GPX analysis, including gpx to elevation profile chart

GPX Analysis for Blog Post.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "76a8bc7b",
   "metadata": {},
   "source": [
    "# Some notes on some GPX tinkering and analysis for a blog post\n",
    "\n",
    "Compared to some other notebooks out there I think this simplifies the charting of an elevation profile from a GPX file a fair bit. Though I had a slightly odd use case (charting gaps when there were large jumps in distance) that might not be suitable for everyone.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f157fe84",
   "metadata": {},
   "outputs": [],
   "source": [
    "!pip install gpxpy\n",
    "!pip install geopy\n",
    "!pip install plotly\n",
    "!pip install kaleido"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "060968f3",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Pre-work was to use GPSBabel to merge all of the gpx files in to 1\n",
    "# https://stackoverflow.com/questions/38554131/merge-all-gpx-files-within-a-folder-into-one-file-with-gpsbabel\n",
    "# This produces All.gpx\n",
    "# As an optimisation for graphing I grep'd out the time/speed elements and produced `All-notimeorspeed.gpx`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b00abf84",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Produce high level stats\n",
    "#sudo apt-get install -y gpxinfo\n",
    "#gpxinfo -m All.gpx > stats"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "06f8684e",
   "metadata": {},
   "outputs": [],
   "source": [
    "import gpxpy\n",
    "import gpxpy.gpx\n",
    "import math\n",
    "import geopy.distance\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import pandas as pd\n",
    "\n",
    "with open('All.gpx', 'r') as gpx_file:\n",
    "    gpx = gpxpy.parse(gpx_file)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "230574b1",
   "metadata": {},
   "outputs": [],
   "source": [
    "elevation = []\n",
    "is_real = []\n",
    "distance = []\n",
    "\n",
    "prev_lat = None\n",
    "prev_lon = None\n",
    "\n",
    "new_track = False\n",
    "\n",
    "for track in gpx.tracks:\n",
    "    \n",
    "\n",
    "    for segment in track.segments:\n",
    "        for point in segment.points:\n",
    "            if point.elevation:\n",
    "                \n",
    "                if(prev_lat is not None):\n",
    "                    diff = geopy.distance.geodesic((prev_lat, prev_lon), (point.latitude, point.longitude)).km\n",
    "                    prev_distance = distance[-1]\n",
    "                else:\n",
    "                    diff = 0\n",
    "                    prev_distance = 0\n",
    "                \n",
    "                # I want to add NaN values for the chart below so that it shows when I jumped (e.g. taking a bus),\n",
    "                # this is the hack to do that.\n",
    "                if new_track:\n",
    "                    if diff > 5:\n",
    "                        elevation.append(point.elevation)\n",
    "                        distance.append(prev_distance + 2)                \n",
    "                        is_real.append(False)\n",
    "                \n",
    "                elevation.append(point.elevation)\n",
    "                distance.append(prev_distance + diff)\n",
    "                is_real.append(True)  \n",
    "                \n",
    "                prev_lat = point.latitude\n",
    "                prev_lon = point.longitude\n",
    "\n",
    "                \n",
    "        new_track = True\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3b429505",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "df = pd.DataFrame({\n",
    "    'distance': distance,\n",
    "    'elevation': elevation,\n",
    "    'is_real': is_real\n",
    "})\n",
    "\n",
    "df = df.set_index('distance')\n",
    "\n",
    "df['elevation'][df.is_real == False] = pd.NA\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "69efac43",
   "metadata": {},
   "outputs": [],
   "source": [
    "import plotly.express as px\n",
    "\n",
    "fig = px.line(\n",
    "    df,\n",
    "    y=\"elevation\",\n",
    "    labels={\"elevation\": \"Elevation (m)\", \"distance\": \"Distance (km)\"},\n",
    ")\n",
    "\n",
    "fig.write_image('/home/andy/git/blog/content/images/vancouver-to-tijuana-elevation-main.png', format=\"png\", width=4000, height=450, scale=1)\n",
    "fig.write_image('/home/andy/git/blog/content/images/vancouver-to-tijuana-elevation-thumb.png', format=\"png\", width=700, height=250, scale=1)\n",
    "\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}