{ "cells": [ { "cell_type": "markdown", "source": [ "# Initialschätzung von Kurswechselpositionen eines Segelboots auf einer Karte anhang von Wind, Start und Zielpunkt\n", "\n", "## Motivation\n", "\n", "Ziel dieser Semesterabschließenden schriftlichen Ausarbeitung im Fach \"Maschine Learning\" an der Fachhochschule Südwestfalen ist das Generieren einer Heatmap von Kurswechselpositionen eines Segelbootes zu einer Karte abhängig von Wind und der Zielpostion. Dies soll das Finden einer guten Route vereinfachen, indem die Qualität der Initialen Route verbessert wird. Diese werden danach über ein Quotientenabstiegsverfahren optimiert. Da ein solches Quotientenabstiegsverfahren sehr gerne in einem lokalen Minimum festhängt, müssen mehrere Routen gefunden und optimiert werden. Hier soll untersucht werden, ob dies durch eine Ersteinschätzung der Lage durch KI verbessert werden kann.\n", "\n", "Eingesetzt werden soll die so erstellte KI in dem Segelroboter des [Sailing Team Darmstadt e.V.](https://www.st-darmstadt.de/)[7], einer Hochschulgruppe an der TU-Darmstadt, welche den [\"roBOOTer\"](https://www.st-darmstadt.de/ueber-uns/boote/prototyp-ii/)[7] ein vollautonomes Segelboot, das eines Tages den Atlantik überqueren soll. [Eine technische Herausforderung, welche zuerst von einem norwegischen Team erfolgreich abgeschlossen wurde](https://www.microtransat.org/)[6]." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "markdown", "source": [ "## Inhaltsverzeichnis\n", "\n", "1. Einleitung\n", "\n", " 1.1. Situation\n", " \n", " 1.2. Vorgehen zur unterstützenden KI\n", " 1.2.1. Eingaben und Ausgeben\n", "2. Vorbereitungen\n", "\n", " 2.1. Imports\n", " \n", " 2.2. Parameter und Settings\n", " \n", "3. Szenarien und Routen generieren\n", "\n", " 3.1. Generieren von Karten\n", " 3.2.1 Parameter zum Generieren der Karte\n", "\n", " 3.2. Generieren des Zieles\n", "\n", " 3.3. Das Normieren eines Szenarios\n", "\n", " 3.4. Massengenerierung von Szenarios\n", "\n", " 3.5. Daten zusammenfassen\n", "\n", "4. Szenarios filtern\n", "\n", " 4.1. Die Route verlässt die Karte\n", "\n", " 4.2. Routen auf Fehler überprüfen\n", "\n", " 4.3. Filtern der Routen nach Kosten\n", "\n", " 4.4. Filtern der Routen nach Komplexität\n", "\n", "5. Das Konvertieren in trainierbare Daten\n", "\n", "6. Das Model\n", "\n", " 6.1. Der Generator\n", " 6.1.1 Model Erstellung\n", " 6.2. Model Training\n", " 6.2.1. Training des Modells mit Routen als Linien\n", " 6.2.2. Betrachtung des trainierten Models mit Routen als Linien\n", " 6.2.3. Training mit Routen als Liste von Wendepunkten\n", " 6.2.4. Betrachtung des trainierten Models mit Routen als Liste von Wendepunkten\n", "\n", "\n", "7. Vergleich der Experimente\n", "\n", "9. Reflektion und Ausblick\n", "\n", "10. Literaturverzeichnis" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "markdown", "source": [ "\n", "## Einleitung\n", "### Situation\n", "Eine Routenplanung für ein Segelboot hat ein Problem, welches man sonst so eher nicht kennt. Eine relativ freie Fläche auf der sich das Schiff bewegen kann. Dies verändert die Wegfindung wie man sie von der Straße kennt, fundamental.\n", "Navigiert man auf Straßen, hat man zumindest nach einer ersten Abstraktion relativ wenige Freiheitsgrade für den Weg. Die Richtung kann nur an Kreuzungen gewechselt werden und dort nur in Richtungen, in die es Straßen gibt. Beim Segeln auf dem freien Meer ist jeder Ort ein potenzieller Wendepunkt, von dem aus potenziell in jede Richtung gesegelt werden kann.\n", "Dennoch ist es auch ohne Hindernisse zwischen Boot und Ziel oft nicht möglich das Ziel direkt anzufahren, da sich die Maximalgeschwindigkeiten relativ zur Windrichtung verändern. Das folgende Diagramm zeigt die Segelgeschwindigkeiten an einem Katamaran.\n", "\n", "Da der roBOOTer anders als ein Katamaran nicht auf Geschwindigkeit, sondern auf mechanische Belastbarkeit ausgelegt wurde hat der Fahrtwind einen geringeren Einfluss auf das Fahrtverhalten des Segelboots. Dies und eine andere Maximalgeschwindigkeit sorgen für ein etwas anderes Fahrverhalten. Die ungefähre Form der Kurven trifft aber auch auf den roBOOTer zu. Man kann deutlich erkennen, dass auch, wenn man nicht direkt gegen den Wind fahren kann man schräg gegen den Wind immer noch erstaunlich schnell ist.\n", "Das aktuelle Verfahren zum Finden einer Route läuft folgendermaßen ab:\n", "Eine direkte Route wird berechnet. Die Route wird an jedem Hindernisse geteilt und rechts und links um jedes Hindernis herum gelegt. Bei folgenden Hindernissen werden die Routen wieder geteilt somit erhält man Vorschläge für Routen, wobei die Anzahl der Hindernisse auf der Route ist. Jeder Abschnitt der Route wird noch einmal zerteilt, um der Route mehr Flexibilität zu geben.\n", "Die Routen werden dann simuliert, um die Kosten der Route zu berechnen. Die so simulierte Route wird danach über die Kosten in einem Gradientenabstiegsverfahren optimiert.\n", "Das ganze oben beschriebene Verfahren ist relativ schnell sehr rechenaufwendig und findet nicht immer ein Ergebnis. Wird kein Ergebnis gefunden wird eine mehr oder weniger zufällige Route optimiert.\n", "Diese Ausarbeitung soll wenigstens bei der alternativen Routenfindung helfen. Im Idealfall kann es aber auch genutzt werden, um die Auswahl der Routen um Hindernisse frühzeitig zu reduzieren und den Rechenaufwand unter zu senken wobei die Anzahl von Hindernissen auf der Route ist.\n", "\n", "### Vorgehen zur unterstützenden KI\n", "\n", "#### Eingaben und Ausgeben\n", "\n", "Die Algorithmen zur Wegfindung vom Sailing Team Darmstadt e.V. arbeiten intern mit Polygonen als Hindernissen. Diese werden durch die Shapely Bibliothek implementiert. Da eine variable Anzahl an Polygonen mit einer variablen Form und Position ein relativ komplexer Eingabewert ist, muss dieser in eine normierte Form gebracht werden. Ein binärfarbiges Bild ist dafür die einfachste Form.\n", "Für den Computer spielen weder Zentrierung noch Skalierung und Ausrichtung der Karte eine Rolle. Wir rotieren also die Karte immer, so das der Wind von Norden kommt und das Boot / die Startposition in der Mitte der Karte liegt. Da Distanz ein linearer Faktor ist, wird davon ausgegangen, dass das Szenario einfach passend skaliert werden kann.\n", "Die nächste Eingabe ist die Zielposition relativ zum Startpunkt. Diese kann entweder durch ein einzelnes Pixel in einem zweiten Farbkanal oder aber in abstrakterer Form an die KI übergeben werden.\n", "Als Ausgabe wird eine Heatmap erwartet. Zwei alternative Heatmaps sind relativ einfach denkbar.\n", "Eine Headmap der Kurswechselpositionen\n", "Eine Headmap des Kursverlaufes\n", "Headmaps sind in gewisser Weise Bilder. Das Problem wird daher wie ein Bild zu Bild KI Problem betrachtet. Diese werden normalerweise durch neuronale Netze gelöst.\n", "Um ein neuronales Netz zu trenntieren, gibt es immer die Wahl zwischen drei primären Prinzipien. Dem unüberwachten Lernen, dem reinforcement Learning und dem überwachten Lernen. Letzteres ist dabei meist am einfachsten wenn, auch nicht immer möglich. Da es Lösungen für die Szenarien gibt, wurde hier das überwachte Lernen gewählt.\n", "Der Wegfindealgorithmus des Sailing Team Darmstadt e.V. ist zwar noch in der Entwicklung, funktioniert aber hinreichend gut, um auf einem normalen PC Szenarios mit Routen zu paaren oder auch diese zu labeln, um beim KI lingo zu bleiben. Um anpassungsfähig an andere Szenarios zu sein wird eine große Menge unterschiedlicher Szenarios und Routen benötigt. Da das Haupteinsatzgebiet das Meer ist gehen wir von einer Insellandschaft oder Küstenlandschaft aus.\n", "Zum Finden von Szenarios gibt es zwei Möglichkeiten.\n", "Das Auswählen von Umgebungen von der Weltkarte und das Bestimmen eines Zielpunktes.\n", "Das Generieren von künstlichen Szenarios.\n", "Hier wird die Annahme getroffen, dass sich neuronale Netzwerke von einem Datensatz auf dem anderen Übertragen lassen. Der Aufwand für künstliche Szenarios wird hierbei als geringer eingestuft und daher gewählt." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "markdown", "source": [ "## Vorbereitungen\n", "\n", "Folgende Python Bibliotheken werden verwendet:\n", "\n", "1. `tensorflow`\\\n", " Die `tensorflow` Bibliothek ist das Werkzeug welches verwendet wurde, um neuronale Netze zu modellieren, zu trainieren, zu analysieren und auszuführen. Tensorflow wird mit den kürzel `tf` abgekürzt.\n", "\n", "2. `pyrate`\\\n", " Die `Pyrate` Bibliothek ist Teil des ROS Operating Systems, welches den roBOOTer betreibt. Kann Routen zu Szenarios finden.\n", "\n", "3. `Shapley`\\\n", " Die `shapley` Bibliothek wird genutzt, um geometrische Körper zu generieren, zu mergen und an den Roboter zum Labeln weiterzugeben.\n", "\n", "4. `pandas`\\\n", " Die `pandas` Bibliothek verwaltet, speichert und analysiert daten. `pandas` wird üblicherweise mit `pd` abgekürzt.\n", "\n", "5. `numpy`\\\n", " Eine Bibliothek um mathematische Operationen an multidimensionalen Arrays auszuführen. `numpy`wir üblicherweise mit `np` abgekürzt.\n", "\n", "6. `matplotlib`\\\n", " Wird genutzt, um Diagramme zu plotten. Das Modul `pyplot` wird hier vermehrt genutzt und mit dem Kürzel `plt` abgekürzt.\n", "\n", "6. `PIL`\\\n", " Eine Library, um Bilder manuell zu zeichnen.\n", "\n", "7. `humanize`\\\n", " Konvertiert Zahlen, Daten und Zeitabstände in ein für Menschen einfach leserliches Format.\n", "\n", "8. `tqdm`\\\n", " Fügt einen Fortschrittsbalken zu vielen Problemen hinzu.\n", "\n", "9. `black`\\\n", " Der `black` code Formatierer wurde genutzt, um den Code in diesem Notebook zu formatieren." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "markdown", "source": [ "#### Imports\n", "Importiert die Imports the necessary packages from python and pypi." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "import sys\n", "\n", "# Pins the python version executing the Jupyter Notebook\n", "assert sys.version_info.major == 3\n", "assert sys.version_info.minor == 10\n", "\n", "import os\n", "from typing import Optional, Final, Literal\n", "import glob\n", "import pickle\n", "\n", "from tqdm.notebook import tqdm\n", "import matplotlib.pyplot as plt\n", "import numpy as np\n", "import pandas as pd\n", "from PIL import ImageDraw, Image\n", "from shapely.geometry import Polygon, Point, LineString\n", "from shapely.ops import unary_union\n", "\n", "# os.environ[\"CUDA_VISIBLE_DEVICES\"] = \"-1\"\n", "import tensorflow as tf\n", "import humanize" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "Definiert den Pfad an dem das Jupyter Notebook ausgeführt werden soll.\n", "Importiert die pyrate module. Wird nur ausgeführt, wenn innerhalb des Pyrate Containers ausgeführt." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "# Import route generation if started in the docker container\n", "if os.getenv(\"PYRATE\"):\n", " %cd /pyrate/\n", " import experiments\n", " from pyrate.plan.nearplanner.timing_frame import TimingFrame\n", "\n", "# Protection against multi execution\n", "if not os.path.exists(\"experiments\"):\n", " %cd ../" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "if os.getenv(\"PYRATE\"):\n", " # Sets the maximum number of optimization steps that can be performed to find a route.\n", " # Significantly lowered for more speed.\n", " experiments.optimization_param.n_iter_grad = 50\n", "\n", " # Disables verbose outputs from the pyrate library.\n", " experiments.optimization_param.verbose = False" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "# Activate pandas for tqdm\n", "tqdm.pandas()" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "#### Parameter und Settings\n", "\n", "In der nachfolgenden Sektion werden verschiedene Parameter gesetzt. Zum Beispiel die Skala auf der Routen generiert werden, das äußere Limit für mögliche Ziele und die Minimaldiestanz von Zielen zum Startpunkt." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "# The scale the route should lie in. Only a mathematical limit.\n", "SIZE_ROUTE: Final[int] = 100\n", "\n", "# The outer limit in with the goal need to be placed.\n", "# Should be smaller than\n", "SIZE_INNER: Final[int] = 75\n", "assert SIZE_ROUTE > SIZE_INNER, \"The goal should be well inside the limit placed \"\n", "\n", "# The minimum distance from the start that should\n", "MIN_DESTINATION_DISTANCE: Final[int] = 25\n", "assert (\n", " SIZE_INNER > MIN_DESTINATION_DISTANCE\n", "), \"The goal should be well closer to the outer limit the\"\n", "\n", "# The size the ANN input has. Equal to the image size. Should be an element of $n^2$ to be easier compatible with ANNs.\n", "IMG_SIZE: Final[int] = 128\n", "\n", "# The size an image should be in to be easily visible by eye.\n", "IMG_SHOW_SIZE: Final[int] = 400\n", "\n", "# The number of Files that should be read to train the ANNs\n", "NUMBER_OF_FILES_LIMIT: Final[int] = 500\n", "\n", "#\n", "NO_SHOW = False\n", "GENERATE_NEW = True\n", "\n", "# The path of all the collected files\n", "DATA_COLLECTION_PATH: Final[str] = \"data/collected.pickle\"\n", "\n", "# The\n", "BATCH_SIZE: Final[int] = 32\n", "\n", "os.environ[\"TF_CPP_MIN_LOG_LEVEL\"] = \"3\"" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "## Szenarien und Routen generieren\n", "\n", "Um das neuronale Netz zu trainieren, werden Datensätze benötigt. Für die Abschätzung der Routen wird eine Karte mit Hindernissen und eine zugehörige Route benötigt. Hier wurde die Designentscheidung getroffen die Karten nicht auszuwählen, sondern zu generieren.\n", "\n", "### Generieren von Karten\n", "\n", "Eine Karte ist für das Sailing Team Darmstadt eine Menge von statischen und dynamischen Hindernissen. Statische Hindernisse sind Inseln, Landmassen und Untiefen und Fahrverbotszonen. Dynamische Hindernisse sind andere Teilnehmer am Schiffsverkehr und Wetterereignisse.\n", "In dieser KI wird sich auf statische Hindernisse beschränkt. Daher kann ein Szenario als eine Menge an Hindernispolygonen gesehen werden.\n", "Um das Generieren der Polygone einfacher zu regeln und größere statistische Kontrolle über den Generationsvorgang zu haben, sind alle generierten Basispolygone als Abschnitte auf einem Umkreis definiert, die zufällig über die Karte verteilt werden.\n", "\n", "Ein einzelnes Polygon wird hier folgendermaßen generiert:\n", "1. Die Anzahl der Kanten/Ecken wird festgelegt.\n", "2. Ein lognormal verteilter Radius wird zufällig ausgewählt.\n", "3. Auf dem Radius werden $n$ Winkel abgetragen.\n", "4. Die Winkel werden sortiert damit sich das Polygon nicht selbst schneidet.\n", "5. Die durch Radius und Winkel entstehenden Punkte werden in das kartesische Koordinatensystem umgewandelt.\n", "6. Der zufällige Offset / Polygon Mittelpunkt wird aufaddiert.\n", "7. Aus den so generierten `np.ndarray` wird ein `shapely.geometry.Polygon` erstellt.\n", "8. Polygonen die den Mittelpunkt berühren oder einschließen werden ersatzlos gelöscht.\n", "\n", "So wird eine festgelegte Anzahl von Polygonen generiert.\n", "Setzt man vor dem Generieren des ersten Polygons eines Scenarios eine random seed über `np.random.seed` so erhält man zu jedem seed ein eindeutiges menge an Polygonen wenn auch alle anderen Parameter übereinstimmen. Diese Polygon-mange hat nun mit hoher Wahrscheinlichkeit überlappende Polygone. Dies ist für den Algorithmus des Sailing Teams Darmstadt e.V. ein Problem. Die Shapley Bibliothek besitzt eine Union function, die Vereinigungsmengen von Polygonen bildet, wenn möglich. So erhält man eine reduzierte Polygonmenge. Diese kann später an einen Solver übergeben werden." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "# https://stackoverflow.com/questions/16444719/python-numpy-complex-numbers-is-there-a-function-for-polar-to-rectangular-co\n", "def polar_to_cartesian(\n", " radii: np.ndarray,\n", " perigons: np.ndarray,\n", "):\n", " \"\"\"Transforms polar coordinates into cartesian coordinates.\n", "\n", " Args:\n", " radii: A array of radii.\n", " perigons: A array of angles in perigons [0, 1[.\n", "\n", " Returns:\n", " An array of cartesian coordinates.\n", " \"\"\"\n", " return radii * np.exp(2j * perigons * np.pi)" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "def random_polygon(\n", " radius_mean: float = 2,\n", " radius_sigma: float = 1.5,\n", "):\n", " \"\"\"Generates the simplest of polygons, a triangle with a size described by a random polygon.\n", "\n", " Args:\n", " radius_mean: The average radius defining a circumcircle of a triangle.\n", " radius_sigma: The variance of a radius defining a circumcircle of a triangle.\n", "\n", " Returns:\n", " A single polygon.\n", " \"\"\"\n", " # define the number of corners\n", " number_of_corners = np.random.randint(3, 10)\n", "\n", " # generate cartesian coordinates from a radius and a sorted list of perigons.\n", " array = polar_to_cartesian(\n", " np.random.lognormal(radius_mean, radius_sigma),\n", " np.sort(np.random.rand(number_of_corners)),\n", " )\n", "\n", " # add an offset\n", " offset = np.random.randint(low=-SIZE_ROUTE, high=SIZE_ROUTE, size=(2,))\n", " return_values = np.zeros((number_of_corners, 2), dtype=float)\n", "\n", " return_values[:] = offset\n", " return_values[:, :] += np.array((np.real(array), np.imag(array))).T\n", " return Polygon(return_values)\n", "\n", "\n", "np.random.seed(42)\n", "random_polygon()" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "### Parameter zum Generieren der Karte\n", "\n", "Die folgenden Parameter wurden für das Generieren von Karten genutzt:\n", "* `radius_mean = 2` \n", "* `radius_sigma = 1`\n", "* `number_of_polygons = 40`" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "def generate_obstacles(\n", " seed: Optional[int] = None,\n", " number_of_polygons: int = 40,\n", " radius_mean: float = 2,\n", " radius_sigma: float = 1,\n", ") -> dict[str, Polygon]:\n", " \"\"\"Generates a set of obstacles from a union of triangles.\n", "\n", " The union of triangles meas that if polygons overlap o polygon containing the union of those polygons is returned.\n", " Args:\n", " seed: A seed to generate a set of obstacles from.\n", " number_of_polygons: The number of polygons that should be drawn.\n", " radius_mean: The average radius defining a circumcircle of an obstacle triangle.\n", " radius_sigma: The variance of a radius defining a circumcircle of an obstacle triangle.\n", "\n", " Returns:\n", " A list of unified obstacles.\n", " \"\"\"\n", " # sets a seed\n", " if seed is not None:\n", " np.random.seed(seed)\n", "\n", " # generate a list of polygons\n", " polygons = []\n", " for _ in range(number_of_polygons):\n", " poly = random_polygon(radius_mean, radius_sigma)\n", " # skip polygons that are to close to the start int point P(0, 0)\n", " if poly.contains(Point(0, 0)):\n", " continue\n", " if poly.exterior.distance(Point(0, 0)) < 1:\n", " continue\n", " # append to polygon list\n", " polygons.append(poly)\n", "\n", " # build unions of all polygons\n", " polygon_list = list(unary_union(polygons).geoms)\n", " return {str(i): p for i, p in enumerate(polygon_list)}" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "### Generieren des Zieles\n", "\n", "Zu jedem Scenario gehört neben einer Situation auch ein Ziel. Auch zum Generieren eines Ziels wurde zuerst der gleiche Seed gesetzt wie für den Kartengenerator. Danach wird eine zufällige Position mit Abstand zum Kartenrand ausgewählt.\n", "Die so generierte Zielposition wird danach auf Plausibilität überprüft. Folgende Prüfungen finden statt:\n", "1. Es wird sichergestellt dass, das Ziel nicht in oder an einem Hindernis liegt.\n", "2. Eine Minimaldistanz in x und y wird sichergestellt. Leider ist hier ein Fehler passiert. Anstelle die Summe der absoluten Distanz zu prüfen wurden die Distanzen für X und Y separat geprüft was verhindert, dass Ziele über, unter und neben dem Startpunkt gefunden werden können. Zielpunkte werden nur in den äußeren vier Quadranten gefunden. Bedauerlicherweise ist dies erst aufgefallen als schon zu viel Zeit vergangen war und die Daten nicht neu generiert werden konnten. Dies sollte aber zumindest das Konzept dieser KI nicht beeinflussen. Wohl aber ihre direkte Anwendbarkeit." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "def generate_destination(\n", " obstacles: dict[str, Polygon],\n", " seed: Optional[int] = None,\n", ") -> Point:\n", " \"\"\"Generates for a map.\n", "\n", " Can be used to generate a valid destination for list of obstacles.\n", " Args:\n", " obstacles: A list of obstacles.\n", " seed: The seed determining the point.\n", "\n", " Returns:\n", " A goal that should be reached by the ship.\n", " \"\"\"\n", " # sets the seed\n", " if seed is not None:\n", " np.random.seed(seed)\n", "\n", " # generates the point\n", " point: Optional[Point] = None\n", " while (\n", " point is None\n", " or abs(point.x) < MIN_DESTINATION_DISTANCE\n", " or abs(point.y) < MIN_DESTINATION_DISTANCE\n", " or any(obstacle.contains(point) for obstacle in obstacles.values())\n", " ):\n", " point = Point(np.random.randint(-SIZE_INNER, SIZE_INNER, size=(2,), dtype=int))\n", " return point\n", "\n", "\n", "print(generate_destination(generate_obstacles(42), 42))" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "def plot_situation(\n", " obstacles: dict[str, Polygon],\n", " destination: Point,\n", " obstacle_color: str | None = \"RED\",\n", " route=None,\n", " legend: bool = True,\n", " title: str | None = None,\n", ") -> None:\n", " \"\"\"PLots the obstacles into a matplotlib plot.\n", "\n", " Args:\n", " obstacles: A list of obstacles.\n", " destination: The destination that should be reached by the boat.\n", " obstacle_color: The color the obstacles should have. Can be None.\n", " If none all obstacles will have different colors.\n", " route: The route that should be plotted.\n", " legend: If true plots a legend.\n", " title: The title of the plot.\n", " Returns:\n", " None\n", " \"\"\"\n", " # Create a plot in the defined size\n", " plt.axis([-SIZE_ROUTE, SIZE_ROUTE, -SIZE_ROUTE, SIZE_ROUTE])\n", "\n", " # Sets a title if one is demanded\n", " if title:\n", " plt.title(title)\n", "\n", " # Plots the obstacles.\n", " if obstacles:\n", " for polygon in obstacles.values():\n", " if obstacle_color is not None:\n", " plt.fill(*polygon.exterior.xy, color=obstacle_color, label=\"Obstacle\")\n", " else:\n", " plt.fill(*polygon.exterior.xy)\n", "\n", " # Plots the wind direction\n", " # The following code for an arrow was taken modeled after:\n", " # https://www.geeksforgeeks.org/matplotlib-pyplot-arrow-in-python/\n", " plt.arrow(\n", " 0,\n", " +int(SIZE_ROUTE * 0.9),\n", " 0,\n", " -int(SIZE_ROUTE * 0.1),\n", " head_width=10,\n", " width=4,\n", " label=\"Wind (3Bft)\",\n", " )\n", "\n", " if route is not None:\n", " if isinstance(route, np.ndarray):\n", " plt.plot(route[:, 0], route[:, 1], color=\"BLUE\", marker=\".\")\n", " else:\n", " if isinstance(route, TimingFrame):\n", " plt.plot(\n", " route.points[:, 0], route.points[:, 1], color=\"BLUE\", marker=\".\"\n", " )\n", " else:\n", " raise TypeError()\n", "\n", " # Plots the estimation\n", " if destination:\n", " plt.scatter(*destination.xy, marker=\"X\", color=\"green\", label=\"Destination\")\n", " plt.scatter(0, 0, marker=\"o\", color=\"green\", label=\"Start\")\n", "\n", " if legend:\n", " # https://stackoverflow.com/questions/13588920/stop-matplotlib-repeating-labels-in-legend\n", " handles, labels = plt.gca().get_legend_handles_labels()\n", " by_label = dict(zip(labels, handles))\n", " plt.legend(by_label.values(), by_label.keys())\n", " return None" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "Die aus den Seeds 0 - 11 generierten Karten werden unten angezeigt um Beispiele der von der KI zu lösenden Scenario zu zeigen.\n", "Wird dieses Notebook im Pyrate Docker Container ausgeführt werden auch die Routen eingezeichnet." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "if not NO_SHOW:\n", " # create a subplot with 12 routes.\n", " plt.figure(figsize=(17.5, 25))\n", " for seed in tqdm(range(12)):\n", " plt.subplot(4, 3, seed + 1)\n", " generated_obstacles = generate_obstacles(seed)\n", " generated_destination = generate_destination(generated_obstacles, seed)\n", " route_generated = None\n", "\n", " # try to generate a route\n", " try:\n", " route_generated, _ = experiments.generate_route(\n", " position=Point(0, 0),\n", " goal=generated_destination,\n", " obstacles=generated_obstacles,\n", " wind=(18, 180),\n", " )\n", " except Exception:\n", " route_generated = None\n", "\n", " # plot the situation\n", " plot_situation(\n", " obstacles=generated_obstacles,\n", " destination=generated_destination,\n", " obstacle_color=\"RED\",\n", " route=route_generated,\n", " title=f\"Seed: {seed}, Cost: {route_generated.cost:.3f}\"\n", " if route_generated\n", " else f\"Seed: {seed}\",\n", " legend=(seed == 0),\n", " )\n", " plt.show()" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "### Das Normieren der Scenarios\n", "\n", "Um für ein neuronales Netz verständlich zu sein, ist es immer einfacher, wenn Eingaben normiert sind. Hier wurde entschieden die Szenarios, als Bilddaten zu normieren. 128 x 128 Pixel sind wesentlich gleichförmiger als eine Mange von maximal 40 Polygonen mit unterschiedlichen Formen. Daher verwandelt die folgende Funktion die mit den oben definierten Funktionen genierten Scenarios Datensätze in eine Bildform. Rot ist dabei das Hindernis. Grün das Ziel und Blau die Route. Entweder als Linie oder als Punkt, wenn die Route sich ändert." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": 1, "outputs": [ { "ename": "NameError", "evalue": "name 'Polygon' is not defined", "output_type": "error", "traceback": [ "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m", "\u001b[1;31mNameError\u001b[0m Traceback (most recent call last)", "Input \u001b[1;32mIn [1]\u001b[0m, in \u001b[0;36m\u001b[1;34m()\u001b[0m\n\u001b[0;32m 1\u001b[0m \u001b[38;5;66;03m# Inspiration: https://programtalk.com/python-examples/PIL.ImageDraw.Draw.polygon/\u001b[39;00m\n\u001b[0;32m 2\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mgenerate_image_from_map\u001b[39m(\n\u001b[1;32m----> 3\u001b[0m obstacles: \u001b[38;5;28mdict\u001b[39m[\u001b[38;5;28mstr\u001b[39m, \u001b[43mPolygon\u001b[49m],\n\u001b[0;32m 4\u001b[0m destination: Point,\n\u001b[0;32m 5\u001b[0m route\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mNone\u001b[39;00m,\n\u001b[0;32m 6\u001b[0m route_type: Literal[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mline\u001b[39m\u001b[38;5;124m\"\u001b[39m, \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mdot\u001b[39m\u001b[38;5;124m\"\u001b[39m] \u001b[38;5;241m=\u001b[39m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mdot\u001b[39m\u001b[38;5;124m\"\u001b[39m,\n\u001b[0;32m 7\u001b[0m ) \u001b[38;5;241m-\u001b[39m\u001b[38;5;241m>\u001b[39m Image:\n\u001b[0;32m 8\u001b[0m \u001b[38;5;124;03m\"\"\"Generate an image from the map.\u001b[39;00m\n\u001b[0;32m 9\u001b[0m \n\u001b[0;32m 10\u001b[0m \u001b[38;5;124;03m Can be used to feed an ANN.\u001b[39;00m\n\u001b[1;32m (...)\u001b[0m\n\u001b[0;32m 20\u001b[0m \u001b[38;5;124;03m If 'dot' is selected the turning points a drawn in.\u001b[39;00m\n\u001b[0;32m 21\u001b[0m \u001b[38;5;124;03m \"\"\"\u001b[39;00m\n\u001b[0;32m 22\u001b[0m \u001b[38;5;66;03m# generate an empty image (All black)\u001b[39;00m\n", "\u001b[1;31mNameError\u001b[0m: name 'Polygon' is not defined" ] } ], "source": [ "# Inspiration: https://programtalk.com/python-examples/PIL.ImageDraw.Draw.polygon/\n", "def generate_image_from_map(\n", " obstacles: dict[str, Polygon],\n", " destination: Point,\n", " route=None,\n", " route_type: Literal[\"line\", \"dot\"] = \"dot\",\n", ") -> Image:\n", " \"\"\"Generate an image from the map.\n", "\n", " Can be used to feed an ANN.\n", " - Obstacles are marked as reed.\n", " - The destination is marked as green.\n", " - The points where the route will likely change are blue.\n", "\n", " Args:\n", " obstacles: A dict of obstacles as shapely Polygons. Keyed as a string.\n", " destination: A destination that should be navigated to.\n", " route: The calculated route that should be followed.\n", " route_type: How the route is drawn. If 'line' is selected the complete route is selected.\n", " If 'dot' is selected the turning points a drawn in.\n", " \"\"\"\n", " # generate an empty image (All black)\n", " img = Image.new(\n", " \"RGB\",\n", " (IMG_SIZE, IMG_SIZE),\n", " \"#000000\",\n", " )\n", " draw = ImageDraw.Draw(img)\n", "\n", " # draw in all obstacles in red\n", " for polygon in obstacles.values():\n", " draw.polygon(\n", " list(\n", " (np.dstack(polygon.exterior.xy).reshape((-1)) + SIZE_ROUTE)\n", " / (2 * SIZE_ROUTE)\n", " * IMG_SIZE\n", " ),\n", " fill=\"#FF0000\",\n", " outline=\"#FF0000\",\n", " )\n", "\n", " # draw in a route if possible. Does so in blue\n", " if os.getenv(\"PYRATE\"):\n", " if isinstance(route, TimingFrame):\n", " route = route.points\n", " if route is not None:\n", " route = ((route + SIZE_ROUTE) / (2 * SIZE_ROUTE) * IMG_SIZE).astype(int)\n", " # draws the route as collection of lines\n", " if route_type == \"line\":\n", " draw.line([tuple(point) for point in route], fill=(0, 0, 0xFF))\n", " # draw the route as a collection of points. The starting point is seen as redundant and left out.\n", " elif route_type == \"dot\":\n", " for point in route[1:]:\n", " img.putpixel(point, (0, 0, 0xFF))\n", " else:\n", " raise ValueError(\"Route type unknown.\")\n", " # draws in the destination in green\n", " img.putpixel(\n", " (\n", " int((destination.x + SIZE_ROUTE) / (2 * SIZE_ROUTE) * IMG_SIZE),\n", " int((destination.y + SIZE_ROUTE) / (2 * SIZE_ROUTE) * IMG_SIZE),\n", " ),\n", " (0, 0xFF, 0),\n", " )\n", " return img" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "def generate_example_image(route_type: Literal[\"line\", \"dot\"]):\n", " \"\"\"\n", " Generates an example image with the seed 42.\n", "\n", " Args:\n", " route_type: How the route is drawn. If 'line' is selected the complete route is selected.\n", " If 'dot' is selected the turning points a drawn in.\n", "\n", " Returns:\n", " The example image.\n", " \"\"\"\n", " # generate obstacles and a destination\n", " obstacles = generate_obstacles(42)\n", " destination = generate_destination(obstacles, 42)\n", " # try to generate a route\n", " try:\n", " route, _ = experiments.generate_route(\n", " position=Point(0, 0),\n", " goal=destination,\n", " obstacles=obstacles,\n", " wind=(18, 180),\n", " )\n", " except Exception:\n", " route = None\n", "\n", " # draw the scenario\n", " return generate_image_from_map(\n", " obstacles=obstacles,\n", " destination=destination,\n", " route=route,\n", " route_type=route_type,\n", " )" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "Nachfolgend werden zwei solcher Scenarios Bilder gezeigt. Zuerst aber wird zum Vergleich das Scenario mit dem Seed 42 als Karte dargestellt, um den Unterschied zu zeigen." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "if not NO_SHOW:\n", " # set the default seed of 42\n", " seed: int = 42\n", " # create a figure\n", " plt.figure(figsize=(8, 8))\n", " wind_dir = 180\n", " # generate obstacles and a destination\n", " generated_obstacles = generate_obstacles(seed)\n", " generated_destination = generate_destination(generated_obstacles, seed)\n", " route_generated = None\n", " # try generating a route\n", " try:\n", " route_generated, _ = experiments.generate_route(\n", " position=Point(0, 0),\n", " goal=generated_destination,\n", " obstacles=generated_obstacles,\n", " wind=(18, wind_dir),\n", " )\n", " except Exception as e:\n", " route_generated = None\n", " # plotting the situation\n", " plot_situation(\n", " obstacles=generated_obstacles,\n", " destination=generated_destination,\n", " obstacle_color=\"RED\",\n", " route=route_generated,\n", " title=f\"Seed: {seed}, Cost: {route_generated.cost:.3f}\"\n", " if route_generated\n", " else f\"Seed: {seed}\",\n", " legend=seed == 0,\n", " )\n", " plt.show()" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "Zeigt das Scenario mit dem Seed 42 mit eingezeichneten Wendepunkten, wenn dieses Notebook im Pyrate Docker Container ausgeführt wurde. Wichtig zu beachten ist in dieser Darstellung die Drehung des Vorzeichens der Y Achse was zu einer Horizontalen Spiegelung der Darstellung führt." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "generate_example_image(route_type=\"dot\").resize(\n", " (IMG_SHOW_SIZE, IMG_SHOW_SIZE), Image.Resampling.BICUBIC\n", ")" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "Zeigt das Scenario mit dem Seed 42 mit einer, durch Linien eingezeichneten Route, wenn dieses Notebook im Pyrate Docker Container ausgeführt wurde." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "generate_example_image(route_type=\"line\").resize(\n", " (IMG_SHOW_SIZE, IMG_SHOW_SIZE), Image.Resampling.BICUBIC\n", ")" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "### Massengenerierung von Szenarien\n", "\n", "Die oben definierten Funktionen generieren immer einen Datensatz.\n", "Die folgenden Funktionen definieren einen einzelnen Datensatz als `pd.Series` einer einzelnen Zeile in einem `pd.DataFrame`. Die so erzeugten Datensatze werden in `pd.DataFrames` zusammengefasst. Hier wurde eine Anzahl von 50 Datensätzen auf einmal gewählt. Diese werden dann gespeichert, um danach mehr Daten zu generieren. Da der Wegfindealgorithmus immer noch experimentell ist, werden für Routen, für die keine Lösung gefunden wurde oder bei deren finden ein Fehler auftritt wird dort `NaN` eingetragen." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "def generate_all_to_series(\n", " seed: Optional[int] = None, image: bool = False\n", ") -> pd.Series:\n", " \"\"\"Generates everything and aggregates all data into a `pd:Series`.\n", "\n", " Args:\n", " seed:The seed that should be used to generate map and destination.\n", " image: If an image should be generated or if that should be postponed to save memory.\n", " Returns:\n", " Contains a `pd.Series`containing the following.\n", " - The seed tha generated the map.\n", " - The destination in x\n", " - The destination in y\n", " - A list of Obstacle polygons.\n", " - The route generated for this map by the roBOOTer navigation system.\n", " - Optionally the image containing all the information.\n", " Can be generated at a later date without the fear for a loss of accuracy.\n", " \"\"\"\n", " # generate obstacles\n", " obstacles = generate_obstacles(seed)\n", " # find a destination\n", " destination = generate_destination(obstacles, seed)\n", "\n", " # find a possible route\n", " try:\n", " route, _ = experiments.generate_route(\n", " position=Point(0, 0),\n", " goal=destination,\n", " obstacles=obstacles,\n", " wind=(18, wind_dir),\n", " )\n", " except Exception:\n", " route = None\n", "\n", " # collect all generated data in a `pd.Series`\n", " return pd.Series(\n", " data={\n", " \"seed\": str(seed),\n", " \"obstacles\": obstacles,\n", " \"destination_x\": destination.x,\n", " \"destination_y\": destination.y,\n", " \"image\": generate_image_from_map(obstacles, destination, route)\n", " if image\n", " else pd.NA,\n", " \"route\": route.points if route else pd.NA,\n", " \"cost\": route.cost if route else pd.NA,\n", " },\n", " name=str(seed),\n", " )" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "Nachfolgend wird ein kurzes Beispiel eines solchen `pd.DataFrame` angezeigt." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "if not NO_SHOW:\n", " df = pd.DataFrame(\n", " [generate_all_to_series(i, image=False) for i in tqdm(range(12))]\n", " ).set_index(\"seed\")\n", " df.to_pickle(\"test.pickle\")\n", "if os.path.exists(\"test.pickle\"):\n", " df = pd.read_pickle(\"test.pickle\")\n", "else:\n", " df = None\n", " print(\"No data generated or cached!\")\n", "df" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "\n" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "markdown", "source": [ "Die folgende Zelle ist Verantwortlich für das massenweise Generieren von Trainingsdaten. Sie kann entweder so eingestellt werden, dass nur eine einzige Batch aus 50 neuen Datensätzen generiert werden soll oder eine ganze Reihe von Batches. Sind nicht alle Anforderungen zum Ausführen der Zelle erfüllt, wird sie automatisch übersprungen." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "# Skips the following cell if the code can't be executed.\n", "if os.getenv(\"PYRATE\"):\n", " save_frequency = int(os.getenv(\"save_frequency\", \"50\"))\n", " start_seed = int(os.getenv(\"seed_start\", \"0\"))\n", " continues = bool(os.getenv(\"continues\", \"false\"))\n", "\n", " # try finding a block of seeds that is not used\n", " files = glob.glob(\"data/*.pickle\")\n", " seed_groups = {int(file[9:-7]) for file in files}\n", " for next_seeds in range(start_seed, 1_000_000, save_frequency):\n", " # skip if the seed block already exists or is generated by another instance if this notebook\n", " if next_seeds in seed_groups:\n", " continue\n", "\n", " # start generating routes for the seed block\n", " print(f\"Start generating routes for seed: {next_seeds}\")\n", "\n", " # reserving the seed block by looking down the seed block with an empty file\n", " tmp_pickle_str: str = f\"data/tmp_{next_seeds:010}.pickle\"\n", " pd.DataFrame().to_pickle(tmp_pickle_str)\n", "\n", " # generate the data\n", " df = pd.DataFrame(\n", " [\n", " generate_all_to_series(i, image=False)\n", " for i in tqdm(range(next_seeds, next_seeds + save_frequency, 1))\n", " ]\n", " ).set_index(\"seed\")\n", "\n", " # saves the data and delete the temporary file\n", " pickle_to_file = f\"data/raw_{next_seeds:010}.pickle\"\n", " df.to_pickle(pickle_to_file)\n", " os.remove(tmp_pickle_str)\n", "\n", " # break the loop if only a single block of data should be generated.\n", " if not continues:\n", " break" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "### Daten zusammenfassen\n", "\n", "Nachdem man den generierenden Teil des Codes für eine Weile hat laufen lassen, erhält man eine Vielzahl einzelner Dateien. Diese werden nachfolgend zusammengefasst. Diese so zusammengefasste Tabelle wird nachfolgend von Ausreißern und Fehlern bereinigt.\n", "Direkt nach dem Zusammenfassen der Daten werden alle Einträge, für die keine Routen gefunden wurde, weggelassen.\n", "\n", "Dies kann folgende Gründe haben:\n", "* Startpunkt $P(0, 0)$ ist von Hindernissen eingeschlossen\n", "* Der Zielpunkt ist von Hindernissen eingeschlossen\n", "* Fehler im Algorithmus der die Routen generiert" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "DATA_COLLECTION_PATH: Final[str] = \"data/collected.pickle\"\n", "\n", "# Load a cached result should it not be demanded to generate all data new.\n", "if os.path.exists(DATA_COLLECTION_PATH) and not GENERATE_NEW:\n", " collected_data = pd.read_pickle(DATA_COLLECTION_PATH)\n", "else:\n", " # Read the first n files\n", " # The number of files read can be defined with the constant: NUMBER_OF_FILES_LIMIT\n", " # The dataframes read are concatenate directly after\n", " collected_data = pd.concat(\n", " [\n", " pd.read_pickle(filename)\n", " for filename in tqdm(glob.glob(\"data/raw_*.pickle\")[:NUMBER_OF_FILES_LIMIT])\n", " ]\n", " )\n", "# Prints a short summary of the data.\n", "number_of_maps = len(collected_data.index)\n", "print(f\"{number_of_maps: 8} maps collected\")\n", "collected_data.dropna(subset=[\"route\"], inplace=True)\n", "number_of_routes = len(collected_data.index)\n", "print(f\"{number_of_routes: 8} routes collected\")\n", "collected_data.to_pickle(DATA_COLLECTION_PATH)\n", "collected_data" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "### Szenarios Filtern\n", "\n", "Die so erzeugten Daten sind ungefiltert. Sie müssen nun überprüft werden. Dazu wurden einige hundert Datensätze geplottet. Einige Muster sind dabei aufgefallen. Die nachfolgenden Filter resultieren aus diesen Mustern.\n", "\n", "#### Die Route verlässt die Karte\n", "\n", "Das Generieren von Heatmaps von Segelrouten erfordert, dass sich das mögliche Ergebnis sinnvoll darstellen lässt. Dazu muss die Route vollständig im definierten Bereich liegen. Alle Routen, die die Karte verlassen, werden weggelassen." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "def check_route_in_bounds(route):\n", " \"\"\"\n", " Check if a route exists and is in bounds.\n", "\n", " Args:\n", " route: An `np.ndarray` of points the builds the route.\n", "\n", " Returns:\n", " A non-existing route or a route that leaves the area routes should stick to return `False` otherwise, `True` is returned.\n", " \"\"\"\n", "\n", " # CHecks if the route exists\n", " if route is None:\n", " return False\n", " if route is pd.NA:\n", " return False\n", " # Checks if the route is of the right data type.\n", " if not isinstance(route, np.ndarray):\n", " return False\n", " # Checks if a position is out of bounds.\n", " if np.array(\n", " abs(route) > SIZE_ROUTE,\n", " ).any():\n", " return False\n", " return True\n", "\n", "\n", "# Count the number of data points there are before this filter is used.\n", "data_before = len(collected_data.index)\n", "\n", "# Filtering\n", "df_filter = collected_data[\"route\"].apply(check_route_in_bounds)\n", "filtered = collected_data[~df_filter]\n", "collected_data = collected_data[df_filter]\n", "\n", "# Count the number of data points there are after this filter is used.\n", "data_after = len(collected_data.index)\n", "\n", "# Print a short report over the changes to the dataset.\n", "print(\n", " f\"{data_before} - {data_before-data_after} = {data_after} sets of data remaining.\"\n", ")\n", "\n", "# delete variables that where only used inside this cell\n", "del data_before, data_after, filtered, df_filter" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "#### Routen auf Fehler überprüfen\n", "\n", "Ein Bug in der Routenfindung hat zu Selbstschneidung der Routen geführt dieser wurde beim Sailing Team Darmstadt e.V. behoben. In den ersten ca. 27000 datensätzen gibt es dennoch Selbstschneidungen der Routen. Routen mit Selbstschneidung werden hier erkannt und da weder representativ noch richtig aus diesem Datensatz herausgenommen." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "def check_route_self_crossing(route):\n", " \"\"\"\n", " Check if a route has self intersections.\n", "\n", " Args:\n", " route: An `np.ndarray` of points the builds the route.\n", "\n", " Returns:\n", " `True` if the route is self intersecting.\n", " \"\"\"\n", " if isinstance(route, float):\n", " print(float)\n", " return not LineString(route).is_simple\n", "\n", "\n", "# count the number of data points before this filter was applied.\n", "data_before = len(collected_data.index)\n", "\n", "# filter the data\n", "collected_data = collected_data[\n", " ~collected_data[\"route\"].apply(check_route_self_crossing)\n", "]\n", "\n", "# count the number of data points after this filter was applied.\n", "data_after = len(collected_data.index)\n", "\n", "# print a short report over the changes to the dataset.\n", "print(\n", " f\"{data_before} - {data_before-data_after} = {data_after} sets of data remaining.\"\n", ")\n", "\n", "# delete variables that where only used inside this cell\n", "del data_before, data_after" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "#### Filtern der Routen nach Kosten\n", "\n", "Einige der Routen haben trotz einer erfolgreichen Routenfindung enorm hohe Kosten. Kosten werden beim Generieren der Route mitberechnet und sind das, was bei dem Routen generierenden Gradientenabstiegsverfahren optimiert wird. Sie setzen sich zusammen aus Segelzeit und Risiken. Außerordentlich hohe Kosten legen daher entwendet nahe, dass keine gute Route gefunden werden konnte oder dass die gefundene Route zu einem schlechten lokalen Minimum konvergiert hat. Daher werden die teuersten $5\\%$ der Routen weggelassen.\n", "\n", "Die folgende Route berechnet das $95\\%$ Quantil und errechnet wie viele Einträge über $95\\%$ liegen." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "QUANTILE_LIMIT: Final[float] = 0.95\n", "if \"DATA_UPPER_LIMIT_QUANTIL\" not in locals():\n", " DATA_UPPER_LIMIT_QUANTIL: Final[float] = collected_data[\"cost\"].quantile(\n", " QUANTILE_LIMIT\n", " )\n", " OVER_QUANTILE: Final[int] = int(len(collected_data.index) * (1 - QUANTILE_LIMIT))\n", "# noinspection PyUnboundLocalVariable\n", "print(\n", " f\"{OVER_QUANTILE} entries over the {QUANTILE_LIMIT} quantile at {DATA_UPPER_LIMIT_QUANTIL:.3f}\"\n", ")" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "Der folgende Codeschnipsel berechnet das Histogramm der Kosten. Wie wenig repräsentativ die höchsten $5\\%$ der Kosten sind, ist direkt ersichtlich." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "collected_data[\"cost\"].plot.hist(bins=15, log=True)\n", "plt.axvline(x=DATA_UPPER_LIMIT_QUANTIL, color=\"RED\", label=\"95% Quantil\")\n", "plt.legend()\n", "plt.show()" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "Nachfolgend werden einige der Routen mit sehr hohen Kosten gezeigt. Die meisten kommen dem Land sehr nahe oder segeln sehr stark gegen den Wind." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "plt.figure(figsize=(15, 25))\n", "for count, (seed, row) in tqdm(\n", " enumerate(\n", " collected_data[collected_data[\"cost\"] > DATA_UPPER_LIMIT_QUANTIL]\n", " .sort_values(\"cost\")\n", " .iloc[0 :: int(OVER_QUANTILE / 12)]\n", " .iloc[:12]\n", " .iterrows()\n", " ),\n", " total=12,\n", "):\n", " plt.subplot(5, 3, count + 1)\n", " plot_situation(\n", " destination=Point(row.destination_x, row.destination_y),\n", " obstacles=row.obstacles,\n", " obstacle_color=\"RED\",\n", " route=row.route,\n", " title=f\"Cost: {row.cost}\",\n", " )\n", "plt.show()" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "Die Daten werden nun beim $95\\%$ Quantil der Kosten gefiltert." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "collected_data = collected_data.loc[collected_data[\"cost\"] < DATA_UPPER_LIMIT_QUANTIL]\n", "collected_data" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "Ein neues Histogramm der Kostenfunktion wird geplottet." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "collected_data[\"cost\"].plot.hist(log=True)\n", "plt.title(\"Route costs cut at the 95% quantile\")\n", "plt.show()" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "#### Filtern der Routen nach Komplexität" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "markdown", "source": [ "Gegenüber den Routen mit zu hohen Kosten stehen die Routen mit zu geringen Kosten. Daher werden als nächstes Routen mit zu niedrigen Kosten betrachtet.\n", "Nachfolgend ist eine Auswahl solcher günstiger Routen angezeigt. Es fällt auf das, all diese Routen direkt sind.\n", "Eine Betrachtung der Verteilung der Routenpunkte ist daher notwendig." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "plt.figure(figsize=(17.5, 25))\n", "for count, (seed, row) in enumerate(\n", " collected_data[collected_data[\"cost\"] < DATA_UPPER_LIMIT_QUANTIL]\n", " .sort_values(\"cost\")\n", " .iloc[1:600:51]\n", " .iterrows()\n", "):\n", " plt.subplot(4, 3, count + 1)\n", " plot_situation(\n", " destination=Point(row.destination_x, row.destination_y),\n", " obstacles=row.obstacles,\n", " obstacle_color=\"RED\",\n", " route=row.route,\n", " title=f\"Cost: {row.cost:.3f}\",\n", " legend=count == 0,\n", " )\n", "plt.show()\n", "del seed" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "def get_route_points(data):\n", " \"\"\"\n", " Counts how many stops are made inbetween.\n", "\n", " Args:\n", " data: a `pd.DataFrame` collecting all the data.\n", " Returns:\n", "\n", " \"\"\"\n", " complexity = data[\"route\"].apply(lambda r: r.shape[0] - 2)\n", " complexity.name = \"route complexity\"\n", " return complexity\n", "\n", "\n", "route_points = get_route_points(collected_data)\n", "route_points.plot.hist()\n", "plt.title(\"Route complexity in intermediate points\")\n", "plt.show()" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "Bei der oben angezeigten Komplexität wird deutlich, das diese teilweise etwas hoch ist. Hier wird ein Limit von 15 Kursänderungen eingeführt." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "routes_before = len(collected_data.index)\n", "collected_data = collected_data[route_points <= 15]\n", "routes_after = len(collected_data.index)\n", "print(\n", " f\"{routes_before} - {routes_before - routes_after} = {routes_after} \"\n", " f\"if only routes with less then 15 course changes remain.\"\n", ")" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "get_route_points(collected_data).plot.hist(bins=15)\n", "plt.title(\"Route complexity in intermediate points\")\n", "plt.show()" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "markdown", "source": [ "Die nun reduzierte Anzahl der Routen enthält eine zwar representative Menge an sehr einfachen Routen. Da das Ergebnis dieser Routen aber eine leere Heat Map für Kursänderungen ist, muss hier deutlich reduziert werden sodas sie nur einen angegebenen Anteil am Gesamtvolumen ausmachen. Dieser Anteil wurde hier auf $5\\%$ gesetzt." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "# Define the upper limit of the percentage easy routes should reach\n", "LIMIT_SIMPLE_CASES = 0.05\n", "values = get_route_points(collected_data).value_counts().sort_index()\n", "chance_limit = (\n", " (len(collected_data.index) * LIMIT_SIMPLE_CASES * (1 - LIMIT_SIMPLE_CASES))\n", " / values.get(0, 1)\n", " if 0 in values.index\n", " else 0\n", ")\n", "print(\n", " f\"Limiting simple cases to {LIMIT_SIMPLE_CASES * 100:.1f}% of the total routes. Reducing simple routes to {(chance_limit * 100):.1f}% of their volume.\"\n", ")" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "Der folgende Abschnitt setzt das oben aufgestellte Limit um." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "np.random.seed = 0\n", "collected_data = collected_data[\n", " (\n", " (get_route_points(collected_data) > 1)\n", " | (np.random.random(len(collected_data.index)) < chance_limit)\n", " )\n", "]\n", "del chance_limit" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "Die so veränderte Distribution der Routenkomplexität sieht dann so aus." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "get_route_points(collected_data).plot.hist(bins=15)\n", "plt.title(\"Complexity Distribution after an enforced limit to trivial solutions.\")\n", "plt.show()" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "Das `pd.DataFrame` welches die gefilterten Daten sammelt, sieht dann wie folgt aus:" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "collected_data" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "### Das Konvertieren in trainierbare Daten\n", "\n", "Die bisher erstellten und gefilterten Daten müssen nun mit der oben definierten Methode zum Generieren von Bildern `generate_image_from_map` transformiert werden. Die so transformierten Daten werden dann zusammengefasst und in ein `np.ndarray` konvertiert werden geeignet ist mehrdimensionale Arrays effizient zu speichern und zu bearbeiten. Durch das Speichern der Pixel als `uint8` war dies relative Speichereffizient." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "def generate_image_maps(row, route_type: Literal[\"dot\", \"line\"]):\n", " \"\"\"Generates the image version of the route.\n", "\n", " Adds another dimension to prepare vor concatenation in a later step.\n", " Divides by 0xFF to contain only 0 and 1 and values.\n", " Color channel zero contains obstacles.\n", " Color channel one contains the destination.\n", " Color channel two contains the route either as course change points or as continues lines.\n", "\n", " Args:\n", " row: The row of the pd.DataFrame that should be used to generate an image.\n", " route_type: Defines if the route should be drawn as a collection of course change points or continues lines.\n", " Returns:\n", " The image modified for concatenation and scaled to be easily used for pandas.\n", " Cast as uint8 for a minimal memory consumption.\n", " \"\"\"\n", " # expands the dimension by one\n", " img = np.expand_dims(\n", " # converts the image into a numpy array\n", " np.asarray(\n", " # generate the situation image form a map\n", " generate_image_from_map(\n", " obstacles=row.obstacles,\n", " destination=Point(row.destination_x, row.destination_y),\n", " route=row.route,\n", " route_type=route_type,\n", " )\n", " ),\n", " axis=0,\n", " )\n", " # integer divide to ensure all values are between 0 and 1\n", " img = img // 0xFF\n", " return img" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "# drop the image column to save some space in the dataset\n", "if \"image\" in collected_data.columns:\n", " del collected_data[\"image\"]" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "# save the collected and filtered data into a pickle file to load again later and flush the ram a bit.\n", "DATA_WITH_IMG_PATH: Final[str] = \"data/collected_and_filtered.pickle\"\n", "collected_data.to_pickle(DATA_WITH_IMG_PATH)" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "# collecting map images with routes as lines.\n", "collected_routes = np.concatenate(\n", " collected_data.progress_apply(generate_image_maps, axis=1, args=(\"line\",))\n", ")" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "Die gesammelten Daten sind relativ groß die nachfolgende Operation zeigt an wie viel RAM dafür gerade belegt ist." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "humanize.naturalsize(sys.getsizeof(collected_routes))" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "Stellt sicher, dass `uint8` als Datentyp verwendet der ein minimum an Speicher verwendet." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "assert (\n", " str(collected_routes.dtype) == \"uint8\"\n", "), \"Dtype needs to be unit8 to fit in the ram.\"" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "COLLECTED_ROUTES_LINE_DUMP = \"data/collected_routes_np_line.pickle\"\n", "with open(COLLECTED_ROUTES_LINE_DUMP, \"wb\") as f:\n", " pickle.dump(collected_routes, f)" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "FGeneriert die Daten für das Line format." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "# collecting map images with routes as dots.\n", "collected_routes_dots = np.concatenate(\n", " collected_data.progress_apply(generate_image_maps, axis=1, args=(\"dot\",))\n", ")\n", "assert (\n", " str(collected_routes_dots.dtype) == \"uint8\"\n", "), \"Dtype needs to be unit8 to fit in the ram.\"\n", "COLLECTED_ROUTES_DOT_DUMP = \"data/collected_routes_np_dot.pickle\"\n", "with open(COLLECTED_ROUTES_DOT_DUMP, \"wb\") as f:\n", " pickle.dump(collected_routes_dots, f)\n", "\n", "# deletes the collection object\n", "del collected_routes_dots" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "## Das Model\n", "\n", "\n", "Jedes neuronale Netz, hat eine bestimmte *Form*, diese wird im Modell definiert. Die Form bestimmt die Anzahl der versteckten Schichten, ob beim Training spezielle Filter wie `Dropout` oder `BatchNormalization` verwendet werden sollen. Auch die Art der Aktivierungsfunktionen wird hier definiert.\n", "\n", "Oft gibt es für bestimmte Probleme schon die eine oder andere Art von Netzwerkstruktur/Modellstruktur, die sich als geeignet erwiesen hat. Für diese Art von Problem, bei dem aus einem Bild eine anderes generiert werden soll, haben sich GAN Netzwerke als besonders Schlagfertig erwiesen [2], [3], [4].\n", "\n", "GAN Netzwerke bestehen immer aus zwei Komponenten, einem `Generator` und einem `Discriminator`. Der `Generator` generiert ein Bild aus einem anderen Bild. Der `Discriminator` versucht die Bilder zu unterscheiden und weist so den `Gernerator` auf Fehler hin, die durch eine normale Lernfunktion mit einer $l_1$ oder $l_2$ Norm nicht hervorgehoben werden, weil es für diese oft reicht, wenn das Ziel ungefähr erreicht wird. Der `Discriminator` sorgt also für klare Kontraste und saubere Farbverläufe.\n", "\n", "Das hier betrachtete Problem erwartet nun eine Heat Map. Da nicht davon auszugehen ist, dass die perfekte Route direkt gefunden wurde, ist ein etwas verwaschenes Ergebnis eine funktion nicht unbedingt ein Problem. Daher wird hier versucht den Routenschätzer ohne `Discriminator` aufzubauen." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "markdown", "source": [ "### Der Generator\n", "\n", "Der `Generator` ist eine Art von Autoencoder. Er nimmt das Bild, abstrahiert es in eine Sammlung von Features und generiert aus diesem Abstraction Format wieder ein Bild.\n", "Der `Generator` besteht daher aus einer Reihe von *Downsamplern* gefolgt von ebenso vielen *Upsamplern*. Der Downsampler fasst alle zwei Pixel, 3 Pixel über ein `tf.keras.layers.Conv2D` Schicht zusammen. Symmetrisch dazu macht der Upsampler dies wieder rückgängig. So wird die Bildgröße mit jedem Downsampler Schicht halbiert und mit jedem Upsampler Schicht wieder verdoppelt. Dafür erhält jeder Downsampler mehr Features während der Upsampler Features reduziert. Up und Downsmapler sind meist Symmetrisch aufgebaut.\n", "Wie in, sowohl dem TensorFlow Tutorial[4], als auch in dem Praxiseinstieg Maschine Learning in der Sektion über GANs[2] zu lesen ist benötigt der Downsampler eine `BatchNormaliszation` leaky Relu als Aktivierungsfunktion im Downsampler. Relu wird für den Upsampler verwendet.\n", "* Relu `y = x if x >= 0 else 0`\n", "* LeakyRelu `y = x if x >= 0 else b * x` wobei $x$ eine Zahl viel kleiner als 1 ist.\n", "\n", "BatchNormalization normalisiert die Ausgabewerte einer Schicht über eine Training Batch, indem der Durchschnitt jeder Ausgangsschicht auf 0 geschoben wird und auf die Varianz 1 skaliert wird[5]. Beim Ausführen des Models wird die in der letzten Epoche festgelegte Gesamtbeschreibung und Skalierung genutzt. Dies sorgt zusammen mit dem DropOut Filter im Upsampler für ein konsistentes Lernen und verhindert das Overfitting.\n", "Interessanterweise erhält jede Upsampling Schicht sowohl das Ergebnis, der vorherigen Schicht als auch das Ergebnis der symmetrischen Downsampling Schicht als Input." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "# Source: https://www.tensorflow.org/tutorials/generative/pix2pix\n", "def downsample(filters, size, apply_batchnorm=True):\n", " \"\"\"Create a downsample layer.\n", "\n", " A downsample layer contains:\n", " * tf.keras.layers.Conv2D\n", " * An activation Function\n", " * Optional a batchnorm\n", " * A activation function (LeakyRelu)\n", " Args:\n", " filters: The number of features that should be gernated.\n", " size: The number of features / pixels should be reduced.\n", " apply_batchnorm: If True the Batchnorm is applied. Batch norms are used by default.\n", " Returns:\n", " A sequential model contain the keras generated layers.\n", " \"\"\"\n", "\n", " initializer = tf.random_normal_initializer(mean=0.0, stddev=0.02)\n", "\n", " result = tf.keras.Sequential()\n", " result.add(\n", " tf.keras.layers.Conv2D(\n", " filters,\n", " size,\n", " strides=2,\n", " padding=\"same\",\n", " kernel_initializer=initializer,\n", " use_bias=False,\n", " )\n", " )\n", "\n", " if apply_batchnorm:\n", " result.add(tf.keras.layers.BatchNormalization())\n", "\n", " result.add(tf.keras.layers.LeakyReLU())\n", "\n", " return result" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "# Source: https://www.tensorflow.org/tutorials/generative/pix2pix\n", "def upsample(filters, size, apply_dropout=False):\n", " \"\"\"Create a upsample layer.\n", "\n", " A downsample layer contains:\n", " * tf.keras.layers.Conv2D\n", " * An activation Function\n", " * Optional a batchnorm\n", " * A activation function (Relu)\n", " Args:\n", " filters: The number of features that should be used to upsample the layer.\n", " size: The number of\n", " apply_dropout: If True a dropout layer ist used.\n", " Returns:\n", " A sequential model contain the keras generated layers.\n", " \"\"\"\n", " initializer = tf.random_normal_initializer(0.0, 0.02)\n", "\n", " result = tf.keras.Sequential()\n", " result.add(\n", " tf.keras.layers.Conv2DTranspose(\n", " filters,\n", " size,\n", " strides=2,\n", " padding=\"same\",\n", " kernel_initializer=initializer,\n", " use_bias=False,\n", " )\n", " )\n", "\n", " result.add(tf.keras.layers.BatchNormalization())\n", "\n", " if apply_dropout:\n", " result.add(tf.keras.layers.Dropout(0.5))\n", "\n", " result.add(tf.keras.layers.ReLU())\n", "\n", " return result" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "#### Model Erstellung\n", "Erstellt ein erstes Model des Generators wie oben beschrieben. Ein schematisches Layout findet sich darunter." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "def model_generator() -> tf.keras.Model:\n", " \"\"\"Creates an initial sequential model.\n", "\n", " Returns:\n", " A Sequential model.\n", " \"\"\"\n", "\n", " inputs = tf.keras.layers.Input(shape=[IMG_SIZE, IMG_SIZE, 2])\n", "\n", " down_stack = [\n", " downsample(64, 4, apply_batchnorm=False), # (batch_size, 64, 64, 128)\n", " downsample(128, 4), # (batch_size, 8, 8, 512)\n", " downsample(128, 4), # (batch_size, 4, 4, 512)\n", " downsample(256, 4), # (batch_size, 2, 2, 512)\n", " downsample(256, 4), # (batch_size, 1, 1, 512)\n", " downsample(512, 4), # (batch_size, 1, 1, 512)\n", " downsample(512, 4), # (batch_size, 1, 1, 512)\n", " ]\n", "\n", " up_stack = [\n", " upsample(512, 4, apply_dropout=True), # (batch_size, 4, 4, 1024)\n", " upsample(256, 4, apply_dropout=True), # (batch_size, 4, 4, 1024)\n", " upsample(256, 4, apply_dropout=True), # (batch_size, 4, 4, 1024)\n", " upsample(128, 4), # (batch_size, 16, 16, 1024)\n", " upsample(128, 4), # (batch_size, 32, 32, 512)\n", " upsample(64, 4), # (batch_size, 64, 64, 256)\n", " ]\n", "\n", " initializer = tf.random_normal_initializer(0.0, 0.02)\n", " last = tf.keras.layers.Conv2DTranspose(\n", " 1,\n", " 4,\n", " strides=2,\n", " padding=\"same\",\n", " kernel_initializer=initializer,\n", " activation=\"sigmoid\", # was tanh\n", " ) # (batch_size, 256, 256, 3)\n", "\n", " x = inputs\n", "\n", " # Down sampling through the model\n", " skips = []\n", " for down in down_stack:\n", " x = down(x)\n", " skips.append(x)\n", "\n", " skips = reversed(skips[:-1])\n", "\n", " # Up sampling and establishing the skip connections\n", " for up, skip in zip(up_stack, skips):\n", " x = up(x)\n", " x = tf.keras.layers.Concatenate()([x, skip])\n", "\n", " x = last(x)\n", "\n", " # drop the chanel dimension\n", " reshaped = tf.keras.layers.Reshape((128, 128))(x)\n", "\n", " return tf.keras.Model(inputs=inputs, outputs=reshaped)\n", "\n", "\n", "tf.keras.utils.plot_model(model_generator(), show_shapes=True, dpi=64)" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "Um mehr Kontrolle über den Lernprozess zu haben werden Callbacks verwendet. Der EarlyStopping Callback verhindert vor allem das Verschwenden von Rechenzeit, indem er den Lernvorgang abbricht, wenn eine definierte Anzahl von Epochen, keine Verbesserung gefunden wurde. Verschlechtert sich die beobachtete Metric wird der Lernvorgang abgebrochen. Das beste Ergebnis wird wieder hergestellt was Overfitting verhindert. Der Callback betrachtet normalerweise den Validaion Loss Parameter." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "early_stop = tf.keras.callbacks.EarlyStopping(\n", " monitor=\"mean_squared_error\",\n", " min_delta=0.00001,\n", " patience=10,\n", " verbose=0,\n", " mode=\"auto\",\n", " restore_best_weights=True,\n", ")" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "Nachfolgendes Diagramm zeigt die verschiedenen Farbkanäle der Eingabe und der Erwarteten ausgabe noch einmal separat und kombiniert." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "plt.figure(figsize=(17.5, 25))\n", "np_array = np.flip(collected_routes[1, :, :, :], axis=0)\n", "\n", "for chanel in tqdm(range(3)):\n", " plt.subplot(1, 4, chanel + 1)\n", " plt.title(f\"Chanel: {chanel}\")\n", " plt.imshow(np_array[:, :, chanel], interpolation=\"nearest\")\n", " plt.axis(\"off\")\n", "plt.subplot(1, 4, 4)\n", "plt.imshow(0x88 * np_array[:, :, 0] + 0xFF * np_array[:, :, 2], interpolation=\"nearest\")\n", "plt.show()" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "### Model Training\n", "\n", "Das oben erstellte Model muss nun trainiert werden. Leider war es mir nicht möglich mehr als ein Model auf einmal im Speicher zu haben. Auch das Löschen von Modellen hat nicht erlaubt mehr als zwei Modelle in diesem Notebook zu trainieren.\n", "\n", "Es wurde versucht, das Modell mit den Fehlerfunktionen `MSE`, `MAE`, `Binary Crossentropy` zu Trainieren. Die Ergebnisse waren sehr unterschiedlich.\n", "* `MAE` Der *M*ean *A*verage *E*rror wird durch den Durchschnitt des absoluten Fehlers berechnet. Das Ergebnis war in diesem Fall ein annähernd blankes Ergebnis. Das Training schlug fehl. Durch die geringe Priorisierung von Ausreißern bei dieser Fehlerfunktion war dies zu erwarten.\n", "* `MSE` Der *M*ean *S*quare *E*rror bildet den Durchschnitt über die Quadrate der Fehlerfunktion. Die dadurch verstärkt priorisierten Ausreißer erlaubten ein wesentlich besseres Lernverhalten.\n", "* `Binary Crossentropy` Die Binary Crossentropy ist eine Fehlerfunktion für Wahrscheinlichkeitsfunktionen. Da wir hier eine Wahrscheinlichkeitsfunktion als Bild darstellen ist es die theoretisch am besten geeignete Funktion. Dies wurde experimentell bestätigt.\n", "\n", "Das Training wurde mit $20\\%$ der Daten als Validierungsdaten durchgeführt.\n", "Experimentell hat sich gezeigt, dass die Ergebnisse schon mit der Standard Lernrate sehr gute Ergebnisse liefern. Diese werden mit einer geringeren Lernrate etwas besser auch, wenn der Lernaufwand so um ein Vielfaches größer ist.\n", "\n", "Leider war es mir nicht möglich mehr als die Routen von 500 pickel Detain auf einmal mit meiner GPU zu trainieren. Es ist zu vermuten, dass ein besseres Ergebnis mit mehr Datensätzen erreicht werden kann. Ohne GPU war der Rechenaufwand dieses Trainings aber nicht praktikabel.\n", "\n", "Als Optimierer wurde RMSprop verwendet." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "markdown", "source": [ "#### Training des Modells mit Routen als Linien\n", "\n", "Erstellt ein neues Model, Compiliert das Model und initialisiert die Schichten.\n", "Danach wird das Model direkt trainiert." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "# Delete the generator variable if already defined. Saves some gpu memory,\n", "if \"generator\" in globals():\n", " print(\"Generator exists. Deleting for reset\")\n", " del generator\n", "\n", "# creates the model\n", "generator = model_generator()\n", "\n", "# compiles the model\n", "generator.compile(\n", " optimizer=tf.keras.optimizers.RMSprop(learning_rate=0.00001), #\n", " loss=\"binary_crossentropy\",\n", " metrics=[\n", " \"binary_crossentropy\",\n", " \"mean_squared_error\",\n", " \"mean_absolute_error\",\n", " ],\n", ")\n", "\n", "# starts the training\n", "history = generator.fit(\n", " collected_routes[:, :, :, :2],\n", " collected_routes[:, :, :, 2],\n", " validation_split=0.2,\n", " epochs=100,\n", " use_multiprocessing=False,\n", " workers=1,\n", " callbacks=[early_stop, tf.keras.callbacks.TerminateOnNaN()],\n", ")" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "def plot_history(history_data, metrics: str | list) -> None:\n", " \"\"\"Plot some metrics from a training history.\n", "\n", " Args:\n", " history_data: The history data to plot.\n", " metrics: The metrics that should be plotted.\n", "\n", " Returns:\n", " None\n", " \"\"\"\n", " if isinstance(metrics, str):\n", " metrics = [metrics]\n", " for metric in metrics:\n", " plt.plot(history_data[metric], label=metric)\n", " plt.plot(history_data[f\"val_{metric}\"], label=f\"val_{metric}\")\n", " if len(metrics) == 1:\n", " plt.ylabel(\"loss\")\n", " plt.xlabel(\"epoch\")\n", " plt.legend(loc=\"best\")\n", "\n", "\n", "# Plots the training history in a single figure\n", "plt.figure(figsize=(20, 5))\n", "\n", "# plots the loss\n", "plt.subplot(1, 3, 1)\n", "plot_history(history.history, \"loss\")\n", "plt.title(\"model loss\")\n", "\n", "# plots the binary cross entropy <-> equal to loss\n", "plt.subplot(1, 3, 2)\n", "plot_history(history.history, \"binary_crossentropy\")\n", "plt.title(\"binary crossentropy\")\n", "\n", "# plots the other metrics\n", "plt.subplot(1, 3, 3)\n", "plot_history(history.history, [\"mean_absolute_error\", \"mean_squared_error\"])\n", "plt.title(\"other metrics\")\n", "\n", "plt.show()" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "#### Betrachtung des trainierten Models mit Routen als Linien\n", "\n", "Die besten Ergebnisse lagen bei einer `binary_crossentropy` von 0.0103 und einer `val_binary_crossentropy` von 0.0127 nach 51 Epochen.\n", "Bei dieser Art von Problem ist jedoch eine visuelle Überprüfung der Prediction notwendig." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "def plot_predicted_data(model: tf.keras.Model, data: np.ndarray, pos=0) -> None:\n", " \"\"\"Plots a single dataset in three versions.\n", "\n", " - The labeled situation.\n", " - The solution in context.\n", " - The prediction as a heat map.\n", "\n", " Flips everything for more consistency.\n", "\n", " Args:\n", " model: The trained model that should be visualised.\n", " data: The raw data that should be used as an example to visualise.\n", " pos: The position of the dataset that should be shown. If negativ test data is shown else training data is shown.\n", "\n", " Returns:\n", " None\n", " \"\"\"\n", " plt.figure(figsize=(15, 5))\n", " tt = \"test\" if pos < 0 else \"train\"\n", " plt.title(f\" for {tt} Nr: {abs(pos)}\")\n", " data = data[pos, :, :, :]\n", " predicted = model.predict(\n", " np.expand_dims(data[:, :, :2], 0),\n", " verbose=\"0\",\n", " steps=None,\n", " callbacks=None,\n", " max_queue_size=10,\n", " workers=1,\n", " use_multiprocessing=False,\n", " )[0]\n", "\n", " # flip for viewer consistency (Wind from north)\n", " data = np.flip(collected_routes[pos, :, :, :], axis=0)\n", " predicted = np.flip(predicted, axis=0)\n", " plt.axis(\"off\")\n", "\n", " # Plots the original labeled data\n", " plt.subplot(1, 3, 1)\n", " plt.title(\"Original map\")\n", " plt.imshow(\n", " data[:, :, 0] * 2 + data[:, :, 1] * 3 + data[:, :, 2], interpolation=\"nearest\"\n", " )\n", "\n", " # Plots the prediction in context\n", " plt.subplot(1, 3, 2)\n", " plt.title(\"Prediction in context\")\n", " plt.imshow(\n", " data[:, :, 0] * 2 + predicted[:, :] / predicted[:, :].max() * 3,\n", " interpolation=\"nearest\",\n", " )\n", "\n", " # Plots the predication as a pure heat map\n", " plt.subplot(1, 3, 3)\n", " plt.title(\"Predicted head map\")\n", " plt.imshow(predicted[:, :], interpolation=\"nearest\")\n", " plt.colorbar()\n", " plt.suptitle(\n", " f\"Plot of the {tt} scenario with the index {pos if pos >= 0 else -pos - 1}\"\n", " )\n", " plt.show()" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "Nachfolgend werden sowohl Vorhersagen mit Scenarios verglichen nebeneinander und gemeinsam dargestellt, um eine Abschätzung über die Vorhersagequalität zu machen.\n", "Erst wurden die Vorhersagen von Trainingsdaten visualisiert. Danach von Validierungsdaten." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "for i in range(10):\n", " plot_predicted_data(generator, collected_routes, i)" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "for i in range(10):\n", " plot_predicted_data(generator, collected_routes, -i - 1)" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "#### Training mit Routen als Liste von Wendepunkten\n", "\n", "Analog wurde das gleiche Model mit einer Liste an Wendepunkten trainiert. Hier wurde mit einem neuen Modell gearbeitet.\n", "Ein Weitertrainieren / Verändern des alten Models wäre vermutlich genauso gut möglich." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "del collected_routes\n", "collected_routes = np.load(COLLECTED_ROUTES_DOT_DUMP, allow_pickle=True)\n", "collected_routes.shape" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "#### Betrachtung des trainierten Models mit Routen als Liste von Wendepunkten\n", "\n", "Darstellung von Hindernissen, Ziel und Label in der Darstellung der Route als Liste von Wendepunkten." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "plt.figure(figsize=(17.5, 25))\n", "np_array = np.flip(collected_routes[1, :, :, :], axis=0)\n", "\n", "for chanel in tqdm(range(3)):\n", " plt.subplot(1, 4, chanel + 1)\n", " plt.imshow(np_array[:, :, chanel], interpolation=\"nearest\")\n", "plt.subplot(1, 4, 4)\n", "plt.imshow(0x88 * np_array[:, :, 0] + 0xFF * np_array[:, :, 2], interpolation=\"nearest\")\n", "plt.show()" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "# Delete the generator variable if allrady defined. Saves some gpu memory,\n", "if \"generator\" in globals():\n", " print(\"Generator exists. Deleting for reset.\")\n", " del generator\n", "\n", "# creates the model\n", "generator = model_generator()\n", "\n", "# compiles the model\n", "generator.compile(\n", " optimizer=tf.keras.optimizers.RMSprop(learning_rate=0.00001), #\n", " loss=\"binary_crossentropy\",\n", " metrics=[\n", " \"binary_crossentropy\",\n", " \"mean_squared_error\",\n", " \"mean_absolute_error\",\n", " ],\n", ")\n", "\n", "# starts the training\n", "history = generator.fit(\n", " collected_routes[:, :, :, :2],\n", " collected_routes[:, :, :, 2],\n", " validation_split=0.2,\n", " epochs=100,\n", " use_multiprocessing=False,\n", " workers=1,\n", " callbacks=[early_stop, tf.keras.callbacks.TerminateOnNaN()],\n", ")" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "plt.figure(figsize=(20, 5))\n", "plt.subplot(1, 3, 1)\n", "plot_history(\"loss\")\n", "plt.title(\"model loss\")\n", "plt.subplot(1, 3, 2)\n", "plot_history(\"binary_crossentropy\")\n", "plt.title(\"binary crossentropy\")\n", "plt.subplot(1, 3, 3)\n", "plot_history([\"mean_absolute_error\", \"mean_squared_error\", \"root_mean_squared_error\"])\n", "plt.title(\"other metrics\")\n", "plt.show()" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "Betrachtung des trainierten Models mit Routen als Liste von Kurswechselpositionen.\n", "Die besten Ergebnisse wurden nach 20 Epochen erreicht und lagen bei einer binary_crossentropy von 0.0023 und einer val_binary_crossentropy von 0.0035. Bei dieser Art von Problem ist jedoch eine Visuelle überprüfung der Prediction notwendig." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "for i in range(10):\n", " plot_predicted_data(generator, collected_routes, i)" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "for i in range(10):\n", " plot_predicted_data(generator, collected_routes, -i - 1)" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "markdown", "source": [ "### Vergleich der Experimente" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "markdown", "source": [ "Wie sehr schnell deutlich wird Skalieren die Ergebnisse der Liniendarstellung viel besser als die Ergebnisse der Punktdarstellung. Der Fehler ist aber bei der Punktdarstellung viel geringer. Dies ist auf die wesentlich mehr Werte die mit Routen markierungen versehen worden zurückzuführen. Welches der Beiden ergebnisse sich nachher besser für den Einsatz geeignet ist will ich nicht sagen. Beide Varianten müssten aber vermutlich neu Skaliert werden.\n", "Beide Modelle Zeigen aber sowohl bei Trainings als auch Validierungsdaten einen guten Instinkt für die Richtung der gefolgt werden muss." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "markdown", "source": [ "## Reflection und Ausblick\n", "\n", "Es konnte gezeigt werden das, das Schätzen einer Segelroute gut Funktionieren kann. Es fehlen sicherlich noch viele Faktoren wie lokale Winde, Strömungen und dynamische Hindernisse. Trotzdem bin ich zufrieden mit den Ergebnissen, die in diesem ersten Schritt erzielt werden konnten. Enttäuschend war wie sehr mich meine eigene GPU limitiert hat. Die Ergebnisse werden sicherlich mit mehr Daten etwas Robuster. Die limits an der Rechenleistung haben leider auch behindert, das ich mit Werkzeugen wie dem `KerasTuner` mit dem ich bisher sehr gute erfahrungen gemacht habe das neuronale Netz noch etwas optimiere und evtl. reduziere. Da das Tensorflow beispiel Netz, an dem sich hier Orientiert wurde, für ein 3 Kanaliges Bild mit höherer Auflösung ausgelegt war. Das Netz wurde darauf zwar angepasst aber nicht auf ein minimum reduziert. Auch war bedauerlicherweise keine Zeit um sich mit Datenaugmentierung zu beschäftigen. Sonst wäre sicher möglich gewesen einen Random Flip einzuführen der zufällig Label/Farbkanal 3 und Situation horizontal spiegelt. In Anbetracht dessen das ich aber mehr Daten besessen habe als genutzt wurden, hätte dies das Ergebnis aber nur unwesentlich verfeinert, auch wenn es zufällige Asymmetrien aus dem Netz genommen hätte. Diese konten zwar nicht beobachtet werden. Sind aber natürlich trotzdem möglich. Auch muss noch geprüft werden, ob der Kurs estimator auch auf einer echten Karte valide Ergebnisse liefert.\n", "Entweder durch mehr Daten oder durch ein Nachgelagertes verfeinern der Ergebnisse durch ein weiteres Netz oder den Discriminator des PIX2PIX Papers [3][4].\n", "\n", "Welches der beiden Modelle sich in der Praxis als Überlegen erweisen wird und ob die Netze einen Mehrwert bringen können, muss sich erst noch zeigen und kann an dieser Stelle von mir nicht beantwortet werden. Sicher ist das sie nicht nur prinzipiell, sondern auch mit dieser Implementierung Funktionieren.\n", "\n", "An dieser Stelle muss ich mich bei den anderen Mitgliedern des Sailing Team Darmstadt e.V. bedanken. Ohne die existierende experimentelle Wegfindung des `Pyrate` projekts wäre diese Arbeit nicht möglich gewesen." ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "markdown", "source": [ "## Literaturverzeichnis\n", "\n", "[1] Jang, Hoyun and Lee, Inwon and Seo, Hyoungseock: *Effectiveness of CFRP rudder aspect ratio for scale model catamaran racing yacht*, 2017\n", "\n", "[2] Aurélien Géron: *Praxiseinstig Machinen Learning mit Scikit-Learn, Keras und TensorFlow*, 2020, O.Reilly Verlag\n", "\n", "[3] Jun-Yan Zhu: *Image-to-Image Translation with Conditional Adversarial Networks*, 2018, Available: https://arxiv.org/abs/1611.07004\n", "\n", "[4] Tensorflow: *pix2pix: Image-to-image translation with a conditional GAN* Available: https://github.com/tensorflow/docs/blob/master/site/en/tutorials/generative/pix2pix.ipynb Commit: df4485e052523e0f852e83cea30ad319808bd97b\n", "\n", "[5] Keras: *Keras* Available: https://keras.io/" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%% md\n" } } }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "## Eigenständigkeitserklärung\n", "\n", "![](Eigenstaendigkeit.jpg)" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\n" } } }, { "cell_type": "code", "execution_count": 61, "metadata": { "pycharm": { "name": "#%%\n" }, "scrolled": false }, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "for i in range(10):\n", " plot_predicted_data(generator, collected_routes, -i - 1)" ] }, { "cell_type": "markdown", "metadata": { "pycharm": { "name": "#%% md\n" } }, "source": [ "### Vergleich der Experimente" ] }, { "cell_type": "markdown", "metadata": { "pycharm": { "name": "#%% md\n" } }, "source": [ "Wie sehr schnell deutlich wird skalieren, die Ergebnisse der Liniendarstellung viel besser als die Ergebnisse der Punktdarstellung. Der Fehler ist aber bei der Punktdarstellung viel geringer. Dies ist auf die wesentlich mehr Werte, die mit Linien markierten Routen zurückzuführen. Welches der beiden Ergebnisse nachher besser für den Einsatz geeignet ist kann ich nicht sagen. Beide Varianten müssten aber vermutlich neu skaliert werden.\n", "Beide Modelle zeigen aber sowohl bei Trainings als auch Validierungsdaten einen guten Instinkt für die Richtung der Route." ] }, { "cell_type": "markdown", "metadata": { "pycharm": { "name": "#%% md\n" } }, "source": [ "## Reflektion und Ausblick\n", "\n", "Es konnte gezeigt werden, dass das Schätzen einer Segelroute gut funktionieren kann. Es fehlen sicherlich noch viele Faktoren wie lokale Winde, Strömungen und dynamische Hindernisse. Trotzdem bin ich zufrieden mit den Ergebnissen, die in diesem ersten Schritt erzielt werden konnten. Enttäuschend war, wie sehr mich meine eigene GPU limitiert hat. Die Ergebnisse werden sicherlich mit mehr Daten etwas robuster. Die Limits an der Rechenleistung haben leider auch behindert, dass ich mit Werkzeugen wie dem `KerasTuner`, mit dem ich bisher sehr gute Erfahrungen gemacht habe das neuronale Netz noch etwas optimiere und evtl. reduziere, da das `Tensorflow` Beispiel Netz, an dem sich hier orientiert wurde, für ein 3 Kanaliges Bild mit höherer auflösung ausgelegt war. Das Netz wurde darauf zwar angepasst aber nicht auf ein Minimum reduziert. Auch war bedauerlicherweise keine Zeit um sich mit Datenaugmentierung zu beschäftigen. Sonst wäre sicher möglich gewesen einen Random Flip einzuführen der zufällig Label und Situation horizontal spiegelt. In Anbetracht dessen, dass ich aber mehr Daten zur verfügung hatte als genutzt wurden, hätte dies das Ergebnis aber nur unwesentlich verfeinert, auch wenn es zufällige Asymmetrien aus dem Netz genommen hätte. Diese konnten zwar nicht beobachtet werden. Sind aber trotzdem möglich und wahrscheinlich. Auch muss noch geprüft werden, ob die Routenschätzer auch auf einer echten Karte valide Ergebnisse liefern.\n", "Eine Nachbearbeitung des Entweder durch mehr Daten oder durch ein Nachgelagertes verfeinern der Ergebnisse durch ein weiteres Netz oder dass Trainieren mit dem Discriminator des PIX2PIX Papers [3][4], kann zu Verbesserungen führen.\n", "\n", "Welches der beiden Modelle sich in der Praxis als überlegen erweisen wird und ob die Netze einen Mehrwert bringen können, muss sich erst noch zeigen und kann an dieser Stelle von mir nicht beantwortet werden. Sicher ist dass, die Modelle nicht nur prinzipiell, sondern auch mit dieser Implementierung funktionieren. Die Eingeschränkte zufälligkeit des Zielpunkts bei der Datengenerierung schränkt die allgemeine Anwendbarkeit ein. Nicht aber den Beweis, dass ein solches Netzwerk funktionieren kann.\n", "\n", "An dieser Stelle muss ich mich bei den anderen Mitgliedern des Sailing Team Darmstadt e.V. bedanken. Ohne die existierende experimentelle Wegfindung des `Pyrate` Projekts wäre diese Arbeit nicht möglich gewesen." ] }, { "cell_type": "markdown", "metadata": { "pycharm": { "name": "#%% md\n" } }, "source": [ "## Literaturverzeichnis\n", "\n", "[1] Jang, Hoyun and Lee, Inwon and Seo, Hyoungseock: *Effectiveness of CFRP rudder aspect ratio for scale model catamaran racing yacht*, 2017\n", "\n", "[2] Aurélien Géron: *Praxiseinstig Machinen Learning mit Scikit-Learn, Keras und TensorFlow*, 2020, O.Reilly Verlag\n", "\n", "[3] Jun-Yan Zhu: *Image-to-Image Translation with Conditional Adversarial Networks*, 2018, Available: https://arxiv.org/abs/1611.07004\n", "\n", "[4] Tensorflow: *pix2pix: Image-to-image translation with a conditional GAN* Available: https://github.com/tensorflow/docs/blob/master/site/en/tutorials/generative/pix2pix.ipynb Commit: df4485e052523e0f852e83cea30ad319808bd97b\n", "\n", "[5] Keras: *Keras* Available: https://keras.io/" ] }, { "cell_type": "code", "execution_count": null, "outputs": [], "source": [ "## Eigenständigkeitserklärung\n", "\n", "![](Eigenstaendigkeit.jpg)" ], "metadata": { "collapsed": false, "pycharm": { "name": "#%%\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": "3.10.5" }, "toc": { "base_numbering": 1, "nav_menu": {}, "number_sections": true, "sideBar": true, "skip_h1_title": false, "title_cell": "Table of Contents", "title_sidebar": "Contents", "toc_cell": false, "toc_position": {}, "toc_section_display": true, "toc_window_display": false } }, "nbformat": 4, "nbformat_minor": 1 }