracy', 'AUC'])\n",
    "K.set_learning_phase(0)\n",
    "\n",
    "print(model.summary())\n",
    "plot_model(model, to_file=f\"model_{label}.png\", dpi=600, show_shapes=True)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "colab_type": "code",
    "id": "SdHvt7jFHfpf",
    "outputId": "503256ea-a8d2-4850-8141-b9562bc2e76b"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/1000\n",
      "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow/python/ops/resource_variable_ops.py:1817: calling BaseResourceVariable.__init__ (from tensorflow.python.ops.resource_variable_ops) with constraint is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "If using Keras pass *_constraint arguments to layers.\n",
      "4/4 [==============================] - 0s 80ms/step - loss: 1.3499 - accuracy: 0.4359 - auc: 0.6502 - val_loss: 1.1876 - val_accuracy: 0.5191 - val_auc: 0.8452\n",
      "Epoch 2/1000\n",
      "4/4 [==============================] - 0s 19ms/step - loss: 1.1454 - accuracy: 0.6410 - auc: 0.7276 - val_loss: 1.0971 - val_accuracy: 0.6183 - val_auc: 0.9378\n",
      "Epoch 3/1000\n",
      "4/4 [==============================] - 0s 17ms/step - loss: 1.1052 - accuracy: 0.5846 - auc: 0.9262 - val_loss: 1.1523 - val_accuracy: 0.5725 - val_auc: 0.9280\n",
      "Epoch 4/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 1.1099 - accuracy: 0.5795 - auc: 0.9425 - val_loss: 1.0796 - val_accuracy: 0.6183 - val_auc: 0.9291\n",
      "Epoch 5/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.9969 - accuracy: 0.7128 - auc: 0.9396 - val_loss: 0.9320 - val_accuracy: 0.8473 - val_auc: 0.9294\n",
      "Epoch 6/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.8889 - accuracy: 0.8923 - auc: 0.9436 - val_loss: 0.8720 - val_accuracy: 0.9237 - val_auc: 0.9294\n",
      "Epoch 7/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.8705 - accuracy: 0.8923 - auc: 0.9503 - val_loss: 0.8551 - val_accuracy: 0.9160 - val_auc: 0.9288\n",
      "Epoch 8/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.8517 - accuracy: 0.8564 - auc: 0.9506 - val_loss: 0.8127 - val_accuracy: 0.9313 - val_auc: 0.9280\n",
      "Epoch 9/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.8022 - accuracy: 0.9026 - auc: 0.9459 - val_loss: 0.7740 - val_accuracy: 0.9160 - val_auc: 0.9280\n",
      "Epoch 10/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.7540 - accuracy: 0.9128 - auc: 0.9486 - val_loss: 0.7740 - val_accuracy: 0.8931 - val_auc: 0.9279\n",
      "Epoch 11/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.7405 - accuracy: 0.9231 - auc: 0.9492 - val_loss: 0.7825 - val_accuracy: 0.8931 - val_auc: 0.9275\n",
      "Epoch 12/1000\n",
      "4/4 [==============================] - 0s 14ms/step - loss: 0.7330 - accuracy: 0.9128 - auc: 0.9480 - val_loss: 0.7714 - val_accuracy: 0.8931 - val_auc: 0.9283\n",
      "Epoch 13/1000\n",
      "4/4 [==============================] - 0s 14ms/step - loss: 0.7136 - accuracy: 0.9179 - auc: 0.9489 - val_loss: 0.7390 - val_accuracy: 0.8931 - val_auc: 0.9301\n",
      "Epoch 14/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.6896 - accuracy: 0.9282 - auc: 0.9496 - val_loss: 0.7056 - val_accuracy: 0.9084 - val_auc: 0.9299\n",
      "Epoch 15/1000\n",
      "4/4 [==============================] - 0s 14ms/step - loss: 0.6688 - accuracy: 0.9282 - auc: 0.9525 - val_loss: 0.6793 - val_accuracy: 0.9237 - val_auc: 0.9324\n",
      "Epoch 16/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.6568 - accuracy: 0.9231 - auc: 0.9510 - val_loss: 0.6582 - val_accuracy: 0.9237 - val_auc: 0.9343\n",
      "Epoch 17/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.6478 - accuracy: 0.9077 - auc: 0.9516 - val_loss: 0.6396 - val_accuracy: 0.9237 - val_auc: 0.9350\n",
      "Epoch 18/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.6326 - accuracy: 0.9128 - auc: 0.9531 - val_loss: 0.6257 - val_accuracy: 0.9237 - val_auc: 0.9363\n",
      "Epoch 19/1000\n",
      "4/4 [==============================] - 0s 17ms/step - loss: 0.6145 - accuracy: 0.9231 - auc: 0.9542 - val_loss: 0.6133 - val_accuracy: 0.9237 - val_auc: 0.9387\n",
      "Epoch 20/1000\n",
      "4/4 [==============================] - 0s 14ms/step - loss: 0.5980 - accuracy: 0.9385 - auc: 0.9563 - val_loss: 0.6029 - val_accuracy: 0.9160 - val_auc: 0.9422\n",
      "Epoch 21/1000\n",
      "4/4 [==============================] - 0s 17ms/step - loss: 0.5834 - accuracy: 0.9333 - auc: 0.9572 - val_loss: 0.5933 - val_accuracy: 0.9160 - val_auc: 0.9448\n",
      "Epoch 22/1000\n",
      "4/4 [==============================] - 0s 14ms/step - loss: 0.5708 - accuracy: 0.9333 - auc: 0.9596 - val_loss: 0.5818 - val_accuracy: 0.9008 - val_auc: 0.9472\n",
      "Epoch 23/1000\n",
      "4/4 [==============================] - 0s 14ms/step - loss: 0.5580 - accuracy: 0.9333 - auc: 0.9616 - val_loss: 0.5654 - val_accuracy: 0.9008 - val_auc: 0.9513\n",
      "Epoch 24/1000\n",
      "4/4 [==============================] - 0s 14ms/step - loss: 0.5441 - accuracy: 0.9333 - auc: 0.9640 - val_loss: 0.5490 - val_accuracy: 0.9160 - val_auc: 0.9538\n",
      "Epoch 25/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.5321 - accuracy: 0.9333 - auc: 0.9635 - val_loss: 0.5304 - val_accuracy: 0.9160 - val_auc: 0.9596\n",
      "Epoch 26/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.5197 - accuracy: 0.9385 - auc: 0.9646 - val_loss: 0.5136 - val_accuracy: 0.9160 - val_auc: 0.9642\n",
      "Epoch 27/1000\n",
      "4/4 [==============================] - 0s 14ms/step - loss: 0.5112 - accuracy: 0.9385 - auc: 0.9626 - val_loss: 0.4972 - val_accuracy: 0.9160 - val_auc: 0.9660\n",
      "Epoch 28/1000\n",
      "4/4 [==============================] - 0s 14ms/step - loss: 0.4990 - accuracy: 0.9385 - auc: 0.9657 - val_loss: 0.4860 - val_accuracy: 0.9160 - val_auc: 0.9687\n",
      "Epoch 29/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.4876 - accuracy: 0.9385 - auc: 0.9669 - val_loss: 0.4761 - val_accuracy: 0.9160 - val_auc: 0.9695\n",
      "Epoch 30/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.4771 - accuracy: 0.9385 - auc: 0.9668 - val_loss: 0.4677 - val_accuracy: 0.9160 - val_auc: 0.9717\n",
      "Epoch 31/1000\n",
      "4/4 [==============================] - 0s 18ms/step - loss: 0.4674 - accuracy: 0.9385 - auc: 0.9666 - val_loss: 0.4617 - val_accuracy: 0.9160 - val_auc: 0.9731\n",
      "Epoch 32/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.4584 - accuracy: 0.9333 - auc: 0.9672 - val_loss: 0.4538 - val_accuracy: 0.9160 - val_auc: 0.9743\n",
      "Epoch 33/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.4501 - accuracy: 0.9333 - auc: 0.9673 - val_loss: 0.4464 - val_accuracy: 0.9160 - val_auc: 0.9748\n",
      "Epoch 34/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.4420 - accuracy: 0.9333 - auc: 0.9680 - val_loss: 0.4360 - val_accuracy: 0.9160 - val_auc: 0.9757\n",
      "Epoch 35/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.4335 - accuracy: 0.9333 - auc: 0.9680 - val_loss: 0.4261 - val_accuracy: 0.9160 - val_auc: 0.9755\n",
      "Epoch 36/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.4261 - accuracy: 0.9385 - auc: 0.9679 - val_loss: 0.4166 - val_accuracy: 0.9160 - val_auc: 0.9748\n",
      "Epoch 37/1000\n",
      "4/4 [==============================] - 0s 14ms/step - loss: 0.4193 - accuracy: 0.9385 - auc: 0.9675 - val_loss: 0.4078 - val_accuracy: 0.9237 - val_auc: 0.9746\n",
      "Epoch 38/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.4118 - accuracy: 0.9385 - auc: 0.9678 - val_loss: 0.4014 - val_accuracy: 0.9160 - val_auc: 0.9739\n",
      "Epoch 39/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.4045 - accuracy: 0.9385 - auc: 0.9681 - val_loss: 0.3956 - val_accuracy: 0.9160 - val_auc: 0.9753\n",
      "Epoch 40/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.3976 - accuracy: 0.9385 - auc: 0.9681 - val_loss: 0.3918 - val_accuracy: 0.9160 - val_auc: 0.9751\n",
      "Epoch 41/1000\n",
      "4/4 [==============================] - 0s 14ms/step - loss: 0.3911 - accuracy: 0.9333 - auc: 0.9683 - val_loss: 0.3862 - val_accuracy: 0.9160 - val_auc: 0.9746\n",
      "Epoch 42/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.3849 - accuracy: 0.9385 - auc: 0.9680 - val_loss: 0.3812 - val_accuracy: 0.9160 - val_auc: 0.9756\n",
      "Epoch 43/1000\n",
      "4/4 [==============================] - 0s 14ms/step - loss: 0.3786 - accuracy: 0.9385 - auc: 0.9680 - val_loss: 0.3738 - val_accuracy: 0.9160 - val_auc: 0.9749\n",
      "Epoch 44/1000\n",
      "4/4 [==============================] - 0s 17ms/step - loss: 0.3726 - accuracy: 0.9385 - auc: 0.9678 - val_loss: 0.3669 - val_accuracy: 0.9160 - val_auc: 0.9748\n",
      "Epoch 45/1000\n",
      "4/4 [==============================] - 0s 17ms/step - loss: 0.3680 - accuracy: 0.9385 - auc: 0.9681 - val_loss: 0.3603 - val_accuracy: 0.9160 - val_auc: 0.9758\n",
      "Epoch 46/1000\n",
      "4/4 [==============================] - 0s 14ms/step - loss: 0.3641 - accuracy: 0.9385 - auc: 0.9682 - val_loss: 0.3559 - val_accuracy: 0.9160 - val_auc: 0.9759\n",
      "Epoch 47/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.3600 - accuracy: 0.9385 - auc: 0.9686 - val_loss: 0.3509 - val_accuracy: 0.9160 - val_auc: 0.9762\n",
      "Epoch 48/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.3559 - accuracy: 0.9385 - auc: 0.9687 - val_loss: 0.3468 - val_accuracy: 0.9160 - val_auc: 0.9758\n",
      "Epoch 49/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.3521 - accuracy: 0.9385 - auc: 0.9688 - val_loss: 0.3445 - val_accuracy: 0.9160 - val_auc: 0.9758\n",
      "Epoch 50/1000\n",
      "4/4 [==============================] - 0s 17ms/step - loss: 0.3488 - accuracy: 0.9385 - auc: 0.9685 - val_loss: 0.3428 - val_accuracy: 0.9160 - val_auc: 0.9772\n",
      "Epoch 51/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.3456 - accuracy: 0.9385 - auc: 0.9690 - val_loss: 0.3400 - val_accuracy: 0.9160 - val_auc: 0.9775\n",
      "Epoch 52/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.3427 - accuracy: 0.9385 - auc: 0.9683 - val_loss: 0.3379 - val_accuracy: 0.9160 - val_auc: 0.9764\n",
      "Epoch 53/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.3404 - accuracy: 0.9385 - auc: 0.9686 - val_loss: 0.3334 - val_accuracy: 0.9160 - val_auc: 0.9761\n",
      "Epoch 54/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.3373 - accuracy: 0.9385 - auc: 0.9684 - val_loss: 0.3310 - val_accuracy: 0.9160 - val_auc: 0.9757\n",
      "Epoch 55/1000\n",
      "4/4 [==============================] - 0s 14ms/step - loss: 0.3347 - accuracy: 0.9385 - auc: 0.9682 - val_loss: 0.3292 - val_accuracy: 0.9160 - val_auc: 0.9748\n",
      "Epoch 56/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.3323 - accuracy: 0.9385 - auc: 0.9683 - val_loss: 0.3269 - val_accuracy: 0.9160 - val_auc: 0.9751\n",
      "Epoch 57/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.3301 - accuracy: 0.9385 - auc: 0.9684 - val_loss: 0.3241 - val_accuracy: 0.9160 - val_auc: 0.9749\n",
      "Epoch 58/1000\n",
      "4/4 [==============================] - 0s 14ms/step - loss: 0.3274 - accuracy: 0.9385 - auc: 0.9684 - val_loss: 0.3224 - val_accuracy: 0.9160 - val_auc: 0.9753\n",
      "Epoch 59/1000\n",
      "4/4 [==============================] - 0s 19ms/step - loss: 0.3250 - accuracy: 0.9385 - auc: 0.9683 - val_loss: 0.3213 - val_accuracy: 0.9160 - val_auc: 0.9745\n",
      "Epoch 60/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.3240 - accuracy: 0.9385 - auc: 0.9676 - val_loss: 0.3224 - val_accuracy: 0.9160 - val_auc: 0.9753\n",
      "Epoch 61/1000\n",
      "4/4 [==============================] - 0s 20ms/step - loss: 0.3214 - accuracy: 0.9385 - auc: 0.9685 - val_loss: 0.3183 - val_accuracy: 0.9160 - val_auc: 0.9756\n",
      "Epoch 62/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.3195 - accuracy: 0.9385 - auc: 0.9682 - val_loss: 0.3137 - val_accuracy: 0.9160 - val_auc: 0.9755\n",
      "Epoch 63/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.3174 - accuracy: 0.9385 - auc: 0.9688 - val_loss: 0.3108 - val_accuracy: 0.9160 - val_auc: 0.9755\n",
      "Epoch 64/1000\n",
      "4/4 [==============================] - 0s 14ms/step - loss: 0.3156 - accuracy: 0.9385 - auc: 0.9684 - val_loss: 0.3093 - val_accuracy: 0.9160 - val_auc: 0.9752\n",
      "Epoch 65/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.3135 - accuracy: 0.9385 - auc: 0.9687 - val_loss: 0.3076 - val_accuracy: 0.9160 - val_auc: 0.9756\n",
      "Epoch 66/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.3125 - accuracy: 0.9385 - auc: 0.9682 - val_loss: 0.3085 - val_accuracy: 0.9160 - val_auc: 0.9756\n",
      "Epoch 67/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.3105 - accuracy: 0.9385 - auc: 0.9676 - val_loss: 0.3082 - val_accuracy: 0.9160 - val_auc: 0.9763\n",
      "Epoch 68/1000\n",
      "4/4 [==============================] - 0s 14ms/step - loss: 0.3087 - accuracy: 0.9385 - auc: 0.9683 - val_loss: 0.3060 - val_accuracy: 0.9160 - val_auc: 0.9751\n",
      "Epoch 69/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.3074 - accuracy: 0.9385 - auc: 0.9683 - val_loss: 0.3019 - val_accuracy: 0.9160 - val_auc: 0.9762\n",
      "Epoch 70/1000\n",
      "4/4 [==============================] - 0s 14ms/step - loss: 0.3057 - accuracy: 0.9436 - auc: 0.9684 - val_loss: 0.2993 - val_accuracy: 0.9160 - val_auc: 0.9762\n",
      "Epoch 71/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.3046 - accuracy: 0.9436 - auc: 0.9687 - val_loss: 0.2989 - val_accuracy: 0.9160 - val_auc: 0.9761\n",
      "Epoch 72/1000\n",
      "4/4 [==============================] - 0s 17ms/step - loss: 0.3033 - accuracy: 0.9385 - auc: 0.9681 - val_loss: 0.2984 - val_accuracy: 0.9160 - val_auc: 0.9753\n",
      "Epoch 73/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.3021 - accuracy: 0.9385 - auc: 0.9685 - val_loss: 0.2978 - val_accuracy: 0.9160 - val_auc: 0.9759\n",
      "Epoch 74/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.3011 - accuracy: 0.9385 - auc: 0.9685 - val_loss: 0.2973 - val_accuracy: 0.9160 - val_auc: 0.9761\n",
      "Epoch 75/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.3016 - accuracy: 0.9385 - auc: 0.9669 - val_loss: 0.2993 - val_accuracy: 0.9160 - val_auc: 0.9734\n",
      "Epoch 76/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2994 - accuracy: 0.9385 - auc: 0.9675 - val_loss: 0.2961 - val_accuracy: 0.9160 - val_auc: 0.9739\n",
      "Epoch 77/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2982 - accuracy: 0.9385 - auc: 0.9684 - val_loss: 0.2940 - val_accuracy: 0.9160 - val_auc: 0.9737\n",
      "Epoch 78/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2972 - accuracy: 0.9385 - auc: 0.9682 - val_loss: 0.2936 - val_accuracy: 0.9160 - val_auc: 0.9736\n",
      "Epoch 79/1000\n",
      "4/4 [==============================] - 0s 18ms/step - loss: 0.2967 - accuracy: 0.9385 - auc: 0.9678 - val_loss: 0.2932 - val_accuracy: 0.9160 - val_auc: 0.9727\n",
      "Epoch 80/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2957 - accuracy: 0.9385 - auc: 0.9681 - val_loss: 0.2927 - val_accuracy: 0.9160 - val_auc: 0.9729\n",
      "Epoch 81/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2950 - accuracy: 0.9385 - auc: 0.9680 - val_loss: 0.2914 - val_accuracy: 0.9160 - val_auc: 0.9730\n",
      "Epoch 82/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2939 - accuracy: 0.9385 - auc: 0.9682 - val_loss: 0.2902 - val_accuracy: 0.9160 - val_auc: 0.9738\n",
      "Epoch 83/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2936 - accuracy: 0.9385 - auc: 0.9683 - val_loss: 0.2918 - val_accuracy: 0.9160 - val_auc: 0.9745\n",
      "Epoch 84/1000\n",
      "4/4 [==============================] - 0s 14ms/step - loss: 0.2925 - accuracy: 0.9385 - auc: 0.9677 - val_loss: 0.2916 - val_accuracy: 0.9160 - val_auc: 0.9751\n",
      "Epoch 85/1000\n",
      "4/4 [==============================] - 0s 17ms/step - loss: 0.2917 - accuracy: 0.9385 - auc: 0.9683 - val_loss: 0.2893 - val_accuracy: 0.9160 - val_auc: 0.9727\n",
      "Epoch 86/1000\n",
      "4/4 [==============================] - 0s 17ms/step - loss: 0.2912 - accuracy: 0.9385 - auc: 0.9679 - val_loss: 0.2876 - val_accuracy: 0.9160 - val_auc: 0.9738\n",
      "Epoch 87/1000\n",
      "4/4 [==============================] - 0s 14ms/step - loss: 0.2903 - accuracy: 0.9385 - auc: 0.9684 - val_loss: 0.2865 - val_accuracy: 0.9160 - val_auc: 0.9745\n",
      "Epoch 88/1000\n",
      "4/4 [==============================] - 0s 17ms/step - loss: 0.2893 - accuracy: 0.9385 - auc: 0.9687 - val_loss: 0.2867 - val_accuracy: 0.9160 - val_auc: 0.9727\n",
      "Epoch 89/1000\n",
      "4/4 [==============================] - 0s 17ms/step - loss: 0.2886 - accuracy: 0.9333 - auc: 0.9687 - val_loss: 0.2857 - val_accuracy: 0.9160 - val_auc: 0.9738\n",
      "Epoch 90/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2883 - accuracy: 0.9333 - auc: 0.9683 - val_loss: 0.2862 - val_accuracy: 0.9160 - val_auc: 0.9727\n",
      "Epoch 91/1000\n",
      "4/4 [==============================] - 0s 14ms/step - loss: 0.2872 - accuracy: 0.9385 - auc: 0.9690 - val_loss: 0.2857 - val_accuracy: 0.9160 - val_auc: 0.9733\n",
      "Epoch 92/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.2866 - accuracy: 0.9385 - auc: 0.9680 - val_loss: 0.2851 - val_accuracy: 0.9160 - val_auc: 0.9739\n",
      "Epoch 93/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2860 - accuracy: 0.9385 - auc: 0.9683 - val_loss: 0.2849 - val_accuracy: 0.9160 - val_auc: 0.9745\n",
      "Epoch 94/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2853 - accuracy: 0.9385 - auc: 0.9685 - val_loss: 0.2829 - val_accuracy: 0.9160 - val_auc: 0.9745\n",
      "Epoch 95/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2844 - accuracy: 0.9385 - auc: 0.9695 - val_loss: 0.2810 - val_accuracy: 0.9160 - val_auc: 0.9744\n",
      "Epoch 96/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2840 - accuracy: 0.9385 - auc: 0.9692 - val_loss: 0.2787 - val_accuracy: 0.9160 - val_auc: 0.9751\n",
      "Epoch 97/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2836 - accuracy: 0.9385 - auc: 0.9692 - val_loss: 0.2774 - val_accuracy: 0.9160 - val_auc: 0.9752\n",
      "Epoch 98/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.2826 - accuracy: 0.9385 - auc: 0.9698 - val_loss: 0.2784 - val_accuracy: 0.9160 - val_auc: 0.9755\n",
      "Epoch 99/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2822 - accuracy: 0.9436 - auc: 0.9691 - val_loss: 0.2809 - val_accuracy: 0.9160 - val_auc: 0.9746\n",
      "Epoch 100/1000\n",
      "4/4 [==============================] - 0s 17ms/step - loss: 0.2816 - accuracy: 0.9385 - auc: 0.9690 - val_loss: 0.2825 - val_accuracy: 0.9160 - val_auc: 0.9748\n",
      "Epoch 101/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2812 - accuracy: 0.9385 - auc: 0.9693 - val_loss: 0.2822 - val_accuracy: 0.9160 - val_auc: 0.9744\n",
      "Epoch 102/1000\n",
      "4/4 [==============================] - 0s 17ms/step - loss: 0.2802 - accuracy: 0.9385 - auc: 0.9693 - val_loss: 0.2796 - val_accuracy: 0.9160 - val_auc: 0.9756\n",
      "Epoch 103/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.2808 - accuracy: 0.9333 - auc: 0.9686 - val_loss: 0.2739 - val_accuracy: 0.9160 - val_auc: 0.9757\n",
      "Epoch 104/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2791 - accuracy: 0.9385 - auc: 0.9696 - val_loss: 0.2736 - val_accuracy: 0.9160 - val_auc: 0.9751\n",
      "Epoch 105/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2785 - accuracy: 0.9385 - auc: 0.9688 - val_loss: 0.2736 - val_accuracy: 0.9160 - val_auc: 0.9748\n",
      "Epoch 106/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2778 - accuracy: 0.9385 - auc: 0.9692 - val_loss: 0.2730 - val_accuracy: 0.9160 - val_auc: 0.9756\n",
      "Epoch 107/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2773 - accuracy: 0.9385 - auc: 0.9699 - val_loss: 0.2738 - val_accuracy: 0.9160 - val_auc: 0.9759\n",
      "Epoch 108/1000\n",
      "4/4 [==============================] - 0s 17ms/step - loss: 0.2769 - accuracy: 0.9333 - auc: 0.9695 - val_loss: 0.2754 - val_accuracy: 0.9160 - val_auc: 0.9756\n",
      "Epoch 109/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2762 - accuracy: 0.9385 - auc: 0.9692 - val_loss: 0.2752 - val_accuracy: 0.9160 - val_auc: 0.9765\n",
      "Epoch 110/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2757 - accuracy: 0.9385 - auc: 0.9693 - val_loss: 0.2739 - val_accuracy: 0.9160 - val_auc: 0.9765\n",
      "Epoch 111/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2751 - accuracy: 0.9333 - auc: 0.9698 - val_loss: 0.2723 - val_accuracy: 0.9160 - val_auc: 0.9761\n",
      "Epoch 112/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2746 - accuracy: 0.9385 - auc: 0.9697 - val_loss: 0.2717 - val_accuracy: 0.9160 - val_auc: 0.9751\n",
      "Epoch 113/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2743 - accuracy: 0.9385 - auc: 0.9696 - val_loss: 0.2706 - val_accuracy: 0.9160 - val_auc: 0.9751\n",
      "Epoch 114/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.2746 - accuracy: 0.9385 - auc: 0.9692 - val_loss: 0.2681 - val_accuracy: 0.9237 - val_auc: 0.9755\n",
      "Epoch 115/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2736 - accuracy: 0.9385 - auc: 0.9698 - val_loss: 0.2687 - val_accuracy: 0.9160 - val_auc: 0.9766\n",
      "Epoch 116/1000\n",
      "4/4 [==============================] - 0s 17ms/step - loss: 0.2727 - accuracy: 0.9385 - auc: 0.9703 - val_loss: 0.2712 - val_accuracy: 0.9160 - val_auc: 0.9777\n",
      "Epoch 117/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2726 - accuracy: 0.9385 - auc: 0.9702 - val_loss: 0.2736 - val_accuracy: 0.9160 - val_auc: 0.9765\n",
      "Epoch 118/1000\n",
      "4/4 [==============================] - 0s 17ms/step - loss: 0.2722 - accuracy: 0.9436 - auc: 0.9701 - val_loss: 0.2736 - val_accuracy: 0.9160 - val_auc: 0.9768\n",
      "Epoch 119/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.2715 - accuracy: 0.9385 - auc: 0.9704 - val_loss: 0.2693 - val_accuracy: 0.9160 - val_auc: 0.9772\n",
      "Epoch 120/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.2722 - accuracy: 0.9385 - auc: 0.9702 - val_loss: 0.2645 - val_accuracy: 0.9313 - val_auc: 0.9776\n",
      "Epoch 121/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2707 - accuracy: 0.9333 - auc: 0.9703 - val_loss: 0.2649 - val_accuracy: 0.9313 - val_auc: 0.9769\n",
      "Epoch 122/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2701 - accuracy: 0.9333 - auc: 0.9703 - val_loss: 0.2655 - val_accuracy: 0.9160 - val_auc: 0.9771\n",
      "Epoch 123/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2696 - accuracy: 0.9385 - auc: 0.9704 - val_loss: 0.2651 - val_accuracy: 0.9160 - val_auc: 0.9778\n",
      "Epoch 124/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.2688 - accuracy: 0.9385 - auc: 0.9709 - val_loss: 0.2661 - val_accuracy: 0.9160 - val_auc: 0.9781\n",
      "Epoch 125/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.2690 - accuracy: 0.9436 - auc: 0.9705 - val_loss: 0.2693 - val_accuracy: 0.9160 - val_auc: 0.9776\n",
      "Epoch 126/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.2686 - accuracy: 0.9385 - auc: 0.9708 - val_loss: 0.2675 - val_accuracy: 0.9160 - val_auc: 0.9772\n",
      "Epoch 127/1000\n",
      "4/4 [==============================] - 0s 20ms/step - loss: 0.2677 - accuracy: 0.9436 - auc: 0.9709 - val_loss: 0.2668 - val_accuracy: 0.9160 - val_auc: 0.9787\n",
      "Epoch 128/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2676 - accuracy: 0.9385 - auc: 0.9712 - val_loss: 0.2641 - val_accuracy: 0.9160 - val_auc: 0.9800\n",
      "Epoch 129/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.2670 - accuracy: 0.9385 - auc: 0.9713 - val_loss: 0.2623 - val_accuracy: 0.9160 - val_auc: 0.9808\n",
      "Epoch 130/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2669 - accuracy: 0.9385 - auc: 0.9714 - val_loss: 0.2641 - val_accuracy: 0.9160 - val_auc: 0.9808\n",
      "Epoch 131/1000\n",
      "4/4 [==============================] - 0s 14ms/step - loss: 0.2663 - accuracy: 0.9385 - auc: 0.9718 - val_loss: 0.2638 - val_accuracy: 0.9160 - val_auc: 0.9809\n",
      "Epoch 132/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2656 - accuracy: 0.9385 - auc: 0.9716 - val_loss: 0.2616 - val_accuracy: 0.9160 - val_auc: 0.9788\n",
      "Epoch 133/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2663 - accuracy: 0.9333 - auc: 0.9714 - val_loss: 0.2586 - val_accuracy: 0.9313 - val_auc: 0.9806\n",
      "Epoch 134/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2654 - accuracy: 0.9333 - auc: 0.9709 - val_loss: 0.2609 - val_accuracy: 0.9160 - val_auc: 0.9790\n",
      "Epoch 135/1000\n",
      "4/4 [==============================] - 0s 14ms/step - loss: 0.2644 - accuracy: 0.9385 - auc: 0.9714 - val_loss: 0.2603 - val_accuracy: 0.9160 - val_auc: 0.9809\n",
      "Epoch 136/1000\n",
      "4/4 [==============================] - 0s 14ms/step - loss: 0.2639 - accuracy: 0.9385 - auc: 0.9723 - val_loss: 0.2608 - val_accuracy: 0.9160 - val_auc: 0.9802\n",
      "Epoch 137/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.2639 - accuracy: 0.9385 - auc: 0.9718 - val_loss: 0.2599 - val_accuracy: 0.9160 - val_auc: 0.9812\n",
      "Epoch 138/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.2638 - accuracy: 0.9385 - auc: 0.9710 - val_loss: 0.2627 - val_accuracy: 0.9160 - val_auc: 0.9808\n",
      "Epoch 139/1000\n",
      "4/4 [==============================] - 0s 14ms/step - loss: 0.2632 - accuracy: 0.9385 - auc: 0.9720 - val_loss: 0.2613 - val_accuracy: 0.9160 - val_auc: 0.9808\n",
      "Epoch 140/1000\n",
      "4/4 [==============================] - 0s 14ms/step - loss: 0.2627 - accuracy: 0.9385 - auc: 0.9720 - val_loss: 0.2601 - val_accuracy: 0.9160 - val_auc: 0.9809\n",
      "Epoch 141/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.2622 - accuracy: 0.9385 - auc: 0.9721 - val_loss: 0.2584 - val_accuracy: 0.9160 - val_auc: 0.9802\n",
      "Epoch 142/1000\n",
      "4/4 [==============================] - 0s 14ms/step - loss: 0.2618 - accuracy: 0.9385 - auc: 0.9719 - val_loss: 0.2579 - val_accuracy: 0.9160 - val_auc: 0.9806\n",
      "Epoch 143/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2617 - accuracy: 0.9385 - auc: 0.9723 - val_loss: 0.2558 - val_accuracy: 0.9313 - val_auc: 0.9806\n",
      "Epoch 144/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.2612 - accuracy: 0.9333 - auc: 0.9725 - val_loss: 0.2556 - val_accuracy: 0.9313 - val_auc: 0.9806\n",
      "Epoch 145/1000\n",
      "4/4 [==============================] - 0s 14ms/step - loss: 0.2609 - accuracy: 0.9333 - auc: 0.9730 - val_loss: 0.2557 - val_accuracy: 0.9237 - val_auc: 0.9804\n",
      "Epoch 146/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2607 - accuracy: 0.9333 - auc: 0.9722 - val_loss: 0.2565 - val_accuracy: 0.9160 - val_auc: 0.9800\n",
      "Epoch 147/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.2602 - accuracy: 0.9385 - auc: 0.9725 - val_loss: 0.2576 - val_accuracy: 0.9160 - val_auc: 0.9808\n",
      "Epoch 148/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2602 - accuracy: 0.9385 - auc: 0.9725 - val_loss: 0.2560 - val_accuracy: 0.9160 - val_auc: 0.9814\n",
      "Epoch 149/1000\n",
      "4/4 [==============================] - 0s 17ms/step - loss: 0.2595 - accuracy: 0.9385 - auc: 0.9731 - val_loss: 0.2559 - val_accuracy: 0.9160 - val_auc: 0.9814\n",
      "Epoch 150/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2597 - accuracy: 0.9385 - auc: 0.9725 - val_loss: 0.2555 - val_accuracy: 0.9160 - val_auc: 0.9810\n",
      "Epoch 151/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2588 - accuracy: 0.9385 - auc: 0.9730 - val_loss: 0.2547 - val_accuracy: 0.9160 - val_auc: 0.9807\n",
      "Epoch 152/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2588 - accuracy: 0.9385 - auc: 0.9729 - val_loss: 0.2538 - val_accuracy: 0.9160 - val_auc: 0.9813\n",
      "Epoch 153/1000\n",
      "4/4 [==============================] - 0s 14ms/step - loss: 0.2591 - accuracy: 0.9385 - auc: 0.9729 - val_loss: 0.2559 - val_accuracy: 0.9160 - val_auc: 0.9815\n",
      "Epoch 154/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.2582 - accuracy: 0.9385 - auc: 0.9736 - val_loss: 0.2557 - val_accuracy: 0.9160 - val_auc: 0.9809\n",
      "Epoch 155/1000\n",
      "4/4 [==============================] - 0s 17ms/step - loss: 0.2578 - accuracy: 0.9385 - auc: 0.9736 - val_loss: 0.2533 - val_accuracy: 0.9160 - val_auc: 0.9809\n",
      "Epoch 156/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.2573 - accuracy: 0.9333 - auc: 0.9732 - val_loss: 0.2514 - val_accuracy: 0.9313 - val_auc: 0.9810\n",
      "Epoch 157/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2584 - accuracy: 0.9333 - auc: 0.9731 - val_loss: 0.2480 - val_accuracy: 0.9313 - val_auc: 0.9814\n",
      "Epoch 158/1000\n",
      "4/4 [==============================] - 0s 17ms/step - loss: 0.2575 - accuracy: 0.9333 - auc: 0.9735 - val_loss: 0.2493 - val_accuracy: 0.9313 - val_auc: 0.9813\n",
      "Epoch 159/1000\n",
      "4/4 [==============================] - 0s 14ms/step - loss: 0.2589 - accuracy: 0.9333 - auc: 0.9717 - val_loss: 0.2548 - val_accuracy: 0.9160 - val_auc: 0.9809\n",
      "Epoch 160/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2565 - accuracy: 0.9385 - auc: 0.9731 - val_loss: 0.2529 - val_accuracy: 0.9160 - val_auc: 0.9813\n",
      "Epoch 161/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.2560 - accuracy: 0.9385 - auc: 0.9733 - val_loss: 0.2523 - val_accuracy: 0.9160 - val_auc: 0.9816\n",
      "Epoch 162/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2559 - accuracy: 0.9333 - auc: 0.9733 - val_loss: 0.2513 - val_accuracy: 0.9160 - val_auc: 0.9820\n",
      "Epoch 163/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2555 - accuracy: 0.9333 - auc: 0.9739 - val_loss: 0.2518 - val_accuracy: 0.9160 - val_auc: 0.9810\n",
      "Epoch 164/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2555 - accuracy: 0.9333 - auc: 0.9734 - val_loss: 0.2497 - val_accuracy: 0.9313 - val_auc: 0.9821\n",
      "Epoch 165/1000\n",
      "4/4 [==============================] - 0s 14ms/step - loss: 0.2553 - accuracy: 0.9333 - auc: 0.9737 - val_loss: 0.2485 - val_accuracy: 0.9313 - val_auc: 0.9817\n",
      "Epoch 166/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.2556 - accuracy: 0.9333 - auc: 0.9738 - val_loss: 0.2478 - val_accuracy: 0.9313 - val_auc: 0.9826\n",
      "Epoch 167/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2543 - accuracy: 0.9333 - auc: 0.9742 - val_loss: 0.2508 - val_accuracy: 0.9160 - val_auc: 0.9822\n",
      "Epoch 168/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.2539 - accuracy: 0.9436 - auc: 0.9743 - val_loss: 0.2527 - val_accuracy: 0.9160 - val_auc: 0.9816\n",
      "Epoch 169/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2541 - accuracy: 0.9436 - auc: 0.9744 - val_loss: 0.2545 - val_accuracy: 0.9160 - val_auc: 0.9809\n",
      "Epoch 170/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2539 - accuracy: 0.9436 - auc: 0.9740 - val_loss: 0.2528 - val_accuracy: 0.9160 - val_auc: 0.9817\n",
      "Epoch 171/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2535 - accuracy: 0.9436 - auc: 0.9742 - val_loss: 0.2493 - val_accuracy: 0.9237 - val_auc: 0.9815\n",
      "Epoch 172/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.2534 - accuracy: 0.9333 - auc: 0.9742 - val_loss: 0.2463 - val_accuracy: 0.9313 - val_auc: 0.9817\n",
      "Epoch 173/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.2527 - accuracy: 0.9333 - auc: 0.9740 - val_loss: 0.2463 - val_accuracy: 0.9313 - val_auc: 0.9819\n",
      "Epoch 174/1000\n",
      "4/4 [==============================] - 0s 14ms/step - loss: 0.2524 - accuracy: 0.9333 - auc: 0.9741 - val_loss: 0.2464 - val_accuracy: 0.9313 - val_auc: 0.9819\n",
      "Epoch 175/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2522 - accuracy: 0.9333 - auc: 0.9738 - val_loss: 0.2474 - val_accuracy: 0.9313 - val_auc: 0.9815\n",
      "Epoch 176/1000\n",
      "4/4 [==============================] - 0s 17ms/step - loss: 0.2525 - accuracy: 0.9333 - auc: 0.9740 - val_loss: 0.2486 - val_accuracy: 0.9237 - val_auc: 0.9817\n",
      "Epoch 177/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2518 - accuracy: 0.9385 - auc: 0.9740 - val_loss: 0.2490 - val_accuracy: 0.9160 - val_auc: 0.9827\n",
      "Epoch 178/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2516 - accuracy: 0.9385 - auc: 0.9747 - val_loss: 0.2482 - val_accuracy: 0.9160 - val_auc: 0.9827\n",
      "Epoch 179/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2512 - accuracy: 0.9333 - auc: 0.9748 - val_loss: 0.2464 - val_accuracy: 0.9313 - val_auc: 0.9829\n",
      "Epoch 180/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2510 - accuracy: 0.9333 - auc: 0.9744 - val_loss: 0.2455 - val_accuracy: 0.9313 - val_auc: 0.9829\n",
      "Epoch 181/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2509 - accuracy: 0.9333 - auc: 0.9743 - val_loss: 0.2452 - val_accuracy: 0.9313 - val_auc: 0.9828\n",
      "Epoch 182/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2506 - accuracy: 0.9333 - auc: 0.9746 - val_loss: 0.2441 - val_accuracy: 0.9313 - val_auc: 0.9828\n",
      "Epoch 183/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.2503 - accuracy: 0.9333 - auc: 0.9747 - val_loss: 0.2448 - val_accuracy: 0.9313 - val_auc: 0.9826\n",
      "Epoch 184/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2499 - accuracy: 0.9333 - auc: 0.9749 - val_loss: 0.2454 - val_accuracy: 0.9237 - val_auc: 0.9834\n",
      "Epoch 185/1000\n",
      "4/4 [==============================] - 0s 17ms/step - loss: 0.2498 - accuracy: 0.9333 - auc: 0.9752 - val_loss: 0.2455 - val_accuracy: 0.9237 - val_auc: 0.9838\n",
      "Epoch 186/1000\n",
      "4/4 [==============================] - 0s 17ms/step - loss: 0.2497 - accuracy: 0.9333 - auc: 0.9752 - val_loss: 0.2463 - val_accuracy: 0.9237 - val_auc: 0.9827\n",
      "Epoch 187/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2495 - accuracy: 0.9385 - auc: 0.9748 - val_loss: 0.2450 - val_accuracy: 0.9237 - val_auc: 0.9836\n",
      "Epoch 188/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2497 - accuracy: 0.9333 - auc: 0.9752 - val_loss: 0.2434 - val_accuracy: 0.9313 - val_auc: 0.9841\n",
      "Epoch 189/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2498 - accuracy: 0.9333 - auc: 0.9746 - val_loss: 0.2463 - val_accuracy: 0.9237 - val_auc: 0.9835\n",
      "Epoch 190/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.2490 - accuracy: 0.9385 - auc: 0.9748 - val_loss: 0.2450 - val_accuracy: 0.9237 - val_auc: 0.9847\n",
      "Epoch 191/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2491 - accuracy: 0.9385 - auc: 0.9752 - val_loss: 0.2453 - val_accuracy: 0.9160 - val_auc: 0.9842\n",
      "Epoch 192/1000\n",
      "4/4 [==============================] - 0s 14ms/step - loss: 0.2488 - accuracy: 0.9385 - auc: 0.9750 - val_loss: 0.2448 - val_accuracy: 0.9237 - val_auc: 0.9841\n",
      "Epoch 193/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2486 - accuracy: 0.9333 - auc: 0.9751 - val_loss: 0.2401 - val_accuracy: 0.9313 - val_auc: 0.9839\n",
      "Epoch 194/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2481 - accuracy: 0.9333 - auc: 0.9750 - val_loss: 0.2392 - val_accuracy: 0.9313 - val_auc: 0.9835\n",
      "Epoch 195/1000\n",
      "4/4 [==============================] - 0s 18ms/step - loss: 0.2482 - accuracy: 0.9333 - auc: 0.9751 - val_loss: 0.2387 - val_accuracy: 0.9313 - val_auc: 0.9840\n",
      "Epoch 196/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.2481 - accuracy: 0.9333 - auc: 0.9752 - val_loss: 0.2394 - val_accuracy: 0.9313 - val_auc: 0.9844\n",
      "Epoch 197/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2476 - accuracy: 0.9333 - auc: 0.9753 - val_loss: 0.2434 - val_accuracy: 0.9237 - val_auc: 0.9841\n",
      "Epoch 198/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2471 - accuracy: 0.9385 - auc: 0.9752 - val_loss: 0.2455 - val_accuracy: 0.9160 - val_auc: 0.9847\n",
      "Epoch 199/1000\n",
      "4/4 [==============================] - 0s 17ms/step - loss: 0.2480 - accuracy: 0.9436 - auc: 0.9758 - val_loss: 0.2486 - val_accuracy: 0.9160 - val_auc: 0.9840\n",
      "Epoch 200/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2470 - accuracy: 0.9436 - auc: 0.9752 - val_loss: 0.2446 - val_accuracy: 0.9160 - val_auc: 0.9842\n",
      "Epoch 201/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2467 - accuracy: 0.9385 - auc: 0.9761 - val_loss: 0.2402 - val_accuracy: 0.9313 - val_auc: 0.9846\n",
      "Epoch 202/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2471 - accuracy: 0.9333 - auc: 0.9748 - val_loss: 0.2363 - val_accuracy: 0.9313 - val_auc: 0.9845\n",
      "Epoch 203/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.2470 - accuracy: 0.9333 - auc: 0.9754 - val_loss: 0.2368 - val_accuracy: 0.9313 - val_auc: 0.9844\n",
      "Epoch 204/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2462 - accuracy: 0.9333 - auc: 0.9759 - val_loss: 0.2381 - val_accuracy: 0.9313 - val_auc: 0.9838\n",
      "Epoch 205/1000\n",
      "4/4 [==============================] - 0s 21ms/step - loss: 0.2462 - accuracy: 0.9333 - auc: 0.9754 - val_loss: 0.2380 - val_accuracy: 0.9313 - val_auc: 0.9855\n",
      "Epoch 206/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.2463 - accuracy: 0.9385 - auc: 0.9755 - val_loss: 0.2435 - val_accuracy: 0.9160 - val_auc: 0.9853\n",
      "Epoch 207/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2463 - accuracy: 0.9436 - auc: 0.9763 - val_loss: 0.2459 - val_accuracy: 0.9160 - val_auc: 0.9849\n",
      "Epoch 208/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2470 - accuracy: 0.9436 - auc: 0.9760 - val_loss: 0.2414 - val_accuracy: 0.9237 - val_auc: 0.9851\n",
      "Epoch 209/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2451 - accuracy: 0.9385 - auc: 0.9763 - val_loss: 0.2398 - val_accuracy: 0.9313 - val_auc: 0.9852\n",
      "Epoch 210/1000\n",
      "4/4 [==============================] - 0s 17ms/step - loss: 0.2450 - accuracy: 0.9333 - auc: 0.9764 - val_loss: 0.2374 - val_accuracy: 0.9313 - val_auc: 0.9855\n",
      "Epoch 211/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2448 - accuracy: 0.9333 - auc: 0.9764 - val_loss: 0.2363 - val_accuracy: 0.9313 - val_auc: 0.9849\n",
      "Epoch 212/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2450 - accuracy: 0.9333 - auc: 0.9758 - val_loss: 0.2379 - val_accuracy: 0.9313 - val_auc: 0.9849\n",
      "Epoch 213/1000\n",
      "4/4 [==============================] - 0s 17ms/step - loss: 0.2448 - accuracy: 0.9333 - auc: 0.9757 - val_loss: 0.2369 - val_accuracy: 0.9313 - val_auc: 0.9848\n",
      "Epoch 214/1000\n",
      "4/4 [==============================] - 0s 17ms/step - loss: 0.2449 - accuracy: 0.9385 - auc: 0.9756 - val_loss: 0.2403 - val_accuracy: 0.9237 - val_auc: 0.9849\n",
      "Epoch 215/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2441 - accuracy: 0.9385 - auc: 0.9760 - val_loss: 0.2407 - val_accuracy: 0.9237 - val_auc: 0.9848\n",
      "Epoch 216/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2439 - accuracy: 0.9385 - auc: 0.9755 - val_loss: 0.2398 - val_accuracy: 0.9237 - val_auc: 0.9845\n",
      "Epoch 217/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.2448 - accuracy: 0.9385 - auc: 0.9753 - val_loss: 0.2363 - val_accuracy: 0.9313 - val_auc: 0.9852\n",
      "Epoch 218/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2445 - accuracy: 0.9385 - auc: 0.9762 - val_loss: 0.2392 - val_accuracy: 0.9313 - val_auc: 0.9851\n",
      "Epoch 219/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2438 - accuracy: 0.9385 - auc: 0.9760 - val_loss: 0.2373 - val_accuracy: 0.9313 - val_auc: 0.9849\n",
      "Epoch 220/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2434 - accuracy: 0.9385 - auc: 0.9758 - val_loss: 0.2372 - val_accuracy: 0.9313 - val_auc: 0.9844\n",
      "Epoch 221/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2432 - accuracy: 0.9385 - auc: 0.9757 - val_loss: 0.2367 - val_accuracy: 0.9313 - val_auc: 0.9844\n",
      "Epoch 222/1000\n",
      "4/4 [==============================] - 0s 15ms/step - loss: 0.2434 - accuracy: 0.9385 - auc: 0.9757 - val_loss: 0.2368 - val_accuracy: 0.9313 - val_auc: 0.9852\n",
      "Epoch 223/1000\n",
      "4/4 [==============================] - 0s 14ms/step - loss: 0.2436 - accuracy: 0.9385 - auc: 0.9761 - val_loss: 0.2377 - val_accuracy: 0.9313 - val_auc: 0.9852\n",
      "Epoch 224/1000\n",
      "4/4 [==============================] - 0s 22ms/step - loss: 0.2428 - accuracy: 0.9385 - auc: 0.9762 - val_loss: 0.2382 - val_accuracy: 0.9313 - val_auc: 0.9851\n",
      "Epoch 225/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.2429 - accuracy: 0.9385 - auc: 0.9760 - val_loss: 0.2366 - val_accuracy: 0.9313 - val_auc: 0.9847\n",
      "Epoch 226/1000\n",
      "4/4 [==============================] - 0s 16ms/step - loss: 0.2426 - accuracy: 0.9385 - auc: 0.9757 - val_loss: 0.2382 - val_accuracy: 0.9313 - val_auc: 0.9844\n",
      "Epoch 227/1000\n",
      "4/4 [==============================] - 0s 17ms/step - loss: 0.2427 - accuracy: 0.9385 - auc: 0.9760 - val_loss: 0.2361 - val_accuracy: 0.9313 - val_auc: 0.9849\n"
     ]
    }
   ],
   "source": [
    "with tf.device(\"GPU\"):\n",
    "    model.fit(X_train, Y_train, epochs=1000, batch_size=50, verbose=1, validation_split=0.4, callbacks=[csv_logger, es], shuffle=True, use_multiprocessing=True, workers=-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 285
    },
    "colab_type": "code",
    "id": "ANz0bgi2Hfpi",
    "outputId": "a6f7f8e6-f910-4617-b701-189ae96e4230"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f164bd442e8>"
      ]
     },
     "execution_count": 17,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "pd.read_csv(\"log.csv\", sep=\",\")[['loss','val_loss']].plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 52
    },
    "colab_type": "code",
    "id": "ie8qRsmeHfpm",
    "outputId": "63d361f5-20a7-443f-c110-f6121b3c9afa"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3/3 [==============================] - 0s 4ms/step - loss: 0.3237 - accuracy: 0.9268 - auc: 0.9493\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[0.32368770241737366, 0.9268292784690857, 0.9492517709732056]"
      ]
     },
     "execution_count": 18,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.evaluate(X_test, Y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "Qcwyj17QHfpp"
   },
   "outputs": [],
   "source": [
    "model.save(f\"model_{label}.h5\")\n",
    "model = load_model(f\"model_{label}.h5\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "id": "eFyBghA0Hfpt",
    "outputId": "9a0f1ff7-b417-4514-9932-8c4a8fd01536"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "36"
      ]
     },
     "execution_count": 9,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gc.collect()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "xIWNWHbwHfpz"
   },
   "source": [
    "# evaluate on non used on topsbm training\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "id": "oydLIuRbG8kJ",
    "outputId": "e59aac55-f66e-4cfe-a451-b4e72484b8d6"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(82, 12) (82,)\n"
     ]
    }
   ],
   "source": [
    "classes = df_files[label].unique()\n",
    "np.savetxt(\"classes.txt\", classes, fmt=\"%s\")\n",
    "np.savetxt(\"X_test.txt\", X_test)\n",
    "np.savetxt(\"Y_test.txt\", Y_test)\n",
    "print(X_test.shape, Y_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 52
    },
    "colab_type": "code",
    "id": "5JLmv_InHfp_",
    "outputId": "85a25360-2d08-428f-e070-8dc4991b8623"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3/3 - 0s - loss: 0.3237 - accuracy: 0.9268 - auc: 0.9493\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[0.32368770241737366, 0.9268292784690857, 0.9492517709732056]"
      ]
     },
     "execution_count": 20,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.evaluate(X_test, Y_test, verbose=2, workers=-1, use_multiprocessing=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "XuS52Or_j_im"
   },
   "outputs": [],
   "source": [
    "from sklearn.metrics import confusion_matrix, roc_curve, roc_auc_score\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "mdOMEPwJkKT0"
   },
   "outputs": [],
   "source": [
    "y_pred_p = model.predict(X_test).ravel()\n",
    "y_true = Y_test\n",
    "y_pred = np.ones(y_pred_p.shape)\n",
    "y_pred[y_pred_p<0.4]=0\n",
    "results = confusion_matrix(y_true, y_pred, normalize=\"true\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 734
    },
    "colab_type": "code",
    "id": "wMC0y97WHfqN",
    "outputId": "61e641c4-6e29-421a-b93f-36b94ebd171d"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 4 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "cm = sns.clustermap(results, \n",
    "                    vmax=1,  \n",
    "                    row_cluster=False, \n",
    "                    col_cluster=False, \n",
    "                    xticklabels=classes, \n",
    "                    yticklabels=classes, \n",
    "                    annot=False,\n",
    "                    annot_kws={\"fontsize\":25})\n",
    "ax = cm.ax_heatmap\n",
    "fig = ax.get_figure()\n",
    "ax.set_ylabel(\"real\", fontsize=35, rotation=90)\n",
    "ax.set_yticklabels(labels=classes, rotation=0)\n",
    "ax.yaxis.tick_left()\n",
    "ax.yaxis.set_label_position(\"left\")\n",
    "\n",
    "ax.set_xticklabels(labels=classes, rotation=45)\n",
    "ax.set_xlabel(\"predicted\",fontsize=35)\n",
    "ax.tick_params(labelsize=35)\n",
    "\n",
    "cax = cm.ax_cbar\n",
    "cax.tick_params(labelsize=30)\n",
    "cax.set_title(\"P()\", fontsize=30)\n",
    "plt.tight_layout()\n",
    "cm.savefig(f\"predict_{label}.pdf\")\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 927
    },
    "colab_type": "code",
    "id": "9_T2el1pHfqT",
    "outputId": "5be6d658-87a7-4cd6-a5e1-f83682d6a4b4"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x1080 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax= plt.subplots(figsize=(18,15))\n",
    "\n",
    "ax.hist(y_pred_p[Y_test==0], histtype=\"step\", lw=15, bins=10, density=True, label=classes[0])\n",
    "ax.hist(y_pred_p[Y_test==1], histtype=\"step\", lw=15, bins=10, density=True, label=classes[1])\n",
    "\n",
    "ax.tick_params(labelsize=35)\n",
    "ax.legend(fontsize=35, loc=\"upper left\")\n",
    "ax.set_title(\"\", fontsize=35)\n",
    "\n",
    "ax.set_xlim(-0.05,1.05)\n",
    "\n",
    "ax.set_xlabel(\"Z\", fontsize=35)\n",
    "ax.set_ylabel(\"pdf\", fontsize=35)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "nc5dF_kK-8YY"
   },
   "outputs": [],
   "source": [
    "fpr, tpr, thresholds = roc_curve(y_true, y_pred_p)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 927
    },
    "colab_type": "code",
    "id": "9F92ZJJ1vWVp",
    "outputId": "d85aa98f-aad5-4dfc-b81f-286ca26ea748"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x1080 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(18,15))\n",
    "\n",
    "ax.plot(fpr,tpr, c=\"gray\", lw=15)\n",
    "\n",
    "ax.plot([0, 1], [0, 1], 'k--', lw=15)\n",
    "ax.set_xlabel('False positive rate', fontsize=35)\n",
    "ax.set_ylabel('True positive rate', fontsize=35)\n",
    "\n",
    "\n",
    "for f, t, thr in zip(fpr[::100], tpr[::100], thresholds[::100]):\n",
    "    ax.annotate(thr, (f, t), fontsize=25)\n",
    "\n",
    "ax.annotate(\"Area Under Curve = %.4f\"%roc_auc_score(y_true, y_pred_p), xy=(0.6, 0.5), fontsize=25)\n",
    "\n",
    "ax.tick_params(labelsize=35)\n",
    "\n",
    "plt.show()\n",
    "fig.savefig(f\"roc_{label}.pdf\")"
   ]
  },
 ],
 "metadata": {
  "accelerator": "GPU",
  "colab": {
   "collapsed_sections": [],
   "name": "predictor_LUAD_LUSC.ipynb",
   "provenance": [],
   "toc_visible": true
  },
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}