neural_train_eval.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Realizar os imports:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "import keras\n",
    "import tensorflow as tf\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Conv1D, MaxPooling1D, Dropout, BatchNormalization, Flatten, Dense\n",
    "from sklearn.preprocessing import StandardScaler, LabelEncoder\n",
    "from sklearn.model_selection import StratifiedKFold\n",
    "from sklearn.metrics import f1_score, confusion_matrix\n",
    "import seaborn as sns\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import os\n",
    "import csv"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Selecionar as features:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "feature_method = \"aug_rms\"\n",
    "dataset = pd.read_csv(f'./features/emoUERJ_features/{feature_method}_features.csv')\n",
    "result_folder = f'./results/{feature_method}/'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Formatar os dados:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classes codificadas:{0: 'felicidade', 1: 'neutro', 2: 'raiva', 3: 'tristeza'}\n"
     ]
    }
   ],
   "source": [
    "num_classes = 4\n",
    "\n",
    "dataset = dataset.iloc[:,1:]\n",
    "X = dataset.drop('label', axis=1)  # Features  # Features\n",
    "# Create a LabelEncoder\n",
    "label_encoder = LabelEncoder()\n",
    "\n",
    "# Fit and transform the \"label\" column\n",
    "y = label_encoder.fit_transform(dataset[\"label\"])\n",
    "\n",
    "text_labels = label_encoder.classes_\n",
    "encoded_to_class_mapping = dict(zip(label_encoder.transform(text_labels), text_labels))\n",
    "print(\"Classes codificadas:\" + str(encoded_to_class_mapping))\n",
    "\n",
    "# Preprocess Data (example: standardizing numerical features)\n",
    "scaler = StandardScaler()\n",
    "X = scaler.fit_transform(X)\n",
    "\n",
    "# # Adjust sequence_length and input_dim based on your actual data\n",
    "X = X.reshape((X.shape[0], X.shape[1], 1))\n",
    "\n",
    "# Define the number of folds\n",
    "num_folds = 5\n",
    "\n",
    "# Initialize StratifiedKFold\n",
    "skf = StratifiedKFold(n_splits=num_folds, shuffle=True, random_state=42)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Escolher a arquitetura:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "# ### My architecture\n",
    "def create_model():\n",
    "    # Define the model\n",
    "    model = Sequential()\n",
    "    # Convolutional layers\n",
    "    model.add(Conv1D(filters=64, kernel_size=3, activation='relu', input_shape=X.shape[1:]))\n",
    "    model.add(MaxPooling1D(pool_size=2))\n",
    "    model.add(BatchNormalization())\n",
    "    model.add(Dropout(0.25))\n",
    "    model.add(Conv1D(filters=128, kernel_size=3, activation='relu'))\n",
    "    model.add(MaxPooling1D(pool_size=2))\n",
    "    model.add(BatchNormalization())\n",
    "    model.add(Dropout(0.25))\n",
    "    model.add(Conv1D(filters=256, kernel_size=3, activation='relu'))\n",
    "    model.add(MaxPooling1D(pool_size=2))\n",
    "    model.add(BatchNormalization())\n",
    "    model.add(Dropout(0.25))\n",
    "    # Flatten layer\n",
    "    model.add(Flatten())\n",
    "    # Dense layers\n",
    "    model.add(Dense(512, activation='relu'))\n",
    "    model.add(BatchNormalization())\n",
    "    model.add(Dropout(0.5))\n",
    "    # Output layer\n",
    "    model.add(Dense(num_classes, activation='softmax'))  # Replace 'num_classes' with the number of output classes\n",
    "    return model\n",
    "\n",
    "# keras.utils.plot_model(model, show_shapes=True)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "## Paper architecture (Overfitting)\n",
    "def create_model():\n",
    "    # Define the model\n",
    "    model = Sequential()\n",
    "    # Convolutional layers\n",
    "    model.add(Conv1D(filters=64, kernel_size=3, activation='relu', input_shape=X.shape[1:]))\n",
    "    model.add(Conv1D(filters=128, kernel_size=3, activation='relu'))\n",
    "    model.add(BatchNormalization())\n",
    "    model.add(MaxPooling1D(pool_size=2))\n",
    "    model.add(Dropout(0.5))\n",
    "    model.add(Conv1D(filters=256, kernel_size=3, activation='relu'))\n",
    "    model.add(BatchNormalization())\n",
    "    model.add(MaxPooling1D(pool_size=2))\n",
    "    model.add(Dropout(0.5))\n",
    "    # Flatten layer\n",
    "    model.add(Flatten())\n",
    "    # Dense layers\n",
    "    model.add(Dense(64, activation='relu'))\n",
    "    # Output layer\n",
    "    model.add(Dense(num_classes, activation='softmax'))  # Replace 'num_classes' with the number of output classes\n",
    "    return model\n",
    "\n",
    "# keras.utils.plot_model(model, show_shapes=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Treinar o modelo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training on fold 1\n",
      "Epoch 1/100\n",
      "76/76 [==============================] - 6s 55ms/step - loss: 1.5825 - f1_metric: 0.5910 - val_loss: 1.9557 - val_f1_metric: 0.2757 - lr: 0.0010\n",
      "Epoch 2/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.9579 - f1_metric: 0.7008 - val_loss: 1.8300 - val_f1_metric: 0.2495 - lr: 0.0010\n",
      "Epoch 3/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.8397 - f1_metric: 0.7327 - val_loss: 2.3847 - val_f1_metric: 0.2046 - lr: 0.0010\n",
      "Epoch 4/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.7238 - f1_metric: 0.7835 - val_loss: 2.3657 - val_f1_metric: 0.2595 - lr: 0.0010\n",
      "Epoch 5/100\n",
      "76/76 [==============================] - 4s 53ms/step - loss: 0.7265 - f1_metric: 0.7726 - val_loss: 1.4658 - val_f1_metric: 0.3837 - lr: 0.0010\n",
      "Epoch 6/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.5498 - f1_metric: 0.8115 - val_loss: 1.0819 - val_f1_metric: 0.4883 - lr: 0.0010\n",
      "Epoch 7/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.5306 - f1_metric: 0.8302 - val_loss: 1.0813 - val_f1_metric: 0.5744 - lr: 0.0010\n",
      "Epoch 8/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.4574 - f1_metric: 0.8541 - val_loss: 0.6755 - val_f1_metric: 0.7114 - lr: 0.0010\n",
      "Epoch 9/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.4139 - f1_metric: 0.8583 - val_loss: 0.6124 - val_f1_metric: 0.7406 - lr: 0.0010\n",
      "Epoch 10/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.3878 - f1_metric: 0.8581 - val_loss: 0.5916 - val_f1_metric: 0.8066 - lr: 0.0010\n",
      "Epoch 11/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.4005 - f1_metric: 0.8870 - val_loss: 0.6992 - val_f1_metric: 0.7272 - lr: 0.0010\n",
      "Epoch 12/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.3404 - f1_metric: 0.8859 - val_loss: 0.5446 - val_f1_metric: 0.7961 - lr: 0.0010\n",
      "Epoch 13/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.3881 - f1_metric: 0.8781 - val_loss: 0.5227 - val_f1_metric: 0.8203 - lr: 0.0010\n",
      "Epoch 14/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.2297 - f1_metric: 0.9220 - val_loss: 0.4912 - val_f1_metric: 0.8360 - lr: 0.0010\n",
      "Epoch 15/100\n",
      "76/76 [==============================] - 4s 53ms/step - loss: 0.2292 - f1_metric: 0.9228 - val_loss: 0.4888 - val_f1_metric: 0.8415 - lr: 0.0010\n",
      "Epoch 16/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.2630 - f1_metric: 0.9169 - val_loss: 0.4735 - val_f1_metric: 0.8269 - lr: 0.0010\n",
      "Epoch 17/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.2893 - f1_metric: 0.9003 - val_loss: 0.5995 - val_f1_metric: 0.8253 - lr: 0.0010\n",
      "Epoch 18/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.3431 - f1_metric: 0.8885 - val_loss: 0.6465 - val_f1_metric: 0.7468 - lr: 0.0010\n",
      "Epoch 19/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.2238 - f1_metric: 0.9174 - val_loss: 0.4856 - val_f1_metric: 0.8406 - lr: 0.0010\n",
      "Epoch 20/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.2190 - f1_metric: 0.9442 - val_loss: 0.6277 - val_f1_metric: 0.7356 - lr: 0.0010\n",
      "Epoch 21/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1714 - f1_metric: 0.9581 - val_loss: 0.4509 - val_f1_metric: 0.8452 - lr: 0.0010\n",
      "Epoch 22/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1546 - f1_metric: 0.9420 - val_loss: 0.5593 - val_f1_metric: 0.8343 - lr: 0.0010\n",
      "Epoch 23/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.2003 - f1_metric: 0.9414 - val_loss: 0.4870 - val_f1_metric: 0.8619 - lr: 0.0010\n",
      "Epoch 24/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1753 - f1_metric: 0.9355 - val_loss: 0.5162 - val_f1_metric: 0.8360 - lr: 0.0010\n",
      "Epoch 25/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.2292 - f1_metric: 0.9333 - val_loss: 0.5343 - val_f1_metric: 0.8456 - lr: 0.0010\n",
      "Epoch 26/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.2117 - f1_metric: 0.9361 - val_loss: 0.6411 - val_f1_metric: 0.7228 - lr: 0.0010\n",
      "Epoch 27/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.2994 - f1_metric: 0.9150 - val_loss: 0.5273 - val_f1_metric: 0.8421 - lr: 0.0010\n",
      "Epoch 28/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.2145 - f1_metric: 0.9320 - val_loss: 0.4352 - val_f1_metric: 0.8408 - lr: 0.0010\n",
      "Epoch 29/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1652 - f1_metric: 0.9481 - val_loss: 0.5028 - val_f1_metric: 0.8493 - lr: 0.0010\n",
      "Epoch 30/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1805 - f1_metric: 0.9478 - val_loss: 0.4660 - val_f1_metric: 0.8506 - lr: 0.0010\n",
      "Epoch 31/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1743 - f1_metric: 0.9496 - val_loss: 0.5153 - val_f1_metric: 0.8421 - lr: 0.0010\n",
      "Epoch 32/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1215 - f1_metric: 0.9616 - val_loss: 0.6120 - val_f1_metric: 0.7861 - lr: 0.0010\n",
      "Epoch 33/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1721 - f1_metric: 0.9567 - val_loss: 0.5412 - val_f1_metric: 0.8110 - lr: 0.0010\n",
      "Epoch 34/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1507 - f1_metric: 0.9638 - val_loss: 0.5598 - val_f1_metric: 0.8358 - lr: 0.0010\n",
      "Epoch 35/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1300 - f1_metric: 0.9587 - val_loss: 0.5696 - val_f1_metric: 0.8228 - lr: 0.0010\n",
      "Epoch 36/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1768 - f1_metric: 0.9423 - val_loss: 0.6128 - val_f1_metric: 0.8252 - lr: 0.0010\n",
      "Epoch 37/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1492 - f1_metric: 0.9438 - val_loss: 0.5393 - val_f1_metric: 0.8133 - lr: 0.0010\n",
      "Epoch 38/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.2039 - f1_metric: 0.9488 - val_loss: 0.6187 - val_f1_metric: 0.8179 - lr: 0.0010\n",
      "Epoch 39/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.2480 - f1_metric: 0.9228 - val_loss: 0.5323 - val_f1_metric: 0.8589 - lr: 0.0010\n",
      "Epoch 40/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.2706 - f1_metric: 0.9151 - val_loss: 0.7143 - val_f1_metric: 0.8063 - lr: 0.0010\n",
      "Epoch 41/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1900 - f1_metric: 0.9480 - val_loss: 0.5279 - val_f1_metric: 0.8164 - lr: 0.0010\n",
      "Epoch 42/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1618 - f1_metric: 0.9527 - val_loss: 0.4578 - val_f1_metric: 0.8533 - lr: 0.0010\n",
      "Epoch 43/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1704 - f1_metric: 0.9469 - val_loss: 0.4766 - val_f1_metric: 0.8496 - lr: 0.0010\n",
      "Epoch 44/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1615 - f1_metric: 0.9613 - val_loss: 0.4980 - val_f1_metric: 0.8432 - lr: 0.0010\n",
      "Epoch 45/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1444 - f1_metric: 0.9617 - val_loss: 0.4801 - val_f1_metric: 0.8465 - lr: 0.0010\n",
      "Epoch 46/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1304 - f1_metric: 0.9602 - val_loss: 0.4735 - val_f1_metric: 0.8730 - lr: 0.0010\n",
      "Epoch 47/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1328 - f1_metric: 0.9772 - val_loss: 0.4906 - val_f1_metric: 0.8531 - lr: 0.0010\n",
      "Epoch 48/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0845 - f1_metric: 0.9744 - val_loss: 0.4820 - val_f1_metric: 0.8469 - lr: 0.0010\n",
      "Epoch 49/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0985 - f1_metric: 0.9732 - val_loss: 0.4772 - val_f1_metric: 0.8693 - lr: 5.0000e-04\n",
      "Epoch 50/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0413 - f1_metric: 0.9841 - val_loss: 0.5199 - val_f1_metric: 0.8898 - lr: 5.0000e-04\n",
      "Epoch 51/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0782 - f1_metric: 0.9794 - val_loss: 0.4868 - val_f1_metric: 0.8576 - lr: 5.0000e-04\n",
      "Epoch 52/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1268 - f1_metric: 0.9686 - val_loss: 0.4940 - val_f1_metric: 0.8839 - lr: 5.0000e-04\n",
      "Epoch 53/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0960 - f1_metric: 0.9697 - val_loss: 0.5650 - val_f1_metric: 0.8557 - lr: 5.0000e-04\n",
      "Epoch 54/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0951 - f1_metric: 0.9735 - val_loss: 0.5518 - val_f1_metric: 0.8762 - lr: 5.0000e-04\n",
      "Epoch 55/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0736 - f1_metric: 0.9728 - val_loss: 0.4955 - val_f1_metric: 0.8795 - lr: 5.0000e-04\n",
      "Epoch 56/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0447 - f1_metric: 0.9850 - val_loss: 0.4952 - val_f1_metric: 0.8662 - lr: 5.0000e-04\n",
      "Epoch 57/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0384 - f1_metric: 0.9926 - val_loss: 0.5167 - val_f1_metric: 0.8690 - lr: 5.0000e-04\n",
      "Epoch 58/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0518 - f1_metric: 0.9858 - val_loss: 0.5099 - val_f1_metric: 0.8657 - lr: 5.0000e-04\n",
      "Epoch 59/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0767 - f1_metric: 0.9772 - val_loss: 0.4786 - val_f1_metric: 0.8822 - lr: 5.0000e-04\n",
      "Epoch 60/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0695 - f1_metric: 0.9816 - val_loss: 0.4851 - val_f1_metric: 0.8610 - lr: 5.0000e-04\n",
      "Epoch 61/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0936 - f1_metric: 0.9776 - val_loss: 0.4639 - val_f1_metric: 0.8702 - lr: 5.0000e-04\n",
      "Epoch 62/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0670 - f1_metric: 0.9801 - val_loss: 0.4749 - val_f1_metric: 0.8743 - lr: 5.0000e-04\n",
      "Epoch 63/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0516 - f1_metric: 0.9838 - val_loss: 0.4592 - val_f1_metric: 0.8895 - lr: 5.0000e-04\n",
      "Epoch 64/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0450 - f1_metric: 0.9910 - val_loss: 0.4799 - val_f1_metric: 0.8765 - lr: 5.0000e-04\n",
      "Epoch 65/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0464 - f1_metric: 0.9894 - val_loss: 0.4696 - val_f1_metric: 0.8888 - lr: 5.0000e-04\n",
      "Epoch 66/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0697 - f1_metric: 0.9844 - val_loss: 0.4643 - val_f1_metric: 0.8823 - lr: 5.0000e-04\n",
      "Epoch 67/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0825 - f1_metric: 0.9734 - val_loss: 0.4611 - val_f1_metric: 0.8800 - lr: 5.0000e-04\n",
      "Epoch 68/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0955 - f1_metric: 0.9722 - val_loss: 0.4825 - val_f1_metric: 0.8650 - lr: 5.0000e-04\n",
      "Epoch 69/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0454 - f1_metric: 0.9874 - val_loss: 0.4754 - val_f1_metric: 0.8661 - lr: 2.5000e-04\n",
      "Epoch 70/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0529 - f1_metric: 0.9908 - val_loss: 0.4787 - val_f1_metric: 0.8701 - lr: 2.5000e-04\n",
      "Epoch 71/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0701 - f1_metric: 0.9780 - val_loss: 0.4559 - val_f1_metric: 0.8624 - lr: 2.5000e-04\n",
      "Epoch 72/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0592 - f1_metric: 0.9874 - val_loss: 0.4470 - val_f1_metric: 0.8820 - lr: 2.5000e-04\n",
      "Epoch 73/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0621 - f1_metric: 0.9843 - val_loss: 0.4632 - val_f1_metric: 0.8612 - lr: 2.5000e-04\n",
      "Epoch 74/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0521 - f1_metric: 0.9842 - val_loss: 0.5178 - val_f1_metric: 0.8678 - lr: 2.5000e-04\n",
      "Epoch 75/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0403 - f1_metric: 0.9905 - val_loss: 0.4703 - val_f1_metric: 0.8850 - lr: 2.5000e-04\n",
      "Epoch 76/100\n",
      "76/76 [==============================] - 4s 53ms/step - loss: 0.0454 - f1_metric: 0.9902 - val_loss: 0.5074 - val_f1_metric: 0.8752 - lr: 2.5000e-04\n",
      "Epoch 77/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0367 - f1_metric: 0.9922 - val_loss: 0.4735 - val_f1_metric: 0.8937 - lr: 2.5000e-04\n",
      "Epoch 78/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0466 - f1_metric: 0.9872 - val_loss: 0.4674 - val_f1_metric: 0.8877 - lr: 2.5000e-04\n",
      "Epoch 78: early stopping\n",
      "5/5 [==============================] - 0s 12ms/step\n",
      "Training on fold 2\n",
      "Epoch 1/100\n",
      "76/76 [==============================] - 6s 55ms/step - loss: 1.5533 - f1_metric: 0.5756 - val_loss: 2.1275 - val_f1_metric: 0.1867 - lr: 0.0010\n",
      "Epoch 2/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.9071 - f1_metric: 0.7326 - val_loss: 1.9681 - val_f1_metric: 0.3640 - lr: 0.0010\n",
      "Epoch 3/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.7957 - f1_metric: 0.7501 - val_loss: 1.5217 - val_f1_metric: 0.3641 - lr: 0.0010\n",
      "Epoch 4/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.7104 - f1_metric: 0.7493 - val_loss: 1.6408 - val_f1_metric: 0.3438 - lr: 0.0010\n",
      "Epoch 5/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.6779 - f1_metric: 0.7647 - val_loss: 1.3915 - val_f1_metric: 0.4239 - lr: 0.0010\n",
      "Epoch 6/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.5469 - f1_metric: 0.8205 - val_loss: 1.1598 - val_f1_metric: 0.5296 - lr: 0.0010\n",
      "Epoch 7/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.4329 - f1_metric: 0.8535 - val_loss: 0.9254 - val_f1_metric: 0.6823 - lr: 0.0010\n",
      "Epoch 8/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.4827 - f1_metric: 0.8413 - val_loss: 0.6070 - val_f1_metric: 0.8040 - lr: 0.0010\n",
      "Epoch 9/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.4086 - f1_metric: 0.8488 - val_loss: 0.4636 - val_f1_metric: 0.8603 - lr: 0.0010\n",
      "Epoch 10/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.4685 - f1_metric: 0.8289 - val_loss: 0.5717 - val_f1_metric: 0.8490 - lr: 0.0010\n",
      "Epoch 11/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.4983 - f1_metric: 0.8144 - val_loss: 0.5432 - val_f1_metric: 0.8261 - lr: 0.0010\n",
      "Epoch 12/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.4141 - f1_metric: 0.8770 - val_loss: 0.5106 - val_f1_metric: 0.8306 - lr: 0.0010\n",
      "Epoch 13/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.3597 - f1_metric: 0.8781 - val_loss: 0.6433 - val_f1_metric: 0.8061 - lr: 0.0010\n",
      "Epoch 14/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.3981 - f1_metric: 0.8648 - val_loss: 0.4778 - val_f1_metric: 0.8567 - lr: 0.0010\n",
      "Epoch 15/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.3598 - f1_metric: 0.8769 - val_loss: 0.4606 - val_f1_metric: 0.8695 - lr: 0.0010\n",
      "Epoch 16/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.2882 - f1_metric: 0.9005 - val_loss: 0.5331 - val_f1_metric: 0.8216 - lr: 0.0010\n",
      "Epoch 17/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.2876 - f1_metric: 0.9031 - val_loss: 0.6601 - val_f1_metric: 0.8277 - lr: 0.0010\n",
      "Epoch 18/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.2266 - f1_metric: 0.9339 - val_loss: 0.4765 - val_f1_metric: 0.8639 - lr: 0.0010\n",
      "Epoch 19/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.2153 - f1_metric: 0.9359 - val_loss: 0.5133 - val_f1_metric: 0.8762 - lr: 0.0010\n",
      "Epoch 20/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1642 - f1_metric: 0.9393 - val_loss: 0.6117 - val_f1_metric: 0.8497 - lr: 0.0010\n",
      "Epoch 21/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1809 - f1_metric: 0.9388 - val_loss: 0.5379 - val_f1_metric: 0.8627 - lr: 0.0010\n",
      "Epoch 22/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1816 - f1_metric: 0.9476 - val_loss: 0.4817 - val_f1_metric: 0.8701 - lr: 0.0010\n",
      "Epoch 23/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.2281 - f1_metric: 0.9319 - val_loss: 0.4705 - val_f1_metric: 0.8793 - lr: 0.0010\n",
      "Epoch 24/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1563 - f1_metric: 0.9448 - val_loss: 0.5222 - val_f1_metric: 0.8760 - lr: 0.0010\n",
      "Epoch 25/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1731 - f1_metric: 0.9452 - val_loss: 0.5596 - val_f1_metric: 0.8839 - lr: 0.0010\n",
      "Epoch 26/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1745 - f1_metric: 0.9498 - val_loss: 0.5304 - val_f1_metric: 0.8829 - lr: 0.0010\n",
      "Epoch 27/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1895 - f1_metric: 0.9442 - val_loss: 0.4957 - val_f1_metric: 0.8932 - lr: 0.0010\n",
      "Epoch 28/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1850 - f1_metric: 0.9343 - val_loss: 0.5272 - val_f1_metric: 0.8280 - lr: 0.0010\n",
      "Epoch 29/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1648 - f1_metric: 0.9582 - val_loss: 0.6044 - val_f1_metric: 0.8576 - lr: 0.0010\n",
      "Epoch 30/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1901 - f1_metric: 0.9408 - val_loss: 0.6159 - val_f1_metric: 0.8773 - lr: 0.0010\n",
      "Epoch 31/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1343 - f1_metric: 0.9537 - val_loss: 0.6246 - val_f1_metric: 0.8491 - lr: 0.0010\n",
      "Epoch 32/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1460 - f1_metric: 0.9532 - val_loss: 0.5368 - val_f1_metric: 0.8823 - lr: 0.0010\n",
      "Epoch 33/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1573 - f1_metric: 0.9481 - val_loss: 0.6039 - val_f1_metric: 0.8707 - lr: 0.0010\n",
      "Epoch 34/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.2025 - f1_metric: 0.9468 - val_loss: 0.6912 - val_f1_metric: 0.8360 - lr: 0.0010\n",
      "Epoch 35/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1837 - f1_metric: 0.9568 - val_loss: 0.7622 - val_f1_metric: 0.8489 - lr: 0.0010\n",
      "Epoch 36/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1622 - f1_metric: 0.9461 - val_loss: 0.6752 - val_f1_metric: 0.8470 - lr: 5.0000e-04\n",
      "Epoch 37/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1545 - f1_metric: 0.9543 - val_loss: 0.6529 - val_f1_metric: 0.8620 - lr: 5.0000e-04\n",
      "Epoch 38/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0955 - f1_metric: 0.9789 - val_loss: 0.7007 - val_f1_metric: 0.8567 - lr: 5.0000e-04\n",
      "Epoch 39/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1279 - f1_metric: 0.9606 - val_loss: 0.6060 - val_f1_metric: 0.8575 - lr: 5.0000e-04\n",
      "Epoch 40/100\n",
      "76/76 [==============================] - 4s 53ms/step - loss: 0.0852 - f1_metric: 0.9747 - val_loss: 0.5792 - val_f1_metric: 0.8712 - lr: 5.0000e-04\n",
      "Epoch 41/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0527 - f1_metric: 0.9877 - val_loss: 0.5756 - val_f1_metric: 0.8686 - lr: 5.0000e-04\n",
      "Epoch 42/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0517 - f1_metric: 0.9873 - val_loss: 0.5693 - val_f1_metric: 0.8654 - lr: 5.0000e-04\n",
      "Epoch 43/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0671 - f1_metric: 0.9861 - val_loss: 0.5947 - val_f1_metric: 0.8441 - lr: 5.0000e-04\n",
      "Epoch 44/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0508 - f1_metric: 0.9882 - val_loss: 0.6103 - val_f1_metric: 0.8711 - lr: 5.0000e-04\n",
      "Epoch 45/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0562 - f1_metric: 0.9875 - val_loss: 0.6384 - val_f1_metric: 0.8512 - lr: 5.0000e-04\n",
      "Epoch 46/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0605 - f1_metric: 0.9840 - val_loss: 0.6411 - val_f1_metric: 0.8545 - lr: 5.0000e-04\n",
      "Epoch 47/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0953 - f1_metric: 0.9740 - val_loss: 0.5875 - val_f1_metric: 0.8286 - lr: 5.0000e-04\n",
      "Epoch 48/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0788 - f1_metric: 0.9729 - val_loss: 0.6113 - val_f1_metric: 0.8811 - lr: 5.0000e-04\n",
      "Epoch 49/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0742 - f1_metric: 0.9797 - val_loss: 0.5345 - val_f1_metric: 0.8520 - lr: 5.0000e-04\n",
      "Epoch 50/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0679 - f1_metric: 0.9822 - val_loss: 0.5517 - val_f1_metric: 0.8675 - lr: 5.0000e-04\n",
      "Epoch 51/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1051 - f1_metric: 0.9592 - val_loss: 0.5860 - val_f1_metric: 0.8705 - lr: 5.0000e-04\n",
      "Epoch 52/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1007 - f1_metric: 0.9699 - val_loss: 0.7111 - val_f1_metric: 0.8460 - lr: 5.0000e-04\n",
      "Epoch 53/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0735 - f1_metric: 0.9737 - val_loss: 0.7181 - val_f1_metric: 0.8335 - lr: 5.0000e-04\n",
      "Epoch 54/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1522 - f1_metric: 0.9668 - val_loss: 0.6564 - val_f1_metric: 0.8277 - lr: 5.0000e-04\n",
      "Epoch 55/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1130 - f1_metric: 0.9721 - val_loss: 0.6186 - val_f1_metric: 0.8488 - lr: 5.0000e-04\n",
      "Epoch 56/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1115 - f1_metric: 0.9704 - val_loss: 0.5889 - val_f1_metric: 0.8690 - lr: 2.5000e-04\n",
      "Epoch 57/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1242 - f1_metric: 0.9736 - val_loss: 0.5233 - val_f1_metric: 0.8745 - lr: 2.5000e-04\n",
      "Epoch 58/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0718 - f1_metric: 0.9791 - val_loss: 0.5268 - val_f1_metric: 0.8593 - lr: 2.5000e-04\n",
      "Epoch 59/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0602 - f1_metric: 0.9878 - val_loss: 0.5077 - val_f1_metric: 0.8708 - lr: 2.5000e-04\n",
      "Epoch 60/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0393 - f1_metric: 0.9930 - val_loss: 0.5402 - val_f1_metric: 0.8665 - lr: 2.5000e-04\n",
      "Epoch 61/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0496 - f1_metric: 0.9886 - val_loss: 0.5616 - val_f1_metric: 0.8418 - lr: 2.5000e-04\n",
      "Epoch 62/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0366 - f1_metric: 0.9928 - val_loss: 0.5535 - val_f1_metric: 0.8629 - lr: 2.5000e-04\n",
      "Epoch 63/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0696 - f1_metric: 0.9796 - val_loss: 0.6016 - val_f1_metric: 0.8837 - lr: 2.5000e-04\n",
      "Epoch 64/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1019 - f1_metric: 0.9791 - val_loss: 0.5561 - val_f1_metric: 0.8746 - lr: 2.5000e-04\n",
      "Epoch 65/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0406 - f1_metric: 0.9908 - val_loss: 0.5516 - val_f1_metric: 0.8687 - lr: 2.5000e-04\n",
      "Epoch 65: early stopping\n",
      "Training on fold 3\n",
      "Epoch 1/100\n",
      "76/76 [==============================] - 6s 55ms/step - loss: 1.6256 - f1_metric: 0.5677 - val_loss: 1.7789 - val_f1_metric: 0.3101 - lr: 0.0010\n",
      "Epoch 2/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.9181 - f1_metric: 0.6902 - val_loss: 2.0334 - val_f1_metric: 0.2309 - lr: 0.0010\n",
      "Epoch 3/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.8215 - f1_metric: 0.7263 - val_loss: 1.4709 - val_f1_metric: 0.2808 - lr: 0.0010\n",
      "Epoch 4/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.7585 - f1_metric: 0.7336 - val_loss: 1.2430 - val_f1_metric: 0.4385 - lr: 0.0010\n",
      "Epoch 5/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.6704 - f1_metric: 0.7561 - val_loss: 1.0963 - val_f1_metric: 0.4969 - lr: 0.0010\n",
      "Epoch 6/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.5573 - f1_metric: 0.8065 - val_loss: 0.7363 - val_f1_metric: 0.6567 - lr: 0.0010\n",
      "Epoch 7/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.4973 - f1_metric: 0.8381 - val_loss: 0.7264 - val_f1_metric: 0.6651 - lr: 0.0010\n",
      "Epoch 8/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.4461 - f1_metric: 0.8621 - val_loss: 0.5749 - val_f1_metric: 0.7489 - lr: 0.0010\n",
      "Epoch 9/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.4237 - f1_metric: 0.8723 - val_loss: 0.6296 - val_f1_metric: 0.7157 - lr: 0.0010\n",
      "Epoch 10/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.3800 - f1_metric: 0.8927 - val_loss: 0.4430 - val_f1_metric: 0.8269 - lr: 0.0010\n",
      "Epoch 11/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.3793 - f1_metric: 0.8837 - val_loss: 0.4523 - val_f1_metric: 0.8127 - lr: 0.0010\n",
      "Epoch 12/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.4010 - f1_metric: 0.8633 - val_loss: 0.4550 - val_f1_metric: 0.8333 - lr: 0.0010\n",
      "Epoch 13/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.3277 - f1_metric: 0.8986 - val_loss: 0.4533 - val_f1_metric: 0.8246 - lr: 0.0010\n",
      "Epoch 14/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.2958 - f1_metric: 0.8949 - val_loss: 0.4100 - val_f1_metric: 0.8655 - lr: 0.0010\n",
      "Epoch 15/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.2671 - f1_metric: 0.9214 - val_loss: 0.4032 - val_f1_metric: 0.8331 - lr: 0.0010\n",
      "Epoch 16/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.3546 - f1_metric: 0.8941 - val_loss: 0.8125 - val_f1_metric: 0.7515 - lr: 0.0010\n",
      "Epoch 17/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.2997 - f1_metric: 0.9131 - val_loss: 0.4457 - val_f1_metric: 0.8181 - lr: 0.0010\n",
      "Epoch 18/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.2891 - f1_metric: 0.9056 - val_loss: 0.4756 - val_f1_metric: 0.8530 - lr: 0.0010\n",
      "Epoch 19/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.4079 - f1_metric: 0.8683 - val_loss: 0.6852 - val_f1_metric: 0.7759 - lr: 0.0010\n",
      "Epoch 20/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.3272 - f1_metric: 0.8812 - val_loss: 0.6424 - val_f1_metric: 0.8059 - lr: 0.0010\n",
      "Epoch 21/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.2611 - f1_metric: 0.9298 - val_loss: 0.4378 - val_f1_metric: 0.8555 - lr: 0.0010\n",
      "Epoch 22/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.2694 - f1_metric: 0.9074 - val_loss: 0.4025 - val_f1_metric: 0.8467 - lr: 0.0010\n",
      "Epoch 23/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.2148 - f1_metric: 0.9303 - val_loss: 0.4084 - val_f1_metric: 0.8394 - lr: 0.0010\n",
      "Epoch 24/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1926 - f1_metric: 0.9428 - val_loss: 0.3901 - val_f1_metric: 0.8613 - lr: 0.0010\n",
      "Epoch 25/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1967 - f1_metric: 0.9415 - val_loss: 0.3802 - val_f1_metric: 0.8812 - lr: 0.0010\n",
      "Epoch 26/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1521 - f1_metric: 0.9463 - val_loss: 0.4043 - val_f1_metric: 0.8456 - lr: 0.0010\n",
      "Epoch 27/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1541 - f1_metric: 0.9564 - val_loss: 0.3472 - val_f1_metric: 0.8504 - lr: 0.0010\n",
      "Epoch 28/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1969 - f1_metric: 0.9303 - val_loss: 0.5553 - val_f1_metric: 0.8500 - lr: 0.0010\n",
      "Epoch 29/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1883 - f1_metric: 0.9440 - val_loss: 0.3406 - val_f1_metric: 0.8592 - lr: 0.0010\n",
      "Epoch 30/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1327 - f1_metric: 0.9596 - val_loss: 0.4770 - val_f1_metric: 0.8715 - lr: 0.0010\n",
      "Epoch 31/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1787 - f1_metric: 0.9530 - val_loss: 0.4174 - val_f1_metric: 0.8672 - lr: 0.0010\n",
      "Epoch 32/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1819 - f1_metric: 0.9431 - val_loss: 0.4677 - val_f1_metric: 0.8672 - lr: 0.0010\n",
      "Epoch 33/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1944 - f1_metric: 0.9454 - val_loss: 0.4948 - val_f1_metric: 0.8238 - lr: 0.0010\n",
      "Epoch 34/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1565 - f1_metric: 0.9407 - val_loss: 0.4266 - val_f1_metric: 0.8360 - lr: 0.0010\n",
      "Epoch 35/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1318 - f1_metric: 0.9555 - val_loss: 0.4893 - val_f1_metric: 0.8313 - lr: 0.0010\n",
      "Epoch 36/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1367 - f1_metric: 0.9661 - val_loss: 0.3884 - val_f1_metric: 0.8693 - lr: 0.0010\n",
      "Epoch 37/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1140 - f1_metric: 0.9759 - val_loss: 0.4784 - val_f1_metric: 0.8352 - lr: 0.0010\n",
      "Epoch 38/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1295 - f1_metric: 0.9693 - val_loss: 0.5396 - val_f1_metric: 0.8026 - lr: 0.0010\n",
      "Epoch 39/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1640 - f1_metric: 0.9609 - val_loss: 0.4975 - val_f1_metric: 0.8297 - lr: 0.0010\n",
      "Epoch 40/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1086 - f1_metric: 0.9682 - val_loss: 0.5276 - val_f1_metric: 0.8437 - lr: 0.0010\n",
      "Epoch 41/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1217 - f1_metric: 0.9585 - val_loss: 0.4698 - val_f1_metric: 0.8383 - lr: 0.0010\n",
      "Epoch 42/100\n",
      "76/76 [==============================] - 4s 53ms/step - loss: 0.1122 - f1_metric: 0.9738 - val_loss: 0.4676 - val_f1_metric: 0.8493 - lr: 0.0010\n",
      "Epoch 43/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0923 - f1_metric: 0.9717 - val_loss: 0.4556 - val_f1_metric: 0.8252 - lr: 0.0010\n",
      "Epoch 44/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1253 - f1_metric: 0.9605 - val_loss: 0.4305 - val_f1_metric: 0.8368 - lr: 0.0010\n",
      "Epoch 45/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.2505 - f1_metric: 0.9357 - val_loss: 0.7821 - val_f1_metric: 0.7762 - lr: 0.0010\n",
      "Epoch 46/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.2569 - f1_metric: 0.9232 - val_loss: 0.4757 - val_f1_metric: 0.8523 - lr: 0.0010\n",
      "Epoch 47/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.2095 - f1_metric: 0.9297 - val_loss: 0.4682 - val_f1_metric: 0.8474 - lr: 0.0010\n",
      "Epoch 48/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1523 - f1_metric: 0.9628 - val_loss: 0.4675 - val_f1_metric: 0.8715 - lr: 0.0010\n",
      "Epoch 49/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1497 - f1_metric: 0.9562 - val_loss: 0.5078 - val_f1_metric: 0.8681 - lr: 0.0010\n",
      "Epoch 50/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1357 - f1_metric: 0.9654 - val_loss: 0.4770 - val_f1_metric: 0.8642 - lr: 5.0000e-04\n",
      "Epoch 51/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1239 - f1_metric: 0.9661 - val_loss: 0.4771 - val_f1_metric: 0.8653 - lr: 5.0000e-04\n",
      "Epoch 52/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0656 - f1_metric: 0.9794 - val_loss: 0.4555 - val_f1_metric: 0.8487 - lr: 5.0000e-04\n",
      "Epoch 53/100\n",
      "76/76 [==============================] - 4s 53ms/step - loss: 0.1000 - f1_metric: 0.9696 - val_loss: 0.5917 - val_f1_metric: 0.8332 - lr: 5.0000e-04\n",
      "Epoch 54/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0983 - f1_metric: 0.9642 - val_loss: 0.4549 - val_f1_metric: 0.8722 - lr: 5.0000e-04\n",
      "Epoch 55/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1684 - f1_metric: 0.9503 - val_loss: 0.5050 - val_f1_metric: 0.8575 - lr: 5.0000e-04\n",
      "Epoch 56/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1024 - f1_metric: 0.9720 - val_loss: 0.4652 - val_f1_metric: 0.8612 - lr: 5.0000e-04\n",
      "Epoch 57/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1013 - f1_metric: 0.9706 - val_loss: 0.4802 - val_f1_metric: 0.8431 - lr: 5.0000e-04\n",
      "Epoch 58/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1581 - f1_metric: 0.9592 - val_loss: 0.5467 - val_f1_metric: 0.8590 - lr: 5.0000e-04\n",
      "Epoch 59/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.1105 - f1_metric: 0.9653 - val_loss: 0.5384 - val_f1_metric: 0.8163 - lr: 5.0000e-04\n",
      "Epoch 60/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0944 - f1_metric: 0.9761 - val_loss: 0.5369 - val_f1_metric: 0.8504 - lr: 5.0000e-04\n",
      "Epoch 61/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0938 - f1_metric: 0.9707 - val_loss: 0.4934 - val_f1_metric: 0.8627 - lr: 5.0000e-04\n",
      "Epoch 62/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0840 - f1_metric: 0.9738 - val_loss: 0.4697 - val_f1_metric: 0.8404 - lr: 5.0000e-04\n",
      "Epoch 63/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0630 - f1_metric: 0.9896 - val_loss: 0.5291 - val_f1_metric: 0.8283 - lr: 5.0000e-04\n",
      "Epoch 64/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0746 - f1_metric: 0.9796 - val_loss: 0.5306 - val_f1_metric: 0.8504 - lr: 5.0000e-04\n",
      "Epoch 65/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0592 - f1_metric: 0.9842 - val_loss: 0.4824 - val_f1_metric: 0.8599 - lr: 5.0000e-04\n",
      "Epoch 66/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0533 - f1_metric: 0.9849 - val_loss: 0.5223 - val_f1_metric: 0.8425 - lr: 5.0000e-04\n",
      "Epoch 67/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0960 - f1_metric: 0.9738 - val_loss: 0.4591 - val_f1_metric: 0.8469 - lr: 5.0000e-04\n",
      "Epoch 68/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0638 - f1_metric: 0.9811 - val_loss: 0.4383 - val_f1_metric: 0.8648 - lr: 5.0000e-04\n",
      "Epoch 69/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0499 - f1_metric: 0.9867 - val_loss: 0.4753 - val_f1_metric: 0.8669 - lr: 5.0000e-04\n",
      "Epoch 70/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0402 - f1_metric: 0.9887 - val_loss: 0.4013 - val_f1_metric: 0.8734 - lr: 2.5000e-04\n",
      "Epoch 71/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0474 - f1_metric: 0.9864 - val_loss: 0.3798 - val_f1_metric: 0.8763 - lr: 2.5000e-04\n",
      "Epoch 72/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0627 - f1_metric: 0.9821 - val_loss: 0.3893 - val_f1_metric: 0.8781 - lr: 2.5000e-04\n",
      "Epoch 73/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0292 - f1_metric: 0.9824 - val_loss: 0.4145 - val_f1_metric: 0.8855 - lr: 2.5000e-04\n",
      "Epoch 74/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0471 - f1_metric: 0.9882 - val_loss: 0.4155 - val_f1_metric: 0.8667 - lr: 2.5000e-04\n",
      "Epoch 75/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0504 - f1_metric: 0.9845 - val_loss: 0.4148 - val_f1_metric: 0.8745 - lr: 2.5000e-04\n",
      "Epoch 76/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0635 - f1_metric: 0.9825 - val_loss: 0.4127 - val_f1_metric: 0.8565 - lr: 2.5000e-04\n",
      "Epoch 77/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0513 - f1_metric: 0.9873 - val_loss: 0.4461 - val_f1_metric: 0.8518 - lr: 2.5000e-04\n",
      "Epoch 78/100\n",
      "76/76 [==============================] - 4s 51ms/step - loss: 0.0613 - f1_metric: 0.9884 - val_loss: 0.4284 - val_f1_metric: 0.8577 - lr: 2.5000e-04\n",
      "Epoch 79/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0490 - f1_metric: 0.9867 - val_loss: 0.4200 - val_f1_metric: 0.8883 - lr: 2.5000e-04\n",
      "Epoch 79: early stopping\n",
      "5/5 [==============================] - 0s 12ms/step\n",
      "Training on fold 4\n",
      "Epoch 1/100\n",
      "76/76 [==============================] - 6s 56ms/step - loss: 1.6194 - f1_metric: 0.6001 - val_loss: 1.1948 - val_f1_metric: 0.3774 - lr: 0.0010\n",
      "Epoch 2/100\n",
      "76/76 [==============================] - 4s 53ms/step - loss: 0.9990 - f1_metric: 0.6727 - val_loss: 1.2462 - val_f1_metric: 0.3246 - lr: 0.0010\n",
      "Epoch 3/100\n",
      "76/76 [==============================] - 4s 53ms/step - loss: 0.8676 - f1_metric: 0.7172 - val_loss: 1.1897 - val_f1_metric: 0.3977 - lr: 0.0010\n",
      "Epoch 4/100\n",
      "76/76 [==============================] - 4s 53ms/step - loss: 0.6443 - f1_metric: 0.7806 - val_loss: 1.3889 - val_f1_metric: 0.4047 - lr: 0.0010\n",
      "Epoch 5/100\n",
      "76/76 [==============================] - 4s 53ms/step - loss: 0.6377 - f1_metric: 0.7971 - val_loss: 0.8983 - val_f1_metric: 0.6255 - lr: 0.0010\n",
      "Epoch 6/100\n",
      "76/76 [==============================] - 4s 53ms/step - loss: 0.6168 - f1_metric: 0.7776 - val_loss: 1.1177 - val_f1_metric: 0.5590 - lr: 0.0010\n",
      "Epoch 7/100\n",
      "76/76 [==============================] - 4s 53ms/step - loss: 0.4378 - f1_metric: 0.8517 - val_loss: 0.7508 - val_f1_metric: 0.7066 - lr: 0.0010\n",
      "Epoch 8/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.3815 - f1_metric: 0.8814 - val_loss: 0.8504 - val_f1_metric: 0.7134 - lr: 0.0010\n",
      "Epoch 9/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.4209 - f1_metric: 0.8661 - val_loss: 0.7785 - val_f1_metric: 0.7425 - lr: 0.0010\n",
      "Epoch 10/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.3148 - f1_metric: 0.8983 - val_loss: 0.7114 - val_f1_metric: 0.7758 - lr: 0.0010\n",
      "Epoch 11/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.3429 - f1_metric: 0.8823 - val_loss: 0.7100 - val_f1_metric: 0.7563 - lr: 0.0010\n",
      "Epoch 12/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.3066 - f1_metric: 0.8984 - val_loss: 0.6920 - val_f1_metric: 0.7672 - lr: 0.0010\n",
      "Epoch 13/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.3237 - f1_metric: 0.8842 - val_loss: 0.6674 - val_f1_metric: 0.7831 - lr: 0.0010\n",
      "Epoch 14/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.2832 - f1_metric: 0.9131 - val_loss: 0.7612 - val_f1_metric: 0.8047 - lr: 0.0010\n",
      "Epoch 15/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.3037 - f1_metric: 0.8963 - val_loss: 0.6862 - val_f1_metric: 0.8044 - lr: 0.0010\n",
      "Epoch 16/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.3455 - f1_metric: 0.8964 - val_loss: 0.8095 - val_f1_metric: 0.7861 - lr: 0.0010\n",
      "Epoch 17/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.2587 - f1_metric: 0.9221 - val_loss: 0.8349 - val_f1_metric: 0.7996 - lr: 0.0010\n",
      "Epoch 18/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.2611 - f1_metric: 0.9166 - val_loss: 0.9091 - val_f1_metric: 0.7250 - lr: 0.0010\n",
      "Epoch 19/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.2767 - f1_metric: 0.9172 - val_loss: 0.7536 - val_f1_metric: 0.7862 - lr: 0.0010\n",
      "Epoch 20/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.2244 - f1_metric: 0.9375 - val_loss: 0.7880 - val_f1_metric: 0.7622 - lr: 0.0010\n",
      "Epoch 21/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.2189 - f1_metric: 0.9329 - val_loss: 0.6942 - val_f1_metric: 0.8000 - lr: 0.0010\n",
      "Epoch 22/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1930 - f1_metric: 0.9373 - val_loss: 0.8036 - val_f1_metric: 0.8129 - lr: 0.0010\n",
      "Epoch 23/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1704 - f1_metric: 0.9478 - val_loss: 0.7408 - val_f1_metric: 0.7860 - lr: 0.0010\n",
      "Epoch 24/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.2236 - f1_metric: 0.9383 - val_loss: 0.7706 - val_f1_metric: 0.8048 - lr: 0.0010\n",
      "Epoch 25/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1717 - f1_metric: 0.9441 - val_loss: 0.7430 - val_f1_metric: 0.8324 - lr: 0.0010\n",
      "Epoch 26/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1499 - f1_metric: 0.9584 - val_loss: 0.8170 - val_f1_metric: 0.7980 - lr: 0.0010\n",
      "Epoch 27/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.2362 - f1_metric: 0.9319 - val_loss: 0.7860 - val_f1_metric: 0.8116 - lr: 0.0010\n",
      "Epoch 28/100\n",
      "76/76 [==============================] - 4s 53ms/step - loss: 0.3306 - f1_metric: 0.9013 - val_loss: 0.9229 - val_f1_metric: 0.7563 - lr: 0.0010\n",
      "Epoch 29/100\n",
      "76/76 [==============================] - 4s 53ms/step - loss: 0.3362 - f1_metric: 0.8892 - val_loss: 0.9470 - val_f1_metric: 0.7744 - lr: 0.0010\n",
      "Epoch 30/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.3061 - f1_metric: 0.9017 - val_loss: 0.7392 - val_f1_metric: 0.7819 - lr: 0.0010\n",
      "Epoch 31/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.2081 - f1_metric: 0.9374 - val_loss: 0.7059 - val_f1_metric: 0.7927 - lr: 0.0010\n",
      "Epoch 32/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1914 - f1_metric: 0.9465 - val_loss: 0.7005 - val_f1_metric: 0.8420 - lr: 0.0010\n",
      "Epoch 33/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1754 - f1_metric: 0.9418 - val_loss: 0.6906 - val_f1_metric: 0.8337 - lr: 0.0010\n",
      "Epoch 34/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1685 - f1_metric: 0.9454 - val_loss: 0.7068 - val_f1_metric: 0.8086 - lr: 5.0000e-04\n",
      "Epoch 35/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1183 - f1_metric: 0.9544 - val_loss: 0.6989 - val_f1_metric: 0.8412 - lr: 5.0000e-04\n",
      "Epoch 36/100\n",
      "76/76 [==============================] - 4s 53ms/step - loss: 0.1180 - f1_metric: 0.9712 - val_loss: 0.7049 - val_f1_metric: 0.8112 - lr: 5.0000e-04\n",
      "Epoch 37/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0949 - f1_metric: 0.9653 - val_loss: 0.6834 - val_f1_metric: 0.8283 - lr: 5.0000e-04\n",
      "Epoch 38/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1308 - f1_metric: 0.9672 - val_loss: 0.6598 - val_f1_metric: 0.8250 - lr: 5.0000e-04\n",
      "Epoch 39/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0985 - f1_metric: 0.9739 - val_loss: 0.6343 - val_f1_metric: 0.8202 - lr: 5.0000e-04\n",
      "Epoch 40/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0920 - f1_metric: 0.9760 - val_loss: 0.6434 - val_f1_metric: 0.8429 - lr: 5.0000e-04\n",
      "Epoch 41/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1209 - f1_metric: 0.9690 - val_loss: 0.6138 - val_f1_metric: 0.8387 - lr: 5.0000e-04\n",
      "Epoch 42/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0728 - f1_metric: 0.9758 - val_loss: 0.6337 - val_f1_metric: 0.8403 - lr: 5.0000e-04\n",
      "Epoch 43/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0924 - f1_metric: 0.9751 - val_loss: 0.6322 - val_f1_metric: 0.8431 - lr: 5.0000e-04\n",
      "Epoch 44/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0800 - f1_metric: 0.9785 - val_loss: 0.6755 - val_f1_metric: 0.8168 - lr: 5.0000e-04\n",
      "Epoch 45/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0784 - f1_metric: 0.9778 - val_loss: 0.6244 - val_f1_metric: 0.8310 - lr: 5.0000e-04\n",
      "Epoch 46/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0665 - f1_metric: 0.9784 - val_loss: 0.7126 - val_f1_metric: 0.8021 - lr: 5.0000e-04\n",
      "Epoch 47/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0898 - f1_metric: 0.9766 - val_loss: 0.6938 - val_f1_metric: 0.8245 - lr: 5.0000e-04\n",
      "Epoch 48/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0670 - f1_metric: 0.9809 - val_loss: 0.7305 - val_f1_metric: 0.8252 - lr: 5.0000e-04\n",
      "Epoch 49/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0687 - f1_metric: 0.9831 - val_loss: 0.6983 - val_f1_metric: 0.8302 - lr: 5.0000e-04\n",
      "Epoch 50/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0600 - f1_metric: 0.9820 - val_loss: 0.6281 - val_f1_metric: 0.8262 - lr: 5.0000e-04\n",
      "Epoch 51/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0573 - f1_metric: 0.9847 - val_loss: 0.6045 - val_f1_metric: 0.8208 - lr: 5.0000e-04\n",
      "Epoch 52/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0847 - f1_metric: 0.9670 - val_loss: 0.7106 - val_f1_metric: 0.7950 - lr: 5.0000e-04\n",
      "Epoch 53/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0590 - f1_metric: 0.9823 - val_loss: 0.6647 - val_f1_metric: 0.8330 - lr: 5.0000e-04\n",
      "Epoch 54/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0604 - f1_metric: 0.9841 - val_loss: 0.6418 - val_f1_metric: 0.8219 - lr: 5.0000e-04\n",
      "Epoch 55/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0712 - f1_metric: 0.9786 - val_loss: 0.6776 - val_f1_metric: 0.8152 - lr: 5.0000e-04\n",
      "Epoch 56/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0686 - f1_metric: 0.9839 - val_loss: 0.7623 - val_f1_metric: 0.8344 - lr: 5.0000e-04\n",
      "Epoch 57/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0496 - f1_metric: 0.9875 - val_loss: 0.7030 - val_f1_metric: 0.8178 - lr: 5.0000e-04\n",
      "Epoch 58/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0792 - f1_metric: 0.9800 - val_loss: 0.6682 - val_f1_metric: 0.8520 - lr: 5.0000e-04\n",
      "Epoch 59/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0731 - f1_metric: 0.9849 - val_loss: 0.6082 - val_f1_metric: 0.8286 - lr: 5.0000e-04\n",
      "Epoch 60/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0907 - f1_metric: 0.9730 - val_loss: 0.7295 - val_f1_metric: 0.8097 - lr: 5.0000e-04\n",
      "Epoch 61/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1060 - f1_metric: 0.9685 - val_loss: 0.6519 - val_f1_metric: 0.8216 - lr: 5.0000e-04\n",
      "Epoch 62/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0764 - f1_metric: 0.9823 - val_loss: 0.6991 - val_f1_metric: 0.8316 - lr: 5.0000e-04\n",
      "Epoch 63/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0615 - f1_metric: 0.9822 - val_loss: 0.6864 - val_f1_metric: 0.8241 - lr: 5.0000e-04\n",
      "Epoch 64/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1217 - f1_metric: 0.9641 - val_loss: 0.6775 - val_f1_metric: 0.8346 - lr: 5.0000e-04\n",
      "Epoch 65/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1372 - f1_metric: 0.9647 - val_loss: 0.6820 - val_f1_metric: 0.8011 - lr: 5.0000e-04\n",
      "Epoch 66/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0975 - f1_metric: 0.9682 - val_loss: 0.6885 - val_f1_metric: 0.8165 - lr: 5.0000e-04\n",
      "Epoch 67/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1243 - f1_metric: 0.9661 - val_loss: 0.7136 - val_f1_metric: 0.8462 - lr: 5.0000e-04\n",
      "Epoch 68/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0989 - f1_metric: 0.9734 - val_loss: 0.6512 - val_f1_metric: 0.7992 - lr: 5.0000e-04\n",
      "Epoch 69/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0483 - f1_metric: 0.9859 - val_loss: 0.5674 - val_f1_metric: 0.8352 - lr: 5.0000e-04\n",
      "Epoch 70/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0536 - f1_metric: 0.9829 - val_loss: 0.5661 - val_f1_metric: 0.8254 - lr: 5.0000e-04\n",
      "Epoch 71/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0621 - f1_metric: 0.9824 - val_loss: 0.6058 - val_f1_metric: 0.8440 - lr: 5.0000e-04\n",
      "Epoch 72/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1100 - f1_metric: 0.9700 - val_loss: 0.5920 - val_f1_metric: 0.8796 - lr: 5.0000e-04\n",
      "Epoch 73/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0738 - f1_metric: 0.9752 - val_loss: 0.7196 - val_f1_metric: 0.8170 - lr: 5.0000e-04\n",
      "Epoch 74/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1298 - f1_metric: 0.9655 - val_loss: 0.6811 - val_f1_metric: 0.8185 - lr: 5.0000e-04\n",
      "Epoch 75/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1165 - f1_metric: 0.9696 - val_loss: 0.7194 - val_f1_metric: 0.8094 - lr: 5.0000e-04\n",
      "Epoch 76/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1042 - f1_metric: 0.9743 - val_loss: 0.6692 - val_f1_metric: 0.8361 - lr: 5.0000e-04\n",
      "Epoch 77/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0811 - f1_metric: 0.9857 - val_loss: 0.6716 - val_f1_metric: 0.8489 - lr: 5.0000e-04\n",
      "Epoch 78/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0554 - f1_metric: 0.9863 - val_loss: 0.6433 - val_f1_metric: 0.8608 - lr: 5.0000e-04\n",
      "Epoch 79/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0715 - f1_metric: 0.9804 - val_loss: 0.6694 - val_f1_metric: 0.8380 - lr: 5.0000e-04\n",
      "Epoch 80/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0745 - f1_metric: 0.9790 - val_loss: 0.6849 - val_f1_metric: 0.8409 - lr: 5.0000e-04\n",
      "Epoch 81/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0445 - f1_metric: 0.9926 - val_loss: 0.6547 - val_f1_metric: 0.8430 - lr: 5.0000e-04\n",
      "Epoch 82/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0456 - f1_metric: 0.9863 - val_loss: 0.6837 - val_f1_metric: 0.8344 - lr: 5.0000e-04\n",
      "Epoch 83/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0572 - f1_metric: 0.9859 - val_loss: 0.6705 - val_f1_metric: 0.8535 - lr: 5.0000e-04\n",
      "Epoch 84/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0401 - f1_metric: 0.9827 - val_loss: 0.7281 - val_f1_metric: 0.8145 - lr: 5.0000e-04\n",
      "Epoch 85/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0653 - f1_metric: 0.9778 - val_loss: 0.7451 - val_f1_metric: 0.8304 - lr: 5.0000e-04\n",
      "Epoch 86/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0715 - f1_metric: 0.9805 - val_loss: 0.7999 - val_f1_metric: 0.7609 - lr: 5.0000e-04\n",
      "Epoch 87/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0805 - f1_metric: 0.9820 - val_loss: 0.7298 - val_f1_metric: 0.8169 - lr: 5.0000e-04\n",
      "Epoch 88/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0619 - f1_metric: 0.9876 - val_loss: 0.7655 - val_f1_metric: 0.7914 - lr: 5.0000e-04\n",
      "Epoch 89/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0663 - f1_metric: 0.9817 - val_loss: 0.7634 - val_f1_metric: 0.7907 - lr: 5.0000e-04\n",
      "Epoch 90/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0680 - f1_metric: 0.9790 - val_loss: 0.8434 - val_f1_metric: 0.8074 - lr: 5.0000e-04\n",
      "Epoch 91/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0519 - f1_metric: 0.9854 - val_loss: 0.7843 - val_f1_metric: 0.7973 - lr: 2.5000e-04\n",
      "Epoch 92/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0684 - f1_metric: 0.9868 - val_loss: 0.7412 - val_f1_metric: 0.8091 - lr: 2.5000e-04\n",
      "Epoch 93/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0652 - f1_metric: 0.9881 - val_loss: 0.7658 - val_f1_metric: 0.7902 - lr: 2.5000e-04\n",
      "Epoch 94/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0748 - f1_metric: 0.9765 - val_loss: 0.7549 - val_f1_metric: 0.7961 - lr: 2.5000e-04\n",
      "Epoch 95/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0468 - f1_metric: 0.9882 - val_loss: 0.7404 - val_f1_metric: 0.7747 - lr: 2.5000e-04\n",
      "Epoch 96/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0488 - f1_metric: 0.9851 - val_loss: 0.7176 - val_f1_metric: 0.8038 - lr: 2.5000e-04\n",
      "Epoch 97/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0343 - f1_metric: 0.9906 - val_loss: 0.7048 - val_f1_metric: 0.8078 - lr: 2.5000e-04\n",
      "Epoch 98/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0400 - f1_metric: 0.9884 - val_loss: 0.7094 - val_f1_metric: 0.8028 - lr: 2.5000e-04\n",
      "Epoch 99/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0262 - f1_metric: 0.9976 - val_loss: 0.6619 - val_f1_metric: 0.7978 - lr: 2.5000e-04\n",
      "Epoch 100/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0294 - f1_metric: 0.9925 - val_loss: 0.6457 - val_f1_metric: 0.8168 - lr: 2.5000e-04\n",
      "Training on fold 5\n",
      "Epoch 1/100\n",
      "76/76 [==============================] - 6s 56ms/step - loss: 1.5720 - f1_metric: 0.5850 - val_loss: 1.2419 - val_f1_metric: 0.3796 - lr: 0.0010\n",
      "Epoch 2/100\n",
      "76/76 [==============================] - 4s 53ms/step - loss: 0.9020 - f1_metric: 0.6898 - val_loss: 1.7330 - val_f1_metric: 0.2368 - lr: 0.0010\n",
      "Epoch 3/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.7640 - f1_metric: 0.7434 - val_loss: 1.7445 - val_f1_metric: 0.3513 - lr: 0.0010\n",
      "Epoch 4/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.6103 - f1_metric: 0.7953 - val_loss: 1.4011 - val_f1_metric: 0.4994 - lr: 0.0010\n",
      "Epoch 5/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.5808 - f1_metric: 0.8052 - val_loss: 1.1481 - val_f1_metric: 0.4966 - lr: 0.0010\n",
      "Epoch 6/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.5328 - f1_metric: 0.8340 - val_loss: 1.0433 - val_f1_metric: 0.6101 - lr: 0.0010\n",
      "Epoch 7/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.4433 - f1_metric: 0.8573 - val_loss: 0.6823 - val_f1_metric: 0.7348 - lr: 0.0010\n",
      "Epoch 8/100\n",
      "76/76 [==============================] - 4s 53ms/step - loss: 0.4237 - f1_metric: 0.8634 - val_loss: 0.6595 - val_f1_metric: 0.7661 - lr: 0.0010\n",
      "Epoch 9/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.4032 - f1_metric: 0.8955 - val_loss: 0.6996 - val_f1_metric: 0.7522 - lr: 0.0010\n",
      "Epoch 10/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.3302 - f1_metric: 0.8944 - val_loss: 0.6966 - val_f1_metric: 0.7375 - lr: 0.0010\n",
      "Epoch 11/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.3034 - f1_metric: 0.8989 - val_loss: 0.8944 - val_f1_metric: 0.7388 - lr: 0.0010\n",
      "Epoch 12/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.3245 - f1_metric: 0.8956 - val_loss: 0.6770 - val_f1_metric: 0.7775 - lr: 0.0010\n",
      "Epoch 13/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.2902 - f1_metric: 0.8964 - val_loss: 0.6579 - val_f1_metric: 0.8095 - lr: 0.0010\n",
      "Epoch 14/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.2935 - f1_metric: 0.9065 - val_loss: 0.7274 - val_f1_metric: 0.7795 - lr: 0.0010\n",
      "Epoch 15/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.2814 - f1_metric: 0.9158 - val_loss: 0.7425 - val_f1_metric: 0.7741 - lr: 0.0010\n",
      "Epoch 16/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.2798 - f1_metric: 0.9086 - val_loss: 0.9130 - val_f1_metric: 0.7198 - lr: 0.0010\n",
      "Epoch 17/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.3080 - f1_metric: 0.9102 - val_loss: 0.8117 - val_f1_metric: 0.7637 - lr: 0.0010\n",
      "Epoch 18/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.2584 - f1_metric: 0.9251 - val_loss: 0.7695 - val_f1_metric: 0.7685 - lr: 0.0010\n",
      "Epoch 19/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.2694 - f1_metric: 0.9040 - val_loss: 0.8737 - val_f1_metric: 0.7429 - lr: 0.0010\n",
      "Epoch 20/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.4152 - f1_metric: 0.8602 - val_loss: 0.8800 - val_f1_metric: 0.7951 - lr: 0.0010\n",
      "Epoch 21/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.2147 - f1_metric: 0.9369 - val_loss: 0.6601 - val_f1_metric: 0.7586 - lr: 0.0010\n",
      "Epoch 22/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.2501 - f1_metric: 0.9172 - val_loss: 0.5698 - val_f1_metric: 0.8149 - lr: 0.0010\n",
      "Epoch 23/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1924 - f1_metric: 0.9384 - val_loss: 0.6289 - val_f1_metric: 0.8097 - lr: 0.0010\n",
      "Epoch 24/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.2192 - f1_metric: 0.9479 - val_loss: 0.8543 - val_f1_metric: 0.7804 - lr: 0.0010\n",
      "Epoch 25/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1835 - f1_metric: 0.9468 - val_loss: 0.7389 - val_f1_metric: 0.7728 - lr: 0.0010\n",
      "Epoch 26/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.2208 - f1_metric: 0.9377 - val_loss: 0.6185 - val_f1_metric: 0.8123 - lr: 0.0010\n",
      "Epoch 27/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1564 - f1_metric: 0.9518 - val_loss: 0.6315 - val_f1_metric: 0.7970 - lr: 0.0010\n",
      "Epoch 28/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1926 - f1_metric: 0.9385 - val_loss: 0.5937 - val_f1_metric: 0.7907 - lr: 0.0010\n",
      "Epoch 29/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1858 - f1_metric: 0.9361 - val_loss: 0.7888 - val_f1_metric: 0.7788 - lr: 0.0010\n",
      "Epoch 30/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1996 - f1_metric: 0.9390 - val_loss: 0.7810 - val_f1_metric: 0.7641 - lr: 0.0010\n",
      "Epoch 31/100\n",
      "76/76 [==============================] - 4s 54ms/step - loss: 0.1406 - f1_metric: 0.9588 - val_loss: 0.6879 - val_f1_metric: 0.7623 - lr: 0.0010\n",
      "Epoch 32/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1108 - f1_metric: 0.9640 - val_loss: 0.7829 - val_f1_metric: 0.7862 - lr: 0.0010\n",
      "Epoch 33/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1293 - f1_metric: 0.9669 - val_loss: 0.6729 - val_f1_metric: 0.7938 - lr: 0.0010\n",
      "Epoch 34/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1574 - f1_metric: 0.9540 - val_loss: 0.8231 - val_f1_metric: 0.7779 - lr: 0.0010\n",
      "Epoch 35/100\n",
      "76/76 [==============================] - 4s 53ms/step - loss: 0.1542 - f1_metric: 0.9592 - val_loss: 0.6867 - val_f1_metric: 0.7912 - lr: 0.0010\n",
      "Epoch 36/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1164 - f1_metric: 0.9634 - val_loss: 0.7177 - val_f1_metric: 0.7941 - lr: 0.0010\n",
      "Epoch 37/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1224 - f1_metric: 0.9579 - val_loss: 0.7184 - val_f1_metric: 0.7660 - lr: 0.0010\n",
      "Epoch 38/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1035 - f1_metric: 0.9675 - val_loss: 0.6847 - val_f1_metric: 0.8138 - lr: 0.0010\n",
      "Epoch 39/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1461 - f1_metric: 0.9590 - val_loss: 0.7513 - val_f1_metric: 0.7597 - lr: 0.0010\n",
      "Epoch 40/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1947 - f1_metric: 0.9428 - val_loss: 0.7414 - val_f1_metric: 0.7664 - lr: 0.0010\n",
      "Epoch 41/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1593 - f1_metric: 0.9494 - val_loss: 0.7109 - val_f1_metric: 0.8080 - lr: 0.0010\n",
      "Epoch 42/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1018 - f1_metric: 0.9724 - val_loss: 0.9486 - val_f1_metric: 0.7600 - lr: 0.0010\n",
      "Epoch 43/100\n",
      "76/76 [==============================] - 4s 53ms/step - loss: 0.1281 - f1_metric: 0.9642 - val_loss: 0.7443 - val_f1_metric: 0.7911 - lr: 5.0000e-04\n",
      "Epoch 44/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0848 - f1_metric: 0.9723 - val_loss: 0.7788 - val_f1_metric: 0.7671 - lr: 5.0000e-04\n",
      "Epoch 45/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0654 - f1_metric: 0.9804 - val_loss: 0.7227 - val_f1_metric: 0.8085 - lr: 5.0000e-04\n",
      "Epoch 46/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1086 - f1_metric: 0.9704 - val_loss: 0.8066 - val_f1_metric: 0.7964 - lr: 5.0000e-04\n",
      "Epoch 47/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0810 - f1_metric: 0.9797 - val_loss: 0.7556 - val_f1_metric: 0.8083 - lr: 5.0000e-04\n",
      "Epoch 48/100\n",
      "76/76 [==============================] - 4s 53ms/step - loss: 0.1180 - f1_metric: 0.9730 - val_loss: 0.8277 - val_f1_metric: 0.7826 - lr: 5.0000e-04\n",
      "Epoch 49/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1189 - f1_metric: 0.9725 - val_loss: 0.7804 - val_f1_metric: 0.7947 - lr: 5.0000e-04\n",
      "Epoch 50/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0978 - f1_metric: 0.9763 - val_loss: 0.7374 - val_f1_metric: 0.8120 - lr: 5.0000e-04\n",
      "Epoch 51/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0852 - f1_metric: 0.9783 - val_loss: 0.7516 - val_f1_metric: 0.8031 - lr: 5.0000e-04\n",
      "Epoch 52/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1007 - f1_metric: 0.9832 - val_loss: 0.7481 - val_f1_metric: 0.7945 - lr: 5.0000e-04\n",
      "Epoch 53/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0782 - f1_metric: 0.9842 - val_loss: 0.8042 - val_f1_metric: 0.7669 - lr: 5.0000e-04\n",
      "Epoch 54/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0633 - f1_metric: 0.9874 - val_loss: 0.7795 - val_f1_metric: 0.7923 - lr: 5.0000e-04\n",
      "Epoch 55/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0834 - f1_metric: 0.9800 - val_loss: 0.7027 - val_f1_metric: 0.8004 - lr: 5.0000e-04\n",
      "Epoch 56/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0501 - f1_metric: 0.9891 - val_loss: 0.7350 - val_f1_metric: 0.7982 - lr: 5.0000e-04\n",
      "Epoch 57/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0671 - f1_metric: 0.9809 - val_loss: 0.6930 - val_f1_metric: 0.8131 - lr: 5.0000e-04\n",
      "Epoch 58/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0288 - f1_metric: 0.9946 - val_loss: 0.7431 - val_f1_metric: 0.8251 - lr: 5.0000e-04\n",
      "Epoch 59/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0329 - f1_metric: 0.9947 - val_loss: 0.7646 - val_f1_metric: 0.8037 - lr: 5.0000e-04\n",
      "Epoch 60/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0499 - f1_metric: 0.9842 - val_loss: 0.7554 - val_f1_metric: 0.8059 - lr: 5.0000e-04\n",
      "Epoch 61/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0384 - f1_metric: 0.9875 - val_loss: 0.7975 - val_f1_metric: 0.8333 - lr: 5.0000e-04\n",
      "Epoch 62/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.1014 - f1_metric: 0.9744 - val_loss: 0.7100 - val_f1_metric: 0.7489 - lr: 5.0000e-04\n",
      "Epoch 63/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0269 - f1_metric: 0.9982 - val_loss: 0.7384 - val_f1_metric: 0.7664 - lr: 2.5000e-04\n",
      "Epoch 64/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0528 - f1_metric: 0.9899 - val_loss: 0.7305 - val_f1_metric: 0.7803 - lr: 2.5000e-04\n",
      "Epoch 65/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0690 - f1_metric: 0.9865 - val_loss: 0.7919 - val_f1_metric: 0.7724 - lr: 2.5000e-04\n",
      "Epoch 66/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0610 - f1_metric: 0.9852 - val_loss: 0.7972 - val_f1_metric: 0.8012 - lr: 2.5000e-04\n",
      "Epoch 67/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0649 - f1_metric: 0.9817 - val_loss: 0.8474 - val_f1_metric: 0.7862 - lr: 2.5000e-04\n",
      "Epoch 68/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0548 - f1_metric: 0.9788 - val_loss: 0.9030 - val_f1_metric: 0.7450 - lr: 2.5000e-04\n",
      "Epoch 69/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0419 - f1_metric: 0.9936 - val_loss: 0.8325 - val_f1_metric: 0.7914 - lr: 2.5000e-04\n",
      "Epoch 70/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0262 - f1_metric: 0.9948 - val_loss: 0.8452 - val_f1_metric: 0.7986 - lr: 2.5000e-04\n",
      "Epoch 71/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0368 - f1_metric: 0.9901 - val_loss: 0.8913 - val_f1_metric: 0.7933 - lr: 2.5000e-04\n",
      "Epoch 72/100\n",
      "76/76 [==============================] - 4s 52ms/step - loss: 0.0395 - f1_metric: 0.9882 - val_loss: 0.8766 - val_f1_metric: 0.7887 - lr: 2.5000e-04\n",
      "Epoch 72: early stopping\n"
     ]
    }
   ],
   "source": [
    "# Define F1 score as a custom metric\n",
    "def f1_metric(y_true, y_pred):\n",
    "    y_pred_labels = tf.argmax(y_pred, axis=1)\n",
    "    y_true_labels = tf.cast(y_true, dtype=tf.int64)\n",
    "    return tf.py_function(lambda y_true, y_pred: f1_score(y_true, y_pred, average='weighted'), \n",
    "                        (y_true_labels, y_pred_labels), \n",
    "                        tf.double)\n",
    "\n",
    "# Encode class labels if necessary\n",
    "epochs = 100\n",
    "batch_size = 8\n",
    "all_histories = []\n",
    "best_model = None\n",
    "best_f1_score = 0.0\n",
    "best_loss = float('inf')\n",
    "best_fold = -1\n",
    "\n",
    "# Iterate over the folds\n",
    "for fold, (train_index, val_index) in enumerate(skf.split(X, y)):\n",
    "    print(f\"Training on fold {fold + 1}\")\n",
    "\n",
    "    model = create_model()\n",
    "\n",
    "    # Compile the model\n",
    "    model.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics=[f1_metric])\n",
    "\n",
    "    callbacks = [\n",
    "        keras.callbacks.ReduceLROnPlateau(\n",
    "            monitor=\"val_loss\", factor=0.5, patience=20, min_lr=0.0001\n",
    "        ),\n",
    "        keras.callbacks.EarlyStopping(monitor=\"val_loss\", patience=50, verbose=1),\n",
    "    ]\n",
    "\n",
    "    # Split the data into training and validation sets\n",
    "    X_fold, X_val_fold = X[train_index], X[val_index]\n",
    "    y_fold, y_val_fold = y[train_index], y[val_index]\n",
    "\n",
    "    history = model.fit(\n",
    "        X_fold,\n",
    "        y_fold,\n",
    "        batch_size=batch_size,\n",
    "        epochs=epochs,\n",
    "        callbacks=callbacks,\n",
    "        validation_data=(X_val_fold, y_val_fold),\n",
    "        verbose=1,\n",
    "    )\n",
    "    \n",
    "    # Track the best model based on validation f1_score\n",
    "    val_f1_score = history.history['val_f1_metric'][-1]\n",
    "    val_loss = history.history['val_loss'][-1]\n",
    "\n",
    "    if val_f1_score > best_f1_score:\n",
    "        best_f1_score = val_f1_score\n",
    "        best_loss = val_loss\n",
    "        best_fold = fold + 1\n",
    "        best_model = model\n",
    "    \n",
    "        y_val_encoded = label_encoder.fit_transform(y_val_fold)\n",
    "        # Transform numeric labels to text labels\n",
    "        y_pred = model.predict(X_val_fold)\n",
    "        y_pred = np.argmax(y_pred, axis=1)  # Convert probabilities to class labels\n",
    "        best_conf_mat = confusion_matrix(y_val_encoded, y_pred)\n",
    "    \n",
    "    # Store the training history for later analysis if needed\n",
    "    all_histories.append(history)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Salvar pesos e métricas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best f1_score in fold 3: 0.8883315920829773\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x1200 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "os.makedirs(result_folder, exist_ok=True)\n",
    "best_model.save(f'{result_folder}best_model_fold{best_fold}.keras')\n",
    "\n",
    "print(f\"Best f1_score in fold {best_fold}: {best_f1_score}\")\n",
    "final_values = []\n",
    "\n",
    "# Plot training and validation f1_score and loss\n",
    "plt.figure(figsize=(12, 12))\n",
    "for i, history in enumerate(all_histories):\n",
    "    \n",
    "    fold_data = {\n",
    "        'Fold': i + 1,\n",
    "        'Final Train Loss': history.history['loss'][-1],\n",
    "        'Final Validation Loss': history.history['val_loss'][-1],\n",
    "        'Final Train F1 Score': history.history['f1_metric'][-1],\n",
    "        'Final Validation F1 Score': history.history['val_f1_metric'][-1],\n",
    "    }\n",
    "    final_values.append(fold_data)\n",
    "    \n",
    "    plt.subplot(2, 2, 1)\n",
    "    plt.plot(history.history['f1_metric'], label=f'Fold {i+1}')\n",
    "    plt.title('Training F1 Score')\n",
    "    plt.xlabel('Epoch')\n",
    "    plt.ylabel('F1_score')\n",
    "    plt.legend()\n",
    "\n",
    "    plt.subplot(2, 2, 2)\n",
    "    plt.plot(history.history['val_f1_metric'], label=f'Fold {i+1}')\n",
    "    plt.title('Validation F1 Score')\n",
    "    plt.xlabel('Epoch')\n",
    "    plt.ylabel('F1_score')\n",
    "    plt.legend()\n",
    "    plt.subplot(2, 2, 3)\n",
    "    plt.plot(history.history['loss'], label=f'Fold {i+1}')\n",
    "    plt.title('Training Loss')\n",
    "    plt.xlabel('Epoch')\n",
    "    plt.ylabel('Loss')\n",
    "    plt.legend()\n",
    "\n",
    "    plt.subplot(2, 2, 4)\n",
    "    plt.plot(history.history['val_loss'], label=f'Fold {i+1}')\n",
    "    plt.title('Validation Loss')\n",
    "    plt.xlabel('Epoch')\n",
    "    plt.ylabel('Loss')\n",
    "    plt.legend()\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.savefig(os.path.join(result_folder, f'f1_and_losses_plot.png'))\n",
    "plt.show()\n",
    "\n",
    "# Plot confusion matrix\n",
    "plt.figure(figsize=(8, 6))\n",
    "sns.heatmap(best_conf_mat, annot=True, fmt=\"d\", cmap=\"Blues\", cbar=False, xticklabels=text_labels, yticklabels=text_labels)\n",
    "plt.title('Confusion Matrix')\n",
    "plt.xlabel('Predicted')\n",
    "plt.ylabel('True')\n",
    "plt.savefig(os.path.join(result_folder, f'confusion_matrix.png'))\n",
    "plt.show()\n",
    "\n",
    "# Save to CSV file\n",
    "csv_file_path = os.path.join(result_folder, 'final_values.csv')\n",
    "fieldnames = ['Fold', 'Final Train Loss', 'Final Validation Loss', 'Final Train F1 Score', 'Final Validation F1 Score']\n",
    "\n",
    "with open(csv_file_path, 'w', newline='') as csv_file:\n",
    "    writer = csv.DictWriter(csv_file, fieldnames=fieldnames)\n",
    "    \n",
    "    # Write header\n",
    "    writer.writeheader()\n",
    "    \n",
    "    # Write data for each fold\n",
    "    for fold_data in final_values:\n",
    "        writer.writerow(fold_data)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": ".venv",
   "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": "3.9.16"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}