{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "_uuid": "0f935bddcc5ca02702578e712bf7f396ba7154ef",
    "execution": {
     "iopub.execute_input": "2022-04-01T19:45:49.419624Z",
     "iopub.status.busy": "2022-04-01T19:45:49.41933Z",
     "iopub.status.idle": "2022-04-01T19:45:51.246319Z",
     "shell.execute_reply": "2022-04-01T19:45:51.245615Z",
     "shell.execute_reply.started": "2022-04-01T19:45:49.419574Z"
    }
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import confusion_matrix\n",
    "\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Conv2D, Lambda, MaxPooling2D # convolution layers\n",
    "from keras.layers import Dense, Dropout, Flatten # core layers\n",
    "\n",
    "from keras.layers.normalization import BatchNormalization\n",
    "\n",
    "from keras.preprocessing.image import ImageDataGenerator\n",
    "\n",
    "from keras.utils.np_utils import to_categorical\n",
    "\n",
    "#from keras.datasets import mnist\n",
    "import numpy as np # linear algebra\n",
    "import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "# import warnings\n",
    "import warnings\n",
    "# filter warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "# Input data files are available in the \"../input/\" directory.\n",
    "# For example, running this (by clicking run or pressing Shift+Enter) will list the files in the input directory\n",
    "\n",
    "import os\n",
    "print(os.listdir(\"../input\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-01T19:45:51.24967Z",
     "iopub.status.busy": "2022-04-01T19:45:51.249411Z",
     "iopub.status.idle": "2022-04-01T19:45:51.254148Z",
     "shell.execute_reply": "2022-04-01T19:45:51.25338Z",
     "shell.execute_reply.started": "2022-04-01T19:45:51.249623Z"
    }
   },
   "outputs": [],
   "source": [
    "import os\n",
    "print(os.listdir('../input'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-01T19:45:51.256377Z",
     "iopub.status.busy": "2022-04-01T19:45:51.255712Z",
     "iopub.status.idle": "2022-04-01T19:45:51.44123Z",
     "shell.execute_reply": "2022-04-01T19:45:51.440366Z",
     "shell.execute_reply.started": "2022-04-01T19:45:51.256253Z"
    }
   },
   "outputs": [],
   "source": [
    "data = pd.read_csv(\"../input/500lukveri/data.csv\") \n",
    "print(data.shape)\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-01T19:45:51.442591Z",
     "iopub.status.busy": "2022-04-01T19:45:51.442308Z",
     "iopub.status.idle": "2022-04-01T19:45:51.450013Z",
     "shell.execute_reply": "2022-04-01T19:45:51.449155Z",
     "shell.execute_reply.started": "2022-04-01T19:45:51.442546Z"
    }
   },
   "outputs": [],
   "source": [
    "X=data.iloc[0:, 1:-1]\n",
    "y=data.iloc[:,-1]\n",
    "print(X.shape)\n",
    "print(y.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-01T19:45:51.504748Z",
     "iopub.status.busy": "2022-04-01T19:45:51.504248Z",
     "iopub.status.idle": "2022-04-01T19:45:51.511105Z",
     "shell.execute_reply": "2022-04-01T19:45:51.510065Z",
     "shell.execute_reply.started": "2022-04-01T19:45:51.504573Z"
    }
   },
   "outputs": [],
   "source": [
    "X = X.values.reshape(500,467,1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-01T19:45:51.521344Z",
     "iopub.status.busy": "2022-04-01T19:45:51.520787Z",
     "iopub.status.idle": "2022-04-01T19:45:51.530732Z",
     "shell.execute_reply": "2022-04-01T19:45:51.53006Z",
     "shell.execute_reply.started": "2022-04-01T19:45:51.521294Z"
    }
   },
   "outputs": [],
   "source": [
    "mean = np.mean(X)\n",
    "std = np.std(X)\n",
    "\n",
    "def standardize(x):\n",
    "    return (x-mean)/std"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-01T19:45:51.533871Z",
     "iopub.status.busy": "2022-04-01T19:45:51.533628Z",
     "iopub.status.idle": "2022-04-01T19:45:51.543723Z",
     "shell.execute_reply": "2022-04-01T19:45:51.542907Z",
     "shell.execute_reply.started": "2022-04-01T19:45:51.533829Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "X_train, X_val, Y_train, Y_val = train_test_split(X, y, test_size = 0.15, random_state=2)\n",
    "print(\"x_train shape\",X_train.shape)\n",
    "print(\"x_test shape\",X_val.shape)\n",
    "print(\"y_train shape\",Y_train.shape)\n",
    "print(\"y_test shape\",Y_val.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-01T19:45:51.553792Z",
     "iopub.status.busy": "2022-04-01T19:45:51.553179Z",
     "iopub.status.idle": "2022-04-01T19:45:51.561178Z",
     "shell.execute_reply": "2022-04-01T19:45:51.560327Z",
     "shell.execute_reply.started": "2022-04-01T19:45:51.553741Z"
    }
   },
   "outputs": [],
   "source": [
    "from keras.utils.np_utils import to_categorical # convert to one-hot-encoding\n",
    "Y_trainn = to_categorical(Y_train, num_classes = 2)\n",
    "Y_vall = to_categorical(Y_val, num_classes = 2)\n",
    "print(Y_trainn.shape)\n",
    "print(Y_vall.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-01T19:45:51.572996Z",
     "iopub.status.busy": "2022-04-01T19:45:51.572522Z",
     "iopub.status.idle": "2022-04-01T19:45:51.722294Z",
     "shell.execute_reply": "2022-04-01T19:45:51.721621Z",
     "shell.execute_reply.started": "2022-04-01T19:45:51.572933Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "import itertools\n",
    "from keras.layers import LeakyReLU\n",
    "from keras.utils.np_utils import to_categorical # convert to one-hot-encoding\n",
    "from keras.models import Sequential, Model\n",
    "from keras.layers import Dense, Dropout, Flatten, Conv2D, MaxPool1D,Conv1D\n",
    "from keras.optimizers import RMSprop,Adam,Nadam\n",
    "from keras.preprocessing.image import ImageDataGenerator\n",
    "from keras.callbacks import ReduceLROnPlateau\n",
    "import tensorflow as tf\n",
    "keras = tf.keras  \n",
    "\n",
    "from tensorflow.keras import layers\n",
    "from tensorflow.keras import initializers\n",
    "\n",
    "initializer = tf.keras.initializers.RandomUniform(minval=0., maxval=1.)\n",
    "\n",
    "model = Sequential()\n",
    "\n",
    "model.add(Conv1D(filters = 3, kernel_size =7,kernel_initializer=initializer, padding = 'Same',activation=\"relu\", \n",
    "                 input_shape = (467,1)))\n",
    "\n",
    "\n",
    "model.add(MaxPool1D(pool_size=2))\n",
    "model.add(Dropout(0.25))\n",
    "#\n",
    "model.add(Conv1D(filters = 1, kernel_size =5,kernel_initializer=initializer,padding = 'Same',activation=\"relu\"))\n",
    "\n",
    "\n",
    "model.add(MaxPool1D(pool_size=2, strides=2))\n",
    "model.add(Dropout(0.25))\n",
    "\n",
    "model.add(Flatten())\n",
    "model.add(Dense(8, activation=\"relu\"))\n",
    "\n",
    "model.add(Dropout(0.5))\n",
    "model.add(Dense(2, activation = \"sigmoid\"))\n",
    "model.summary()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "_uuid": "d00d9b96668d11a33c272b2b64cbb4397dcbba89",
    "execution": {
     "iopub.execute_input": "2022-04-01T19:45:51.724158Z",
     "iopub.status.busy": "2022-04-01T19:45:51.72387Z",
     "iopub.status.idle": "2022-04-01T19:45:51.745249Z",
     "shell.execute_reply": "2022-04-01T19:45:51.74452Z",
     "shell.execute_reply.started": "2022-04-01T19:45:51.724112Z"
    },
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "optimizer = Adam(lr=0.001, beta_1=0.9, beta_2=0.999)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-01T19:45:51.767248Z",
     "iopub.status.busy": "2022-04-01T19:45:51.767022Z",
     "iopub.status.idle": "2022-04-01T19:45:51.793763Z",
     "shell.execute_reply": "2022-04-01T19:45:51.793182Z",
     "shell.execute_reply.started": "2022-04-01T19:45:51.767205Z"
    }
   },
   "outputs": [],
   "source": [
    "model.compile(optimizer = optimizer , loss = \"categorical_crossentropy\", metrics=[\"accuracy\"])\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-01T19:45:51.868278Z",
     "iopub.status.busy": "2022-04-01T19:45:51.868048Z",
     "iopub.status.idle": "2022-04-01T19:45:51.873976Z",
     "shell.execute_reply": "2022-04-01T19:45:51.873294Z",
     "shell.execute_reply.started": "2022-04-01T19:45:51.868236Z"
    }
   },
   "outputs": [],
   "source": [
    "epochs = 1 # for better result increase the epochs\n",
    "batch_size = 32"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-01T19:45:52.084503Z",
     "iopub.status.busy": "2022-04-01T19:45:52.084247Z",
     "iopub.status.idle": "2022-04-01T19:45:54.788016Z",
     "shell.execute_reply": "2022-04-01T19:45:54.78732Z",
     "shell.execute_reply.started": "2022-04-01T19:45:52.084457Z"
    }
   },
   "outputs": [],
   "source": [
    "history = model.fit(X_train,Y_trainn,epochs = epochs, validation_data = (X_val,Y_vall),batch_size=32)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "_uuid": "82bddf135bfad1d3f91b52cd0cadb62e55db964b",
    "execution": {
     "iopub.execute_input": "2022-04-01T19:45:54.790267Z",
     "iopub.status.busy": "2022-04-01T19:45:54.789977Z",
     "iopub.status.idle": "2022-04-01T19:45:54.930069Z",
     "shell.execute_reply": "2022-04-01T19:45:54.929008Z",
     "shell.execute_reply.started": "2022-04-01T19:45:54.790218Z"
    }
   },
   "outputs": [],
   "source": [
    "from keras.utils.vis_utils import plot_model\n",
    "plot_model(model, to_file='model_plot.png', show_shapes=True, show_layer_names=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-01T19:45:54.932875Z",
     "iopub.status.busy": "2022-04-01T19:45:54.932314Z",
     "iopub.status.idle": "2022-04-01T19:45:55.452768Z",
     "shell.execute_reply": "2022-04-01T19:45:55.451994Z",
     "shell.execute_reply.started": "2022-04-01T19:45:54.932822Z"
    }
   },
   "outputs": [],
   "source": [
    "fig, ax = plt.subplots(2,1, figsize=(18, 10))\n",
    "ax[0].plot(history.history['loss'], color='b', label=\"Training loss\")\n",
    "ax[0].plot(history.history['val_loss'], color='r', label=\"validation loss\",axes =ax[0])\n",
    "legend = ax[0].legend(loc='best', shadow=True)\n",
    "\n",
    "ax[1].plot(history.history['acc'], color='b', label=\"Training accuracy\")\n",
    "ax[1].plot(history.history['val_acc'], color='r',label=\"Validation accuracy\")\n",
    "legend = ax[1].legend(loc='best', shadow=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-01T19:45:55.454835Z",
     "iopub.status.busy": "2022-04-01T19:45:55.454309Z",
     "iopub.status.idle": "2022-04-01T19:45:55.727878Z",
     "shell.execute_reply": "2022-04-01T19:45:55.72707Z",
     "shell.execute_reply.started": "2022-04-01T19:45:55.454784Z"
    }
   },
   "outputs": [],
   "source": [
    "fig = plt.figure(figsize=(10, 10)) # Set Figure\n",
    "\n",
    "yy_pred = model.predict(X_val) # Predict encoded label as 2 => [0, 0, 1, 0, 0, 0, 0, 0, 0, 0]\n",
    "\n",
    "YY_pred = np.argmax(yy_pred, 1) # Decode Predicted labels\n",
    "Y_valc = np.argmax(Y_vall, 1) # Decode labels\n",
    "\n",
    "mat = confusion_matrix(Y_valc, YY_pred) # Confusion matrix\n",
    "\n",
    "# Plot Confusion matrix\n",
    "sns.heatmap(mat.T, square=True, annot=True, cbar=False, cmap=plt.cm.Blues)\n",
    "plt.xlabel('Predicted Values')\n",
    "plt.ylabel('True Values');\n",
    "plt.show();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-01T19:45:55.731467Z",
     "iopub.status.busy": "2022-04-01T19:45:55.730859Z",
     "iopub.status.idle": "2022-04-01T19:45:55.844571Z",
     "shell.execute_reply": "2022-04-01T19:45:55.842932Z",
     "shell.execute_reply.started": "2022-04-01T19:45:55.731417Z"
    }
   },
   "outputs": [],
   "source": [
    "model_feat = Model(inputs=model.input,outputs=model.get_layer('dense_1').output)\n",
    "\n",
    "feat_trainf = model_feat.predict(X_train)\n",
    "print(feat_trainf.shape)\n",
    "\n",
    "feat_valf = model_feat.predict(X_val)\n",
    "print(feat_valf.shape)\n",
    "\n",
    "print(feat_valf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-01T19:45:55.936629Z",
     "iopub.status.busy": "2022-04-01T19:45:55.936354Z",
     "iopub.status.idle": "2022-04-01T19:45:56.086744Z",
     "shell.execute_reply": "2022-04-01T19:45:56.086084Z",
     "shell.execute_reply.started": "2022-04-01T19:45:55.936583Z"
    }
   },
   "outputs": [],
   "source": [
    "featval=pd.DataFrame(feat_vald)\n",
    "featval.to_excel('X_val.xlsx')\n",
    "featvall=pd.DataFrame(feat_traind)\n",
    "featvall.to_excel('X_train.xlsx')\n",
    "featvalll=pd.DataFrame(Y_train)\n",
    "featvalll.to_excel('Y_train.xlsx')\n",
    "featvallll=pd.DataFrame(Y_val)\n",
    "featvallll.to_excel('Y_val.xlsx')"
   ]
  }
 ],
 "metadata": {
  "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": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}