{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Advanced Filtering"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Kalman Filter"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from vitalDSP.filtering.advanced_signal_filtering import AdvancedSignalFiltering\n",
    "import plotly.io as pio\n",
    "pio.renderers.default = \"sphinx_gallery\"\n",
    "# pio.renderers.default = \"plotly_mimetype\"  # or \"jupyterlab\"\n",
    "# from IPython.display import display, HTML\n",
    "# display(HTML('<script src=\"https://cdnjs.cloudflare.com/ajax/libs/require.js/2.3.6/require.min.js\"></script>'))\n",
    "import numpy as np\n",
    "from plotly import graph_objects as go\n",
    "import os\n",
    "from vitalDSP.notebooks import plot_trace, load_sample_ecg_small\n",
    "\n",
    "\n",
    "# FILE_PATH = os.path.join(\"..\",\"..\",\"..\",\"sample_data\",\"public\",\"ecg.csv\")\n",
    "# signal_col, date_col = process_in_chunks(FILE_PATH,data_type='ecg', fs=128)\n",
    "signal_col, date_col = load_sample_ecg_small()\n",
    "signal_col = np.array(signal_col)\n",
    "\n",
    "af = AdvancedSignalFiltering(signal_col)\n",
    "filtered_signal = af.kalman_filter(R=0.1, Q=0.01)\n",
    "\n",
    "plot_trace(signal_col,filtered_signal)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Optimization-based filter"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from vitalDSP.filtering.advanced_signal_filtering import AdvancedSignalFiltering\n",
    "from vitalDSP.utils.data_processing.synthesize_data import generate_ecg_signal\n",
    "import numpy as np\n",
    "from vitalDSP.notebooks import load_sample_ecg_small, plot_trace\n",
    "\n",
    "fs = 128\n",
    "signal_col, date_col = load_sample_ecg_small()\n",
    "signal_col = np.array(signal_col)\n",
    "\n",
    "target = generate_ecg_signal(sfecg=128, N=30, Anoise=0.01, hrmean=70, sfint=512)\n",
    "target = np.array(target[:min(len(target), len(signal_col))])\n",
    "signal_col = signal_col[:len(target)]\n",
    "\n",
    "af = AdvancedSignalFiltering(signal_col)\n",
    "filtered_signal = af.optimization_based_filtering(target, loss_type=\"mse\",\n",
    "                                                   initial_guess=0, learning_rate=0.01,\n",
    "                                                   iterations=20)\n",
    "plot_trace(signal_col, filtered_signal)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Gradient Descent Filter"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from vitalDSP.filtering.advanced_signal_filtering import AdvancedSignalFiltering\n",
    "from vitalDSP.utils.data_processing.synthesize_data import generate_ecg_signal\n",
    "import numpy as np\n",
    "from plotly import graph_objects as go\n",
    "import os\n",
    "from vitalDSP.notebooks import load_sample_ecg_small, plot_trace\n",
    "\n",
    "fs = 128\n",
    "signal_col, date_col = load_sample_ecg_small()\n",
    "signal_col = np.array(signal_col)\n",
    "\n",
    "# Create a target signal\n",
    "target = generate_ecg_signal(sfecg=128, N=30, Anoise=0.01, hrmean=70, sfint=512)\n",
    "target = target[:min(len(target),len(signal_col))]\n",
    "\n",
    "signal_col = signal_col[:min(len(target),len(signal_col))]\n",
    "af = AdvancedSignalFiltering(signal_col)\n",
    "filtered_signal = af.gradient_descent_filter(target, learning_rate=0.1, \n",
    "                                                iterations=20)\n",
    "\n",
    "plot_trace(signal_col,filtered_signal)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Convolution-based filter"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from vitalDSP.filtering.advanced_signal_filtering import AdvancedSignalFiltering\n",
    "import numpy as np\n",
    "from vitalDSP.notebooks import plot_trace, load_sample_ecg_small\n",
    "\n",
    "signal_col, date_col = load_sample_ecg_small()\n",
    "signal_col = np.array(signal_col)\n",
    "\n",
    "af = AdvancedSignalFiltering(signal_col)\n",
    "\n",
    "smoothed = af.convolution_based_filter(kernel_type=\"smoothing\", kernel_size=5)\n",
    "plot_trace(signal_col, smoothed, title=\"Smoothing kernel\")\n",
    "\n",
    "edge = af.convolution_based_filter(kernel_type=\"edge_detection\")\n",
    "plot_trace(signal_col, edge, title=\"Edge detection kernel\")\n",
    "\n",
    "sharpened = af.convolution_based_filter(kernel_type=\"sharpening\")\n",
    "plot_trace(signal_col, sharpened, title=\"Sharpening kernel\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Attention-based filter"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from vitalDSP.filtering.advanced_signal_filtering import AdvancedSignalFiltering\n",
    "import numpy as np\n",
    "from vitalDSP.notebooks import load_sample_ecg_small, plot_trace\n",
    "\n",
    "signal_col, date_col = load_sample_ecg_small()\n",
    "signal_col = np.array(signal_col)\n",
    "\n",
    "af = AdvancedSignalFiltering(signal_col)\n",
    "\n",
    "gaussian_out = af.attention_based_filter(attention_type=\"gaussian\", size=11, sigma=2.0)\n",
    "plot_trace(signal_col, gaussian_out, title=\"Gaussian attention\")\n",
    "\n",
    "linear_out = af.attention_based_filter(attention_type=\"linear\", size=11, ascending=True)\n",
    "plot_trace(signal_col, linear_out, title=\"Linear attention\")\n",
    "\n",
    "exp_out = af.attention_based_filter(attention_type=\"exponential\", size=7, base=2.0)\n",
    "plot_trace(signal_col, exp_out, title=\"Exponential attention\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## LMS Adaptive Filtering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from vitalDSP.filtering.advanced_signal_filtering import AdvancedSignalFiltering\n",
    "from vitalDSP.utils.data_processing.synthesize_data import generate_ecg_signal\n",
    "import numpy as np\n",
    "from vitalDSP.notebooks import load_sample_ecg_small, plot_trace\n",
    "\n",
    "fs = 128\n",
    "signal_col, date_col = load_sample_ecg_small()\n",
    "signal_col = np.array(signal_col)\n",
    "\n",
    "target = generate_ecg_signal(sfecg=128, N=30, Anoise=0.01, hrmean=70, sfint=512)\n",
    "target = np.array(target[:min(len(target), len(signal_col))])\n",
    "signal_col = signal_col[:len(target)]\n",
    "\n",
    "af = AdvancedSignalFiltering(signal_col)\n",
    "# mu must satisfy mu < 1/(filter_order * max_signal_power) for LMS stability\n",
    "filtered_signal = af.adaptive_filtering(target, mu=0.005, filter_order=5)\n",
    "\n",
    "plot_trace(signal_col, filtered_signal)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "wearables",
   "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": "3.9.17"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}