{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Download the data\n",
    "\n",
    "https://drive.switch.ch/index.php/s/4cL6G5iN7PamGiM\n"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Artificial Example: A Gaussian Distribution"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from scipy.stats import multivariate_normal\n",
    "\n",
    "# Define the mean and covariance matrix\n",
    "mean = [0, 0]\n",
    "cov = [[7, 4], [4, 7]]\n",
    "\n",
    "\n",
    "# Create a grid of points\n",
    "prange = 9\n",
    "x, y = np.meshgrid(np.linspace(-prange, prange, 100), np.linspace(-prange, prange, 100))\n",
    "\n",
    "\n",
    "# Create the 2D Gaussian\n",
    "pos = np.empty(x.shape + (2,))\n",
    "pos[:, :, 0] = x\n",
    "pos[:, :, 1] = y\n",
    "\n",
    "# TODO: Create a multivariate normal distribution and sample 100 points from it (variable sample),\n",
    "# create z for the meshgrid of the normal distribution to show the 2D Gaussian density (variable z).\n",
    "\n",
    "\n",
    "# Plot the 2D Gaussian\n",
    "fig, ax = plt.subplots(figsize=(8, 8))\n",
    "ax.contourf(x, y, z, cmap='Blues')\n",
    "ax.scatter(sample[:, 0], sample[:, 1], color='red')\n",
    "ax.set_aspect('equal')\n",
    "ax.set_xlabel('x')\n",
    "ax.set_ylabel('y')\n",
    "ax.set_title('2D Gaussian')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "# TODO: It is important to scale the data for PCA, we do that here (rescale the variable sample):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.decomposition import PCA\n",
    "\n",
    "# TODO: Perform PCA on the sample using sklearn.decomposition.PCA\n",
    "\n",
    "# TODO: Compute the eigenvalues and eigenvectors of the covariance matrix using (np.linalg.eigh)\n",
    "# eigh returns The normalized (unit \"length\") eigenvectors, \n",
    "# such that the column v[:,i] is the eigenvector corresponding to the eigenvalue w[i].\n",
    "\n",
    "# Print the explained variance ratio and the principal components and compare the results\n",
    "print(\"Explained variance ratio:\", pca.explained_variance_ratio_)\n",
    "print(\"Principal components:\")\n",
    "print(pca.components_)\n",
    "print(\"------------------\")\n",
    "print(\"Eigenvalues and Eigenvectors:\")\n",
    "print(eigenvalues, eigenvectors)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "PC_values = np.arange(pca.n_components_) + 1\n",
    "plt.plot(PC_values, pca.explained_variance_ratio_, 'ro-', linewidth=2)\n",
    "plt.title('Scree Plot')\n",
    "plt.xlabel('Principal Component')\n",
    "plt.ylabel('Proportion of Variance Explained')\n",
    "plt.show()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# A Dataset, the CMAPSS Engine"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Aircraft Engine Run-to-Failure Dataset under Real Flight Conditions for Prognostics and Diagnostics\n",
    "\n",
    "https://drive.switch.ch/index.php/s/FAfONeeRGr4ENC2\n",
    "\n",
    "<img src=\"5u4eg1n3.bmp\">"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "OC_MAPPING = {\n",
    "    'steady': 1,\n",
    "    'ascending': 0,\n",
    "    'descending': 2,\n",
    "}\n",
    "\n",
    "def label_oc(df):\n",
    "    '''\n",
    "    Function to label all operating conditions.\n",
    "    '''\n",
    "    \n",
    "    df.loc[:, 'Operating Condition'] = 4\n",
    "    \n",
    "    if 'ds' in df.columns:\n",
    "        df['ds_unit'] = df['ds'].astype(str) + '_' + df['unit'].astype(str)\n",
    "        unique_engines = df['ds_unit'].unique()\n",
    "    else:\n",
    "        unique_engines = df['unit'].unique()\n",
    "    \n",
    "    # Loop through each engine unit\n",
    "    for engine in unique_engines:\n",
    "        \n",
    "        # Select only the flight data\n",
    "        if 'ds' in df.columns:\n",
    "            df_engine = df[df['ds_unit'] == engine]\n",
    "        else:\n",
    "            df_engine = df[df['unit'] == engine]\n",
    "        unique_cycle = df_engine['cycle'].unique()\n",
    "        \n",
    "        # Loop through each flight within one unit\n",
    "        for cycle in unique_cycle:\n",
    "            \n",
    "            df_cycle = df_engine[df_engine['cycle'] == cycle]\n",
    "            # df_cycle.reset_index(inplace = True)\n",
    "            \n",
    "            # Shift altitude forward in time by one (set initial value to zero)\n",
    "            alt_shift = df_cycle['alt'].shift(periods = 1, fill_value = 0)\n",
    "            alt_change = df_cycle['alt'] - alt_shift\n",
    "            \n",
    "            \n",
    "            # Name operating condition based on altitude change\n",
    "            oc = np.zeros(len(df_cycle))\n",
    "            oc[alt_change >= 5] = 0 # Ascending\n",
    "            oc[alt_change.abs() < 5] = 1 # Steady\n",
    "            oc[alt_change <= -5] = 2 # Descending\n",
    "            \n",
    "            # Smooth predictions with a centered median filter (fill nan values in the beginning and end)\n",
    "            oc = pd.Series(oc).rolling(51, center = True).median()\n",
    "            oc.ffill(inplace = True)\n",
    "            oc.bfill(inplace = True)\n",
    "            \n",
    "            # Transfer results to original dataframe\n",
    "            df.loc[df_cycle.index, 'oc'] = oc.values\n",
    "    \n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 135,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ncmapss_u2 = pd.read_csv('pca_lecture/ncmapps_ds02.csv')\n",
    "ncmapss_u2 = label_oc(ncmapss_u2)\n",
    "run_slice = slice(None, None, None)\n",
    "ncmapss_u2 = ncmapss_u2[run_slice]\n",
    "ncmapss_u2['RUL'].plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 137,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 137,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1080x360 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "index_key = \"oc\"\n",
    "\n",
    "control = ['alt', 'Mach', 'TRA', 'T2', 'Fc', 'unit', 'hs', 'cycle']\n",
    "sensor = ['T24', 'T30', 'T48', 'T50', 'P15', 'P2', 'P21', 'P24', 'Ps30', 'P40', 'P50', 'Nf', 'Nc', 'Wf']\n",
    "\n",
    "\n",
    "ds = ncmapss_u2[sensor].reset_index(drop=True)\n",
    "label = ncmapss_u2[index_key].reset_index(drop=True)\n",
    "rul = ncmapss_u2['RUL'].reset_index(drop=True)\n",
    "\n",
    "fig, ax = plt.subplots(1, 3, figsize=(15, 5))\n",
    "ds[::1000].plot(ax=ax[0])\n",
    "rul.plot(ax=ax[1], legend=True)\n",
    "ncmapss_u2['oc'].plot(ax=ax[2], marker='o', linestyle='None', markersize=1,  legend=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "# TODO: again, scale and transform the data using StandardScaler and PCA and create your own Scree plot of the sensor signals"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 141,
   "metadata": {},
   "outputs": [],
   "source": [
    "# TODO: Now select an appropriate number of components and transform the data and try to to relate the PCA components to the RUL. use pandas df.plot to visualize the results!"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Clustering With PCA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# TODO visualize two princial components and color them by the operating condition \"oc\"\n",
    "# Ask yourself how you could use this representation to categorize the operating condition of unseen data?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# We can visualize how much the original data contributes to each PCA Principal axes in feature space.\n",
    "# Use one of your transformations and the following code to visualize the contribution as arrows:\n",
    "\n",
    "\n",
    "# (Principal axes in feature space, representing the directions of maximum variance in the data.)\n",
    "loadings = pca.components_[:2].T\n",
    "arrows = loadings * np.abs(dss_pca.values[:,:2]).max(axis=0)\n",
    "\n",
    "# proportions of variance explained by axes\n",
    "pvars = pca.explained_variance_ratio_[:2] * 100\n",
    "\n",
    "\n",
    "fig, ax = plt.subplots(1,1,figsize=(10,10))\n",
    "\n",
    "pd.concat([dss_pca, label], axis=1) \\\n",
    "  .plot.scatter(x=0, y=1, c=index_key, colormap='viridis' , ax=ax)\n",
    "\n",
    "# features as arrows\n",
    "for i, arrow in enumerate(arrows):\n",
    "    ax.arrow(0, 0, *arrow, color='k', alpha=0.5, width=0.1, ec='none',\n",
    "              length_includes_head=True)\n",
    "    plt.text(*(arrow * 1.05), ds.columns[i], ha='center', va='center')\n",
    "\n"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# PCA as Encoder / Decoder Bottleneck"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Likewise to e.g. an Autoencoder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Optional exercise: reconstruct the original data from the PCA components and plot the difference of the reconstructed data and the original data.\n",
    "# Can you see any anomaly in the data?\n",
    "\n",
    "# this is a sketch how you can reconstruct x_hat\n",
    "nComp = 3\n",
    "x_hat = np.dot(dss_pca.values[:,:nComp], pca.components_[:nComp,:])"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Calculation of the Q and T2-Statistics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 177,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Optional exercise 2: can you spot strong degradation of the engine using Q and T2-Statistics?"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "torch2_cuda12.1",
   "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": "3.10.12"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}
