{ "cells": [ { "cell_type": "markdown", "id": "b426e0b3", "metadata": { "editable": true, "slideshow": { "slide_type": "slide" }, "tags": [ "remove-cell" ] }, "source": [ "
\n", "

Machine
Learning

\n", "

Joern Ploennigs

\n", "

Reinforcement Learning

" ] }, { "cell_type": "markdown", "id": "4ea7e1b3", "metadata": { "editable": true, "slideshow": { "slide_type": "skip" }, "tags": [] }, "source": [ "# Reinforcement Learning" ] }, { "cell_type": "markdown", "id": "0e0ebea4", "metadata": { "editable": true, "slideshow": { "slide_type": "skip" }, "tags": [] }, "source": [ "![](images/15_Reinforcement_Learning/mj_title_band.png)\n", "\n", "> The way positive reinforcement is carried out is more important than the amount.\n", "> \n", "> — Burrhus Frederic Skinner" ] }, { "cell_type": "markdown", "id": "95a25f29-aa53-4e1c-9386-0cba3d5edca0", "metadata": { "editable": true, "slideshow": { "slide_type": "skip" }, "tags": [], "vscode": { "languageId": "plaintext" } }, "source": [ "Reinforcement Learning (RL) ist ein Bereich des maschinellen Lernens, bei dem ein Agent lernt, in einer Umgebung durch Interaktionen zu handeln, um eine maximale Belohnung zu erzielen. Der Agent nimmt Aktionen basierend auf seinem aktuellen Zustand vor und erhält Belohnungen oder Bestrafungen, die er verwendet, um seine Strategie zu verbessern. \n", "\n", "Während Supervised Learning aus Beispielen lernt („Was ist das?“), lernt Reinforcement Learning durch Versuch und Irrtum („Was soll ich tun?“), indem es für gute Entscheidungen belohnt wird. Es benötigt keine gelabelten Daten, sondern eine Umgebung und Rückmeldung (Reward).\n", "\n", "RL eignet sich besonders für Aufgaben, bei denen ein Agent eigenständig durch Interaktion mit einer Umgebung lernen muss, optimale Entscheidungen zu treffen. Typische Einsatzgebiete sind dynamische Steuerungs- und Optimierungsprobleme, bei denen klassische regelbasierte Ansätze an ihre Grenzen stoßen. RL wird häufig verwendet, wenn die Umgebung komplex, unsicher oder sich verändernd ist und keine gelabelten Trainingsdaten vorliegen.\n", "\n", "Beispiele im Bauwesen sind: Ein Heiz-/Kühlsystem in Gebäuden lernt, abhängig von Wetter und Nutzung, effizient zu regeln. Beim 3D-Druck von Betonelementen kann RL den Materialeinsatz optimieren. Ein autonomer Bagger lernt, wie er Erdbewegungen möglichst ressourcenschonend und schnell ausführt." ] }, { "cell_type": "markdown", "id": "86c0d77e-80cc-4853-9e76-788dafb198b6", "metadata": { "editable": true, "slideshow": { "slide_type": "skip" }, "tags": [], "vscode": { "languageId": "plaintext" } }, "source": [ "Das macht auf der einen Seite RL vom Konzept her einfach zu verstehen und zu implementieren. Die Spezifikation der Belohnungsstrategie ist allerdings meist schwierig, da dies Verständnis des Problems erfordert. Der Ansatz benötigt keine gelabelte Trainingsdaten, ist also kein Überwachter ML-Ansatz. Er ist auch kein unüberwachter Ansatz, da ja durchaus Wissen in Form der Belohnungsstrategie notwendig ist." ] }, { "cell_type": "markdown", "id": "c4f7f9e4-2994-44aa-b7ea-2ac3f65ff0e8", "metadata": {}, "source": [ "
\n", "RL ist in der Praxis oft datenintensiv und empfindlich gegenüber Hyperparametern, Belohnungsdefinitionen und Zufall. Im Vergleich zu Supervised Learning benötigt es häufig deutlich mehr Interaktionen mit der Umgebung und das Training kann instabiler sein.\n", "
" ] }, { "cell_type": "markdown", "id": "fb624a37-c262-46a9-86d6-74bee41f6925", "metadata": { "editable": true, "slideshow": { "slide_type": "skip" }, "tags": [], "vscode": { "languageId": "plaintext" } }, "source": [ "RL eignet sich insbesondere für dynamische Entscheidungen und wird häufig in der Robotik verwendet, aber wird auch immer beliebter in traditionellen Gebieten." ] }, { "cell_type": "markdown", "id": "d1c65ecc", "metadata": { "editable": true, "slideshow": { "slide_type": "skip" }, "tags": [] }, "source": [ "## Folien\n", "" ] }, { "cell_type": "markdown", "id": "b49da939-24e5-4b16-acfc-660476aec1e4", "metadata": { "editable": true, "slideshow": { "slide_type": "slide" }, "tags": [] }, "source": [ "## Grundlegende Konzepte" ] }, { "cell_type": "markdown", "id": "04a474db-44f0-4618-8b08-27fd1ca5b2b1", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "Reinforcement Learning basiert auf einer iterativen Lernstrategie, bei der die Qualität der Lösung durch positives oder negatives Feedback bewertet wird. Das RL-Modell wird hierbei als Agent gesehen, der mit der Umgebung interagiert. Die Begriffe in diesem Kontext sind:\n", "\n", "- **Agent (Agent)**: Der Lernende oder Entscheidungsträger, der in der Umgebung agiert.\n", "- **Umgebung (Environment)**: Alles, mit dem der Agent interagiert.\n", "- **Aktion (Action)**: Eine Entscheidung oder Bewegung, die der Agent treffen kann.\n", "- **Belohnung (Reward)**: Rückmeldung aus der Umgebung, die angibt, wie gut eine Aktion im gegebenen Zustand war.\n", "- **Wertfunktion (Value Function)**: Eine Funktion, die angibt, wie gut ein bestimmter Zustand oder eine Aktion ist.\n", "- **Strategie (Policy)**: Eine Strategie, die der Agent verwendet, um Aktionen zu wählen basierend auf dem Zustand und der Wertfunktion.\n", "- **Zustand (State)**: Eine Repräsentation der aktuellen Situation der Umgebung und ggf. der Strategie." ] }, { "cell_type": "markdown", "id": "4ba94565-e4ce-42ed-bba4-9c97c0aaa8cb", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "source": [ "
\n", "\n", "\"Komponenten\n", "\n", "
" ] }, { "cell_type": "markdown", "id": "b2c3b12f-29fb-4d87-a984-735a636c832b", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "Zur Modellierung der Umgebung und aktuellen Strategie verwendet man im RL oft ein _Markov Decision Process (MDP)_. Das ist ein diskretes Zustandsmodell, bei dem das System sich immer nur in einem einzigen Zustand befinden kann, der die Umgebung widerspiegelt. Jeder Zustand beschreibt eine bestimmte Bedingung oder Position im Verhalten des Systems. Auf Basis des aktuellen Zustands und der gewählten Aktion erzeugt die Umgebung den nächsten Zustand sowie eine Belohnung." ] }, { "cell_type": "markdown", "id": "0dcff468-c7c1-4835-b345-e06ff350b043", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "source": [ "_Markov-Modelle_ sind eine sehr beliebte Modelltyp im Maschinellem Lernen. Sie basieren alle auf der _Markov-Bedingung_, dass die Übergangswahrscheinlichkeit von einem Zustand in den anderen nur von dem aktuellen Zustand abhängt und nicht vorhergehenden Zuständen (es gibt also keine Autokorrelation). Man spricht dabei auch von der _Gedächtnislosigkeit_. Dies basiert auf der Idee, dass der Zustand des Systems zu einem bestimmten Zeitpunkt alle Informationen enthält, die notwendig sind, um sein zukünftiges Verhalten vorherzusagen." ] }, { "cell_type": "markdown", "id": "dfd65fc5-6a13-4cd1-9fb0-3ac70b29c117", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "Ein MDP besteht aus:\n", "- $ S $: Menge aller möglichen Zustände.\n", "- $ A $: Menge aller möglichen Aktionen.\n", "- $ P(s'|s, a) $: Übergangswahrscheinlichkeit vom Zustand $ s $ zum Zustand $ s' $ bei Aktion $ a $.\n", "- $ R(s, a) $: Belohnungsfunktion, die die Belohnung angibt, die der Agent erhält, wenn er im Zustand $ s $ die Aktion $ a $ ausführt.\n", "- $ \\gamma $: Diskontierungsfaktor, der zukünftige Belohnungen abwertet." ] }, { "cell_type": "markdown", "id": "95303a93-6c50-4f60-bd53-667d93f6aad0", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "source": [ "
\n", "\n", "\"Beispiel\n", "\n", "
" ] }, { "cell_type": "markdown", "id": "095bb240-440a-42c6-b796-d229840131f5", "metadata": { "editable": true, "slideshow": { "slide_type": "skip" }, "tags": [] }, "source": [ "Der MDP modelliert also die Zustände und die bisherige Übergangswahrscheinlichkeit sowie die erhaltenen Belohnungen. Die entscheidende Frage im Reinforcement Learning ist nun, wie man daraus die beste Aktion und somit den nächsten Zustand ableiten kann. Hierbei gibt es das Dilemma, das aufgrund der Markov-Bedingung das Modell zwar einfach ist, aber wir auch gedächtnislos sind, wir also nicht die Historie der Zustände mit in unserer Entscheidung betrachten können. Das ist problematisch, wenn das Ziel nur erreicht werden kann, wenn eine bestimmte Zustandsfolge eintritt, wie in dem Gridworld-Beispiel unten diskutiert." ] }, { "cell_type": "markdown", "id": "960b36b3-4c0d-4af1-b0bb-297f64f5b202", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "Eine wichtige Gleichung ist hierfür die _Bellman-Gleichung_. Sie beschreibt die Beziehung zwischen dem aktuellen Zustands-Aktionspaar $(s,a)$, der beobachteten Belohnung und den möglichen Nachfolge-Zustands-Aktionspaaren $s',a'$. Diese Beziehung wird verwendet, um die optimale Wertfunktion zu finden." ] }, { "cell_type": "markdown", "id": "46f0bdd5-c98b-4c69-bb70-7fef5edca1f7", "metadata": { "editable": true, "slideshow": { "slide_type": "skip" }, "tags": [] }, "source": [ "Die Bellman-Gleichung basiert auf dem Prinzip der Optimalität, das besagt, dass ein optimaler Policy eine rekursive Struktur aufweist. Das bedeutet, dass der Wert eines Zustands unter einer optimalen Policy aus der sofortigen Belohnung und dem diskontierten Wert der zukünftigen Zustände besteht. Die Bellman-Gleichung für die _Wertfunktion_ $V(s)$ eines Zustands $s$ definiert den Wert eines Zustands als die erwartete Belohnung für die beste Aktion in diesem Zustand plus den diskontierten Wert des besten nächsten Zustands, unter der Annahme, dass der Agent die optimale Policy verfolgt. Mathematisch formuliert lautet die Bellman-Optimalitätsgleichung:" ] }, { "cell_type": "markdown", "id": "52bb1c28-6280-436c-b71a-2de82a2291b8", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "source": [ "$$\n", "V(s) = \\max_a \\left( R(s, a) + \\gamma \\sum_{s'} P(s'|s, a) V(s') \\right)\n", "$$" ] }, { "cell_type": "markdown", "id": "1766153d-ab22-40ab-921f-7d930692b16e", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "source": [ "Diese Gleichung besagt, dass der optimale Wert eines Zustands $s$ die maximale erwartete Belohnung ist, die der Agent erhalten kann, wenn er im Zustand $s$ startet, die Aktion $a$ wählt und danach der optimalen Policy folgt. Dadurch berücksichtigen wir bei der Bewertung des Zustandsüberganges durch die Aktion $a$ nicht nur den aktuellen Zustand, sondern auch die durch Aktion $a$ ermöglichten zukünftigen Zustandsübergänge. Damit lösen wir das Dilemma der Gedächtnislosigkeit, das aus der Markov-Bedingung folgt." ] }, { "cell_type": "markdown", "id": "ef8a4d63-58a9-4ab8-baec-fa9c09173472", "metadata": {}, "source": [ "Die Wertfunktion $V(s)$ bewertet einen Zustand insgesamt. Die Q-Funktion $Q(s,a)$ bewertet dagegen eine konkrete Aktion in einem Zustand. $Q(s,a)$ ist damit für die Aktionsauswahl oft direkter nutzbar, weil man die beste Aktion über das Maximum der Q-Werte bestimmen kann." ] }, { "cell_type": "markdown", "id": "a99a605b-60e1-4e38-b512-4605b15bb5e5", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "Die Bellman-Gleichung wird genutzt, um die optimale Strategie in Form einer _Q-Funktion_ $Q(s,a)$ im _Q-Learning_ zu erlernen. Das ist ein populärer Off-Policy-Algorithmus, bei dem der Agent eine Q-Funktion $ Q(s, a) $ lernt, welche Belohnungen einer Aktion $ a $ in einem Zustand $ s $ erwartet wird. Off-Policy bedeutet hier, dass das Update den aktuell besten geschätzten nächsten Q-Wert verwendet, auch wenn die tatsächlich ausgeführte Aktion aufgrund von Exploration eine andere war." ] }, { "cell_type": "markdown", "id": "bcaba3e2-551a-48b3-93fa-1608d5a6f9ae", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "source": [ "$$\n", "Q(s, a) \\leftarrow Q(s, a) + \\alpha \\left( R(s, a) + \\gamma \\max_{a'} Q(s', a') - Q(s, a) \\right)\n", "$$" ] }, { "cell_type": "markdown", "id": "a160d934-c610-4a0e-87a9-10b16056d5f7", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "source": [ "hierbei stellt $\\alpha$ die Lernrate dar. Die ist ein sehr wichtiger Parameter, da er entscheidet, wie schnell das Modell auf eine Lösung konvergiert. Ein niedriger Wert sorgt dazu, dass der Algorithmus mehr positive Beispiele zum Lernen benötigt, was also längeres Training erfordert, insbesondere bei seltenen, positiven Feedback. Ein hoher Wert kann dazu sorgen, dass sich das Modell schnell in einer schlechteren Lösung (lokales Optimum) verrennt." ] }, { "cell_type": "markdown", "id": "1683e8a4-fe25-45f1-8587-7449b2967e7f", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "Deshalb kombiniert man den Algorithmus man meist mit einer Erkundungsstrategie. Statt immer nur die geschätzte optimale Strategie $Q(s,a)$ zu nehmen, wählt man mit der Wahrscheinlichkeit $\\epsilon$ zufällige Strategien aus, um somit alternative Strategien zu entdecken. Ein hoher $\\epsilon$-Wert sorgt für eine hohe Erkundungsrate (Exploration), während ein niedriger ein verlässlichere Vorhersage erlaubt. Um hier einen Trade-Off zu finden, macht man oft beides und wählt am Anfang des Trainings ein hohes $\\epsilon$, um den Lösungsbereich zu erkunden und am Ende des Trainings ein niedriges $\\epsilon$ um schneller und verlässlicher zu konvergieren." ] }, { "cell_type": "markdown", "id": "a0c38c20-b23c-4c81-b39c-f2294bcc65f0", "metadata": { "editable": true, "slideshow": { "slide_type": "slide" }, "tags": [] }, "source": [ "## Beispiel: Gridworld-Umgebung" ] }, { "cell_type": "markdown", "id": "60ab042b-9061-4382-8545-2a9f18411655", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "Wir werden eine einfache Gridworld-Umgebung verwenden, um die Grundprinzipien von Reinforcement Learning zu veranschaulichen. In dieser Umgebung versucht ein Agent, in einem Grid von einem Startzustand zu einem Zielzustand zu gelangen und dabei den kürzesten Weg zu finden.\n", "\n", "Dies kann zum Beispiel zur Steuerung eines Materialtransports auf einer Baustelle dienen. Ein Agent (z.B. ein Bagger) befindet sich auf einer vereinfachten Baustelle und muss Material von einer Quelle zu einem Zielbereich bringen. \n", "Wir simulieren dieses Szenario in einer Gitterwelt mit folgenden Regeln:\n", "\n", "- Der Agent kann sich **hoch, runter, links, rechts** bewegen.\n", "- Er erhält **+1 Punkt**, wenn er das Ziel (Z) erreicht.\n", "- Für jeden Schritt wird er mit **-0.1 Punkten** bestraft.\n", "- Die Umgebung ist sehr einfach und zufallsbasiert." ] }, { "cell_type": "markdown", "id": "5d43ed73", "metadata": {}, "source": [ "Wir erstellen uns als erstes eine Klasse, welche die Umgebung repräsentiert und die Zustände enthält. Die Umgebung initialisiert zuerst unser Raster (Grid) mit der Start- und Endposition. Dieses Raster definiert unseren Zustandsraum, da der Agent, wenn er sich fort bewegt sich in jeder dieser Zellen im Raster aufhalten kann.\n", "\n", "Damit wir Experimente beim Lernen wiederholen können gibt es eine `reset`-Funktion. Mit der `step`-Funktion kann der Agent sich fortbewegen. Wir übergeben der Funktion als `action` die Richtung, in der wir uns bewegen wollen. Dadurch wechselt sich die aktuelle Position, also der aktuelle Zustand in unserem Zustandsmodell." ] }, { "cell_type": "code", "execution_count": 1, "id": "dcddd300-82f0-4d8c-a3c5-2b021ac92f36", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "outputs": [], "source": [ "import numpy as np # Import von NumPy\n", "\n", "# Definition der Gridworld-Umgebung\n", "class Gridworld:\n", " def __init__(self, size, start, goal):\n", " self.size = size\n", " self.start = start\n", " self.goal = goal\n", " self.state = start\n", " self.actions = ['up', 'down', 'left', 'right']\n", " self.grid=np.zeros((self.size,self.size))\n", " self.grid[start]=1\n", " self.grid[goal]=0\n", "\n", " def reset(self):\n", " self.state = self.start\n", " return self.state\n", "\n", " def step(self, action):\n", " x, y = self.state\n", " if action == 'up':\n", " x = max(0, x - 1)\n", " elif action == 'down':\n", " x = min(self.size - 1, x + 1)\n", " elif action == 'left':\n", " y = max(0, y - 1)\n", " elif action == 'right':\n", " y = min(self.size - 1, y + 1)\n", " self.state = (x, y)\n", " self.grid[self.state]+=1\n", " reward = 1 if self.state == self.goal else -0.1\n", " done = self.state == self.goal\n", " return self.state, reward, done" ] }, { "cell_type": "markdown", "id": "377b4a1c-c020-4a87-82ce-bb21fa727f0f", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "Erstellen wir uns als Beispiel ein 3x3 Gridworld und wollen uns vom Punkt $[0,0]$ zum Punkt $[3,3]$ bewegen." ] }, { "cell_type": "code", "execution_count": 2, "id": "c6dc6fd2-881e-4664-8f65-42c095d6eab4", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Startzustand: (0, 0)\n" ] }, { "data": { "text/plain": [ "array([[1., 0., 0., 0.],\n", " [0., 0., 0., 0.],\n", " [0., 0., 0., 0.],\n", " [0., 0., 0., 0.]])" ] }, "execution_count": 2, "metadata": {}, "output_type": "execute_result" } ], "source": [ "env = Gridworld(size=4, start=(0, 0), goal=(3, 3))\n", "state = env.reset()\n", "print(\"Startzustand:\", state)\n", "env.grid" ] }, { "cell_type": "markdown", "id": "deeab4a4-9002-449e-80bf-e3b90096ac1d", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "Der Zustandsraum ist hierbei ein 2D Grid" ] }, { "cell_type": "markdown", "id": "27812877-f2ad-4d8e-9de0-fc3aba2abde1", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "source": [ "
\n", "\n", "\"Zustandsraum\n", "\n", "
" ] }, { "cell_type": "markdown", "id": "090787bf-a7c2-46a5-a469-f96d01091367", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "Bewegen wir uns einmal manuell durch das Grid, so sehen wir wie die Zustände (unsere Position) sich ändern und welche Belohnungen wir erhalten. Wichtig hierbei ist, dass wir eine positive Belohnung erst beim letzten Schritt erhalten, vorher immer nur bestraft werden (z.B. Erschöpfung)." ] }, { "cell_type": "code", "execution_count": 3, "id": "be9cfdc5-c4a6-4900-bac8-50b6367dd204", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Nächster Zustand: (0, 1) Belohnung: -0.1 Ziel erreicht: False\n", "Nächster Zustand: (1, 1) Belohnung: -0.1 Ziel erreicht: False\n", "Nächster Zustand: (1, 2) Belohnung: -0.1 Ziel erreicht: False\n", "Nächster Zustand: (2, 2) Belohnung: -0.1 Ziel erreicht: False\n", "Nächster Zustand: (2, 3) Belohnung: -0.1 Ziel erreicht: False\n", "Nächster Zustand: (3, 3) Belohnung: 1 Ziel erreicht: True\n" ] } ], "source": [ "# Beispiel für einen Schritt in der Umgebung\n", "steps=['right','down','right','down','right','down']\n", "for step in steps:\n", " next_state, reward, done = env.step(step)\n", " print(\"Nächster Zustand:\", next_state, \"Belohnung:\", reward, \"Ziel erreicht:\", done)" ] }, { "cell_type": "code", "execution_count": 4, "id": "8fd847a0-2b60-439f-a5ec-6d52dc50dcd8", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "outputs": [ { "data": { "application/vnd.plotly.v1+json": { "config": { "plotlyServerURL": "https://plot.ly" }, "data": [ { "coloraxis": "coloraxis", "hovertemplate": "x: %{x}
y: %{y}
color: %{z}", "name": "0", "type": "heatmap", "xaxis": "x", "yaxis": "y", "z": [ [ 1, 1, 0, 0 ], [ 0, 1, 1, 0 ], [ 0, 0, 1, 1 ], [ 0, 0, 0, 1 ] ] } ], "layout": { "coloraxis": { "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ] }, "margin": { "t": 60 }, "template": { "data": { "bar": [ { "error_x": { "color": "#2a3f5f" }, "error_y": { "color": "#2a3f5f" }, "marker": { "line": { "color": "#E5ECF6", "width": 0.5 }, "pattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 } }, "type": "bar" } ], "barpolar": [ { "marker": { "line": { "color": "#E5ECF6", "width": 0.5 }, "pattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 } }, "type": "barpolar" } ], "carpet": [ { "aaxis": { "endlinecolor": "#2a3f5f", "gridcolor": "white", "linecolor": "white", "minorgridcolor": "white", "startlinecolor": "#2a3f5f" }, "baxis": { "endlinecolor": "#2a3f5f", "gridcolor": "white", "linecolor": "white", "minorgridcolor": "white", "startlinecolor": "#2a3f5f" }, "type": "carpet" } ], "choropleth": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "choropleth" } ], "contour": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "contour" } ], "contourcarpet": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "contourcarpet" } ], "heatmap": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "heatmap" } ], "heatmapgl": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "heatmapgl" } ], "histogram": [ { "marker": { "pattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 } }, "type": "histogram" } ], "histogram2d": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "histogram2d" } ], "histogram2dcontour": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "histogram2dcontour" } ], "mesh3d": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "mesh3d" } ], "parcoords": [ { "line": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "parcoords" } ], "pie": [ { "automargin": true, "type": "pie" } ], "scatter": [ { "fillpattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 }, "type": "scatter" } ], "scatter3d": [ { "line": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatter3d" } ], "scattercarpet": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattercarpet" } ], "scattergeo": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattergeo" } ], "scattergl": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattergl" } ], "scattermapbox": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattermapbox" } ], "scatterpolar": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterpolar" } ], "scatterpolargl": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterpolargl" } ], "scatterternary": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterternary" } ], "surface": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "surface" } ], "table": [ { "cells": { "fill": { "color": "#EBF0F8" }, "line": { "color": "white" } }, "header": { "fill": { "color": "#C8D4E3" }, "line": { "color": "white" } }, "type": "table" } ] }, "layout": { "annotationdefaults": { "arrowcolor": "#2a3f5f", "arrowhead": 0, "arrowwidth": 1 }, "autotypenumbers": "strict", "coloraxis": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "colorscale": { "diverging": [ [ 0, "#8e0152" ], [ 0.1, "#c51b7d" ], [ 0.2, "#de77ae" ], [ 0.3, "#f1b6da" ], [ 0.4, "#fde0ef" ], [ 0.5, "#f7f7f7" ], [ 0.6, "#e6f5d0" ], [ 0.7, "#b8e186" ], [ 0.8, "#7fbc41" ], [ 0.9, "#4d9221" ], [ 1, "#276419" ] ], "sequential": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "sequentialminus": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ] }, "colorway": [ "#636efa", "#EF553B", "#00cc96", "#ab63fa", "#FFA15A", "#19d3f3", "#FF6692", "#B6E880", "#FF97FF", "#FECB52" ], "font": { "color": "#2a3f5f" }, "geo": { "bgcolor": "white", "lakecolor": "white", "landcolor": "#E5ECF6", "showlakes": true, "showland": true, "subunitcolor": "white" }, "hoverlabel": { "align": "left" }, "hovermode": "closest", "mapbox": { "style": "light" }, "paper_bgcolor": "white", "plot_bgcolor": "#E5ECF6", "polar": { "angularaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "bgcolor": "#E5ECF6", "radialaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" } }, "scene": { "xaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" }, "yaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" }, "zaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" } }, "shapedefaults": { "line": { "color": "#2a3f5f" } }, "ternary": { "aaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "baxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "bgcolor": "#E5ECF6", "caxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" } }, "title": { "x": 0.05 }, "xaxis": { "automargin": true, "gridcolor": "white", "linecolor": "white", "ticks": "", "title": { "standoff": 15 }, "zerolinecolor": "white", "zerolinewidth": 2 }, "yaxis": { "automargin": true, "gridcolor": "white", "linecolor": "white", "ticks": "", "title": { "standoff": 15 }, "zerolinecolor": "white", "zerolinewidth": 2 } } }, "xaxis": { "anchor": "y", "constrain": "domain", "domain": [ 0, 1 ], "scaleanchor": "y" }, "yaxis": { "anchor": "x", "autorange": "reversed", "constrain": "domain", "domain": [ 0, 1 ] } } } }, "metadata": {}, "output_type": "display_data" } ], "source": [ "import plotly.express as px\n", "px.imshow(env.grid)" ] }, { "cell_type": "markdown", "id": "a01187a1-1645-4953-b338-fd2085697fe1", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "## Random Walk" ] }, { "cell_type": "markdown", "id": "d63b31e6-45c6-4fc9-b5f0-90e7c21ef535", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "Ein naiver Lösungsansatz ist der so genannte Random Walk (Zufallsweg), bei dem man bei jedem Schritt in eine zufällig gewählte Richtung läuft, was auch bedeutet, dass man ggf. zurück läuft. Untersuchen wir einmal in 500 Experimenten die Erfolgschancen dieser Lösungsstrategie." ] }, { "cell_type": "code", "execution_count": 5, "id": "a64b61ab-08b6-4e62-96da-d53dd552055b", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "outputs": [], "source": [ "env = Gridworld(size=4, start=(0, 0), goal=(3, 3))\n", "trials = 500\n", "\n", "rewards = []\n", "retries = []\n", "for trial in range(trials):\n", " env.reset()\n", " n, r_sum = 0, 0\n", " done = False\n", " while not done:\n", " # Wir wählen eine zufällige Aktion\n", " action = np.random.choice(env.actions)\n", " # Führe die Aktion aus und erhalte von der Umgebung den neuen Zustand und die Belohnung\n", " new_state, reward, done = env.step(action)\n", " # Sammel Erfolgsstatistik\n", " r_sum += reward\n", " n += 1\n", " rewards.append(r_sum)\n", " retries.append(n)" ] }, { "cell_type": "markdown", "id": "85465dd0-c0aa-4a48-bcd9-66a249806dfe", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "In allen Experimenten ist unser Agent zum Ziel gekommen, was zeigt, dass diese zufällige Explorations-Strategie durchaus erfolgreich ist. Schauen wir auf die Anzahl der Versuche bis zum Ziel über unsere Experimente, so sehen wir, dass diese gleichbleibend stark variiert. Wir haben also kein Lerneffekt." ] }, { "cell_type": "code", "execution_count": 6, "id": "f220ba95-2526-4ee8-9d42-4918c9899941", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "outputs": [ { "data": { "application/vnd.plotly.v1+json": { "config": { "plotlyServerURL": "https://plot.ly" }, "data": [ { "hovertemplate": "variable=0
index=%{x}
value=%{y}", "legendgroup": "0", "line": { "color": "#636efa", "dash": "solid" }, "marker": { "symbol": "circle" }, "mode": "lines", "name": "0", "orientation": "v", "showlegend": true, "type": "scatter", "x": [ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 191, 192, 193, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203, 204, 205, 206, 207, 208, 209, 210, 211, 212, 213, 214, 215, 216, 217, 218, 219, 220, 221, 222, 223, 224, 225, 226, 227, 228, 229, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 285, 286, 287, 288, 289, 290, 291, 292, 293, 294, 295, 296, 297, 298, 299, 300, 301, 302, 303, 304, 305, 306, 307, 308, 309, 310, 311, 312, 313, 314, 315, 316, 317, 318, 319, 320, 321, 322, 323, 324, 325, 326, 327, 328, 329, 330, 331, 332, 333, 334, 335, 336, 337, 338, 339, 340, 341, 342, 343, 344, 345, 346, 347, 348, 349, 350, 351, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 473, 474, 475, 476, 477, 478, 479, 480, 481, 482, 483, 484, 485, 486, 487, 488, 489, 490, 491, 492, 493, 494, 495, 496, 497, 498, 499 ], "xaxis": "x", "y": [ 99, 173, 58, 65, 48, 180, 182, 73, 20, 21, 50, 66, 9, 83, 152, 53, 127, 17, 88, 25, 26, 21, 104, 102, 55, 18, 128, 76, 12, 140, 56, 22, 61, 19, 37, 14, 16, 20, 127, 13, 204, 33, 145, 11, 203, 374, 109, 21, 82, 47, 21, 13, 56, 21, 37, 12, 19, 43, 60, 30, 26, 184, 67, 14, 70, 43, 14, 156, 17, 43, 149, 31, 132, 39, 75, 44, 16, 62, 40, 23, 19, 15, 147, 76, 39, 141, 56, 234, 76, 74, 14, 31, 99, 39, 15, 15, 47, 89, 145, 46, 129, 105, 39, 33, 28, 12, 47, 46, 13, 26, 31, 136, 59, 20, 86, 13, 88, 40, 12, 44, 36, 23, 13, 50, 64, 80, 36, 49, 177, 104, 86, 12, 71, 44, 154, 22, 54, 12, 47, 13, 11, 50, 10, 127, 68, 17, 89, 52, 17, 89, 28, 21, 20, 27, 27, 87, 30, 73, 94, 49, 16, 22, 48, 87, 73, 24, 39, 35, 37, 56, 18, 90, 76, 54, 23, 63, 21, 75, 61, 114, 104, 31, 46, 14, 11, 65, 58, 91, 61, 27, 199, 110, 16, 89, 131, 20, 15, 54, 24, 15, 231, 38, 22, 107, 45, 38, 166, 243, 20, 42, 91, 46, 14, 34, 46, 78, 104, 46, 28, 9, 13, 19, 58, 134, 37, 95, 99, 36, 53, 73, 87, 84, 16, 24, 116, 69, 15, 38, 60, 45, 53, 44, 134, 49, 56, 11, 13, 34, 30, 40, 97, 57, 38, 34, 9, 36, 73, 43, 34, 104, 63, 86, 71, 57, 46, 186, 76, 30, 25, 26, 67, 19, 62, 48, 30, 91, 60, 66, 89, 40, 71, 15, 43, 72, 67, 22, 35, 24, 23, 28, 175, 110, 18, 74, 61, 11, 32, 39, 38, 11, 93, 40, 46, 68, 12, 29, 64, 19, 95, 10, 75, 39, 22, 28, 100, 15, 217, 41, 38, 252, 8, 37, 44, 24, 51, 8, 72, 14, 231, 36, 35, 16, 16, 38, 90, 29, 15, 39, 54, 42, 61, 43, 33, 16, 17, 34, 43, 39, 203, 50, 9, 18, 86, 224, 56, 33, 48, 72, 30, 64, 8, 80, 37, 90, 30, 48, 124, 31, 47, 91, 24, 20, 23, 83, 26, 16, 42, 72, 253, 136, 44, 72, 16, 13, 143, 77, 64, 124, 37, 9, 47, 12, 59, 80, 14, 48, 13, 177, 19, 175, 115, 15, 104, 39, 93, 58, 14, 45, 69, 56, 6, 27, 31, 79, 107, 14, 92, 51, 59, 52, 11, 76, 46, 50, 55, 44, 52, 23, 16, 86, 42, 276, 68, 39, 9, 119, 58, 13, 11, 99, 7, 65, 31, 37, 74, 117, 75, 80, 185, 62, 79, 52, 31, 24, 12, 31, 113, 42, 58, 14, 89, 25, 11, 83, 34, 30, 111, 58, 60, 60, 46, 30, 20, 14, 84, 73, 76, 152, 9, 18, 141, 108, 22, 33, 71, 59, 125, 135, 23, 14, 22, 11, 54, 40, 190, 18, 53, 31, 18, 125 ], "yaxis": "y" } ], "layout": { "autosize": true, "legend": { "title": { "text": "variable" }, "tracegroupgap": 0 }, "margin": { "t": 60 }, "template": { "data": { "bar": [ { "error_x": { "color": "#2a3f5f" }, "error_y": { "color": "#2a3f5f" }, "marker": { "line": { "color": "#E5ECF6", "width": 0.5 }, "pattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 } }, "type": "bar" } ], "barpolar": [ { "marker": { "line": { "color": "#E5ECF6", "width": 0.5 }, "pattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 } }, "type": "barpolar" } ], "carpet": [ { "aaxis": { "endlinecolor": "#2a3f5f", "gridcolor": "white", "linecolor": "white", "minorgridcolor": "white", "startlinecolor": "#2a3f5f" }, "baxis": { "endlinecolor": "#2a3f5f", "gridcolor": "white", "linecolor": "white", "minorgridcolor": "white", "startlinecolor": "#2a3f5f" }, "type": "carpet" } ], "choropleth": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "choropleth" } ], "contour": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "contour" } ], "contourcarpet": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "contourcarpet" } ], "heatmap": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "heatmap" } ], "heatmapgl": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "heatmapgl" } ], "histogram": [ { "marker": { "pattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 } }, "type": "histogram" } ], "histogram2d": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "histogram2d" } ], "histogram2dcontour": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "histogram2dcontour" } ], "mesh3d": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "mesh3d" } ], "parcoords": [ { "line": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "parcoords" } ], "pie": [ { "automargin": true, "type": "pie" } ], "scatter": [ { "fillpattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 }, "type": "scatter" } ], "scatter3d": [ { "line": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatter3d" } ], "scattercarpet": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattercarpet" } ], "scattergeo": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattergeo" } ], "scattergl": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattergl" } ], "scattermapbox": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattermapbox" } ], "scatterpolar": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterpolar" } ], "scatterpolargl": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterpolargl" } ], "scatterternary": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterternary" } ], "surface": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "surface" } ], "table": [ { "cells": { "fill": { "color": "#EBF0F8" }, "line": { "color": "white" } }, "header": { "fill": { "color": "#C8D4E3" }, "line": { "color": "white" } }, "type": "table" } ] }, "layout": { "annotationdefaults": { "arrowcolor": "#2a3f5f", "arrowhead": 0, "arrowwidth": 1 }, "autotypenumbers": "strict", "coloraxis": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "colorscale": { "diverging": [ [ 0, "#8e0152" ], [ 0.1, "#c51b7d" ], [ 0.2, "#de77ae" ], [ 0.3, "#f1b6da" ], [ 0.4, "#fde0ef" ], [ 0.5, "#f7f7f7" ], [ 0.6, "#e6f5d0" ], [ 0.7, "#b8e186" ], [ 0.8, "#7fbc41" ], [ 0.9, "#4d9221" ], [ 1, "#276419" ] ], "sequential": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "sequentialminus": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ] }, "colorway": [ "#636efa", "#EF553B", "#00cc96", "#ab63fa", "#FFA15A", "#19d3f3", "#FF6692", "#B6E880", "#FF97FF", "#FECB52" ], "font": { "color": "#2a3f5f" }, "geo": { "bgcolor": "white", "lakecolor": "white", "landcolor": "#E5ECF6", "showlakes": true, "showland": true, "subunitcolor": "white" }, "hoverlabel": { "align": "left" }, "hovermode": "closest", "mapbox": { "style": "light" }, "paper_bgcolor": "white", "plot_bgcolor": "#E5ECF6", "polar": { "angularaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "bgcolor": "#E5ECF6", "radialaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" } }, "scene": { "xaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" }, "yaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" }, "zaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" } }, "shapedefaults": { "line": { "color": "#2a3f5f" } }, "ternary": { "aaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "baxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "bgcolor": "#E5ECF6", "caxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" } }, "title": { "x": 0.05 }, "xaxis": { "automargin": true, "gridcolor": "white", "linecolor": "white", "ticks": "", "title": { "standoff": 15 }, "zerolinecolor": "white", "zerolinewidth": 2 }, "yaxis": { "automargin": true, "gridcolor": "white", "linecolor": "white", "ticks": "", "title": { "standoff": 15 }, "zerolinecolor": "white", "zerolinewidth": 2 } } }, "xaxis": { "anchor": "y", "autorange": true, "domain": [ 0, 1 ], "range": [ 0, 499 ], "title": { "text": "index" }, "type": "linear" }, "yaxis": { "anchor": "x", "autorange": true, "domain": [ 0, 1 ], "range": [ -14.444444444444443, 394.44444444444446 ], "title": { "text": "value" }, "type": "linear" } } }, "image/png": "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", "text/html": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "import plotly.express as px\n", "px.line(retries)" ] }, { "cell_type": "markdown", "id": "1b8d0d49-886e-43ff-b222-b39eba19af94", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "Entsprechend niedrig sind auch die Belohnungen, die das Programm pro Experiment sammelt." ] }, { "cell_type": "code", "execution_count": 7, "id": "711159f5-3646-4ca3-acbb-e37dc4ff8531", "metadata": {}, "outputs": [ { "data": { "application/vnd.plotly.v1+json": { "config": { "plotlyServerURL": "https://plot.ly" }, "data": [ { "hovertemplate": "variable=0
index=%{x}
value=%{y}", "legendgroup": "0", "line": { "color": "#636efa", "dash": "solid" }, "marker": { "symbol": "circle" }, "mode": "lines", "name": "0", "orientation": "v", "showlegend": true, "type": "scatter", "x": [ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 191, 192, 193, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203, 204, 205, 206, 207, 208, 209, 210, 211, 212, 213, 214, 215, 216, 217, 218, 219, 220, 221, 222, 223, 224, 225, 226, 227, 228, 229, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 285, 286, 287, 288, 289, 290, 291, 292, 293, 294, 295, 296, 297, 298, 299, 300, 301, 302, 303, 304, 305, 306, 307, 308, 309, 310, 311, 312, 313, 314, 315, 316, 317, 318, 319, 320, 321, 322, 323, 324, 325, 326, 327, 328, 329, 330, 331, 332, 333, 334, 335, 336, 337, 338, 339, 340, 341, 342, 343, 344, 345, 346, 347, 348, 349, 350, 351, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 473, 474, 475, 476, 477, 478, 479, 480, 481, 482, 483, 484, 485, 486, 487, 488, 489, 490, 491, 492, 493, 494, 495, 496, 497, 498, 499 ], "xaxis": "x", "y": [ -8.799999999999981, -16.199999999999974, -4.699999999999996, -5.399999999999993, -3.6999999999999993, -16.899999999999984, -17.099999999999987, -6.19999999999999, -0.9000000000000006, -1.0000000000000004, -3.8999999999999986, -5.499999999999993, 0.20000000000000007, -7.199999999999987, -14.099999999999962, -4.1999999999999975, -11.599999999999971, -0.6000000000000003, -7.699999999999985, -1.4000000000000008, -1.5000000000000009, -1.0000000000000004, -9.29999999999998, -9.09999999999998, -4.399999999999997, -0.7000000000000004, -11.69999999999997, -6.499999999999989, -0.09999999999999987, -12.899999999999967, -4.4999999999999964, -1.1000000000000005, -4.999999999999995, -0.8000000000000005, -2.600000000000002, -0.30000000000000004, -0.5000000000000002, -0.9000000000000006, -11.599999999999971, -0.19999999999999996, -19.30000000000002, -2.2000000000000015, -13.399999999999965, 1.1102230246251565e-16, -19.200000000000017, -36.30000000000026, -9.799999999999978, -1.0000000000000004, -7.099999999999987, -3.5999999999999996, -1.0000000000000004, -0.19999999999999996, -4.4999999999999964, -1.0000000000000004, -2.600000000000002, -0.09999999999999987, -0.8000000000000005, -3.200000000000001, -4.899999999999995, -1.9000000000000012, -1.5000000000000009, -17.29999999999999, -5.5999999999999925, -0.30000000000000004, -5.8999999999999915, -3.200000000000001, -0.30000000000000004, -14.499999999999961, -0.6000000000000003, -3.200000000000001, -13.799999999999963, -2.0000000000000013, -12.09999999999997, -2.800000000000002, -6.39999999999999, -3.3000000000000007, -0.5000000000000002, -5.099999999999994, -2.900000000000002, -1.2000000000000006, -0.8000000000000005, -0.40000000000000013, -13.599999999999964, -6.499999999999989, -2.800000000000002, -12.999999999999966, -4.4999999999999964, -22.30000000000006, -6.499999999999989, -6.29999999999999, -0.30000000000000004, -2.0000000000000013, -8.799999999999981, -2.800000000000002, -0.40000000000000013, -0.40000000000000013, -3.5999999999999996, -7.799999999999985, -13.399999999999965, -3.5, -11.79999999999997, -9.399999999999979, -2.800000000000002, -2.2000000000000015, -1.700000000000001, -0.09999999999999987, -3.5999999999999996, -3.5, -0.19999999999999996, -1.5000000000000009, -2.0000000000000013, -12.499999999999968, -4.799999999999995, -0.9000000000000006, -7.499999999999986, -0.19999999999999996, -7.699999999999985, -2.900000000000002, -0.09999999999999987, -3.3000000000000007, -2.5000000000000018, -1.2000000000000006, -0.19999999999999996, -3.8999999999999986, -5.299999999999994, -6.899999999999988, -2.5000000000000018, -3.799999999999999, -16.59999999999998, -9.29999999999998, -7.499999999999986, -0.09999999999999987, -5.999999999999991, -3.3000000000000007, -14.299999999999962, -1.1000000000000005, -4.299999999999997, -0.09999999999999987, -3.5999999999999996, -0.19999999999999996, 1.1102230246251565e-16, -3.8999999999999986, 0.10000000000000009, -11.599999999999971, -5.699999999999992, -0.6000000000000003, -7.799999999999985, -4.099999999999998, -0.6000000000000003, -7.799999999999985, -1.700000000000001, -1.0000000000000004, -0.9000000000000006, -1.600000000000001, -1.600000000000001, -7.599999999999985, -1.9000000000000012, -6.19999999999999, -8.299999999999983, -3.799999999999999, -0.5000000000000002, -1.1000000000000005, -3.6999999999999993, -7.599999999999985, -6.19999999999999, -1.3000000000000007, -2.800000000000002, -2.4000000000000017, -2.600000000000002, -4.4999999999999964, -0.7000000000000004, -7.899999999999984, -6.499999999999989, -4.299999999999997, -1.2000000000000006, -5.199999999999994, -1.0000000000000004, -6.39999999999999, -4.999999999999995, -10.299999999999976, -9.29999999999998, -2.0000000000000013, -3.5, -0.30000000000000004, 1.1102230246251565e-16, -5.399999999999993, -4.699999999999996, -7.999999999999984, -4.999999999999995, -1.600000000000001, -18.80000000000001, -9.899999999999977, -0.5000000000000002, -7.799999999999985, -11.99999999999997, -0.9000000000000006, -0.40000000000000013, -4.299999999999997, -1.3000000000000007, -0.40000000000000013, -22.000000000000057, -2.700000000000002, -1.1000000000000005, -9.599999999999978, -3.4000000000000004, -2.700000000000002, -15.499999999999964, -23.200000000000074, -0.9000000000000006, -3.1000000000000014, -7.999999999999984, -3.5, -0.30000000000000004, -2.3000000000000016, -3.5, -6.699999999999989, -9.29999999999998, -3.5, -1.700000000000001, 0.20000000000000007, -0.19999999999999996, -0.8000000000000005, -4.699999999999996, -12.299999999999969, -2.600000000000002, -8.399999999999983, -8.799999999999981, -2.5000000000000018, -4.1999999999999975, -6.19999999999999, -7.599999999999985, -7.2999999999999865, -0.5000000000000002, -1.3000000000000007, -10.499999999999975, -5.799999999999992, -0.40000000000000013, -2.700000000000002, -4.899999999999995, -3.4000000000000004, -4.1999999999999975, -3.3000000000000007, -12.299999999999969, -3.799999999999999, -4.4999999999999964, 1.1102230246251565e-16, -0.19999999999999996, -2.3000000000000016, -1.9000000000000012, -2.900000000000002, -8.599999999999982, -4.599999999999996, -2.700000000000002, -2.3000000000000016, 0.20000000000000007, -2.5000000000000018, -6.19999999999999, -3.200000000000001, -2.3000000000000016, -9.29999999999998, -5.199999999999994, -7.499999999999986, -5.999999999999991, -4.599999999999996, -3.5, -17.499999999999993, -6.499999999999989, -1.9000000000000012, -1.4000000000000008, -1.5000000000000009, -5.5999999999999925, -0.8000000000000005, -5.099999999999994, -3.6999999999999993, -1.9000000000000012, -7.999999999999984, -4.899999999999995, -5.499999999999993, -7.799999999999985, -2.900000000000002, -5.999999999999991, -0.40000000000000013, -3.200000000000001, -6.099999999999991, -5.5999999999999925, -1.1000000000000005, -2.4000000000000017, -1.3000000000000007, -1.2000000000000006, -1.700000000000001, -16.399999999999977, -9.899999999999977, -0.7000000000000004, -6.29999999999999, -4.999999999999995, 1.1102230246251565e-16, -2.1000000000000014, -2.800000000000002, -2.700000000000002, 1.1102230246251565e-16, -8.199999999999983, -2.900000000000002, -3.5, -5.699999999999992, -0.09999999999999987, -1.8000000000000012, -5.299999999999994, -0.8000000000000005, -8.399999999999983, 0.10000000000000009, -6.39999999999999, -2.800000000000002, -1.1000000000000005, -1.700000000000001, -8.89999999999998, -0.40000000000000013, -20.600000000000037, -3.0000000000000018, -2.700000000000002, -24.100000000000087, 0.30000000000000004, -2.600000000000002, -3.3000000000000007, -1.3000000000000007, -3.9999999999999982, 0.30000000000000004, -6.099999999999991, -0.30000000000000004, -22.000000000000057, -2.5000000000000018, -2.4000000000000017, -0.5000000000000002, -0.5000000000000002, -2.700000000000002, -7.899999999999984, -1.8000000000000012, -0.40000000000000013, -2.800000000000002, -4.299999999999997, -3.1000000000000014, -4.999999999999995, -3.200000000000001, -2.2000000000000015, -0.5000000000000002, -0.6000000000000003, -2.3000000000000016, -3.200000000000001, -2.800000000000002, -19.200000000000017, -3.8999999999999986, 0.20000000000000007, -0.7000000000000004, -7.499999999999986, -21.300000000000047, -4.4999999999999964, -2.2000000000000015, -3.6999999999999993, -6.099999999999991, -1.9000000000000012, -5.299999999999994, 0.30000000000000004, -6.899999999999988, -2.600000000000002, -7.899999999999984, -1.9000000000000012, -3.6999999999999993, -11.299999999999972, -2.0000000000000013, -3.5999999999999996, -7.999999999999984, -1.3000000000000007, -0.9000000000000006, -1.2000000000000006, -7.199999999999987, -1.5000000000000009, -0.5000000000000002, -3.1000000000000014, -6.099999999999991, -24.200000000000088, -12.499999999999968, -3.3000000000000007, -6.099999999999991, -0.5000000000000002, -0.19999999999999996, -13.199999999999966, -6.599999999999989, -5.299999999999994, -11.299999999999972, -2.600000000000002, 0.20000000000000007, -3.5999999999999996, -0.09999999999999987, -4.799999999999995, -6.899999999999988, -0.30000000000000004, -3.6999999999999993, -0.19999999999999996, -16.59999999999998, -0.8000000000000005, -16.399999999999977, -10.399999999999975, -0.40000000000000013, -9.29999999999998, -2.800000000000002, -8.199999999999983, -4.699999999999996, -0.30000000000000004, -3.4000000000000004, -5.799999999999992, -4.4999999999999964, 0.5, -1.600000000000001, -2.0000000000000013, -6.799999999999988, -9.599999999999978, -0.30000000000000004, -8.099999999999984, -3.9999999999999982, -4.799999999999995, -4.099999999999998, 1.1102230246251565e-16, -6.499999999999989, -3.5, -3.8999999999999986, -4.399999999999997, -3.3000000000000007, -4.099999999999998, -1.2000000000000006, -0.5000000000000002, -7.499999999999986, -3.1000000000000014, -26.50000000000012, -5.699999999999992, -2.800000000000002, 0.20000000000000007, -10.799999999999974, -4.699999999999996, -0.19999999999999996, 1.1102230246251565e-16, -8.799999999999981, 0.4, -5.399999999999993, -2.0000000000000013, -2.600000000000002, -6.29999999999999, -10.599999999999975, -6.39999999999999, -6.899999999999988, -17.39999999999999, -5.099999999999994, -6.799999999999988, -4.099999999999998, -2.0000000000000013, -1.3000000000000007, -0.09999999999999987, -2.0000000000000013, -10.199999999999976, -3.1000000000000014, -4.699999999999996, -0.30000000000000004, -7.799999999999985, -1.4000000000000008, 1.1102230246251565e-16, -7.199999999999987, -2.3000000000000016, -1.9000000000000012, -9.999999999999977, -4.699999999999996, -4.899999999999995, -4.899999999999995, -3.5, -1.9000000000000012, -0.9000000000000006, -0.30000000000000004, -7.2999999999999865, -6.19999999999999, -6.499999999999989, -14.099999999999962, 0.20000000000000007, -0.7000000000000004, -12.999999999999966, -9.699999999999978, -1.1000000000000005, -2.2000000000000015, -5.999999999999991, -4.799999999999995, -11.399999999999972, -12.399999999999968, -1.2000000000000006, -0.30000000000000004, -1.1000000000000005, 1.1102230246251565e-16, -4.299999999999997, -2.900000000000002, -17.9, -0.7000000000000004, -4.1999999999999975, -2.0000000000000013, -0.7000000000000004, -11.399999999999972 ], "yaxis": "y" } ], "layout": { "autosize": true, "legend": { "title": { "text": "variable" }, "tracegroupgap": 0 }, "margin": { "t": 60 }, "template": { "data": { "bar": [ { "error_x": { "color": "#2a3f5f" }, "error_y": { "color": "#2a3f5f" }, "marker": { "line": { "color": "#E5ECF6", "width": 0.5 }, "pattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 } }, "type": "bar" } ], "barpolar": [ { "marker": { "line": { "color": "#E5ECF6", "width": 0.5 }, "pattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 } }, "type": "barpolar" } ], "carpet": [ { "aaxis": { "endlinecolor": "#2a3f5f", "gridcolor": "white", "linecolor": "white", "minorgridcolor": "white", "startlinecolor": "#2a3f5f" }, "baxis": { "endlinecolor": "#2a3f5f", "gridcolor": "white", "linecolor": "white", "minorgridcolor": "white", "startlinecolor": "#2a3f5f" }, "type": "carpet" } ], "choropleth": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "choropleth" } ], "contour": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "contour" } ], "contourcarpet": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "contourcarpet" } ], "heatmap": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "heatmap" } ], "heatmapgl": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "heatmapgl" } ], "histogram": [ { "marker": { "pattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 } }, "type": "histogram" } ], "histogram2d": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "histogram2d" } ], "histogram2dcontour": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "histogram2dcontour" } ], "mesh3d": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "mesh3d" } ], "parcoords": [ { "line": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "parcoords" } ], "pie": [ { "automargin": true, "type": "pie" } ], "scatter": [ { "fillpattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 }, "type": "scatter" } ], "scatter3d": [ { "line": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatter3d" } ], "scattercarpet": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattercarpet" } ], "scattergeo": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattergeo" } ], "scattergl": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattergl" } ], "scattermapbox": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattermapbox" } ], "scatterpolar": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterpolar" } ], "scatterpolargl": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterpolargl" } ], "scatterternary": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterternary" } ], "surface": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "surface" } ], "table": [ { "cells": { "fill": { "color": "#EBF0F8" }, "line": { "color": "white" } }, "header": { "fill": { "color": "#C8D4E3" }, "line": { "color": "white" } }, "type": "table" } ] }, "layout": { "annotationdefaults": { "arrowcolor": "#2a3f5f", "arrowhead": 0, "arrowwidth": 1 }, "autotypenumbers": "strict", "coloraxis": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "colorscale": { "diverging": [ [ 0, "#8e0152" ], [ 0.1, "#c51b7d" ], [ 0.2, "#de77ae" ], [ 0.3, "#f1b6da" ], [ 0.4, "#fde0ef" ], [ 0.5, "#f7f7f7" ], [ 0.6, "#e6f5d0" ], [ 0.7, "#b8e186" ], [ 0.8, "#7fbc41" ], [ 0.9, "#4d9221" ], [ 1, "#276419" ] ], "sequential": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "sequentialminus": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ] }, "colorway": [ "#636efa", "#EF553B", "#00cc96", "#ab63fa", "#FFA15A", "#19d3f3", "#FF6692", "#B6E880", "#FF97FF", "#FECB52" ], "font": { "color": "#2a3f5f" }, "geo": { "bgcolor": "white", "lakecolor": "white", "landcolor": "#E5ECF6", "showlakes": true, "showland": true, "subunitcolor": "white" }, "hoverlabel": { "align": "left" }, "hovermode": "closest", "mapbox": { "style": "light" }, "paper_bgcolor": "white", "plot_bgcolor": "#E5ECF6", "polar": { "angularaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "bgcolor": "#E5ECF6", "radialaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" } }, "scene": { "xaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" }, "yaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" }, "zaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" } }, "shapedefaults": { "line": { "color": "#2a3f5f" } }, "ternary": { "aaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "baxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "bgcolor": "#E5ECF6", "caxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" } }, "title": { "x": 0.05 }, "xaxis": { "automargin": true, "gridcolor": "white", "linecolor": "white", "ticks": "", "title": { "standoff": 15 }, "zerolinecolor": "white", "zerolinewidth": 2 }, "yaxis": { "automargin": true, "gridcolor": "white", "linecolor": "white", "ticks": "", "title": { "standoff": 15 }, "zerolinecolor": "white", "zerolinewidth": 2 } } }, "xaxis": { "anchor": "y", "autorange": true, "domain": [ 0, 1 ], "range": [ 0, 499 ], "title": { "text": "index" }, "type": "linear" }, "yaxis": { "anchor": "x", "autorange": true, "domain": [ 0, 1 ], "range": [ -38.34444444444472, 2.544444444444459 ], "title": { "text": "value" }, "type": "linear" } } }, "image/png": "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", "text/html": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "px.line(rewards)" ] }, { "cell_type": "markdown", "id": "823bb2c9-f8ef-48d0-8c83-b64e916e36f2", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "Wenn wir uns die Häufigkeit ansehen, der besuchten Rasterpunkte, so zeigt sich, dass der Großteil der Versuche um dem Starpunkt herum hängen bleibt, also oft auch wieder sich zurückbewegt." ] }, { "cell_type": "code", "execution_count": 8, "id": "a1a0df15-520f-43da-9e8f-09ebe5e9e8d4", "metadata": {}, "outputs": [ { "data": { "application/vnd.plotly.v1+json": { "config": { "plotlyServerURL": "https://plot.ly" }, "data": [ { "coloraxis": "coloraxis", "hovertemplate": "x: %{x}
y: %{y}
color: %{z}", "name": "0", "type": "heatmap", "xaxis": "x", "yaxis": "y", "z": [ [ 3267, 2757, 2155, 1773 ], [ 2791, 2316, 1869, 1525 ], [ 2260, 1877, 1370, 930 ], [ 1916, 1547, 964, 500 ] ] } ], "layout": { "autosize": true, "coloraxis": { "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ] }, "margin": { "t": 60 }, "template": { "data": { "bar": [ { "error_x": { "color": "#2a3f5f" }, "error_y": { "color": "#2a3f5f" }, "marker": { "line": { "color": "#E5ECF6", "width": 0.5 }, "pattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 } }, "type": "bar" } ], "barpolar": [ { "marker": { "line": { "color": "#E5ECF6", "width": 0.5 }, "pattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 } }, "type": "barpolar" } ], "carpet": [ { "aaxis": { "endlinecolor": "#2a3f5f", "gridcolor": "white", "linecolor": "white", "minorgridcolor": "white", "startlinecolor": "#2a3f5f" }, "baxis": { "endlinecolor": "#2a3f5f", "gridcolor": "white", "linecolor": "white", "minorgridcolor": "white", "startlinecolor": "#2a3f5f" }, "type": "carpet" } ], "choropleth": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "choropleth" } ], "contour": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "contour" } ], "contourcarpet": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "contourcarpet" } ], "heatmap": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "heatmap" } ], "heatmapgl": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "heatmapgl" } ], "histogram": [ { "marker": { "pattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 } }, "type": "histogram" } ], "histogram2d": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "histogram2d" } ], "histogram2dcontour": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "histogram2dcontour" } ], "mesh3d": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "mesh3d" } ], "parcoords": [ { "line": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "parcoords" } ], "pie": [ { "automargin": true, "type": "pie" } ], "scatter": [ { "fillpattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 }, "type": "scatter" } ], "scatter3d": [ { "line": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatter3d" } ], "scattercarpet": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattercarpet" } ], "scattergeo": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattergeo" } ], "scattergl": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattergl" } ], "scattermapbox": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattermapbox" } ], "scatterpolar": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterpolar" } ], "scatterpolargl": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterpolargl" } ], "scatterternary": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterternary" } ], "surface": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "surface" } ], "table": [ { "cells": { "fill": { "color": "#EBF0F8" }, "line": { "color": "white" } }, "header": { "fill": { "color": "#C8D4E3" }, "line": { "color": "white" } }, "type": "table" } ] }, "layout": { "annotationdefaults": { "arrowcolor": "#2a3f5f", "arrowhead": 0, "arrowwidth": 1 }, "autotypenumbers": "strict", "coloraxis": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "colorscale": { "diverging": [ [ 0, "#8e0152" ], [ 0.1, "#c51b7d" ], [ 0.2, "#de77ae" ], [ 0.3, "#f1b6da" ], [ 0.4, "#fde0ef" ], [ 0.5, "#f7f7f7" ], [ 0.6, "#e6f5d0" ], [ 0.7, "#b8e186" ], [ 0.8, "#7fbc41" ], [ 0.9, "#4d9221" ], [ 1, "#276419" ] ], "sequential": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "sequentialminus": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ] }, "colorway": [ "#636efa", "#EF553B", "#00cc96", "#ab63fa", "#FFA15A", "#19d3f3", "#FF6692", "#B6E880", "#FF97FF", "#FECB52" ], "font": { "color": "#2a3f5f" }, "geo": { "bgcolor": "white", "lakecolor": "white", "landcolor": "#E5ECF6", "showlakes": true, "showland": true, "subunitcolor": "white" }, "hoverlabel": { "align": "left" }, "hovermode": "closest", "mapbox": { "style": "light" }, "paper_bgcolor": "white", "plot_bgcolor": "#E5ECF6", "polar": { "angularaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "bgcolor": "#E5ECF6", "radialaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" } }, "scene": { "xaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" }, "yaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" }, "zaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" } }, "shapedefaults": { "line": { "color": "#2a3f5f" } }, "ternary": { "aaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "baxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "bgcolor": "#E5ECF6", "caxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" } }, "title": { "x": 0.05 }, "xaxis": { "automargin": true, "gridcolor": "white", "linecolor": "white", "ticks": "", "title": { "standoff": 15 }, "zerolinecolor": "white", "zerolinewidth": 2 }, "yaxis": { "automargin": true, "gridcolor": "white", "linecolor": "white", "ticks": "", "title": { "standoff": 15 }, "zerolinecolor": "white", "zerolinewidth": 2 } } }, "xaxis": { "anchor": "y", "autorange": true, "constrain": "domain", "domain": [ 0.39065606361829025, 0.6093439363817097 ], "range": [ -0.5, 3.5 ], "scaleanchor": "y" }, "yaxis": { "anchor": "x", "autorange": true, "constrain": "domain", "domain": [ 0, 1 ], "range": [ 3.5, -0.5 ] } } }, "image/png": "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", "text/html": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "px.imshow(env.grid)" ] }, { "cell_type": "markdown", "id": "c5b6f8a3-f860-42c1-a726-aee6c894ab27", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "## Q-Learning" ] }, { "cell_type": "markdown", "id": "2f0cc10f-c9fc-499a-8eff-15f32df8985c", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "Wir implementieren als nächstes den Q-Learning Ansatz wie er oben beschrieben worden ist. Hierfür initialisieren wir als erstes die Q-Matrix, in welcher wir die erlernten Q-Werte speichern. Da unser Zustandsraum die Große 3x3 hat und wir 4 Aktionen haben, hat die Q-Matrix eine Größe von 3x3x4.\n", "\n", "Der Q-Learning Algorithmus besteht nun zum einen in der Möglichkeit mit der Wahrscheinlichkeit $\\epsilon$ ein explorativen zufälligen Schritt zu machen oder den vielversprechendsten Schritt unseres aktuellen Zustands $s_1,s_2$ mit dem maximalen Q-Wert (argmax)." ] }, { "cell_type": "code", "execution_count": 9, "id": "5282d9ab-cafa-46fc-ae7a-e64cf42df1d7", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "outputs": [], "source": [ "env = Gridworld(size=4, start=(0, 0), goal=(3, 3))\n", "\n", "# Initialisiere Q-Werte\n", "Q = np.zeros((env.size, env.size, len(env.actions)))\n", "\n", "# Hyperparameter\n", "alpha = 0.2 # Lernrate\n", "gamma = 0.9 # Diskontierungsfaktor\n", "epsilon = 0.1 # Epsilon für die Epsilon-Greedy-Strategie\n", "\n", "rewardsQ=[]\n", "retriesQ=[]\n", "for trial in range(trials):\n", " state = env.reset()\n", " done = False\n", " n, r_sum=0, 0\n", " while not done:\n", " if np.random.uniform(0,1) < epsilon:\n", " # Wähle Aktion entweder Zufällig aus\n", " action = np.random.randint(0, len(env.actions))\n", " else:\n", " # Wähle Aktion mit maximaler erwarteten Belohnung Q(s)\n", " action = np.argmax(Q[state[0], state[1], :])\n", " # Führe die Aktion aus und erhalte von der Umgebung den neuen Zustand und die Belohnung\n", " new_state, reward, done = env.step(env.actions[action])\n", " # Aktualisiere Q-Werte auf Basis der erhaltenen Belohnung\n", " Q[state[0], state[1], action] += alpha * (reward + gamma * np.max(Q[new_state[0], new_state[1], :]) - Q[state[0], state[1], action])\n", " # Aktualisiere den Zustand\n", " state = new_state\n", " # Sammel Erfolgsstatistik\n", " r_sum+=reward\n", " n += 1\n", " rewardsQ.append(r_sum)\n", " retriesQ.append(n)" ] }, { "cell_type": "markdown", "id": "0c240015-3459-4c16-b52c-1cbc7597b879", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "Wenn wir nun die Wiederholungsversuche uns ansehen, dann sehen wir, dass diese schnell auf das Minimum von nur 6 Schritte abfallen. Es gibt immer noch Schwankungen, die an dem Zufallsanteil $\\epsilon$ liegen." ] }, { "cell_type": "code", "execution_count": 10, "id": "ee4fb42b-38a9-42d4-8fb6-9553afa27d01", "metadata": {}, "outputs": [ { "data": { "application/vnd.plotly.v1+json": { "config": { "plotlyServerURL": "https://plot.ly" }, "data": [ { "hovertemplate": "variable=0
index=%{x}
value=%{y}", "legendgroup": "0", "line": { "color": "#636efa", "dash": "solid" }, "marker": { "symbol": "circle" }, "mode": "lines", "name": "0", "orientation": "v", "showlegend": true, "type": "scatter", "x": [ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 191, 192, 193, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203, 204, 205, 206, 207, 208, 209, 210, 211, 212, 213, 214, 215, 216, 217, 218, 219, 220, 221, 222, 223, 224, 225, 226, 227, 228, 229, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 285, 286, 287, 288, 289, 290, 291, 292, 293, 294, 295, 296, 297, 298, 299, 300, 301, 302, 303, 304, 305, 306, 307, 308, 309, 310, 311, 312, 313, 314, 315, 316, 317, 318, 319, 320, 321, 322, 323, 324, 325, 326, 327, 328, 329, 330, 331, 332, 333, 334, 335, 336, 337, 338, 339, 340, 341, 342, 343, 344, 345, 346, 347, 348, 349, 350, 351, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 473, 474, 475, 476, 477, 478, 479, 480, 481, 482, 483, 484, 485, 486, 487, 488, 489, 490, 491, 492, 493, 494, 495, 496, 497, 498, 499 ], "xaxis": "x", "y": [ 70, 40, 7, 38, 8, 9, 12, 14, 13, 7, 8, 17, 7, 6, 11, 7, 10, 7, 10, 8, 6, 7, 7, 7, 7, 6, 6, 6, 6, 8, 6, 6, 8, 8, 6, 6, 6, 6, 7, 6, 6, 6, 11, 6, 6, 8, 6, 8, 6, 8, 10, 6, 7, 11, 6, 8, 6, 6, 6, 6, 6, 10, 6, 8, 8, 8, 6, 7, 6, 8, 6, 6, 6, 6, 6, 6, 6, 8, 6, 10, 6, 7, 6, 11, 6, 6, 8, 6, 6, 6, 6, 6, 6, 6, 6, 8, 8, 6, 10, 6, 8, 6, 6, 6, 6, 8, 7, 6, 6, 7, 8, 8, 7, 6, 6, 6, 8, 8, 6, 6, 10, 8, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 8, 6, 6, 6, 6, 6, 8, 6, 10, 6, 6, 6, 9, 6, 6, 6, 6, 7, 8, 7, 6, 6, 6, 6, 6, 6, 6, 8, 6, 8, 6, 7, 6, 10, 8, 6, 6, 8, 6, 8, 8, 8, 6, 6, 6, 6, 6, 6, 10, 8, 6, 6, 8, 7, 6, 7, 7, 6, 8, 6, 6, 9, 8, 8, 6, 7, 7, 10, 6, 8, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 7, 8, 10, 10, 6, 8, 6, 6, 6, 6, 6, 6, 6, 8, 7, 6, 6, 6, 6, 6, 6, 6, 9, 11, 6, 6, 6, 9, 6, 6, 6, 6, 6, 6, 6, 7, 6, 8, 6, 7, 6, 8, 6, 8, 6, 6, 7, 6, 6, 8, 6, 6, 6, 6, 6, 6, 6, 6, 8, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 8, 6, 6, 6, 10, 6, 6, 6, 11, 8, 6, 6, 6, 6, 6, 7, 6, 6, 6, 6, 6, 6, 8, 6, 6, 6, 8, 8, 6, 8, 8, 7, 6, 6, 6, 6, 6, 10, 8, 6, 6, 6, 7, 6, 8, 6, 6, 6, 8, 8, 8, 6, 6, 6, 8, 10, 6, 8, 6, 6, 6, 6, 8, 6, 6, 6, 8, 10, 6, 6, 6, 6, 9, 11, 10, 6, 6, 7, 6, 8, 8, 7, 8, 7, 8, 6, 6, 6, 7, 6, 6, 8, 6, 6, 8, 9, 7, 8, 8, 6, 7, 6, 6, 6, 6, 7, 6, 6, 8, 10, 7, 6, 6, 6, 6, 6, 6, 6, 8, 8, 6, 6, 6, 6, 6, 6, 6, 10, 8, 6, 6, 8, 6, 10, 6, 8, 6, 8, 6, 10, 6, 7, 6, 6, 6, 8, 6, 9, 6, 6, 6, 7, 7, 8, 8, 8, 8, 6, 6, 6, 6, 6, 6, 8, 6, 6, 6, 6, 6, 6, 6, 6, 6, 9, 6, 6, 8, 6, 6, 8, 6, 6, 6, 6, 6, 6, 6, 6, 6, 8, 6, 6, 6, 9, 6, 8, 6, 6, 6, 6, 6, 7, 8, 6, 6, 6, 6, 7 ], "yaxis": "y" } ], "layout": { "autosize": true, "legend": { "title": { "text": "variable" }, "tracegroupgap": 0 }, "margin": { "t": 60 }, "template": { "data": { "bar": [ { "error_x": { "color": "#2a3f5f" }, "error_y": { "color": "#2a3f5f" }, "marker": { "line": { "color": "#E5ECF6", "width": 0.5 }, "pattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 } }, "type": "bar" } ], "barpolar": [ { "marker": { "line": { "color": "#E5ECF6", "width": 0.5 }, "pattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 } }, "type": "barpolar" } ], "carpet": [ { "aaxis": { "endlinecolor": "#2a3f5f", "gridcolor": "white", "linecolor": "white", "minorgridcolor": "white", "startlinecolor": "#2a3f5f" }, "baxis": { "endlinecolor": "#2a3f5f", "gridcolor": "white", "linecolor": "white", "minorgridcolor": "white", "startlinecolor": "#2a3f5f" }, "type": "carpet" } ], "choropleth": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "choropleth" } ], "contour": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "contour" } ], "contourcarpet": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "contourcarpet" } ], "heatmap": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "heatmap" } ], "heatmapgl": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "heatmapgl" } ], "histogram": [ { "marker": { "pattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 } }, "type": "histogram" } ], "histogram2d": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "histogram2d" } ], "histogram2dcontour": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "histogram2dcontour" } ], "mesh3d": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "mesh3d" } ], "parcoords": [ { "line": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "parcoords" } ], "pie": [ { "automargin": true, "type": "pie" } ], "scatter": [ { "fillpattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 }, "type": "scatter" } ], "scatter3d": [ { "line": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatter3d" } ], "scattercarpet": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattercarpet" } ], "scattergeo": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattergeo" } ], "scattergl": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattergl" } ], "scattermapbox": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattermapbox" } ], "scatterpolar": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterpolar" } ], "scatterpolargl": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterpolargl" } ], "scatterternary": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterternary" } ], "surface": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "surface" } ], "table": [ { "cells": { "fill": { "color": "#EBF0F8" }, "line": { "color": "white" } }, "header": { "fill": { "color": "#C8D4E3" }, "line": { "color": "white" } }, "type": "table" } ] }, "layout": { "annotationdefaults": { "arrowcolor": "#2a3f5f", "arrowhead": 0, "arrowwidth": 1 }, "autotypenumbers": "strict", "coloraxis": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "colorscale": { "diverging": [ [ 0, "#8e0152" ], [ 0.1, "#c51b7d" ], [ 0.2, "#de77ae" ], [ 0.3, "#f1b6da" ], [ 0.4, "#fde0ef" ], [ 0.5, "#f7f7f7" ], [ 0.6, "#e6f5d0" ], [ 0.7, "#b8e186" ], [ 0.8, "#7fbc41" ], [ 0.9, "#4d9221" ], [ 1, "#276419" ] ], "sequential": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "sequentialminus": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ] }, "colorway": [ "#636efa", "#EF553B", "#00cc96", "#ab63fa", "#FFA15A", "#19d3f3", "#FF6692", "#B6E880", "#FF97FF", "#FECB52" ], "font": { "color": "#2a3f5f" }, "geo": { "bgcolor": "white", "lakecolor": "white", "landcolor": "#E5ECF6", "showlakes": true, "showland": true, "subunitcolor": "white" }, "hoverlabel": { "align": "left" }, "hovermode": "closest", "mapbox": { "style": "light" }, "paper_bgcolor": "white", "plot_bgcolor": "#E5ECF6", "polar": { "angularaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "bgcolor": "#E5ECF6", "radialaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" } }, "scene": { "xaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" }, "yaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" }, "zaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" } }, "shapedefaults": { "line": { "color": "#2a3f5f" } }, "ternary": { "aaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "baxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "bgcolor": "#E5ECF6", "caxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" } }, "title": { "x": 0.05 }, "xaxis": { "automargin": true, "gridcolor": "white", "linecolor": "white", "ticks": "", "title": { "standoff": 15 }, "zerolinecolor": "white", "zerolinewidth": 2 }, "yaxis": { "automargin": true, "gridcolor": "white", "linecolor": "white", "ticks": "", "title": { "standoff": 15 }, "zerolinecolor": "white", "zerolinewidth": 2 } } }, "xaxis": { "anchor": "y", "autorange": true, "domain": [ 0, 1 ], "range": [ 0, 499 ], "title": { "text": "index" }, "type": "linear" }, "yaxis": { "anchor": "x", "autorange": true, "domain": [ 0, 1 ], "range": [ 2.444444444444444, 73.55555555555556 ], "title": { "text": "value" }, "type": "linear" } } }, "image/png": "iVBORw0KGgoAAAANSUhEUgAABLIAAAFoCAYAAABDgdIXAAAgAElEQVR4XuydCZgdVZm/v+5OyMISErZEFII6A+7iqIAzIgQ3An8VHBZRkUWHxQEEAo4gOy5sbiyiI6AoKi4sso0LgriB4+g4OAIyyqLIDglLQpbu/te5zemcVO7tW1Xn+849nbz3eeYZSdf5nXPfr27Vqfeeqts3XLyEFwQgAAEIQAACEIAABCAAAQhAAAIQgAAEMifQh8jKvEIMDwIQgAAEIAABCEAAAhCAAAQgAAEIQKBFAJHFjgABCEAAAhCAAAQgAAEIQAACEIAABCAwLgggssZFmRgkBCAAAQhAAAIQgAAEIAABCEAAAhCAACKLfQACEIAABCAAAQhAAAIQgAAEIAABCEBgXBBAZI2LMjFICEAAAhCAAAQgAAEIQAACEIAABCAAAUQW+wAEIAABCEAAAhCAAAQgAAEIQAACEIDAuCCAyBoXZWKQEIAABCAAAQhAAAIQgAAEIAABCEAAAogs9gEIQAACEIAABCAAAQhAAAIQgAAEIACBcUEAkTUuysQgIQABCEAAAhCAAAQgAAEIQAACEIAABBBZ7AMQgAAEIAABCEAAAhCAAAQgAAEIQAAC44IAImtclIlBQgACEIAABCAAAQhAAAIQgAAEIAABCCCy2AcgAAEIQAACEIAABCAAAQhAAAIQgAAExgUBRNa4KBODhAAEIAABCEAAAhCAAAQgAAEIQAACEEBksQ9AAAIQgAAEIAABCEAAAhCAAAQgAAEIjAsCiKxxUSYGCQEIQAACEIAABCAAAQhAAAIQgAAEIIDIYh+AAAQgAAEIQAACEIAABCAAAQhAAAIQGBcEEFnjokwMEgIQgAAEIAABCEAAAhCAAAQgAAEIQACRxT4AAQhAAAIQgAAEIAABCEAAAhCAAAQgMC4IILLGRZkYJAQgAAEIQAACEIAABCAAAQhAAAIQgAAii30AAhCAAAQgAAEIQAACEIAABCAAAQhAYFwQQGSNizIxSAhAAAIQgAAEIAABCEAAAhCAAAQgAAFEFvsABCAAAQhAAAIQgAAEIAABCEAAAhCAwLgggMgaF2VikBCAAAQgAAEIQAACEIAABCAAAQhAAAKILPYBCEAAAhCAAAQgAAEIQAACEIAABCAAgXFBAJE1LsrEICEAAQhAAAIQgAAEIAABCEAAAhCAAAQQWewDEIAABCAAAQhAAAIQgAAEIAABCEAAAuOCACJrXJSJQUIAAhCAAAQgAAEIQAACEIAABCAAAQggstgHIAABCEAAAhCAAAQgAAEIQAACEIAABMYFAUTWuCgTg4QABCAAAQhAAAIQgAAEIAABCEAAAhBAZLEPQAACEIAABCAAAQhAAAIQgAAEIAABCIwLAoiscVEmBgkBCEAAAhCAAAQgAAEIQAACEIAABCCAyGIfgAAEIAABCEAAAhCAAAQgAAEIQAACEBgXBBBZ46JMDBICEIAABCAAAQhAAAIQgAAEIAABCEAAkcU+AAEIQAACEIAABCAAAQhAAAIQgAAEIDAuCCCyxkWZGCQEIAABCEAAAhCAAAQgAAEIQAACEIAAIot9AAIQgAAEIAABCEAAAhCAAAQgAAEIQGBcEEBkjYsyMUgIQAACEIAABCAAAQhAAAIQgAAEIAABRBb7AAQgAAEIQAACEIAABCAAAQhAAAIQgMC4IIDIGhdlYpAQgAAEIAABCEAAAhCAAAQgAAEIQAACiCz2AQhAAAIQgAAEIAABCEAAAhCAAAQgAIFxQQCRNS7KxCAhAAEIQAACEIAABCAAAQhAAAIQgAAEEFnsAxCAAAQgAAEIQAACEIAABCAAAQhAAALjggAia1yUiUFCAAIQgAAEIAABCEAAAhCAAAQgAAEIILLYByAAAQhAAAIQgAAEIAABCEAAAhCAAATGBQFE1rgoE4OEAAQgAAEIQAACEIAABCAAAQhAAAIQQGSxD0AAAhCAAAQgAAEIQAACEIAABCAAAQiMCwKIrHFRJgYJAQhAAAIQgAAEIAABCEAAAhCAAAQggMhiH4AABCAAAQhAAAIQgAAEIAABCEAAAhAYFwQQWeOiTAwSAhCAAAQgAAEIQAACEIAABCAAAQhAAJHFPgABCEAAAhCAAAQgAAEIQAACEIAABCAwLgggssZFmRgkBCAAAQhAAAIQgAAEIAABCEAAAhCAACKLfQACEIAABCAAAQhAAAIQgAAEIAABCEBgXBBAZI2LMjFICEAAAhCAAAQgAAEIQAACEIAABCAAAUQW+wAEIAABCEAAAhCAAAQgAAEIQAACEIDAuCCAyBoXZWKQEIAABCAAAQhAAAIQgAAEIAABCEAAAogs9gEIQAACEIAABCAAAQhAAAIQgAAEIACBcUEAkTUuysQgIQABCEAAAhCAAAQgAAEIQAACEIAABBBZ7AMQgAAEIAABCEAAAhCAAAQgAAEIQAAC44IAImtclIlBQgACEIAABCAAAQhAAAIQgAAEIAABCCCyFPaBvz26SCGFCE0CM9ZeQxY+s0yeWTqkGUuWAoGNpk+WhxcslqGhYYU0IrQI9PeJbDh9ijzwGMczLaZaOZMm9staUybKo08s1ookR4nAWlMmSF9fnzy5cKlSIjFaBFrzgMWD8sySQa1IcpQIuHnAI8U8YJB5gBJRnZjiUCYzi3nA/cwDdIAqpqwxoV/WWXNi63PDKx8Cz1lvSj6DWQ1HgshSKDoiSwGicgQiSxmoYhwiSxGmYhQiSxGmchQiSxmoYhwiSxGmchQiSxmoYhwiSxGmYhQiSxGmchQiSxmoUhwiSwlkwxhEVkNwYTNElgJE5QhEljJQxThEliJMxShEliJM5ShEljJQxThEliJM5ShEljJQxThEliJMxShEliJM5ShEljJQpThElhLIhjGIrIbgEFkK4AwjEFmGcCOjEVmRAI2aI7KMwCrEIrIUIBpFILKMwCrEIrIUIBpFILKMwEbGIrIiARo2R2QZwo2IRmRFwFNoishSgMiKLAWIyhGILGWginGILEWYilGILEWYylGILGWginGILEWYylGILGWginGILEWYilGILEWYylGILGWgSnGILCWQDWMQWQ3Bhc0QWQoQlSMQWcpAFeMQWYowFaMQWYowlaMQWcpAFeMQWYowlaMQWcpAFeMQWYowFaMQWYowlaMQWcpAleIQWUogG8YgshqCQ2QpgDOMQGQZwo2MRmRFAjRqjsgyAqsQi8hSgGgUgcgyAqsQi8hSgGgUgcgyAhsZi8iKBGjYHJFlCDciGpEVAU+hKSJLASIrshQgKkcgspSBKsYhshRhKkYhshRhKkchspSBKsYhshRhKkchspSBKsYhshRhKkYhshRhKkchspSBKsWNV5F12bU3ybrrrCVz/ulVlUh84atXyc9+dat89exj2m6/cNFiece+x8qRB+4ub9nutZUyNTZCZClQRGQpQFSOQGQpA1WMQ2QpwlSMQmQpwlSOQmQpA1WMQ2QpwlSOQmQpA1WMQ2QpwlSMQmQpwlSOQmQpA1WKG68i6zU7HiDP3+Q5cukXTqhE4qOnXSA/vOnXcss1n2+7/dMLn5HXzj1Qjjn0PfLuXd9YKVNjI0SWAkVElgJE5QhEljJQxThEliJMxShEliJM5ShEljJQxThEliJM5ShEljJQxThEliJMxShEliJM5ShEljJQpbjxKrLmL3hK+gf6ZZ21plYigciqhGl8boTIyq9uiKz8auJHhMjKszaIrDzr4kaFyMq3NoisfGuDyMq3NoisPGuDyMqzLm5UiKw8a2MlstwKp93+5QR5505vkP3fNXf0zd9511/l0I9+rlj59F7ZbJOZcuCHPyX3PfCILFmyVKZOmSQ77bCNHHvYe2TixAmtNgd/5NPywtkby4v/frZ8+6ob5cFHHpeLP3eMnHjWRfK852woRx20Z2u79x32Cbn9/+6Vp55eJGusMVFe/fLN5cR5+8jGM9dv/d2JrGuuv1n22f2t8p2rb5TH5j8pL3vR8+VzpxwqG66/rrRbkbXgiaflpE99RX56y+/E3Xr4gmIcpxy9n7zixS9QKyYrsiJRvv+wpXLy8csiU2iuTQCRpU1ULw+RpcdSMwmRpUlTNwuRpctTMw2RpUlTNwuRpctTMw2RpUlTLwuRpcdSOwmRpU1UJ89KZLnR7XXwKeLE1S3XnC/9bpJevE4888ty2XU3yc1Xnyf3P/hoIaS+Ittu/XLZYL115Q9/vFsuuexHLdl01MEjgmrObofLgw8/3vrfL918MxkoVmGd/bHD5F0HnSwv3GxjOe8Th7f+5vp69Ss2L+TYLHnksQVy3leulE033kiuuOjU1t+dyLr8up/KtHXWlF3e+nr545//Kr/49e9l8xc8Ty674JSVRNbg4JC85V3z5MlCjLlbDdebPk0u+ua18tCj8+Wnl5/dytF4IbIiKSKyIgEaNUdkGYFViEVkKUA0iEBkGUBVikRkKYE0iEFkGUBVikRkKYE0iEFkGUBViERkKUA0ikBkGYGNjLUUWe6ZVB86/hw59+Mfku1e90pZunSZvHang4qHqb9GPnnMv6ww8ieeWiiPFgLKrcCaOmWyfPdLJ7f+7kTWhIEBueTcj7Zkl3+9ec95K4gs/++Lnlkijz6+QM696Ar53g9+Lv9z/YUt+eVXZN1SCDS3Ysu9Tv3MV+UbV1wv111yeiGq1lnhGVlX/+iX8uFTvyAXfOpo2fpVL25t//s77pI9DjhJPvZv75d3vPWfIsmPNEdkRWJEZEUCNGqOyDICqxCLyFKAaBCByDKAqhSJyFICaRCDyDKAqhSJyFICaRCDyDKAqhCJyFKAaBSByDICGxlrKbLcqqatdz5IXl7cinfBWUe3bu07+pTzW5JqixduIkuXDcoZ532zWCl1U+vWPf/a9LkbybVfO631n05kvXSLzVq3AIavssj6ztU/kbMvvKy1Git8/fYH/94SV+2ekfUfN/xKjjzpPPniGfPklS954Qoi65PnfF2++p0fyEYbTB+Nc+/H5R+499vkkP12jSQ/0hyRFYkRkRUJ0Kg5IssIrEIsIksBokEEIssAqlIkIksJpEEMIssAqlIkIksJpEEMIssAqkIkIksBolEEIssIbGSspchyQzvz/EuLW/Kuk59c9ln512M/K088+fSopDrprC/Lt4rnXu2351x5y/avkefO3ECOKkTXfQ88XEtk+ZVfr9/qZUXWTq1nb33vB7+QT33hWzKWyLrux7fIvJM/35Js7nlZ4a8WnnDmRcWztH4iZ51w8EqE/+75z5UXbPqcSPKILBWAiCwVjOohiCx1pGqBiCw1lKpBiCxVnKphiCxVnKphiCxVnKphiCxVnKphiCxVnGphiCw1lOpBiCx1pCqB1iLLPQfrjXscKbvs+PrWM6rcA9h323m71tjfuPsRss7aa7aeUeVfBxx9lvzlbw/VElleOv33D780+pD4i7/9fTnt3G+MKbL+7eNflKsK4XXjdz/Tup0xFFmXXPZD+fjnLpFvf/HE1oPmw9fw8LD0uYONwosVWZEQEVmRAI2aI7KMwCrEIrIUIBpEILIMoCpFIrKUQBrEILIMoCpFIrKUQBrEILIMoCpEIrIUIBpFILKMwEbGWossN7z3HvJx+c2tf2w9q8o9+H3K5DVao3a3+7nnWJ1y9P6tW/i+f+N/yre+d4PUvbXw2utvKVZyfV4+8O6d5fVbvbzVl3vYu/slxHBFluvrjOMOktnPmylX/fAX4mTXa16xRes5WOVfLXTP7NqhuK3R/XricR/au/U8rv+76z75+uXXyx5v3152fuM2keRHmiOyIjE6kbXf3kMye/ZQZBLNNQkgsjRp6mYhsnR5aqUhsrRI6ucgsvSZaiUisrRI6ucgsvSZaiUisrRI6uYgsnR5aqYhsjRp6mWlEFk//tlv5JCPfk52nbttIa32Gx383X95oPVw93v++mDr3zaeub4MDQ3J5MmT5OqLP9H6N7dq68Wbzx7zGVnLBgflgKPOkpt/84dWG/dMLPdrhLfe9udRkXXc6RfKZdfe1JJp7llX7uVWWp1/2hGtB70vXPSMvGbHA+XYw94re+2yQ+vvt//fvXJ08cD3P9193+iYnXD77MmHtG5F1HghsiIpIrIiARo1R2QZgVWIRWQpQDSIQGQZQFWKRGQpgTSIQWQZQFWKRGQpgTSIQWQZQFWIRGQpQDSKQGQZgY2MTSGyug3x3vseagkmJ7JiXo8+/oTMX/BkseJqViuv3cut0rrnvgdl2tpryYbrL/8VxLH6dQ+if/Dhx2S9GdNknbWmxgxxpbaIrEiciKxIgEbNEVlGYBViEVkKEA0iEFkGUJUiEVlKIA1iEFkGUJUiEVlKIA1iEFkGUBUiEVkKEI0iEFlGYCNjcxBZkW9hXDdHZEWWD5EVCdCoOSLLCKxCLCJLAaJBBCLLAKpSJCJLCaRBDCLLAKpSJCJLCaRBDCLLAKpCJCJLAaJRBCLLCGxkLCIrEmBkc0RWJEBEViRAo+aILCOwCrGILAWIBhGILAOoSpGILCWQBjGILAOoSpGILCWQBjGILAOoCpGILAWIRhGILCOwkbGIrEiAkc0RWZEAEVmRAI2aI7KMwCrEIrIUIBpEILIMoCpFIrKUQBrEILIMoCpFIrKUQBrEILIMoCpEIrIUIBpFILKMwEbGIrIiAUY2R2RFAkRkRQI0ao7IMgKrEIvIUoBoEIHIMoCqFInIUgJpEIPIMoCqFInIUgJpEIPIMoCqEInIUoBoFIHIMgIbGYvIigQY2RyRFQkQkRUJ0Kg5IssIrEIsIksBokEEIssAqlIkIksJpEEMIssAqlIkIksJpEEMIssAqkIkIksBolEEIssIbGQsIisSYGRzRFYkQERWJECj5ogsI7AKsYgsBYgGEYgsA6hKkYgsJZAGMYgsA6hKkYgsJZAGMYgsA6gKkYgsBYhGEYgsI7CRsYisSICRzRFZkQARWZEAjZojsozAKsQishQgGkQgsgygKkUispRAGsQgsgygKkUispRAGsQgsgygKkQishQgGkUgsozARsYisiIBRjZHZEUCRGRFAjRqjsgyAqsQi8hSgGgQgcgygKoUichSAmkQg8gygKoUichSAmkQg8gygKoQichSgGgUgcgyAhsZuzqLrMVLlsojjy2Q52y0nvS5g0cPXoisSOiIrEiARs0RWUZgFWIRWQoQDSIQWQZQlSIRWUogDWIQWQZQlSIRWUogDWIQWQZQFSIRWQoQjSIQWUZgI2NXR5E1PDwsnzj7Ernksh+16K2xxkQ5/7QjZKstXxRJs35zRFZ9Ziu0cCJru20HZc52w5FJNNckgMjSpKmbhcjS5amVhsjSIqmfg8jSZ6qViMjSIqmfg8jSZ6qViMjSIqmbg8jS5amZhsjSpKmXtTqKrJv/6w+y/5GnyxfPmCevecXmcvKnL5bv3/grueWa86XfXUwkfCGyImEjsiIBGjVHZBmBVYhFZClANIhAZBlAVYpEZCmBNIhBZBlAVYpEZCmBNIhBZBlAVYhEZClANIpAZBmBjYxdHUXWR0+7QH5/+11yxUWntujd/+Cj8sY9jpSvnXOsbPnSv4skWq85Iqser5W2RmRFAjRqjsgyAqsQi8hSgGgQgcgygKoUichSAmkQg8gygKoUichSAmkQg8gygKoQichSgGgUgcgyAhsZm0JkPfqYyCOPpr/za70ZfbL+eisD2vfwT8r0aevIp048ePSPL9luHznjuINk7g5bRRKt1xyRVY8XIiuSV6rmiKxUpOv3g8iqzyxFC0RWCsrN+kBkNeOWohUiKwXlZn0gsppxS9EKkZWCcv0+EFn1maVqgchKRbpePylE1tU/GJIrrhmsNzCFrXd6c7/sstPASkm77n+cvPjvZ8upH95/9G8v32E/OfbQ98geb5+j0HP1CERWdVZtt2RFViRAo+aILCOwCrGILAWIBhGILAOoSpGILCWQBjGILAOoSpGILCWQBjGILAOoCpGILAWIRhGILCOwkbEpRNYv/3NIfnbzUORI6zff+tX98vpt+ldq6FZkzVh3HTnrBFZk1aeaWQtEVmYFeXY4iKw86+JGhcjKszaIrDzr4kaFyMq3NoisfGuDyMq3NoisPGuDyMqzLm5UiKw8a5NCZOX2zt0zsv7wx7vlsgtOaQ3tbw88Im/acx7PyMqtUFXGg8iqQin9Nois9Myr9ojIqkoq7XaIrLS86/SGyKpDK+22iKy0vOv0hsiqQyvttoistLyr9obIqkoq/XaIrPTMq/S4OoqsX/76f+X9885o/Wrha1+5hZxw5kXyw5t+za8WVtlhrLd56JH5rS42XH/dFbpa8MTTsnTZMll/xrQV/h2RZV2RZvmIrGbcUrRCZKWgXL8PRFZ9ZqlaILJSka7fDyKrPrNULRBZqUjX7weRVZ9ZihaIrBSUm/WByGrGzbrV6iiyhoeH5eRPXyzf+t4NLbwDA/3yhdOOlG1e/RJr3Cvl84ysAsnQ0LCcdf6l8vUrrpclS5a2CvI/11/YgvXU04ta1vHW2/7c+u+NZ64vF599jMzcYEbrvxFZyffZSh0isiph6slGiKyeYO/aKSKrK6KebYDI6hn6rh0jsroi6tkGiKyeoe/aMSKrK6KebIDI6gn2Sp0isiphSr7R6iiyPOSFixbLI4/Nl+fO2lD63UVED16IrAK6u9fzmutvlg+8e2fZbec3yOJCZj131gatcpxZCK5vX3WjXF7cB7rm1Cmy50EnyWabzJLzPnE4IqsHO2zVLhFZVUml3w6RlZ55lR4RWVUo9WYbRFZvuFfpFZFVhVJvtkFk9YZ7lV4RWVUopd8GkZWeedUeEVlVSaXdbnUWWWlJt+9ttRdZ9z/0mLxx9yPk6A++S96321tWojRnt8NlxzlbyVEH7dn623evuUmOP+NC+f0NF0lfccRnRVYOu/HKY0Bk5VkXNypEVp61QWTlWRc3KkRWvrVBZOVbG0RWvrVBZOVZG0RWnnVxo0Jk5VkbRFZv67Lai6wrv/9zOeYT/y5vfsOr5Y4//UXWmDhRdvt/b5B37/qmVmVevsN+cuKR+8iuc7dt/fdvbr1T3nvIx+RnV54t06et3RJZW792SOa+Nf3PYvZ218m7d0RWvvVBZOVZG0RWnnVBZOVbFzcyRFa+9UFk5VsbRFaetUFk5VkXRFa+dUFk9bY2q73IOu8rV8q5F13eElcv22IzufX2u+SSy34oxx++t+z+tu3lpdvvK2ccd5DM3WGrVqVuu/Me+ecPnCDXXXK6bLLxhi2R9YLNRA45oDf3hvZ298m398lr9MuyZcOyrHj+Ga+8CEydPCCLFg9K8azAcf9yb2FV+eS7CezUSRPk6WeWjfu6rGpvYKCwjO7b2EVLBle1tzbu38/ECX2t1dlLlvJlVm7FbM0DBot5QPF/vPIi4OYBzxTzgPE+RVuV5gBuD2EekNfnJBxNax4wsZgHFJ8bXvkQcF9m8eodAURWIbK+cfmP5KdXnD1ahQ/MO1MWPbNYvnbOsa0VWSfN21d22fH1rb+3W5H1/Nki/3pA74pIzysTcBfkS5YNFRNYLi5y2z/WnjJRnipkifvVi/H+cm/BTfxWhZd7G2tNnShPLly6KrydVeo9TCh+gMTdXohkzK+skyYOtAa1eCkXF7lVh3lAbhVZPh43D3DHs6FxPg9YleYAvjrrFPOAJ5gHZPfhGejvFyfnmQfkVRr3eeHVOwKrvci69vpb5KhTPi///cMvycSJI1Z138M/KQsXLpZLv3CCuGdkzd1ha5l34B6tv33n6p/ICWdetMIzsjbdZFj234dJbO9245V75tbCnKqx4li4tTDP2nBrYZ51caPiGVn51oZbC/OtDbcW5lsbbi3MszbcWphnXdyoeEZWnrXh1sLe1mW1F1nzFzwl2+56aPFrhdvJMYe+R/7zd7fL/kecLofst6scuPfb5IzPf7Mlr6648FSZOnWy7Hngyr9aiMjq7U7crndEVn418SNCZOVZG0RWnnVBZOVbFzcyRFa+9UFk5VsbRFaetUFk5VkXRFa+dUFk9bY2q73Icviv/+lv5PATz5HBZ29De8t2r5XTjztAJgwMyBNPLWyJrT/88e5WpWZtOEO+evaxMmuj9Vr/7Z6Rhcjq7U6MyMqP/1gjQmTlWS9EVp51QWTlWxdEVt61QWTlWx9EVp61QWTlWRdEVr51QWT1tjaIrGf5LxsclL/+7WFZf8Y0WWvNKStV5bH5TxYPc10qMzeYscLfEFm93YE79c6KrDzr4kaFyMqzNoisPOuCyMq3LoisvGuDyMq3PoisPGuDyMqzLoisfOuCyOptbRBZkfwRWZEAjZojsozAKsQishQgGkQgsgygKkXyjCwlkAYx3FpoAFUpEpGlBNIgBpFlAFUhEpGlANEogmdkGYGNjEVkRQKMbI7IigSIyIoEaNQckWUEViEWkaUA0SACkWUAVSkSkaUE0iAGkWUAVSkSkaUE0iAGkWUAVSESkaUA0SgCkWUENjIWkRUJMLI5IisSICIrEqBRc0SWEViFWESWAkSDCESWAVSlSESWEkiDGESWAVSlSESWEkiDGESWAVSFSESWAkSjCESWEdjIWERWJMDI5oisSICIrEiARs0RWUZgFWIRWQoQDSIQWQZQlSIRWUogDWIQWQZQlSIRWUogDWIQWQZQFSIRWQoQjSIQWUZgI2MRWZEAI5sjsiIBIrIiARo1R2QZgVWIRWQpQDSIQGQZQFWKRGQpgTSIQWQZQFWKRGQpgTSIQWQZQFWIRGQpQDSKQGQZgY2MRWRFAoxsjsiKBOhE1syNhuXgAwYjk2iuSQCRpUlTNwuRpctTKw2RpUVSPweRpc9UKxGRpUVSPweRpc9UKxGRpUVSNweRpctTMw2RpUlTLwuRpceySRIiqwm1oI0TWe518vHLIpNorkkAkaVJUzcLkaXLUysNkaVFUj8HkaXPVCsRkaVFUj8HkaXPVCsRkaVFUjcHkaXLUzMNkaVJU7MK17oAACAASURBVC8LkaXHskkSIqsJNURWJDX75ogse8ZNe0BkNSVn2w6RZcs3Jh2RFUPPti0iy5ZvTDoiK4aebVtEli3fpumIrKbk7NshsuwZN+kBkdWEml4bRFYkS1ZkRQI0ao7IMgKrEIvIUoBoEIHIMoCqFInIUgJpEIPIMoCqFInIUgJpEIPIMoCqEInIUoBoFIHIMgIbGYvIigQY2RyRFQkQkRUJ0Kg5IssIrEIsIksBokEEIssAqlIkIksJpEEMIssAqlIkIksJpEEMIssAqkIkIksBolEEIssIbGQsIisSYGRzRFYkQERWJECj5ogsI7AKsYgsBYgGEYgsA6hKkYgsJZAGMYgsA6hKkYgsJZAGMYgsA6gKkYgsBYhGEYgsI7CRsYisSICRzRFZkQARWZEAjZojsozAKsQishQgGkQgsgygKkUispRAGsQgsgygKkUispRAGsQgsgygKkQishQgGkUgsozARsYisiIBRjZHZEUCRGRFAjRqjsgyAqsQi8hSgGgQgcgygKoUichSAmkQg8gygKoUichSAmkQg8gygKoQichSgGgUgcgyAhsZi8iKBBjZHJEVCRCRFQnQqDkiywisQiwiSwGiQQQiywCqUiQiSwmkQQwiywCqUiQiSwmkQQwiywCqQiQiSwGiUQQiywhsZCwiKxJgZHNEViRARFYkQKPmiCwjsAqxiCwFiAYRiCwDqEqRiCwlkAYxiCwDqEqRiCwlkAYxiCwDqAqRiCwFiEYRiCwjsJGxiKxIgJHNEVmRABFZkQCNmiOyjMAqxCKyFCAaRCCyDKAqRSKylEAaxCCyDKAqRSKylEAaxCCyDKAqRCKyFCAaRSCyjMBGxiKyIgFGNkdkRQJEZEUCNGqOyDICqxCLyFKAaBCByDKAqhSJyFICaRCDyDKAqhSJyFICaRCDyDKAqhCJyFKAaBSByDICGxmLyIoEGNkckRUJ8MTTlslf/zYshx82KNOnDUem0VyLACJLi6R+DiJLn6lGIiJLg6JNBiLLhqtGKiJLg6JNBiLLhqtGKiJLg6J+BiJLn6lWIiJLi6RuDiJLl2fdNERWXWKl7U//3DL545+GZb+9h2T27KHINJprEUBkaZHUz0Fk6TPVSERkaVC0yUBk2XDVSEVkaVC0yUBk2XDVSEVkaVDUz0Bk6TPVSkRkaZHUzUFk6fKsm4bIqksMkRVJLE1zRFYazk16QWQ1oWbfBpFlz7hpD4ispuTs2yGy7Bk37QGR1ZScfTtElj3jJj0gsppQS9MGkZWGc91eEFl1ieluj8iK5MmKrEiARs0RWUZgFWIRWQoQDSIQWQZQlSIRWUogDWIQWQZQlSIRWUogDWIQWQZQFSIRWQoQjSIQWUZgI2MRWZEAI5sjsiIBIrIiARo1R2QZgVWIRWQpQDSIQGQZQFWKRGQpgTSIQWQZQFWKRGQpgTSIQWQZQFWIRGQpQDSKQGQZgY2MRWRFAoxsjsiKBIjIigRo1ByRZQRWIRaRpQDRIAKRZQBVKRKRpQTSIAaRZQBVKRKRpQTSIAaRZQBVIRKRpQDRKAKRZQQ2MhaRFQkwsjkiKxIgIisSoFFzRJYRWIVYRJYCRIMIRJYBVKVIRJYSSIMYRJYBVKVIRJYSSIMYRJYBVIVIRJYCRKMIRJYR2MhYRFYkwMjmiKxIgIisSIBGzRFZRmAVYhFZChANIhBZBlCVIhFZSiANYhBZBlCVIhFZSiANYhBZBlAVIhFZChCNIhBZRmAjYxFZkQAjmyOyIgEisiIBGjVHZBmBVYhFZClANIhAZBlAVYpEZCmBNIhBZBlAVYpEZCmBNIhBZBlAVYhEZClANIpAZBmBjYxFZEUCjGyOyIoEiMiKBGjUHJFlBFYhFpGlANEgApFlAFUpEpGlBNIgBpFlAFUpEpGlBNIgBpFlAFUhEpGlANEoApFlBDYyFpEVCTCyOSIrEiAiKxKgUXNElhFYhVhElgJEgwhElgFUpUhElhJIgxhElgFUpUhElhJIgxhElgFUhUhElgJEowhElhHYyFhEViTAyOaIrEiAiKxIgEbNEVlGYBViEVkKEA0iEFkGUJUiEVlKIA1iEFkGUJUiEVlKIA1iEFkGUBUiEVkKEI0iEFlGYCNjEVmRACObI7IiAX7zskH50U+GZJe3D8mWrxiKTKO5FgFElhZJ/RxElj5TjURElgZFmwxElg1XjVRElgZFmwxElg1XjVRElgZF/QxElj5TrURElhZJ3RxEli7PummIrLrESttfed2gXPUfQ7LdtoMyZ7vhyDSaaxFAZGmR1M9BZOkz1UhEZGlQtMlAZNlw1UhFZGlQtMlAZNlw1UhFZGlQ1M9AZOkz1UpEZGmR1M1BZOnyrJuGyKpLDJEVSSxNc0RWGs5NekFkNaFm3waRZc+4aQ+IrKbk7NshsuwZN+0BkdWUnH07RJY94yY9ILKaUEvTBpGVhnPdXhBZdYnpbo/IiuTJiqxIgEbNEVlGYBViEVkKEA0iEFkGUJUiEVlKIA1iEFkGUJUiEVlKIA1iEFkGUBUiEVkKEI0iEFlGYCNjEVmRACObI7IiASKyIgEaNUdkGYFViEVkKUA0iEBkGUBVikRkKYE0iEFkGUBVikRkKYE0iEFkGUBViERkKUA0ikBkGYGNjEVkRQKMbI7IigSIyIoEaNQckWUEViEWkaUA0SACkWUAVSkSkaUE0iAGkWUAVSkSkaUE0iAGkWUAVSESkaUA0SgCkWUENjIWkRUJMLI5IisSICIrEqBRc0SWEViFWESWAkSDCESWAVSlSESWEkiDGESWAVSlSESWEkiDGESWAVSFSESWAkSjCESWEdjIWERWJMDI5oisSICIrEiARs0RWUZgFWIRWQoQDSIQWQZQlSIRWUogDWIQWQZQlSIRWUogDWIQWQZQFSIRWQoQjSIQWUZgI2MRWZEAI5sjsiIBIrIiARo1R2QZgVWIRWQpQDSIQGQZQFWKRGQpgTSIQWQZQFWKRGQpgTSIQWQZQFWIRGQpQDSKQGQZgY2MRWRFAoxsjsiKBIjIigRo1ByRZQRWIRaRpQDRIAKRZQBVKRKRpQTSIAaRZQBVKRKRpQTSIAaRZQBVIRKRpQDRKAKRZQQ2MhaRFQkwsjkiKxIgIisSoFFzRJYRWIVYRJYCRIMIRJYBVKVIRJYSSIMYRJYBVKVIRJYSSIMYRJYBVIVIRJYCRKMIRJYR2MhYRFYkwMjmiKxIgIisSIBGzRFZRmAVYhFZChANIhBZBlCVIhFZSiANYhBZBlCVIhFZSiANYhBZBlAVIhFZChCNIhBZRmAjYxFZkQAjmyOyIgHeceewnHHOMtli8yHZa4+hyDSaaxFAZGmR1M9BZOkz1UhEZGlQtMlAZNlw1UhFZGlQtMlAZNlw1UhFZGlQ1M9AZOkz1UpEZGmR1M1BZOnyrJuGyKpLrLS9F1mbbjIs++8zGJlGcy0CiCwtkvo5iCx9phqJiCwNijYZiCwbrhqpiCwNijYZiCwbrhqpiCwNivoZiCx9plqJiCwtkro5iCxdnnXTEFl1iSGyIomlaY7ISsO5SS+IrCbU7NsgsuwZN+0BkdWUnH07RJY946Y9ILKakrNvh8iyZ9ykB0RWE2pp2iCy0nCu2wsiqy4x3e17KrKGh4flzrvuk7/c95Bs8XebyMYz15c/3X2frLnmFJm5wQzdd2qUxoosI7CRsYisSICGzRFZhnAjohFZEfCMmyKyjAFHxCOyIuAZN0VkGQOOiEdkRcAzbIrIMoQbGY3IigRo1ByRZQS2YmzPRNYTTy2UPQ88Se7564OtoR5z6Hvk3bu+UfY6+BT5y98ekp9ecXbFt9DbzRBZveXfqXdEVp51caNCZOVZG0RWnnVxo0Jk5VsbRFa+tUFk5VsbRFaetUFk5VkXNypEVp61QWT1ti49E1lf+vo1cu6Xr5CjDtpTLvzGNbLvnnNbIusnv/ydHPyRT8uPLj1LZm20Xm/pVOgdkVUBUg82QWT1AHrFLhFZFUEl3gyRlRh4je4QWTVgJd4UkZUYeI3uEFk1YCXeFJGVGHjF7hBZFUH1YDNEVg+gV+gSkVUBkuEmPRNZb9j1MNlxzlbyb/+6l+y6/3Hyzp3e0BJZDz86X7Z754fkK5/9iLz6FZsbvnWdaESWDkftFESWNlG9PESWHkvNJESWJk3dLESWLk/NNESWJk3dLESWLk/NNESWJk29LESWHkvtJESWNlGdPESWDsemKT0TWa9/xyGyy46vlyMO2H0FkXXbnffIP3/gBLnmq5+U2c+b2fR9JWuHyEqGulZHiKxauJJujMhKirtyZ4isyqiSb4jISo68coeIrMqokm+IyEqOvHKHiKzKqJJuiMhKirtWZ4isWriSbYzISoa6bUc9E1mHn3CO/OLX/yvf/dLJcuhHP9dakfWOt/6T7Hf4afLHu/4qv77uCzIw0N9bOhV6R2RVgNSDTRBZPYBesUtEVkVQiTdDZCUGXqM7RFYNWIk3RWQlBl6jO0RWDViJN0VkJQZesTtEVkVQPdgMkdUD6BW6RGRVgGS4Sc9E1iOPLZAd3320LFy0uPX21ip+qXDRM4tlcHBITvvoAbLzG7cxfNt60YgsPZaaSYgsTZq6WYgsXZ5aaYgsLZL6OYgsfaZaiYgsLZL6OYgsfaZaiYgsLZK6OYgsXZ6aaYgsTZp6WYgsPZZNknomstxgFy56RtxD33/3v38S9yuGL5j9HHnPO98kL918sybvxbTNgieelqXLlsn6M6at0A8iyxR743BEVmN05g0RWeaIG3WAyGqELUkjRFYSzI06QWQ1wpakESIrCeZGnSCyGmEzb4TIMkfcuANEVmN0pg0RWaZ4u4b3VGR1HV3iDe7+ywPytn2Okbe9+R/l1A/v3+r9qacXyfvnnSG33vbn1n9vPHN9ufjsY2TmBjNa/43ISlykit0hsiqC6sFmiKweQK/QJSKrAqQebYLI6hH4Ct0isipA6tEmiKwega/QLSKrAqQebILI6gH0il0isiqCSrwZIisx8FJ3PRNZt//fvfL4/Cc7vvvXbLmFTBgYSEbHrbjaee9/k8eKMbmH0HuRdeb5l8q3r7pRLr/gFFlz6hTZ86CTZLNNZsl5nzh8dGzvP2xp63+ffPyyZOOlo7EJILLy3UMQWXnWBpGVZ13cqBBZ+dYGkZVvbRBZ+dYGkZVnbRBZedbFjQqRlWdtEFm9rUvPRNaeB508usqpHYKfXnG2zFh37SR0li4blD0PPEmeU6y2euLJp+V5z9lwVGTN2e1w2XHOVnLUQXu2xvLda26S48+4UH5/w0XS5474xQuRlaRMtTpBZNXClXRjRFZS3JU7Q2RVRpV8Q0RWcuSVO0RkVUaVfENEVnLklTtEZFVGlXRDRFZS3LU6Q2TVwpVsY0RWMtRtO+qZyLr3voeK2/YWrjQo9wuGz9t4Q/nSmUcn+9XCeSd/Xv7457/Kd754onzgqDNXEFkv32E/OfHIfWTXudu2xvqbW++U9x7yMfnZlWfL9Gkjog2R1duduF3viKz8auJHhMjKszaIrDzr4kaFyMq3NoisfGuDyMq3NoisPGuDyMqzLm5UiKw8a4PI6m1deiayOr3t/7jhV3LkSefJL753rkxbZ01zOl/46lVy4TevlWu/dpqsN30ded9hnxgVWcPDw/LS7feVM447SObusFVrLLfdeY/88wdOkOsuOV02KYRbKLLOOaPffLx0UI3AxAn9rV/AHBqutj1bpSPgLsqXLBuS4uM17l+DxQ424AzQKvByE9g1JgzI4qWDq8C7WbXeQn+xj00o/s99bnjlRWDCgPv898my4nzDKy8Cbh4wVByj3XGaV14EVpV5wKo0B/B7yKSJzAPy+rSMjKa/mKS58w3zgLyqM3mNdI9Byuud5zGa7ETWn+6+r3jg+rHFiqyjZJtXv8Sc0lY7HdR6gPvfv+B5rb5u+Plvi2dhTZYdty9uJzx4T3Ersk6at2/ruVnuNdaKrNNOYSJrXrCKHaw9ZUJxQT7EAb8ir5SbrbvWGrLg6aWFyBr/FxfuHbjJxarwcu9i3bXXkMefXLIqvJ1V6j1MLCavkydNkCcXjjyPkVc+BNwk1h0CFi1GAOdTlZGRuHnAM8U8YCkCOLfSiJsHPFHMA4bG+TxgVZjHhDuHO5ZNX2uSPPbk4uz2mdV9QE5iTZ08ofW54ZUPgfXWmZTPYFbDkfRMZN3/0GOycOGiUeTuXPbEU0/LF792lfz0llsl1TOy/v2Sq1d46PwV3/+ZTFt7zdathB94987inpE1d4etZd6Be7TG+p2rfyInnHkRz8jK/MPCrYX5FohbC/OsDbcW5lkXNypuLcy3NtxamG9tuLUw39pwa2GeteHWwjzr4kbFrYV51oZbC3tbl56JrE4Pex8Y6Jf377WTHLr/O3tCJry10A3gjM9/syWvrrjwVJlarNRyD4XnVwt7UppanSKyauFKujEiKynuyp0hsiqjSr4hIis58sodIrIqo0q+ISIrOfLKHSKyKqNKuiEiKynuWp0hsmrhSrYxIisZ6rYd9Uxk3Xrbn+XhR+evMKi115oqW77s74p7gHt3v2lZZD3x1ELZ/4jT5Q9/vLs11lkbzpCvnn2szNpovdGx87D33u7E7XpHZOVXEz8iRFaetUFk5VkXNypEVr61QWTlWxtEVr61QWTlWRtEVp51caNCZOVZG0RWb+vSM5HV27ddv/fH5j8pS5YulZkbzFipMSKrPk/rFogsa8LN8xFZzdlZtkRkWdKNy0ZkxfGzbI3IsqQbl43IiuNn2RqRZUm3eTYiqzk765aILGvCzfIRWc24abVKKrKu/uEv5a6/3F9p7O/fa2eZMnmNStv2eiNEVq8rsHL/iKz8auJHhMjKszaIrDzr4kaFyMq3NoisfGuDyMq3NoisPGuDyMqzLm5UiKw8a4PI6m1dkoosd9ver393R6V3/LMrz5bp09autG2vNyqLrAu+PCD33Nsn++09JLNn80uGvagPIqsX1Kv1iciqxin1Vois1MSr94fIqs4q9ZaIrNTEq/eHyKrOKvWWiKzUxKv1h8iqxqkXWyGyekG9e5+IrO6MLLdIKrIs30gvsz949BJZvLhPTj5+WWsYiKxeVmOkb0RW72vQaQSIrDxrg8jKsy5uVIisfGuDyMq3NoisfGuDyMqzNoisPOviRoXIyrM2iKze1gWRpcD/lLOWrLAC6/iTJ7RSWZGlALdhBCKrIbgEzRBZCSA36AKR1QBaoiaIrESgG3SDyGoALVETRFYi0A26QWQ1gJagCSIrAeSGXSCyGoIzbobIMgbcJb6nIuuHN/1afvarW+XJpxatNMxTP7yfTJ0yubd0KvaOyKoIKuFmiKyEsGt2hciqCSzR5oisRKAbdIPIagAtURNEViLQDbpBZDWAlqgJIisR6JrdILJqAku4OSIrIewaXSGyasAy2LRnIuvSK38sJ3/6YhkY6JfBwSGZts6aMnHCBHnksQWyxhoT5cbvfkamrb2mwVvWj0Rk6TONTURkxRK0a4/IsmMbk4zIiqFn2xaRZcs3Jh2RFUPPti0iy5ZvTDoiK4aeXVtElh3b2GREVixBm/aILBuuVVN7JrJ23vsjrYe5f/aUQ+T17zhErrvkNNlk443kxDO/LL/49e/lB988s+p76Pl2iKyel2ClASCy8quJHxEiK8/aILLyrIsbFSIr39ogsvKtDSIr39ogsvKsDSIrz7q4USGy8qwNIqu3demZyNpqp4PkwPe+Td63+1vlZXP2lQs+dbRs/aoXy+/vuEv2OOAkufzCU+Xvn//c3tKp2DsiqyKohJshshLCrtkVIqsmsESbI7ISgW7QDSKrAbRETRBZiUA36AaR1QBaoiaIrESga3aDyKoJLOHmiKyEsGt0hciqActg056JLLcK6507vUE+9IF/ljm7HS477bCNHHng7vKbW++U9x7ysVGxZfCe1SMRWepIowMRWdEIzQIQWWZoo4IRWVH4TBsjskzxRoUjsqLwmTZGZJnijQpHZEXhM2uMyDJDGx2MyIpGaBKAyDLBWjm0ZyLLrbpyr0u/cIJ8/HNfk0su+5Hs/MZt5Obf/EGeeGqh/Ora84tnZg1UfiO93BCR1Uv67ftGZOVXEz8iRFaetUFk5VkXNypEVr61QWTlWxtEVr61QWTlWRtEVp51caNCZOVZG0RWb+vSM5H13//7f/K3Bx6VuTtsJYuXLJWDP/Jpufm//iCbPncjmXfgHjLnn17VWzI1eu8ksnZ5+5Bs+YqhGklsqkUAkaVFUj8HkaXPVCMRkaVB0SYDkWXDVSMVkaVB0SYDkWXDVSMVkaVBUT8DkaXPVCsRkaVFUjcHkaXLs25az0TWvfc9JM97zgbS546az76Ghoal311NjbNXJ5G13baDMme74XH2blaN4SKy8q0jIivP2iCy8qyLGxUiK9/aILLyrQ0iK9/aILLyrA0iK8+6uFEhsvKsDSKrt3Xpmcja86CT5b77H5b3vPPNsvvbtmv9guF4fSGy8qscIiu/mvgRIbLyrA0iK8+6ILLyrYsbGSIr3/ogsvKtDSIrz9ogsvKsCyIr37ogsnpbm56JrJ//5+/l/Iu/Vzzc/Y8tAtu/bsviFwzfIq955Ra9JdKgd0RWA2jGTRBZxoAj4hFZEfAMmyKyDOFGRrMiKxKgYXNEliHcyGhEViRAw+aILEO4EdGIrAh4xk1ZkWUMuGE8IqshOKVmPRNZfvwPPTJfvvW9G+TrV/xIFjzxtKw/Y5rss/tb5b27vVkmDIyPh72fed5iuf2OfnnX7kOy2exh+fjpI+Pm1kKlvbRBDCKrAbRETRBZiUDX7AaRVRNYws0RWQlh1+wKkVUTWMLNEVkJYdfsCpFVE1iizRFZiUA36AaR1QBagiaIrASQx+ii5yLLj809H+sz//5tueAb17b+6adXnC0z1h0ftxt+7bvPyI03DbTE1fNn98mFF/cjsnq7Xwsiq8cFGKN7RFaetUFk5VkXNypEVr61QWTlWxtEVr61QWTlWRtEVp51caNCZOVZG0RWb+vSc5H14MOPy7euukG+ccX1rRVZU6dMkj3fvoMcuv+uMnHihN7Sqdg7IqsiqISbIbISwq7ZFSKrJrBEmyOyEoFu0A0iqwG0RE0QWYlAN+gGkdUAWqImiKxEoGt2g8iqCSzh5oishLBrdIXIqgHLYNOeiazyM7K2fOnftZ6RNecfXyUDAyMrmsbLC5GVX6UQWfnVxI8IkZVnbRBZedbFjQqRlW9tEFn51gaRlW9tEFl51gaRlWdd3KgQWXnWBpHV27r0TGS5Xy380933Fb9YuL28Z9c3yayN1ustiYjeEVkR8IyaIrKMwCrEIrIUIBpEILIMoCpFIrKUQBrEILIMoCpFIrKUQBrEILIMoCpEIrIUIBpFILKMwEbGIrIiAUY275nI+uOf/yov2PQ54271VTveiKzIvdCgOSLLAKpSJCJLCaRyDCJLGahiHCJLEaZyFCJLGahiHCJLEaZyFCJLGahSHCJLCaRBDCLLAKpCJCJLAWJERM9EVsSYs2uKyMquJDzsPb+SjI4IkZVncRBZedbFjQqRlW9tEFn51gaRlW9tEFl51gaRlWdd3KgQWXnWBpHV27ogshT4dxJZW792SOa+dUihByLqEmBFVl1i6bZHZKVjXacnRFYdWmm3RWSl5V2nN0RWHVppt0VkpeVdpzdEVh1a6bZFZKVjXbcnRFZdYmm2R2Sl4dypF0SWAv9OImvTTYZl/30GFXogoi4BRFZdYum2R2SlY12nJ0RWHVppt0VkpeVdpzdEVh1aabdFZKXlXac3RFYdWum2RWSlY123J0RWXWJptkdkpeGMyDLkjMgyhNswGpHVEFyCZoisBJAbdIHIagAtURNEViLQDbpBZDWAlqgJIisR6AbdILIaQEvQBJGVAHLDLhBZDcEZN0NkGQPuEs+KLAX+iCwFiMoRiCxloIpxiCxFmIpRiCxFmMpRiCxloIpxiCxFmMpRiCxloIpxiCxFmIpRiCxFmMpRiCxloEpxiCwlkA1jEFkNwYXNEFkKEJUjEFnKQBXjEFmKMBWjEFmKMJWjEFnKQBXjEFmKMJWjEFnKQBXjEFmKMBWjEFmKMJWjEFnKQJXiEFlKIBvGILIaggubXfPjxXL5lf3iHu4+a5a0/rd78YwsBbgNIxBZDcElaIbISgC5QReIrAbQEjVBZCUC3aAbRFYDaImaILISgW7QDSKrAbQETRBZCSA37AKR1RCccTNEljHgLvGILAX+v/ivxXLhxf0tcbXZ7CG58aYBRJYC15gIRFYMPdu2iCxbvk3TEVlNydm3Q2TZM27aAyKrKTn7dogse8ZNe0BkNSVn2w6RZcs3Jh2RFUPPri0iy45tlWREVhVKXbZBZClAVI5AZCkDVYxDZCnCVIxCZCnCVI5CZCkDVYxDZCnCVI5CZCkDVYxDZCnCVIxCZCnCVI5CZCkDVYpDZCmBbBiDyGoILmyGyFKAqByByFIGqhiHyFKEqRiFyFKEqRyFyFIGqhiHyFKEqRyFyFIGqhiHyFKEqRiFyFKEqRyFyFIGqhSHyFIC2TAGkdUQHCJLAZxhBCLLEG5kNCIrEqBRc0SWEViFWESWAkSjCESWEViFWESWAkSjCESWEdjIWERWJEDD5ogsQ7gR0YisCHgKTRFZChBZkaUAUTkCkaUMVDEOkaUIUzEKkaUIUzkKkaUMVDEOkaUIUzkKkaUMVDEOkaUIUzEKkaUIUzkKkaUMVCkOkaUEsmEMIqshuLAZIksBonIEIksZqGIcIksRpmIUIksRpnIUIksZqGIcIksRpnIUIksZqGIcIksRpmIUIksRpnIUIksZqFIcIksJZMMYRFZDcFVEltvm5OOXKfRARF0CiKy6xNJtj8hKx7pOT4isOrTSbovISsu7Tm+IrDq00m6LyErLu05viKw6tNJti8hKx7puT4is3+t9rwAAIABJREFUusTSbI/ISsO5Uy+ILAX+nVZkIbIU4DaMQGQ1BJegGSIrAeQGXSCyGkBL1ASRlQh0g24QWQ2gJWqCyEoEukE3iKwG0BI0QWQlgNywC0RWQ3DGzRBZxoC7xCOyFPgjshQgKkcgspSBKsYhshRhKkYhshRhKkchspSBKsYhshRhKkchspSBKsYhshRhKkYhshRhKkchspSBKsUhspRANoxBZDUEFzZDZClAVI5AZCkDVYxDZCnCVIxCZCnCVI5CZCkDVYxDZCnCVI5CZCkDVYxDZCnCVIxCZCnCVI5CZCkDVYpDZCmBbBiDyGoIDpGlAM4wApFlCDcyGpEVCdCoOSLLCKxCLCJLAaJRBCLLCKxCLCJLAaJRBCLLCGxkLCIrEqBhc0SWIdyIaERWBDyFpogsBYj/++dn5NOfHZB11x2WaeuI3HNv32gqD3tXANwgApHVAFqiJoisRKBrdoPIqgks4eaIrISwa3aFyKoJLOHmiKyEsGt2hciqCSzR5oisRKAbdIPIagAtQRNEVgLIY3SByFLg/7dHF8nxJ09oJW26yTAiS4FpbAQiK5agXXtElh3bmGREVgw927aILFu+MemIrBh6tm0RWbZ8Y9IRWTH07NoisuzYxiYjsmIJ2rRHZNlwrZqKyKpKaoztEFkKEJUjEFnKQBXjEFmKMBWjEFmKMJWjEFnKQBXjEFmKMJWjEFnKQBXjEFmKMBWjEFmKMJWjEFnKQJXiEFlKIBvGILIaggubIbIUICpHILKUgSrGIbIUYSpGIbIUYSpHIbKUgSrGIbIUYSpHIbKUgSrGIbIUYSpGIbIUYSpHIbKUgSrFIbKUQDaMQWQ1BIfIUgBnGIHIMoQbGY3IigRo1ByRZQRWIRaRpQDRKAKRZQRWIRaRpQDRKAKRZQQ2MhaRFQnQsDkiyxBuRDQiKwKeQlNElgJEVmQpQFSOQGQpA1WMQ2QpwlSMQmQpwlSOQmQpA1WMQ2QpwlSOQmQpA1WMQ2QpwlSMQmQpwlSOQmQpA1WKQ2QpgWwYg8hqCC5sNpbIOvywQZk+bVihFyLqEEBk1aGVdltEVlreVXtDZFUllX47RFZ65lV7RGRVJZV+O0RWeuZVe0RkVSWVdjtEVlredXpDZNWhlW5bRFY61u16QmQp8B9LZO2395DMnj2k0AsRdQggsurQSrstIist76q9IbKqkkq/HSIrPfOqPSKyqpJKvx0iKz3zqj0isqqSSrsdIist7zq9IbLq0Eq3LSIrHWtElhFrRJYR2IhYRFYEPOOmiCxjwA3jEVkNwSVohshKALlhF4ishuASNENkJYDcsAtEVkNwxs0QWcaAI+IRWRHwDJsisgzhVohmRVYFSN02QWR1I5T+74is9Myr9ojIqkoq7XaIrLS86/SGyKpDK+22iKy0vOv0hsiqQyvttoistLyr9obIqkoq/XaIrPTMq/SIyKpCyW4bRJYCW0SWAkTlCESWMlDFOESWIkzFKESWIkzlKESWMlDFOESWIkzlKESWMlDFOESWIkzFKESWIkzlKESWMlClOESWEsiGMYishuDCZqHImrnRsDzwYJ+sWzzgff6CPuEZWQqAG0QgshpAS9QEkZUIdM1uEFk1gSXcHJGVEHbNrhBZNYEl3ByRlRB2za4QWTWBJdockZUIdINuEFkNoCVogshKAHmMLhBZCvydyDr3/AF58KG+0bRNNxmWe+5FZCngbRSByGqELUkjRFYSzLU7QWTVRpasASIrGeraHSGyaiNL1gCRlQx17Y4QWbWRJWmAyEqCuVEniKxG2MwbIbLMEY/ZASLrWTyPL3hSFi1aLM+ZuX5bYAueeFqWLlsm68+YttLfnci64MsDLXHlX4is3u7YiKze8h+rd0RWnrVBZOVZFzcqRFa+tUFk5VsbRFa+tUFk5VkbRFaedXGjQmTlWRtEVm/rstqLrAcefkx23f84caLKvWasu7Yc9L53yF677ND676eeXiTvn3eG3Hrbn1v/vXEhui4++xiZucGM0cohsnq7E7frHZGVX038iBBZedYGkZVnXRBZ+dbFjQyRlW99EFn51gaRlWdtEFl51gWRlW9dEFm9rc1qL7Luf/BR+ep3fiB77fpGWXvNqXLORZfJpd+7QW6++jyZOmWynHn+pfLtq26Uyy84RdacOkX2POgk2WyTWXLeJw5HZPV23x2zd0RWvsVBZOVZG0RWnnVBZOVbF0RW3rVBZOVbH0RWnrVBZOVZF0RWvnVBZPW2Nqu9yCrjv+ve+2XnvT8iXzxjnvzja14qc3Y7XHacs5UcddCerU2/e81NcvwZF8rvb7hI+twRv3ixIqu3O3G73hFZ+dXEjwiRlWdtEFl51gWRlW9dEFl51waRlW99EFl51gaRlWddEFn51gWR1dvaILJK/L/y7e/L6ed+Q2787mdkg/XWlZfvsJ+ceOQ+suvcbVtb/ubWO+W9h3xMfnbl2TJ92todRdYWmw/J7Xf0y7t2H5IXbTHU2yqvhr0jsvItOiIrz9ogsvKsCyIr37ogsvKuDSIr3/ogsvKsDSIrz7ogsvKtCyKrt7VBZAX8f3/HXbLXwafIbjtvJ8cdvrcMDw/LS7ffV8447iCZu8NWrS1vu/Me+ecPnCDXXXK6bLLxhq1/W7psSD517qDcOfIYrdZr7pv75NofDLf+//97S39vq7wa9j4w0CdDhT90NeSVF4EJA/2yrFWcvMbVZDTLhoZlgjNAq8hrwoSiNsXxjFdeBNzq3+JjI8sGV4EPTV5oo0fT/+znf6g4FvDKiwDzgLzqEY7GzQMGi3nAeJ+irWpzAFejicU8wF3X8MqLAPOAvOrhR+M+L7x6RwCR9Sz7u//ygOz2LyfIy1/8Avni6fNkwF01FC+3IuukefvKLju+vvXf7VZkPbxgsXzxgj65657lF7RzthuSH9/YL+7/v2lO7wq8uvY8bepEeWbJoCzmZJzdLrDeOmvI408ulaHxPoPNjmzcgNz1+Iy1J8kjTyyOC6K1OgH3a0VTJ02Q+U8vUc8mMI7A1EkDrccMPP3MsrggWqsTaM0DlhbzgKVclKvDjQx084D5xTxgkHlAJEnd5m5F1nrMA3ShKqVNLK5L15xSzAOeYh6ghFQlZoNpk1RyCGlGAJFVcPvDH++Wd//rx+QfX/1S+cwp/yoTBgZGabpnZM3dYWuZd+AerX/7ztU/kRPOvKjrM7K223ZQbrxpQNz/n7Md39Q22z2bt+LWwubsrFtya6E14Wb53FrYjFuKVpMm9he/jjdRHkUypsBdqw9+tbAWrqQbc2thUty1OuPWwlq4km3MrYXJUNfuyH2htc6aE+WRYvEEr3wIcGthb2ux2ousW2/7c/FLhCfL1q96sRz7ofdK/7MPcF9rzSmy/oxpcsbnv9mSV1dceKpMnTpZ9jyw2q8WIrJ6u2MjsnrLf6zeEVl51gaRlWdd3KgQWfnWBpGVb20QWfnWBpGVZ20QWXnWxY0KkZVnbRBZva3Lai+yvn31jXLimV9eqQqvK1Zn/fuZ8+SJpxbK/kec3lq15V6zNpwhXz37WJm10Xqjbdr9aiEiq7c7NiKrt/wRWfny7zQyRFa+NUNk5VsbRFa+tUFk5VsbRFaetUFk5VkXRFa+dUFk9bY2q73Iqor/sflPypKlS2XmBjNWalIWWRttONz6pUJuLaxKV387RJY+U61EVmRpkdTNQWTp8tRMQ2Rp0tTNQmTp8tRMQ2Rp0tTNQmTp8tRKQ2RpkdTPYUWWPlONRESWBsXmGYis5uxGWzqRde1/9MvNvxp5QPymmwzLZrMRWQpoG0cgshqjM2+IyDJH3KgDRFYjbEkaIbKSYG7UCSKrEbYkjRBZSTA36gSR1QibeSNEljnixh0gshqjM22IyDLF2zUckdUVUfcNnMj68Y19rRVYiKzuvFJsgchKQblZH4isZtysWyGyrAk3z0dkNWdn3RKRZU24eT4iqzk765aILGvCzfIRWc24pWiFyEpBuX4fiKz6zDRbILIUaCKyFCAqRyCylIEqxiGyFGEqRiGyFGEqRyGylIEqxiGyFGEqRyGylIEqxiGyFGEqRiGyFGEqRyGylIEqxSGylEA2jEFkNQQXNkNkKUBUjkBkKQNVjENkKcJUjEJkKcJUjkJkKQNVjENkKcJUjkJkKQNVjENkKcJUjEJkKcJUjkJkKQNVikNkKYFsGIPIagjOUmT52xTds7b232dQYYSrXwQiK9+aI7LyrA0iK8+6uFEhsvKtDSIr39ogsvKtDSIrz9ogsvKsixsVIivP2iCyelsXRJYC/3YrsnbYblguvLi/9eD3ujIKkRVfFERWPEOrBESWFdm4XERWHD/L1ogsS7px2YisOH6WrRFZlnTjshFZcfysWiOyrMjG5yKy4hlaJCCyLKhWz0RkVWfVcUtElgJE5QhEljJQxThEliJMxShEliJM5ShEljJQxThEliJM5ShEljJQxThEliJMxShEliJM5ShEljJQpThElhLIhjGIrIbgwmbaIuvrl/bL7Xf0y8yNhuXgA7i1sEmJEFlNqKVpg8hKw7luL4isusTSbY/ISse6bk+IrLrE0m2PyErHum5PiKy6xNJsj8hKw7lJL4isJtTs2yCy7BmP1QMiS4G/tsi64MsDcs+9fa2RnXz8MoURrn4RiKx8a47IyrM2iKw86+JGhcjKtzaIrHxrg8jKtzaIrDxrg8jKsy5uVIisPGuDyOptXRBZCvwRWQoQlSMQWcpAFeMQWYowFaMQWYowlaMQWcpAFeMQWYowlaMQWcpAFeMQWYowFaMQWYowlaMQWcpAleIQWUogG8YgshqCC5uVRdYWmw/J67aSxg97/9TnBmT+fFZkxZQGkRVDz7YtIsuWb9N0RFZTcvbtEFn2jJv2gMhqSs6+HSLLnnHTHhBZTcnZtkNk2fKNSUdkxdCza4vIsmNbJRmRVYVSl23KImu7bQfl+bP7Gous40+eMNrj4YcNyvRpwwqjXL0iEFn51huRlWdtEFl51sWNCpGVb20QWfnWBpGVb20QWXnWBpGVZ13cqBBZedYGkdXbuiCyFPg7kXX33f0tceVeVUTW/Q/2FYJKZPLklSVVKLL223tIZs8eUhjl6hWByMq33oisPGuDyMqzLoisfOviRobIyrc+iKx8a4PIyrM2iKw864LIyrcuiKze1gaRpcC/rsjyv0o4e9Nh2e99K/8qISIrviiIrHiGVgmILCuycbmIrDh+lq1ZkWVJNy4bkRXHz7I1IsuSblw2IiuOn1VrRJYV2fhcVmTFM7RIQGRZUK2eiciqzqrjlnVFViiqdnzLkGyz1fIVV+HKLtchK7KaFQiR1YxbilaIrBSU6/eByKrPLFULRFYq0vX7QWTVZ5aqBSIrFen6/SCy6jNL0QKRlYJysz4QWc24WbdCZFkTHjsfkaXAv47I8qJq0qRhWby4TyYVtxYefMDQ6HOwyiJrl7cPyZav4NbCumVCZNUllm57RFY61nV6QmTVoZV2W0RWWt51ekNk1aGVdltEVlredXpDZNWhlW5bRFY61nV7QmTVJZZme0RWGs6dekFkKfBvJ7Jet7XIx08fkHXXHZYjDl1+++CPb+yTG28akK1fOyTzF4jcfkd/65lac7YbeVZWWWSFf1MY6moTgcjKt9SIrDxrg8jKsy5uVIisfGuDyMq3NoisfGuDyMqzNoisPOviRoXIyrM2iKze1gWRpcC/nchyYsrfQnjy8ctGeznvCwPyQPGg93ftPiT3PzDcklqhrPrlLf1y3fdHHhrvXoisZgVCZDXjlqIVIisF5fp9ILLqM0vVApGVinT9fhBZ9ZmlaoHISkW6fj+IrPrMUrRAZKWg3KwPRFYzbtatEFnWhMfOR2Qp8K8qsp55pq+1Ssu9jjl6UB54oK/1S4ebbjIs++8zsmrLr9jyw0JkNSsQIqsZtxStEFkpKNfvA5FVn1mqFoisVKTr94PIqs8sVQtEVirS9ftBZNVnlqIFIisF5WZ9ILKacbNuhciyJozIMidcVWTddnu/fONby8WVv42wncjaaMNhefChPtli8yHZaw+ekVW3iIisusTSbY/ISse6Tk+IrDq00m6LyErLu05viKw6tNJui8hKy7tOb4isOrTSbYvISse6bk+IrLrE0myPyErDuVMvrMhS4F9VZF37H/1y86+WPxPLr9ByD3w/tlih5V5+Gyew3POzQsmlMNTVJgKRlW+pEVl51gaRlWdd3KgQWfnWBpGVb20QWfnWBpGVZ20QWXnWxY0KkZVnbRBZva0LIkuBf1WR5Z+Ptd/eQzJ79sgqq/JztC748oDcc29f69lY7vlZiKxmBUJkNeOWohUiKwXl+n0gsuozS9UCkZWKdP1+EFn1maVqgchKRbp+P4is+sxStEBkpaDcrA9EVjNu1q0QWdaEx85HZCnwL4usXd4+JFu+YmgFSfX4gj759GcHZNKkYvXVh5f/iuGnin+bX/zt8MMGZfq0YRlLZLmM//5dvwwPDY3+yqHC8LOL+G3xHh2TzYpnh3nhV3eQnUTWZVf2y/TpfbJ9IQpXl9f9xY8LuNV96xb7l9sve/1CZPW6Au37R2TlWRc3KkRWb2rjjp3XFSupZ2/a+ZyLyOpNbbr16uZL1/9oQAaHhmWP3Vaf8303Lrn8vVciy/2g0o9/0ifP33RY3sVjO1baHRBZuXxCVh4HIivP2iCyelsXRJYCfyey3IT388UvErqXX3EVrrZycubyQqKUn3nlxZVvE/63exC8e/lfPfTP2Ar/TWH42UWEMs/9+mOTVzuR5Z9J5vLcw/YnF7d0rg4v/wMCuazuQ2TludchsvKsCyKrd3Xxt/qPdc5FZPWuPmP1HJ7v/ReFeY509RxVr0TWpz5XfHk8v68F3f16+Iu26P2XezntAYisnKqx4lgQWXnWBpHV27ogshT4O5HlXl5ctRNZbiWQW02141uGZJutlp84v35pf2u1jD+h+gwnWvwvHHqRFf6i4ao8MftY8cuOi4tfeHxlsXpo12J1W5NXO5HlZaLLC2/vbJI/ntr4fW+si7GU7weRlZJ29b4QWdVZpd5yVVqR5Z4N+cCDywlOW1daq5FzfIUXvf48XB4nIivHyq34C9B+lXyeI109R9ULkRV+4eyou+fTHnno0GrzpWaVPQ2RVYVSb7ZBZPWGe7deEVndCNn+HZGlwLeKyPIT4oMOGJRZGy2ftHs55Z6J5VYfhau4ys/PCoWEEzEzZw7LL4uHx7uXv1XOfYP8QLE6zP3S4XhdceTfd9MVRG6y8vOfDRSr34blpS9dfktBKAKbTmzdqrgHil+TjLntUWGXWyHCffN8V/FctU5j8ivceiGy2o0td5Hlb+FdtHBY5r61mUjVrnE5z49xSjER3/q1OmPspchykvmuu0Ve9QppfDtxU+Y3FM8ifOD+YXnrW4ezEyrux0EWFeJnq38Qed6sCfLoE4ubvs1s2oVfKLhB5frLvOWL3k5ffiCyerNr+XPLzOIXntutqgnP9zFfivXm3dXv1fH4w+0is2ZJFo8QcO/A3cb3+OMiL95i5eO6lsj6RvFl8MxZI198dhPifoWlO2fe/0Df6PNom678r1+l/Ft4kXXZdYvl7uKcvM1W6c/JnSj5fXz6dDeu7vOeJvMk/2iTKvtT6moislITr9YfIqsaJ6utEFkKZLuJLCev3G2H7hlFRxTPwgpfY4msj51WrExa3Dd6G1woJJz4mjWzT77xrRGR5VZoLXi8T/ztiE1FjQKOqIjwdoDw1xzrhHa6HaTMr8nkxWe4iVAukqPbmMJVBalXovkVh+HFau4iy02+r/v+yOcq11tQw89Jp5UidT4zbtteiiyNVZh136/f3ovz8mrZpnla7fyv2rq8vfcS2epVA6uEyPLHBHc+dM9CnFl8sXNwcY7M7RWeR9zYOt2GhMjqTeW63TIfnu+bziV6886a9erfb06fJz/3aCcSNURWu0d6jEUv/EJ5fjFfdvPnpl+YNqtS/q28yDrqxCWtWzBz4jN67li3uJY6tPs5o8k8yc8HcrztFJGV5+cHkdXbuiCyFPh3ElleRPlfIGx3MvcHWney2HWXodYD4b3wKj8/KxQSLstt537Z0E+y5y+Q0QvwXL/l7oY7PPG4bZtcpIecwlswy/ya3LboL7hzOrn799VpTP7EPNbFWLe6NP27/6XOcHKdu8gKL2BzFcLhaoPyKs+mteqVyAqFTeoLzvB4k9MFoKthOLYdti9us95p4rgXWWGt290+33TftWjnj6vuXOpu//erpst9IbIs6HfP9POjTscMf+7xSVrHye4j680WneY9vRmNrPDc2HZzEw2RFZ6rO30+/fv30svPr0f/u6IU6RXH1P06kbVk4WQ55Yxlra5Tn5PHer/hZ7rK57nuPMn/KJcbQ7f9KXVdXH+IrF5Q794nIqs7I8stEFkKdDuJrHCi5Z751O6iOBRZOxS3FroVVf6kXxZZoZBw27jXPcUtZe7lVgg5keUm3P6V62qSsZCHJx63XZWTVZg31u0gZX7779P9G50wu5cX3GMx8++r3YSjzCP1yTlk7qVk7iIr/CY/VyEc3mbc6ZtDd8uce21THBuq3GbcK5FVltcpvwkt3+Zm+exBN0m+4ca+4vl/UpwLhrvWJDwWvniLYfng+8f/rYXlHz3xv9pb9zivcNoeMyK86N1m6+HWF0SdVuH+/BcTxV38vW6bpdbDIj8gEIqbdnMdf+75hy2H5b9+27fS80lXNZjhuTaH1aXlFY3lLyU1RFadLyfD2wr9avry4zty3SfcIy1+eUvfmL+eqjV2dyy78YY15Pqblt+6l8vxOdzHq9wVUWWeFHIL5yJN536WvxLeS5FVdz6ptT/WyenVbaGIrDpV0t8WkaXAtJvIcl1MmlQ8VPKwlS8ovRxxEuLdu3cWWZOmDI/enuhuxyi/nNha8IS0lgK7vtwtibmuJhkLeXnyU/dWuHJ7L278CcrfztLkW6byBXcOorDbmMp/r3LyV/hItCLKEs1PhnIXWX7VneeQQ53LNel2m2z4zWLVz1CvRFZ4K6d7nymfZ1MW55YXgKE0q/I5DGs8a6bIR+eN/1sL/a0h/tzk32NKeVnl+Ob3Sbcvuue2hV8whe3r3tpUpW+2qUYgvKgtH+P8vGryZJG3zRX51mX5Pout2rsde6vyeT6H1aVeMvn5aPlLgliRVZ5fdFslX/5i2BE99/wBebB45mnVc6RGrepmuH35rM/1t34Ayb2s5yNOZH3m7OJLk8dGrltyuZYo13vdCivpus2TyrUI5wNNP0P+GqRp+7H2j16JrCbzybr7ucb2vXpMBCJLo3rNMxBZzdmNtqwissaaqIf3ZIf37PtvE9ykf/q05ZNp94tP7uTiXhsVDzp1J2L/cice9+wn9w1y028U6iK5+57l/cf+ApU/8fj3VffC0k+etvqHPrnlv4ZHL4rDlQB+1Vrd2xbLF9z+gfvhL3B5drEcqtagvKKkPCHzY/Y8u032qvZb3s6d6BbMX/6vMzeS4uHdy5/h5v7iPwPaIivse1Jx4RL+mEK791P+1TQ3Vr9iaVQsF5+jzWYPt1Y45iiEw4u4dp9z9w2uf35e1VV4S4pjyvU/nig77rik6W7QqJ2f+PlbuKpI5nINm37u/PHGCQv3q7IWk08/trJkd7c2u8l4p322LFTPPXN8i6zybYXuM1d+RmSjHcigUSjc3I+qtJ5x2ebCKTz+1j1XGQx7tYksi5sye//3F24m8t53D8sJp47MUeqe8y2AhvOl8LPvLtTdak33qnIeC8cWzm/cedfNDzVXl4ZjC/udvWn7XxsNVzROK+au7s6B8twkVmT546k/dnerb/iL4P58XxbrnerdaY7rzrPu1e7HBrT2Hfcw+9uCOy38fCSsieZ80/2Y0XmF4AuvJWK/XIrZtz1HP6dxc1h/DdRtpVi3eVK5RuEKrm77U6f6hvKsqnQsz5/L2f5z1iuR1WQ+qbX/V80JRWfs/lq1T78dIqsuMd3tEVkKPLuJrG5Cyd9eUX4WRzjJnzKlb/T2Bv9rK27o7gN7//19ozIrfNaW+3vVA2k3DH65bPkXgkJT7zJiLwT9PfDtnktS/gW88hLecPL0vncNyGfPHxq9TTNk6Q7KTv65k6CXWlVuvypfiLrJs/vG9/Irl9/O6Tl2OpC6izn3S5Mxvzbnl886RrfdViwDf/YWMtd3eULv37fnGdanyS+6dNpPyhMAN9l6/PHlz3Bz7ZxQef7sPnn4oYmyzrrLZPO/73xrp+Pkxr6guF22221Y5WehdPt2tSwkw9Ux4a2+rypuSXG17fb57fbZaff3ur9+6WvlVhTOLgSbe5ae/7a03Wcu3Feryksnvx2b2Atyv4+79+1/TdUvS/f/HTIJvym/7Mq+1gPAy+K//NkvC9xun7tONQof/utkhXuVLwBdX7cXvwb2ymJlTpULlk63Fvj91NXD3xLu+msnSMJjmdvGMfnIEQNy+5+Wtv53jr+o1O1z4D934ecpvADfsfjVSCcT3T6+ZXFea/LytYr9pa3yrwx3ugUpPO5pHSe6/Rpf+bjd7vPWjZ3WLRjhcalpzTqN1R8jHdfylxN+v/HHwPIKR//3l72kEFnvGi5WtEiSlTed5kn+PZbnS+FnPxTXTuYffED3X+HzueH8xh0f3Oeo6nG8yjwgvIUvrFcnWRbexue2d7++6s//7heW3Txyu23WkEcWLJbBoZVlWJ0xuXnchV8uViyNIe/8ed19offBA1f+JeuxvuzpNMf1K6Xc+zvy0JHj1eXF+cuJO/dlcpXb+T3LTr887cft9nP3K31uruc+D9sX+f585TJi591hTf08IFyNGpuvscq93T5e/ty3myf59+bfQ3kuEt4yd8k3+1c4N3cTZS67XLuxVoq2O9aVV9y128Z/zqqKrG6/ZN7t/FD++1jzyU7nkrpz3CpjGmseGc7r/XFVGtdCAAAgAElEQVQ1vFbq9gV32L/j98tbilW8xa+tVjmvIbKqVM9uG0SWAttOIstFu4mNW0011kktNPiujb+ICw/c7t/dScydcN2Bz00M3MsJA/ez9W7i4l7+hOyXTGutJvET9vJFlz94uIuPRcW3ibHfBHZanebeW/nB4X5MfgVHOHl6584T5KjjRyYX7lvY8BaWX9w88rPLbqLnf52uCief4S9Ew+eSuQmSk1ru5bI7rSoJL8CbfmMa/sLb48Uv77j+wjGFv6YYvm+/Qsd/Kx0e+JuOxX98/Jj8yi832XIvJwr92MJ/67YsvOptWP4z4iZ66xafMycoXfbB/9L5uVB+P/K3mYZj8Uxcbd1k9OOnj8gNLSHseYUPkt5rj+4X7eFEwh0f/MpNL0TKKw3q/mJX+ddC61xElQ+hYe3cRHDxouW/ptpuZarfd9w++MtfjhzbyiK4/Nn336RX/dx1OsyHgsJnli8A/fjcZ9pdsHS7QCkfl3zf5b4WFVw6fbMcCh63navze/fol69eOrKvVLk1sdN77sW/uwuM874wcntMuA/4/c5dZGz5ypGVxE0/b+GKr27HgLEY+AtXd0w59sMjF73+y6bycTKU6N2OaVW5d/tlrvJx++5iBY7/MqXKhVfIKXY/Co9LVsfIdl8Khb9Y6M9/4TMv/d93fFMh098wLJd9b3hUpjT5teKqtes0T/Lt282XXM38L+i5fW5KMY9wMsqtwtjvfdWe4xme511f7vxQVT50mwd4qe7G5lYvu5f7gsmNsZMsC0Ww2z/d59rV8ZlijujmBO6zctoJEzuKrKpj6vTDSOV6hbcKhz/yE67y6fTM1HY1c/PsB4u5vV8p9aJivuMEVjgvr3IB7MfZ7kdx3N/8/uT623JLaX2B5V6utg8U/bvzn3s2buy8O+Tla+eP0+1WslX9PLjtPGO3/7hX01sVw9VzkyeN7OPlY263eZKfO7lxuM+de3kh6K4B/HHUz1mr3PIe1m7uW0YeD+NfVVbD+xV3bl92+1D4Kn/OqoqsTvtTnbqF23aaT451Lqk7x+02tnBVWLu6lL9Md+cjf0tunRVa7j2d98X+kcf0FHO+Y4ucbi9EVjdCtn9HZCnwHUtkVYn3B2i3bfitbiiy7ioMsZuwuQ9w+OuE7mD8wAPLJ7J+NUq7b7+rjKXTNuE3KuFk2R8w3bhuu2NEqHX7JtBNjNzkxn3DFL7CC4h377Hi88LK34q5Cwp/YeQy3Pu+7HvFao7i4OPG95IXTpR5Hy0mTotHVlhc9JWRA5P722+LB7/6CYfvv8q36Z6BP+H5yYTLCCfx/qLHH2xdLbZ8xcg3dGGtu3FqV4vwYO4Osq1Jb/G+/Jj86ht3QndCK1ztcsmlK95yEF6ENRmLH58fk5tU7fWukV/edC/Px/9qZ8jL/X2si66Qk6/v7Nkr7i/tnk/jBW6nC7TyRar/ZVE/Fj8R8jyq3npQ57NVftZDlQvA8BtxN3lzNXdcw9WF4TdOfvLpZd1YrMMTt9ufnJCucxFVfu9h7Vwd3Mt/3vyEwtXBHQNetPlyWehk3OhFUzCBKH/23Xtpt3rK177K5NONKVx95y5ivDwKLwDDz5tr4y5Y3tVFPIbHSj+WTisC2j2A2PXjJ2VuP1y0aGRl44zpIo89PkJbS5rU2W9jtr3wKwPibs9pd5z1+2p4fKjyxUJ5POVVek0lTbuL23bP1wkn8e7Cyp9rphcXJHVf7ptmt1oxzAyPkW4u8LqtR74QKx+33RdZfmVx+T27z5OTJOFKwpBT7H4UHpea1KwTp/AY2e5iovwFTXkb//l5TyF/X/6y4rP9P8O15I6rw29/t/IcpVtdO82TfLt28yVXM7fvuLmT/wLlrM+OrDCqel4uryD057VOX1CFc5Ju84B2xyj/GWkny8q/DhjKaidf/OujRw3IlDWXtF2RVXdMfoydxEGn83inXy5s93lzx3J3fgxFRShnwn2j03yy3X7Vbm7rjyHhlzzu3/w5rnUOKP7bfeF07fdXXIHneLtXeb5U3nfdWH5x88jqMf/yPNz7/MjRI79a2O7Y1+1zEP49PJe5f2/62BO/T/hrnPLczWV3mie5z5YTr+E5pjw38X9zXN08xS8ccMfdMif//sq1c5lurlP1UR6h5Dv4wJVXYJbnJFVE1lj7U526hdt2mk92Opc0meN2G1soqtqJqfIz+cJbjqsKKTeG8p03VeaTiKxu1bP9OyJLgW9ZZFW5MA27DVeVhA+E9wc5d1L038q7g7h7uZOp/8Y4PGj4vsODmfs3dyIP7/Mvv+2xbscqX8z5yXJZCow1uQn7cyekx4tvkcqrPsoXluEKhvLtYO5CO3w//r/9N3Qz1l5DzjpnUO4qLp7Cb2HcxXL5Ac9+bGPVLbzAcJNDL2tc2/KkJXx2g/vFK3fR7S6A3S1yfoWPa1f1G9N2k4Jy/fyY3AF7TvENtJsseIHktg1XpblaT5s+cnuafzUZi28bTlScnAwnW77vcLm1b9fpQrP8LY+bGLQbn5/YhDndHr5c/mbWj91nlCdt5V9ZUzhcrHSi7HYBWJ4U+DG4drcVt7y5i9h2q1zcZGp6Ueduz/ny++tGxbe8H/nQGvLhE5fUuogKmZQvxN2Fsns58eZefkLhxMb9xbP+3DeY7lvQ8PbHdiLYr9Txnxv/bXR4m0inb9071czX1k+Kyp9xd+EQ7h/uwtZdXIaCOhyX+7zNmrnic+F8tj/mtBMN/vlLbhWh/ybQj9l9Vt2xst3ty1VW32jsr7EZvi6dfvDE1zvsp8oXC+Vx+YtVfzHh/t7tNuN2763dRXH5GOfa+fPV82dL61cL/3TX8tXUdZh5Pu1uVXfv5cXF7Q3ufO/OIW8tbr8sH7dDOVAWU/5cG64kLH9J0HQ/Kh+XmtSsE6duFxOhuPG3lYXnb38cP+zAftn4uYUgXDIodVaW+FUSddh0mif59xjOydz52G/vatZazV6IBd/f6AVuxVsMy7e+tttf/Th83TrtTwc/u1LFb1+WZP7fO8mysvgqnxN8+znb9skOc5atJLKqPK6iPKZux/5O78GNpczOf65DPuHqzHDfdJ9Z9/LngfDY024+6duGx6Xy3NbLy/ALQn+eC7f1GeG8e7/3jZxD3GusVenu7+32cT++bV5TfCH97LMyO33ZUvUYF8o418Yfv+peJ3Xax/05dax50m+KL6/D2/ndOMpzE/9+3FwkfKyEm7m4lXftziXl2rn5jfsc+7l3O4ESrk72fXYS1uXnSq6zVp+ss+bISsZOr077U9V6lbcLv4Qrzyc7nUvKx+9uc9wqYyvfeh2ulAqfbetvwS1nVhFS4Z0JXoSNtZrL9Xv7HX0yd07xTRavnhFAZCmg9yIrJqrdtyjlE7rL7/Sw0nbtvVAoS5924xzrVgw/KfIfbD9Zbjd56PZNYCiRyqs+yheW4cTzwmJFlZuw+2c9+fcQWnf3b/6k5kTWN787KD8rbiMMT1Bu5UV4sHLiyz3Mt9vFfifJ5rLLB+lwVYm7rdRfaLiJkTshuvfQ9IGs/mAecvASwPPyJ9Owzm6/CX88wC3x99+M+W/zm95eWP7WcIV71Qu+RxTiL7xY3eY1xXPC/rO4FbDDr86U5ZHfj8NvW/1+5L+VDG/3Gv1bm1sMw2/E3SqF8GLiiEMHpfxe2j2gOuZz7tr6SbXfd7tdAJYfhu77dxOrP989slonZBN+Lt2v3fnbOsJbKnxG+Flwk+VXbDFZbvjF4tbqhbrPaXGZvnatX1F99vYT9++uTpOKc7279dN/Dty/+301nCyEItiNufzcPD/28uRvVKxXXA7e7kHj5dsLw/3h9ttHbpFxY3YXDO6Y5H9Jyo/JH2v9BY2fyI61ss9/Ntodp93ntvxga//Zb7riKHb/rdO+0y2FYUZ420J4XKtzoVP+nLpv0N3noskthu1WILTbV/y/vf51xcrY4hmWP7i+/i2fIR+3r/jzhd9/yhdbfh8pnwedtF5cXNu4VQdehoTnWr+SMOQUux+1Oy7VqdlY+1H5GFm+mAgvatvVy39uTzqmWPUzdVlLZFX9hcxQSNX5gqfTPMm/z3ar5P18yR8j3bnSv/wxo9vq2HarPcf6UjFc7RTuT+3mJOXVVWHNOsmydtIofJ9+315vhog755afkdXtwfXtxlSen4XjbHercPj38JcL3Vww/DLB8wn3v9YK5uI2aXcbmL8d0e1bbk7lzqGdjvUrzDuDuUn5/Ob3uXZ8Wz9yUtyB4b7cCG9d9HzDc+tY54dO+7gfywf3nyAbbTzyywPhuOuIXde2nYxr8tiTdjUvz92qzJPcmMJjZXlu4v7bP8uttVjgWTHV2m6M+WT5eOw4faN43lZ4PPb7nF+d7P+72zNMw/1pq3+QriKr0/5U57wdbttpPum+hPRfzPu5rN/nwtsK3fVFtzlut7GF+5E/z4ViKlxF/bpi8YB/jIrL9bXpdntheGdCeCvvWKu5/D73pc9O7PYW+LshAUSWAlwNkdVpGOXVQ3V+dads5jsZ6W63Y4UXcxcVD9X0B+fLrxiRS2FulW8C3Xv1D2kNL0bLFwt+4hneu9667/nZZfcup3Xb4LNjcv/tT7ROZH3/hiG56toRsq4/v3Q3PDG7A++sWSMrHsY62JalXbdfJmm3ysDXOHyuWdVbB8qTgl13GRq9vcofoFe4j/3Zn0127fyJMuTrTi6+dr/93ciqnjpj8e+l3UQlFLC+73BsHz9ugpz+uZGHVrebGJUnguE3beXnGnRacdFuny5/I+7fg58Euix/20D4OdO8vTCckO27z/LbMMe6AAwvDC67fOSHCtwr/FY/3HfD8fpfO213QVY+cb9x+2HZcPoUeeCxRaOT8W4XUeXjViiCHi9ug/O3FLrP2fTi1rhwBZM/BriMUMSFItit2vQPtnerVcMVje3Ea53bC9tdAIdL+d2zddyEKHxAcPl5hp57+RvI8Ljkjo/utmcnvdqNudy23WfCX7i7sfjPfuxtYQqnvq4RY91S6BuH5zjHyh+P6nyL227lZLfzWqfBt1u5E66O9s+0G91/3tMnU6eKnPPF4RVWFnaFU2zg+YSfBdfOHQ/cxbI7RobnS58ZcnL/5o7d/vPmPmtbFj9U4W+/9dmujbuFzZ3rNPaj8Fk6TWrWiU+7Y2R4MVEWN+XbwV2ur+E5Z/TLwsUjK7I6rYoMxxE+eNlzq/KcG5fRaZ7kb/n2F5jhfh3eLlMWD62xVLjFsNOq4XZfKnZajd5pTjLWapx2sqyT+CrPl/y+/a/F7VQbbrjiIwPC85d//ms4N2k3plDQlufI3Z6DFfZ3//0j56zy57Hdj4+4Ve3+FkA3t3Aiy9W6XT1WeO7Os3MzV+9ttln+wy3h+S18dEaVLxjLzwfy+3S7+VGnfdw/g8u993NOW0PuL+YB/uWZ1xG7rm34jC9/C2OTx550qmF4e6FbYeYfH1KeJ4W3wIXHSjfG8tzEfz7Duwja/ahE+a4X/1n1q/fazRu7rU5ud0wM96d93j22yOo0pir7UKfjcaf5pLvbxK+mn7vjyPWIm5O4Ry+4/+1XYGs8Zza8rvTnuVBMhdc27lZQ32d4nut2e+HonQnBD0KMNZ8MryMRWVVmO3bbILIU2FqKLDc8/2HqZu7LbyU8qI0laTotyQ3z/MVc+aIrXHLtti8vr2+H1504Nps98rDGdi9/IikvW/XvwR/UymPytxW6TCeyfn/boJx/4ciFQFnS+JOUu2h0F/vhBfJYu4Sf1PoxdOIacnKTieuLZ5z4pc3uIsV9+9np/XfbJf178bKsPCYv7bzg82MsP0PG16787936b/f3Mt/yKip/onE1O/kjE+XL31zS+oW8sV6h3Gk3AR/rG8ex9ulyzcqTwPLnTINP+X36sZdvw+zEw+/b4YNn3bf35dU6YXsvRNrd1tnus91ffFS8yAovoprsD27i5Cb2/mLajcX9GIK/raB8DChPuMsi2E9awoe8h7cV+jGW5X2VsYeTvHa3wYT7dmsFzfkjz68Jb5Xz39I78VD1WOnHFu6rnS6aHTfXd/mzX+X99XqbTrcU+nGFv0DnHq4e83kLJUGV81onNuVfNxvrc/bRo0dE1jEn1n82luvffbad1Pb7VTs56j4/11438pzMdsft8ufNv692Ajk8H471hUuV/UbzHFL3GOk5dfrMuxoeVzyHyYusOvuDy37dVis+D6kKj06f/bBteF4L50vtBHaV+ZTPLh87OskNt33dOUmnlTjhSqvwPZbPzeFqGSeCy/PIdmy7zZPKY+o0lk58/L+3m1u0+7xVqb/bptOtlO5v4YV1mFc+v/m/lY9DncYQ7iduP3Cv8Nes27Vz8xwvIMpj+eC+k1YQWeH5rSqHcLuwVu3uNKma2W0frzJP6jY38XMRf2zsxMmPuXxd4uePY81Fqtzm5vPH2p86cSuPqSrfsbbrNJ/sNCfptE/HjKXTec5neq6+dk3mS+G+WmU+6fbJ97zz2V/6inlztG1MAJHVGN3yhtYiyz8cffbs4ZV+hrrb8N1FweOPD48+KLbT9p2+qfPb+wNCeTLYbrnmWBOK8FawsqjyffkTSXlM/iLFnQgXFA+w9d+I+eXWk6Ys5+NE1qPzB+X2O0d+1aX8YPnyrZidxlLm5cdWHkN5O/9QT3fx7paAu+1/+1uR588ufo3o2YeWd5t4daqVv/B2tXC/COeWw7tb6/wDtJ2Yc324v7tl6G6/cd8cli/G2j0bqNv+1G1M/u9lPuF/bzF7kvzPHUsKQdtZZLUThO0edDrWL8iN9e1zuCy/fLFQ3qebTCS6cfQnyyonSpflLwza7evtRFUomMuriMpj82MJRZbbps5FVJgZTr7Ln7Py8ci9f/drk5sV+2hYy07PyOn2uWsySS5/g18+FpS/zfQ1K09G3eftttuKz3vx61Lu81bnGUKeS6dfVHv00QHpG5ogU9de0uJU5UKw2z6Y6u/dJu3lfTrm81Ze2djtvNaJQdVjgDufHf9vfcUzsvrkk58aGl0tWYetP6e4z4q7VdjvP/4c4j4f7jbo8jnEc3N9+XNKKKbCc235GOD36dj9SPMc0um41OkY6S9qO0lGd0F51CHLRZbLr3LODcVrXT6d5kn+vbU7r3V7OHfVuUn5c9bp+B0+47TKnCT8grBco06yrCyZysftblIx5NSuZu3G1O0812n1dpmT36/azdmqfq471azdbb8u09euLPHrrJIvn0PG+pIsvEOhvI870bj9NiuKLDfGsWT+WFza1arqF3jl3PIq3U7PpWs3Typ/zrr9t7/u2mKLkflzt/lkeR/vxKvJbXZVjwGel+fUdA5X5j7WfLLTuaTTPl31M1TeLpxXdvoCxp//y9dGVY/jZVHabT7px8TD3ptWVacdIkuBo7XIUhjiahfhRNbCZ4pnYyxdcdn6agciwze80fTJ8nDxsMqhoWYrGDJ8S6vEkMoia5V4U0Zvwl1cl+W4UVet2EkT+2WtKRPl0Sc6P+TVsn+yOxNYa8qElsh6cuFSMGVGoDUPePbWwsyGttoPx80D3EOry8/IWu3B9BiA++GKmcUjBsJbC3s8JLp/lkCVXy0EVnoCiKz0zMMeEVkK/BFZChCVIxBZykAV4xBZijAVoxBZijCVoxBZykAV4xBZijCVoxBZykAV4xBZijAVoxBZijCVoxBZykCV4hBZSiAbxiCyGoILmyGyFCAqRyCylIEqxiGyFGEqRiGyFGEqRyGylIEqxiGyFGEqRyGylIEqxiGyFGEqRiGyFGEqRyGylIEqxSGylEA2jEFkNQSHyFIAZxiByDKEGxmNyIoEaNQckWUEViEWkaUA0SgCkWUEViEWkaUA0SgCkWUENjIWkRUJ0LA5IssQbkQ0IisCnkJTRJYCRFZkKUBUjkBkKQNVjENkKcJUjEJkKcJUjkJkKQNVjENkKcJUjkJkKQNVjENkKcJUjEJkKcJUjkJkKQNVikNkKYFsGIPIaggubIbIUoCoHIHIUgaqGIfIUoSpGIXIUoSpHIXIUgaqGIfIUoSpHIXIUgaqGIfIUoSpGIXIUoSpHIXIUgaqFIfIUgLZMAaR1RAcIksBnGEEIssQbmQ0IisSoFFzRJYRWIVYRJYCRKMIRJYRWIVYRJYCRKMIRJYR2MhYRFYkQMPmiCxDuBHRiKwIeApNEVkKEFmRpQBROQKRpQxUMQ6RpQhTMQqRpQhTOQqRpQxUMQ6RpQhTOQqRpQxUMQ6RpQhTMQqRpQhTOQqRpQxUKQ6RpQSyYQwiqyG4sBkiSwGicgQiSxmoYhwiSxGmYhQiSxGmchQiSxmoYhwiSxGmchQiSxmoYhwiSxGmYhQiSxGmchQiSxmoUhwiSwlkwxhEVkNwiCwFcIYRiCxDuJHRiKxIgEbNEVlGYBViEVkKEI0iEFlGYBViEVkKEI0iEFlGYCNjEVmRAA2bI7IM4UZEI7Ii4Ck0RWQpQCQCAhCAAAQgAAEIQAACEIAABCAAAQhAwJ4AIsueMT1AAAIQgAAEIAABCEAAAhCAAAQgAAEIKBBAZClAJAICEIAABCAAAQhAAAIQgAAEIAABCEDAngAiK4LxgieelqXLlsn6M6ZFpNDUmsDQ0LD87cFH5P+3dydwOpV7AMf/DNm3iCyXSqmEokKia62rpJAlIkuWsY4xUqbsuyE7Y5cU2XeKbCEUukqlhVsJ2XdC7vk/7rzXjPGe97zvvHOG+Z3Pp8+HnOd5zvk+7znnOf/zLLnuulNSpwoJdnHJPv+rV6/K5St/x2ttVxcX/7okR46dlDy5sksKnayBLUEF1F/rJyQkpeN8ud85JvM5wfGTp+X8+YuS5+4c8aaxuy4OHj4mmTOml/Tp0vpcJjvaC+j1cvjoCUmdOpXcmTWTfYI4e9jd7xxnSAKPgN7H9Flx6sw563mRQ9KlveMGHW/XBXUTvB/T5StX5OCfx0SN81r3NKfPG7v7XfCOnJzt7GkHJN3fCO2ApFs3HFnwBAhk+WF75ux5eSNisOz67heTWh/U74/sKndbgRK2xBVYuHKjdO0/4YZCv1wx3jRsl3+2Rbr0jZYrVmBFt44tassb9V9I3INMZqV9OH+1vDf+Y9m2PDrWmXurC30p6T9yhsyYt8qkueOO1DJuYLiUKv5wMtML3umqcbvIEaaAUf06eArSF/XytcJuKFj3qVCmuHC/C16daMOzZrN3RV8OdNNgSejrL0v9GpXM3+2ui5/27pdGHfp50v/zqUdleK92JvDCFpjAZ59vl7DuozzPjnvz55aeEU3k8WKFTMY8ewLzDST11h3fS8suQ+Qv68OHbhoo6dSqrrxe+znzd7vrgnZBIPre006YsUSGTZjj2Sl9ujTWs7yT57op93I7OXbidKxM9H4X2aGh7f0ueEedvHLWYFWNpu/IhYt/yWez3/PpWUM7IHF+I/t+OyjVG3eV6s8+LX26NDOF2rXR7O53iXPklIKAOwIEsvxwjxo3S2YvXivzJ/WWDOnTSb3QnqKN3DH9O/qRG0kCEViw4nPpNniyzJ3YK1Y299+TV85fuCilq7U2gSt9OVy6arNEDpgoS97vb+qLLWEFft63X+q36WMCH9p4vT6Qde78Ba918cVXu6VZp0EyfnCEPPnog9Lrvfdl5dqtsmXpOEmpy+mxBSQwd+l66TN8unnx0+DU9YGsP4+ckAqvhMnQHm3kvgL/vy40QK89fLjfBUTvNfGBQ0dl+pxPpH7NypIpQ3oZNWWezFq0Rr5YMsbY210XGgTTZ1D0oHD57Y/DUrtFd3mrbQNPICx4R37757xm0w754+BReb5SKTln9ZYL7z5a/raCwbPH9zAnz7PHvd/Alh3fye4f9km1Kk9J1swZpZ/1EeRj67qJ+YDl7bqwexa5d1a3R8n6ISt7tkxStmQxuXTpsjQNH2h6aC+a2tecoAayqlUpIzWfL+c54WxZMpmRDXb3u9tDyN2z0I8joW8NlQ1bdlmjFLJ5All29rQDgl9v+kGrWqO3TKC3RtVynkCWXRuNdkDw64YSkq4AgSw/6qZi7Y5StWIp6Rxaz6TWl0QNpnyzZgrDofzwDCSJvkz0HDpNdnxyY6+sZau3SOfeY2W79W9prB4+upWp3kZeq1lFWjd+OZBiSRuPwKXLV+SQ1cNk8aebZPJHy2IFsuzq4p2Bk+Sb7/fKgil9TM76gl+5bif5YFSkFC/yAN4BCpw9d0F0+Fr3qCmSLk2aeANZ8yf3kUL35buhJO53AeI7SL731wNWQ/ZtE9B9+ski4u26KJDvbvNSODGqszz1xCOmlPAeY8ww6pljuzkolV19EdDeov1GfCBfr54kqUJCTCCLZ48vcsHfZ+KHS2X01AWydelYOW19SPF2Xdg9i4J/tMmrhCYdB5ghhtOGv21OXOummfVxsXGdf90AQTsg+L+NwWNmyhLro672+Fm6erMnkGVnTzsguHWj7ed6rXqa6QVOnT4r/8iT84ZAVnxtNA160Q4Ibt2Qe9IWIJDlR/0Uq9RUenRqbH1Resak3r7rR2nYrq98vnCk6JcltsQT0JcJ7WVVtmRRSZMmtTz9RBGp+cI/zfxM2ridPHOZbFo02nNA9UJ7ifbWiumym3hHmnxKmr1krQwa/VGsQJZdXWhjN1uWzFavoNYeqEfKN5bB74aaHhFsCSMQ1m2UXLYaTPH1yCr68H3m/lW4UAF5rVYVz72M+13C2PuSy7TZK821s3buMLkre1bxdl3cm/9ueaV5d1n18VDJnfPasPZRk+fLvOXrPS8nvpTJPr4JNI+Ikp//s99jy7PHN7dg7qW9SmYt/Ey27NgtEdbQwrovVZTvfvyP1+vC7lkUzONNTnnPmPepNbXDVvnl1z9k7IBwebRwQXP6+tKdLm0aKWi1w/LlziH1rDrTP+tGOyC4v5D5yzeY3u6Lp/Uz0258tGC1535mZ087ILh1E9FrrOz55XeZY/X4bd45Kt5AVnxtNLv7XXqASNMAABIaSURBVHCPmtwRcF+AQJbDOtBuuUUqNIn1gh1zI1k+Y5Dkz5vTYY7sHojAl1//IPOWrZds1twyv/3xp6zesF2eK1/SBES0K/Sy1V/EeqnTh3VGawjPyD7tAymWtF4E4gtk2dWFdo0uXOieWAFGbThFtn/NvJywJYxAfIGsk9bXv55DpplgiP55qXXN6MThn8yMkjusuZa43yWMvV0u3/ywV+q37i21q5WXdzs2Mrt7uy60R5YOx73+A4q+pEdPX3TD/HR2ZfPv3gV0uFRfa2juiN7tpVK5EmZnnj3u/2r0RXzesg3ykzWsvbnVy0d7WscMkbrZdWH3LHL/rG6PI9Dev9t2fi9Hj5+SQe+0Ep2/T7c+w6Zb0wXoYiNXZdWGr8yE/fMm9pb7783r9X5HOyCw34Xer3SY5+ShXeQJa/oGncvs+kCWt2dNneoVaAcExu81dfT0xeaj+7IPBlrDcjPL6x36xwpkeWuj7bA6UtAOCGLlkHWSFyCQ5UcV6Qu2TvqqY5h1o0eWH4hBSjL14xWiXad3rpok06w/0yMrSNBesvW3R9adWTPLkO70yApmjcUXyIpbnn4V1Ilgp7z3lpQs/pBwvwtmjVzLWyd41fmtilm9FsYPivCs8qWB95tdFzE9slbPHupZaIQeWQlfV5+u/1L0uglr/oo0b1DtpgXw7El4e19z1J5ZrazJ3xdb819etCaw1p6KN7su6JHlq2rC7DfQ6mE60+o1F9/0Dzpn49MvtZUG1nQPen15u9/RMzuw+uhoLVyhc8s9U/paQFGncvh1/yGrx3tpeduaVzGs+8ibPmvUnnZAYP7eUpd6IdQsGlao4D/Mbms27rDmvkwrVStYU9i0vjaFzfXb9W20TBnTeb3fBe+oyRmBpCFAIMuPetCx4nrz167sus1Zss7MPcMcWX5gJnCSlWu3WfPEjDY9EtZu2mnmyNIGlK6Cp5s+MHRlI+bISmD467KLL5AVMy/JzepC52fYvWefzLMWUNDtj4NHpEq9CObISuBq8iWQpcvZP2UtkjC6X5iUL/OYcL9L4EqIk53+7hu07WuGRQ/r3dbMvxSzebsuYubImjTkTSn9eGGTRF9WDljL3jNHVsLUWcz8l13avCqN/rci3s1y5tmTMOb+5KKrf1aqHW7mi3vw/vxm+NrNrgu7Z5E/5ZPm5gI6nE3vYzFzy8Xds3KdcKlYtoR0tXpf0w4I3i9JF8/5+tufPQV89e89smfv7/Kq1eNd28MDRn3otQ1GOyB4daO9445ft5LngpWfS5ZMGcz0NfF9PLm+jaYfv7zd74J31OSMQNIQIJDlRz0MHjvTBK8WWJMjp7ei5jpBH6sW+gGZAEnGvr9Qijx4n7W08wNmpY8W1tjyVKlSmRVydILrks+3klaNqlv/vcSqhQng7S0LHXarqxTNtq6NYRNmy8aFoySltSy6vpjb1cXmL7+VNyIGm0muSz72kAkMa08IVi1MmEq7Yq0adeXKFTMZ+OUrl80QKb1OdEVInStDV2Wr8HRxM5Swx5Cp8sm6bbJu3nAzTxb3u4Spg/hy2fXdL9aqt72kdInCEhnWUFKmuLZCZ8YM6cwqXnbXxctN3pEsmTNYc9B0lN8PHLG+zHaTLm3qWz0cKgfvoJNJzjrHT78RMyTUenbo6ngxm9aL1g/PHvd+CLrSp65WWLZUUeuaSWmeF59t3C4b5o8014O368LuWeTeWd0eJesQ3ErlHpdiDxeUP48cN22ytNacWNom05WNl6z6Qmq98IzkypFN5ixdZ4YajhvYScpZdWl3v7s9hJLGWcQdWmhnTzsg8eot7tBCuzYa7YDEqxtKSnoCBLL8qBONhjcLH2S+Xuimc8tMHxkpuXNl9yM3kgQioF/w9ItfzKbdc6MHdTKBRd2WfLpZuvSN9vx7+2a1pGXDFwMpkrQ3EdDroXaLHrH+tYzVy2RCVIRtXWgQTCch1SXUdQuxAmDRVuM2ZjU20AMT0EbrsAlzYmWiXdZ15SidtFpXXdVgl27ae3FIt1DzlVw37neB2XtLrb0Xe0RNvWGXmOvG7rrQIQa60MgZa6U23fRlUIOUMT1Qg3fkt3/O2ntRg+lxtzet3lnaq5dnj3u/AR0e+N742Z4DSJ8ujQyMbOm5Z9ldF7QLgld3OsRTh3rGbDpv7Jj+HU2bTANZGrjXDycxm35obNe0pvmr3f0ueEed/HKOG8iys6cdkHi/kbiBLLs2mt39LvGOnJIQSHwBAlkBmGsPoL8uXfLMTxJAViQNQEAbRTq0QCeo1q/lcTd9QdeJ4PNYgUZe8AKAToCkdnWhdXnk2AlrNaOcprcQW+II6NLP+vVcN71OUvyvZ9D1pXO/S5y6iK8Uu+tivzUUN5N1/9N7IFviCfDsSTzruCVdtnqYHjp83AQ/8uTKEe/zwtt1Yfcscu/Mbv2SL1pzX+n0AHpPitsm0/o6fPSknDl33kxorStMx93s7ne3vlDSPQM7e9oB7tSdL2002gHu1A2luitAIMtdf0pHAAEEEEAAAQQQQAABBBBAAAEEEPBRgECWj1DshgACCCCAAAIIIIAAAggggAACCCDgrgCBLHf9KR0BBBBAAAEEEEAAAQQQQAABBBBAwEcBAlk+QrEbAggggAACCCCAAAIIIIAAAggggIC7AgSy3PWndAQQQAABBBBAAAEEEEAAAQQQQAABHwUIZPkIxW4IIIAAAggggAACCCCAAAIIIIAAAu4KEMhy15/SEUAAAQQQQAABBBBAAAEEEEAAAQR8FCCQ5SMUuyGAAAIIIIAAAggggAACCCCAAAIIuCtAIMtdf0pHAAEEEEAAAQQQQAABBBBAAAEEEPBRgECWj1DshgACCCCAAAIIIIAAAggggAACCCDgrgCBLHf9KR0BBBBAAAEEEEAAAQQQQAABBBBAwEcBAlk+QrEbAggggAACCCCAAAIIIIAAAggggIC7AgSy3PWndAQQQAABBBBAAAEEEEAAAQQQQAABHwUIZPkIxW4IIIAAAggggAACCCCAAAIIIIAAAu4KEMhy15/SEUAAAQQQQAABBBBAAAEEEEAAAQR8FCCQ5SMUuyGAAAIIIIDA7SEwb9l6yZo5o1QsW8LvE9p/8Ig0CRsg/bu2kMeLFfI7HxIigAACCCCAAAIIOBMgkOXMi70RQAABBBBA4BYXeLJqS7kvfx6ZFd3d7zP5ed9+qd44Ukb2aR9QQMzvAyAhAggggAACCCCQTAUIZCXTiue0EUAAAQQQSK4CJ06ekZQhKSVzxvR+ExDI8puOhAgggAACCCCAQEACBLIC4iMxAggggAACCNxqAu3fHSH/yJNTOofWk3PnL0itN7pJjarlZPNX38pX/94j+XLfJc0bVDP/L2Y7evyURA6YIJu+/Nb8r/x5c8neXw/E6pH1497f5d1Bk2X3nn0SEhIiJR97SPp0aSZ3Zc8qPYdOky3bd8vIvh2kYIE8Jo9d3/0ib/YZJw1feU7q16h0qzFyvAgggAACCCCAgCsCBLJcYadQBBBAAAEEEHBL4Nl6EXL/vXllTP+OcvLUWSlTvY05lNIlCkvJ4g/LJ+u2yfc//SobF46SrFkyyt9/X5Vn63WSA38ekwplikuZJ4vI+i92yoYtuzyBrAOHjkrlup2kQL5c0qBmFTl6/KRMmbXCGsKYW+ZO7CUHDx+Tqg26SDYrv+UzBsnFvy7Jv+p3NnN1LZzSV1KnTuUWB+UigAACCCCAAAK3lACBrFuqujhYBBBAAAEEEAhUIL5AVtumNSS00Usm68NHT0j5WmHS+82mUvP5Z2Tl2q0S3mOMvBPWUF59+VrPqbhDC9/uN0GWrt4smxaNlowZ0pl9xn+wWIZPnCtr5gyTnDmyysZt30iLzlFStWIpOX7ytGzf9aOssIJaue7KFugpkR4BBBBAAAEEEEg2AgSykk1Vc6IIIIAAAgggoALxBbK6h78udapX8AA9Ur6xtHjtRenwRi2JGjdLpsxc7glIxRfIqtbobTPU8Pqg1NlzF+TM2fMyfWSklCj6gMl7xKS5Ej19sfnz+yO6suIhP0kEEEAAAQQQQMChAIEsh2DsjgACCCCAAAK3toDTQFaPqKkye8la2fnpRM8QwLg9sirXCZeUKVNKeMs6N+DokEUdoqjbijVbpVPPMebPC6b0kQfuzXdrY3L0CCCAAAIIIIBAIgsQyEpkcIpDAAEEEEAAAXcFnAayJsxYIsMmzInVs+qnvfvlpSaRnjmyWnUZYiaKXzdvuKRPl9ZzglevXpUUKVKYv2uPLU1TtmRR2fntTxJiBb5WfDhYMqT///7uylA6AggggAACCCCQ9AUIZCX9OuIIEUAAAQQQQCABBZwGsn4/cFiee7WzGTbYrmlN0eDUqCnz5dDh455A1o5vfpTX2vaVwoXukYjQupIlUwb52gpWRVvzZH0c3cMKbqUxk72HhKQ0k73rZPL1W/eWMk8UkQlREQl4dmSFAAIIIIAAAgjc3gIEsm7v+uXsEEAAAQQQQCCOQKxA1mlr1cIX20j3To2lzovlPXvqHFktG74o7ZvVMv/vw/mrpe/w6Z5/L1eqqFm1cFS/DmYlQ910Uvhug6eYebFitofuz2/15Ooqb/aOljWbdsiiqX2l4D15zT/HTAYf0aquNKlXlXpCAAEEEEAAAQQQ8EGAQJYPSOyCAAIIIIAAAgicO39RtHdWgXy5JM0dqW8KoisSnjx1VnLnyu51P0QRQAABBBBAAAEEnAsQyHJuRgoEEEAAAQQQQAABBBBAAAEEEEAAARcECGS5gE6RCCCAAAIIIIAAAggggAACCCCAAALOBQhkOTcjBQIIIIAAAggggAACCCCAAAIIIICACwIEslxAp0gEEEAAAQQQQAABBBBAAAEEEEAAAecCBLKcm5ECAQQQQAABBBBAAAEEEEAAAQQQQMAFAQJZLqBTJAIIIIAAAggggAACCCCAAAIIIICAcwECWc7NSIEAAggggAACCCCAAAIIIIAAAggg4IIAgSwX0CkSAQQQQAABBBBAAAEEEEAAAQQQQMC5AIEs52akQAABBBBAAAEEEEAAAQQQQAABBBBwQYBAlgvoFIkAAggggAACCCCAAAIIIIAAAggg4FyAQJZzM1IggAACCCCAAAIIIIAAAggggAACCLggQCDLBXSKRAABBBBAAAEEEEAAAQQQQAABBBBwLkAgy7kZKRBAAAEEEEAAAQQQQAABBBBAAAEEXBAgkOUCOkUigAACCCCAAAIIIIAAAggggAACCDgXIJDl3IwUCCCAAAIIIIAAAggggAACCCCAAAIuCBDIcgGdIhFAAAEEEEAAAQQQQAABBBBAAAEEnAsQyHJuRgoEEEAAAQQQQAABBBBAAAEEEEAAARcECGS5gE6RCCCAAAIIIIAAAggggAACCCCAAALOBQhkOTcjBQIIIIAAAggggAACCCCAAAIIIICACwIEslxAp0gEEEAAAQQQQAABBBBAAAEEEEAAAecCBLKcm5ECAQQQQAABBBBAAAEEEEAAAQQQQMAFAQJZLqBTJAIIIIAAAggggAACCCCAAAIIIICAcwECWc7NSIEAAggggAACCCCAAAIIIIAAAggg4IIAgSwX0CkSAQQQQAABBBBAAAEEEEAAAQQQQMC5AIEs52akQAABBBBAAAEEEEAAAQQQQAABBBBwQYBAlgvoFIkAAggggAACCCCAAAIIIIAAAggg4FyAQJZzM1IggAACCCCAAAIIIIAAAggggAACCLggQCDLBXSKRAABBBBAAAEEEEAAAQQQQAABBBBwLkAgy7kZKRBAAAEEEEAAAQQQQAABBBBAAAEEXBAgkOUCOkUigAACCCCAAAIIIIAAAggggAACCDgXIJDl3IwUCCCAAAIIIIAAAggggAACCCCAAAIuCBDIcgGdIhFAAAEEEEAAAQQQQAABBBBAAAEEnAsQyHJuRgoEEEAAAQQQQAABBBBAAAEEEEAAARcECGS5gE6RCCCAAAIIIIAAAggggAACCCCAAALOBQhkOTcjBQIIIIAAAggggAACCCCAAAIIIICACwIEslxAp0gEEEAAAQQQQAABBBBAAAEEEEAAAecCBLKcm5ECAQQQQAABBBBAAAEEEEAAAQQQQMAFAQJZLqBTJAIIIIAAAggggAACCCCAAAIIIICAcwECWc7NSIEAAggggAACCCCAAAIIIIAAAggg4ILAfwELA7AAKmQ2bQAAAABJRU5ErkJggg==", "text/html": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "px.line(retriesQ)" ] }, { "cell_type": "markdown", "id": "4828a683-6bc6-451b-b60b-e5890254fdfd", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "Betrachten wir die Belohnungen, so sehen wir, dass der Ansatz sehr schnell nur noch positive Belohnungen erzieht." ] }, { "cell_type": "code", "execution_count": 11, "id": "df889aee-f65d-4438-a3a8-bf2fdf4a2d7b", "metadata": {}, "outputs": [ { "data": { "application/vnd.plotly.v1+json": { "config": { "plotlyServerURL": "https://plot.ly" }, "data": [ { "hovertemplate": "variable=0
index=%{x}
value=%{y}", "legendgroup": "0", "line": { "color": "#636efa", "dash": "solid" }, "marker": { "symbol": "circle" }, "mode": "lines", "name": "0", "orientation": "v", "showlegend": true, "type": "scatter", "x": [ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 191, 192, 193, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203, 204, 205, 206, 207, 208, 209, 210, 211, 212, 213, 214, 215, 216, 217, 218, 219, 220, 221, 222, 223, 224, 225, 226, 227, 228, 229, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 285, 286, 287, 288, 289, 290, 291, 292, 293, 294, 295, 296, 297, 298, 299, 300, 301, 302, 303, 304, 305, 306, 307, 308, 309, 310, 311, 312, 313, 314, 315, 316, 317, 318, 319, 320, 321, 322, 323, 324, 325, 326, 327, 328, 329, 330, 331, 332, 333, 334, 335, 336, 337, 338, 339, 340, 341, 342, 343, 344, 345, 346, 347, 348, 349, 350, 351, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442, 443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468, 469, 470, 471, 472, 473, 474, 475, 476, 477, 478, 479, 480, 481, 482, 483, 484, 485, 486, 487, 488, 489, 490, 491, 492, 493, 494, 495, 496, 497, 498, 499 ], "xaxis": "x", "y": [ -5.8999999999999915, -2.900000000000002, 0.4, -2.700000000000002, 0.30000000000000004, 0.20000000000000007, -0.09999999999999987, -0.30000000000000004, -0.19999999999999996, 0.4, 0.30000000000000004, -0.6000000000000003, 0.4, 0.5, 1.1102230246251565e-16, 0.4, 0.10000000000000009, 0.4, 0.10000000000000009, 0.30000000000000004, 0.5, 0.4, 0.4, 0.4, 0.4, 0.5, 0.5, 0.5, 0.5, 0.30000000000000004, 0.5, 0.5, 0.30000000000000004, 0.30000000000000004, 0.5, 0.5, 0.5, 0.5, 0.4, 0.5, 0.5, 0.5, 1.1102230246251565e-16, 0.5, 0.5, 0.30000000000000004, 0.5, 0.30000000000000004, 0.5, 0.30000000000000004, 0.10000000000000009, 0.5, 0.4, 1.1102230246251565e-16, 0.5, 0.30000000000000004, 0.5, 0.5, 0.5, 0.5, 0.5, 0.10000000000000009, 0.5, 0.30000000000000004, 0.30000000000000004, 0.30000000000000004, 0.5, 0.4, 0.5, 0.30000000000000004, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.30000000000000004, 0.5, 0.10000000000000009, 0.5, 0.4, 0.5, 1.1102230246251565e-16, 0.5, 0.5, 0.30000000000000004, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.30000000000000004, 0.30000000000000004, 0.5, 0.10000000000000009, 0.5, 0.30000000000000004, 0.5, 0.5, 0.5, 0.5, 0.30000000000000004, 0.4, 0.5, 0.5, 0.4, 0.30000000000000004, 0.30000000000000004, 0.4, 0.5, 0.5, 0.5, 0.30000000000000004, 0.30000000000000004, 0.5, 0.5, 0.10000000000000009, 0.30000000000000004, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.30000000000000004, 0.5, 0.5, 0.5, 0.5, 0.5, 0.30000000000000004, 0.5, 0.10000000000000009, 0.5, 0.5, 0.5, 0.20000000000000007, 0.5, 0.5, 0.5, 0.5, 0.4, 0.30000000000000004, 0.4, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.30000000000000004, 0.5, 0.30000000000000004, 0.5, 0.4, 0.5, 0.10000000000000009, 0.30000000000000004, 0.5, 0.5, 0.30000000000000004, 0.5, 0.30000000000000004, 0.30000000000000004, 0.30000000000000004, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.10000000000000009, 0.30000000000000004, 0.5, 0.5, 0.30000000000000004, 0.4, 0.5, 0.4, 0.4, 0.5, 0.30000000000000004, 0.5, 0.5, 0.20000000000000007, 0.30000000000000004, 0.30000000000000004, 0.5, 0.4, 0.4, 0.10000000000000009, 0.5, 0.30000000000000004, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.4, 0.30000000000000004, 0.10000000000000009, 0.10000000000000009, 0.5, 0.30000000000000004, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.30000000000000004, 0.4, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.20000000000000007, 1.1102230246251565e-16, 0.5, 0.5, 0.5, 0.20000000000000007, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.4, 0.5, 0.30000000000000004, 0.5, 0.4, 0.5, 0.30000000000000004, 0.5, 0.30000000000000004, 0.5, 0.5, 0.4, 0.5, 0.5, 0.30000000000000004, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.30000000000000004, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.30000000000000004, 0.5, 0.5, 0.5, 0.10000000000000009, 0.5, 0.5, 0.5, 1.1102230246251565e-16, 0.30000000000000004, 0.5, 0.5, 0.5, 0.5, 0.5, 0.4, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.30000000000000004, 0.5, 0.5, 0.5, 0.30000000000000004, 0.30000000000000004, 0.5, 0.30000000000000004, 0.30000000000000004, 0.4, 0.5, 0.5, 0.5, 0.5, 0.5, 0.10000000000000009, 0.30000000000000004, 0.5, 0.5, 0.5, 0.4, 0.5, 0.30000000000000004, 0.5, 0.5, 0.5, 0.30000000000000004, 0.30000000000000004, 0.30000000000000004, 0.5, 0.5, 0.5, 0.30000000000000004, 0.10000000000000009, 0.5, 0.30000000000000004, 0.5, 0.5, 0.5, 0.5, 0.30000000000000004, 0.5, 0.5, 0.5, 0.30000000000000004, 0.10000000000000009, 0.5, 0.5, 0.5, 0.5, 0.20000000000000007, 1.1102230246251565e-16, 0.10000000000000009, 0.5, 0.5, 0.4, 0.5, 0.30000000000000004, 0.30000000000000004, 0.4, 0.30000000000000004, 0.4, 0.30000000000000004, 0.5, 0.5, 0.5, 0.4, 0.5, 0.5, 0.30000000000000004, 0.5, 0.5, 0.30000000000000004, 0.20000000000000007, 0.4, 0.30000000000000004, 0.30000000000000004, 0.5, 0.4, 0.5, 0.5, 0.5, 0.5, 0.4, 0.5, 0.5, 0.30000000000000004, 0.10000000000000009, 0.4, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.30000000000000004, 0.30000000000000004, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.10000000000000009, 0.30000000000000004, 0.5, 0.5, 0.30000000000000004, 0.5, 0.10000000000000009, 0.5, 0.30000000000000004, 0.5, 0.30000000000000004, 0.5, 0.10000000000000009, 0.5, 0.4, 0.5, 0.5, 0.5, 0.30000000000000004, 0.5, 0.20000000000000007, 0.5, 0.5, 0.5, 0.4, 0.4, 0.30000000000000004, 0.30000000000000004, 0.30000000000000004, 0.30000000000000004, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.30000000000000004, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.20000000000000007, 0.5, 0.5, 0.30000000000000004, 0.5, 0.5, 0.30000000000000004, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.30000000000000004, 0.5, 0.5, 0.5, 0.20000000000000007, 0.5, 0.30000000000000004, 0.5, 0.5, 0.5, 0.5, 0.5, 0.4, 0.30000000000000004, 0.5, 0.5, 0.5, 0.5, 0.4 ], "yaxis": "y" } ], "layout": { "autosize": true, "legend": { "title": { "text": "variable" }, "tracegroupgap": 0 }, "margin": { "t": 60 }, "template": { "data": { "bar": [ { "error_x": { "color": "#2a3f5f" }, "error_y": { "color": "#2a3f5f" }, "marker": { "line": { "color": "#E5ECF6", "width": 0.5 }, "pattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 } }, "type": "bar" } ], "barpolar": [ { "marker": { "line": { "color": "#E5ECF6", "width": 0.5 }, "pattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 } }, "type": "barpolar" } ], "carpet": [ { "aaxis": { "endlinecolor": "#2a3f5f", "gridcolor": "white", "linecolor": "white", "minorgridcolor": "white", "startlinecolor": "#2a3f5f" }, "baxis": { "endlinecolor": "#2a3f5f", "gridcolor": "white", "linecolor": "white", "minorgridcolor": "white", "startlinecolor": "#2a3f5f" }, "type": "carpet" } ], "choropleth": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "choropleth" } ], "contour": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "contour" } ], "contourcarpet": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "contourcarpet" } ], "heatmap": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "heatmap" } ], "heatmapgl": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "heatmapgl" } ], "histogram": [ { "marker": { "pattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 } }, "type": "histogram" } ], "histogram2d": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "histogram2d" } ], "histogram2dcontour": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "histogram2dcontour" } ], "mesh3d": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "mesh3d" } ], "parcoords": [ { "line": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "parcoords" } ], "pie": [ { "automargin": true, "type": "pie" } ], "scatter": [ { "fillpattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 }, "type": "scatter" } ], "scatter3d": [ { "line": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatter3d" } ], "scattercarpet": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattercarpet" } ], "scattergeo": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattergeo" } ], "scattergl": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattergl" } ], "scattermapbox": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattermapbox" } ], "scatterpolar": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterpolar" } ], "scatterpolargl": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterpolargl" } ], "scatterternary": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterternary" } ], "surface": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "surface" } ], "table": [ { "cells": { "fill": { "color": "#EBF0F8" }, "line": { "color": "white" } }, "header": { "fill": { "color": "#C8D4E3" }, "line": { "color": "white" } }, "type": "table" } ] }, "layout": { "annotationdefaults": { "arrowcolor": "#2a3f5f", "arrowhead": 0, "arrowwidth": 1 }, "autotypenumbers": "strict", "coloraxis": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "colorscale": { "diverging": [ [ 0, "#8e0152" ], [ 0.1, "#c51b7d" ], [ 0.2, "#de77ae" ], [ 0.3, "#f1b6da" ], [ 0.4, "#fde0ef" ], [ 0.5, "#f7f7f7" ], [ 0.6, "#e6f5d0" ], [ 0.7, "#b8e186" ], [ 0.8, "#7fbc41" ], [ 0.9, "#4d9221" ], [ 1, "#276419" ] ], "sequential": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "sequentialminus": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ] }, "colorway": [ "#636efa", "#EF553B", "#00cc96", "#ab63fa", "#FFA15A", "#19d3f3", "#FF6692", "#B6E880", "#FF97FF", "#FECB52" ], "font": { "color": "#2a3f5f" }, "geo": { "bgcolor": "white", "lakecolor": "white", "landcolor": "#E5ECF6", "showlakes": true, "showland": true, "subunitcolor": "white" }, "hoverlabel": { "align": "left" }, "hovermode": "closest", "mapbox": { "style": "light" }, "paper_bgcolor": "white", "plot_bgcolor": "#E5ECF6", "polar": { "angularaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "bgcolor": "#E5ECF6", "radialaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" } }, "scene": { "xaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" }, "yaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" }, "zaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" } }, "shapedefaults": { "line": { "color": "#2a3f5f" } }, "ternary": { "aaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "baxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "bgcolor": "#E5ECF6", "caxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" } }, "title": { "x": 0.05 }, "xaxis": { "automargin": true, "gridcolor": "white", "linecolor": "white", "ticks": "", "title": { "standoff": 15 }, "zerolinecolor": "white", "zerolinewidth": 2 }, "yaxis": { "automargin": true, "gridcolor": "white", "linecolor": "white", "ticks": "", "title": { "standoff": 15 }, "zerolinecolor": "white", "zerolinewidth": 2 } } }, "xaxis": { "anchor": "y", "autorange": true, "domain": [ 0, 1 ], "range": [ 0, 499 ], "title": { "text": "index" }, "type": "linear" }, "yaxis": { "anchor": "x", "autorange": true, "domain": [ 0, 1 ], "range": [ -6.255555555555547, 0.8555555555555551 ], "title": { "text": "value" }, "type": "linear" } } }, "image/png": "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", "text/html": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "px.line(rewardsQ)" ] }, { "cell_type": "markdown", "id": "e5f6161b-fc63-4549-aef3-5bbc70e38d73", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "Der resultierende Weg ist hierbei einer von vielen optimalen Wegen." ] }, { "cell_type": "code", "execution_count": 12, "id": "f20a43bb-d227-44ef-b50a-db710e465e0c", "metadata": {}, "outputs": [ { "data": { "application/vnd.plotly.v1+json": { "config": { "plotlyServerURL": "https://plot.ly" }, "data": [ { "coloraxis": "coloraxis", "hovertemplate": "x: %{x}
y: %{y}
color: %{z}", "name": "0", "type": "heatmap", "xaxis": "x", "yaxis": "y", "z": [ [ 80, 60, 66, 28 ], [ 547, 508, 515, 512 ], [ 42, 51, 63, 484 ], [ 14, 10, 63, 500 ] ] } ], "layout": { "autosize": true, "coloraxis": { "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ] }, "margin": { "t": 60 }, "template": { "data": { "bar": [ { "error_x": { "color": "#2a3f5f" }, "error_y": { "color": "#2a3f5f" }, "marker": { "line": { "color": "#E5ECF6", "width": 0.5 }, "pattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 } }, "type": "bar" } ], "barpolar": [ { "marker": { "line": { "color": "#E5ECF6", "width": 0.5 }, "pattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 } }, "type": "barpolar" } ], "carpet": [ { "aaxis": { "endlinecolor": "#2a3f5f", "gridcolor": "white", "linecolor": "white", "minorgridcolor": "white", "startlinecolor": "#2a3f5f" }, "baxis": { "endlinecolor": "#2a3f5f", "gridcolor": "white", "linecolor": "white", "minorgridcolor": "white", "startlinecolor": "#2a3f5f" }, "type": "carpet" } ], "choropleth": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "choropleth" } ], "contour": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "contour" } ], "contourcarpet": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "contourcarpet" } ], "heatmap": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "heatmap" } ], "heatmapgl": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "heatmapgl" } ], "histogram": [ { "marker": { "pattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 } }, "type": "histogram" } ], "histogram2d": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "histogram2d" } ], "histogram2dcontour": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "histogram2dcontour" } ], "mesh3d": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "mesh3d" } ], "parcoords": [ { "line": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "parcoords" } ], "pie": [ { "automargin": true, "type": "pie" } ], "scatter": [ { "fillpattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 }, "type": "scatter" } ], "scatter3d": [ { "line": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatter3d" } ], "scattercarpet": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattercarpet" } ], "scattergeo": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattergeo" } ], "scattergl": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattergl" } ], "scattermapbox": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattermapbox" } ], "scatterpolar": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterpolar" } ], "scatterpolargl": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterpolargl" } ], "scatterternary": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterternary" } ], "surface": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "surface" } ], "table": [ { "cells": { "fill": { "color": "#EBF0F8" }, "line": { "color": "white" } }, "header": { "fill": { "color": "#C8D4E3" }, "line": { "color": "white" } }, "type": "table" } ] }, "layout": { "annotationdefaults": { "arrowcolor": "#2a3f5f", "arrowhead": 0, "arrowwidth": 1 }, "autotypenumbers": "strict", "coloraxis": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "colorscale": { "diverging": [ [ 0, "#8e0152" ], [ 0.1, "#c51b7d" ], [ 0.2, "#de77ae" ], [ 0.3, "#f1b6da" ], [ 0.4, "#fde0ef" ], [ 0.5, "#f7f7f7" ], [ 0.6, "#e6f5d0" ], [ 0.7, "#b8e186" ], [ 0.8, "#7fbc41" ], [ 0.9, "#4d9221" ], [ 1, "#276419" ] ], "sequential": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "sequentialminus": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ] }, "colorway": [ "#636efa", "#EF553B", "#00cc96", "#ab63fa", "#FFA15A", "#19d3f3", "#FF6692", "#B6E880", "#FF97FF", "#FECB52" ], "font": { "color": "#2a3f5f" }, "geo": { "bgcolor": "white", "lakecolor": "white", "landcolor": "#E5ECF6", "showlakes": true, "showland": true, "subunitcolor": "white" }, "hoverlabel": { "align": "left" }, "hovermode": "closest", "mapbox": { "style": "light" }, "paper_bgcolor": "white", "plot_bgcolor": "#E5ECF6", "polar": { "angularaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "bgcolor": "#E5ECF6", "radialaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" } }, "scene": { "xaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" }, "yaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" }, "zaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" } }, "shapedefaults": { "line": { "color": "#2a3f5f" } }, "ternary": { "aaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "baxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "bgcolor": "#E5ECF6", "caxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" } }, "title": { "x": 0.05 }, "xaxis": { "automargin": true, "gridcolor": "white", "linecolor": "white", "ticks": "", "title": { "standoff": 15 }, "zerolinecolor": "white", "zerolinewidth": 2 }, "yaxis": { "automargin": true, "gridcolor": "white", "linecolor": "white", "ticks": "", "title": { "standoff": 15 }, "zerolinecolor": "white", "zerolinewidth": 2 } } }, "xaxis": { "anchor": "y", "autorange": true, "constrain": "domain", "domain": [ 0.3915187376725838, 0.6084812623274162 ], "range": [ -0.5, 3.5 ], "scaleanchor": "y" }, "yaxis": { "anchor": "x", "autorange": true, "constrain": "domain", "domain": [ 0, 1 ], "range": [ 3.5, -0.5 ] } } }, "image/png": "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", "text/html": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "px.imshow(env.grid)" ] }, { "cell_type": "markdown", "id": "d7571526", "metadata": { "editable": true, "slideshow": { "slide_type": "slide" }, "tags": [] }, "source": [ "## Zeitreihe" ] }, { "cell_type": "markdown", "id": "fed37c82-3efd-4bd4-83ce-fdc95d40b3ed", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "Mit Reinforcement Learning kann man auch Modelle auf Zeitreihen trainieren. Wir wollen uns zum Beispiel einen Entscheidungsalgorithmus trainieren, welcher lernt, wann er Aktien oder Shorts auf diese kaufen soll und wann er sie verkaufen soll. Hier ein zufälliger Aktienkurs." ] }, { "cell_type": "code", "execution_count": 13, "id": "0536c5df-3b7d-41bd-8cfe-a64186a09395", "metadata": {}, "outputs": [ { "data": { "application/vnd.plotly.v1+json": { "config": { "plotlyServerURL": "https://plot.ly" }, "data": [ { "hovertemplate": "variable=0
index=%{x}
value=%{y}", "legendgroup": "0", "line": { "color": "#636efa", "dash": "solid" }, "marker": { "symbol": "circle" }, "mode": "lines", "name": "0", "orientation": "v", "showlegend": true, "type": "scatter", "x": [ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 191, 192, 193, 194, 195, 196, 197, 198, 199 ], "xaxis": "x", "y": [ 0.9967141530112327, 1.358449851840048, 2.5061383899407406, 4.529168246348766, 4.79501487162543, 5.060877914676249, 7.1400907301836405, 8.407525459336549, 8.438051073401597, 9.480611116987562, 9.517193424175101, 9.551463670604845, 10.29342594217088, 8.88014569751308, 7.6552278650000485, 7.592940335759076, 7.0801092154246525, 7.8943565480199265, 7.486332472498716, 6.574028771163424, 8.539677540084979, 8.813901239598444, 9.381429444286368, 8.456681258072912, 8.41229853354773, 9.023221123257596, 8.372227545835292, 9.247925564180964, 9.147286874262159, 9.355593124468882, 9.253886512239486, 11.606164696748424, 12.09266747201049, 11.53495654305459, 12.85750145515778, 12.136657805186758, 12.845521400191513, 11.385851276311739, 10.557665227413308, 11.254526463282431, 12.492993043277842, 13.164361324467812, 13.548713042079571, 13.747609346490282, 12.769087356122855, 12.549243147728147, 12.588604376768359, 14.145726602987274, 14.989344892555735, 13.726304737193, 14.550388706587796, 14.66530642617148, 14.48838442586552, 15.600060714706387, 17.131060237202338, 18.562340356318536, 18.223122833095896, 18.413910457244683, 19.245173888648246, 20.720719015770605, 20.741544777925316, 21.0558858012615, 20.44955082725547, 19.7533442031748, 21.065870025568998, 22.92211005413982, 23.350099932559488, 24.853632830451513, 25.715268855499147, 25.570149100894024, 26.431544706402438, 28.469581272868407, 28.933755233758454, 30.99839888957246, 28.878653785482715, 30.20055628985794, 30.78760335809611, 30.988596007630242, 31.580356784165744, 30.09278786956485, 30.373115981727338, 31.230228553239083, 33.2081225979806, 33.18985237970695, 32.88135877681376, 32.87960173322922, 34.295003850931295, 35.12375496059098, 35.09399475682394, 36.107262189937295, 36.70433973928534, 38.172984729818225, 37.97093163594087, 38.143269489343105, 38.25116133621095, 37.28764638807883, 38.0837666651434, 38.84482193732329, 39.34993539396575, 39.6153482605906, 38.69997751854019, 38.77933219577483, 38.93661767924806, 38.63434041002644, 38.97305469836043, 39.877105555174964, 42.263291456385495, 42.93786926921734, 43.6954196599401, 44.12097374417393, 42.70220252887489, 43.175688653425674, 43.7359188633667, 46.69916097585198, 47.00680001107086, 47.80834735340447, 48.273635583699225, 47.604957546079696, 49.24778036059472, 50.49971339328149, 51.790745340324534, 51.381357885529795, 53.2841521964659, 52.382301133673614, 53.46915822747388, 56.159613853283865, 55.669077528153174, 55.6027797985504, 56.20243116363804, 56.19895550952184, 55.148292078455704, 55.71685505326173, 55.15455133953563, 56.12814377017081, 55.70871953593701, 57.758653940954545, 57.47540064861831, 57.65333913241263, 58.966856349782304, 58.235992033348346, 58.96345196795247, 60.770594722234904, 59.663111487673675, 60.347745346205976, 61.1076281404544, 62.38945101223171, 61.652500301353626, 60.83204368826935, 61.85398525388625, 62.650969927119434, 63.40146277746531, 64.24791098696228, 64.06788626538379, 64.80013996254479, 65.59321243584347, 65.3788610178171, 67.74463552896185, 68.71846844987364, 68.02716495267099, 69.18371856130481, 68.70903689107749, 69.99612149481995, 71.65471707382736, 71.33403475547564, 72.79741088471997, 73.71019181165646, 75.03225197165095, 77.4290449543049, 77.68365683830203, 77.42992067394454, 77.04040624431902, 76.72459595935358, 77.14749424993947, 77.98864622475611, 78.76533702408612, 80.09252027312215, 80.60552216500005, 82.55905624215737, 82.79439940891942, 86.01456857550903, 87.14023592327403, 86.78307836685775, 86.21218586879664, 87.19465828403982, 87.47119549871397, 88.68519599280606, 89.6584336173796, 90.08560470472273, 89.73881098665431, 88.72396376196845, 88.77744880990143, 90.1338476042249, 90.8479413483551, 90.10220256964313, 90.77538349549431, 91.66070087522314, 91.276843439022, 91.93056854496753, 92.48877726341352, 91.8458069655829 ], "yaxis": "y" } ], "layout": { "autosize": true, "legend": { "title": { "text": "variable" }, "tracegroupgap": 0 }, "margin": { "t": 60 }, "template": { "data": { "bar": [ { "error_x": { "color": "#2a3f5f" }, "error_y": { "color": "#2a3f5f" }, "marker": { "line": { "color": "#E5ECF6", "width": 0.5 }, "pattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 } }, "type": "bar" } ], "barpolar": [ { "marker": { "line": { "color": "#E5ECF6", "width": 0.5 }, "pattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 } }, "type": "barpolar" } ], "carpet": [ { "aaxis": { "endlinecolor": "#2a3f5f", "gridcolor": "white", "linecolor": "white", "minorgridcolor": "white", "startlinecolor": "#2a3f5f" }, "baxis": { "endlinecolor": "#2a3f5f", "gridcolor": "white", "linecolor": "white", "minorgridcolor": "white", "startlinecolor": "#2a3f5f" }, "type": "carpet" } ], "choropleth": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "choropleth" } ], "contour": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "contour" } ], "contourcarpet": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "contourcarpet" } ], "heatmap": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "heatmap" } ], "heatmapgl": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "heatmapgl" } ], "histogram": [ { "marker": { "pattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 } }, "type": "histogram" } ], "histogram2d": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "histogram2d" } ], "histogram2dcontour": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "histogram2dcontour" } ], "mesh3d": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "mesh3d" } ], "parcoords": [ { "line": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "parcoords" } ], "pie": [ { "automargin": true, "type": "pie" } ], "scatter": [ { "fillpattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 }, "type": "scatter" } ], "scatter3d": [ { "line": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatter3d" } ], "scattercarpet": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattercarpet" } ], "scattergeo": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattergeo" } ], "scattergl": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattergl" } ], "scattermapbox": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattermapbox" } ], "scatterpolar": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterpolar" } ], "scatterpolargl": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterpolargl" } ], "scatterternary": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterternary" } ], "surface": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "surface" } ], "table": [ { "cells": { "fill": { "color": "#EBF0F8" }, "line": { "color": "white" } }, "header": { "fill": { "color": "#C8D4E3" }, "line": { "color": "white" } }, "type": "table" } ] }, "layout": { "annotationdefaults": { "arrowcolor": "#2a3f5f", "arrowhead": 0, "arrowwidth": 1 }, "autotypenumbers": "strict", "coloraxis": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "colorscale": { "diverging": [ [ 0, "#8e0152" ], [ 0.1, "#c51b7d" ], [ 0.2, "#de77ae" ], [ 0.3, "#f1b6da" ], [ 0.4, "#fde0ef" ], [ 0.5, "#f7f7f7" ], [ 0.6, "#e6f5d0" ], [ 0.7, "#b8e186" ], [ 0.8, "#7fbc41" ], [ 0.9, "#4d9221" ], [ 1, "#276419" ] ], "sequential": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "sequentialminus": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ] }, "colorway": [ "#636efa", "#EF553B", "#00cc96", "#ab63fa", "#FFA15A", "#19d3f3", "#FF6692", "#B6E880", "#FF97FF", "#FECB52" ], "font": { "color": "#2a3f5f" }, "geo": { "bgcolor": "white", "lakecolor": "white", "landcolor": "#E5ECF6", "showlakes": true, "showland": true, "subunitcolor": "white" }, "hoverlabel": { "align": "left" }, "hovermode": "closest", "mapbox": { "style": "light" }, "paper_bgcolor": "white", "plot_bgcolor": "#E5ECF6", "polar": { "angularaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "bgcolor": "#E5ECF6", "radialaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" } }, "scene": { "xaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" }, "yaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" }, "zaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" } }, "shapedefaults": { "line": { "color": "#2a3f5f" } }, "ternary": { "aaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "baxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "bgcolor": "#E5ECF6", "caxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" } }, "title": { "x": 0.05 }, "xaxis": { "automargin": true, "gridcolor": "white", "linecolor": "white", "ticks": "", "title": { "standoff": 15 }, "zerolinecolor": "white", "zerolinewidth": 2 }, "yaxis": { "automargin": true, "gridcolor": "white", "linecolor": "white", "ticks": "", "title": { "standoff": 15 }, "zerolinecolor": "white", "zerolinewidth": 2 } } }, "xaxis": { "anchor": "y", "autorange": true, "domain": [ 0, 1 ], "range": [ 0, 199 ], "title": { "text": "index" }, "type": "linear" }, "yaxis": { "anchor": "x", "autorange": true, "domain": [ 0, 1 ], "range": [ -4.086178242011117, 97.57166965843588 ], "title": { "text": "value" }, "type": "linear" } } }, "image/png": "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", "text/html": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "import numpy as np\n", "import pandas as pd\n", "\n", "# Simulierte Zeitreihe (z.B. Aktienpreise)\n", "np.random.seed(42)\n", "data = np.cumsum(np.random.randn(200) + 0.5)\n", "\n", "# Daten visualisieren\n", "px.line(data)" ] }, { "cell_type": "markdown", "id": "975a7592-88fe-4c32-b1a9-a35a3a358590", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "Wir definieren uns wieder eine Trainingsumgebung. Dieses Mal haben wir allerdings die Schwierigkeit, dass der Wert kontinuierlich sind und nicht diskret, wir also keine natürliche Zustandsraumdarstellung haben. Deshalb müssen wir die kontinuierliche Zeitreihe des Aktienwertes zuerst diskretisieren." ] }, { "cell_type": "code", "execution_count": 14, "id": "0e244a42-e706-46e6-88a4-fb0ed96de67a", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "outputs": [], "source": [ "class StockTradingEnv:\n", " def __init__(self, data, num_states=40):\n", " self.data = data\n", " self.current_step = 0\n", " self.state = None\n", " self.states = np.linspace(min(data), max(data), num_states)\n", " self.done = False\n", " self.actions = ['kaufen', 'verkaufen', 'halten']\n", " self.position = 0 # 1 = long, -1 = short, 0 = neutral\n", " self.balance = 0\n", " self.current_price=0\n", " self.old_price = 0\n", "\n", " def discretize(self):\n", " return np.digitize(self.data[self.current_step], self.states) - 1\n", "\n", " def reset(self):\n", " self.current_step = 0\n", " #self.state = self.data[self.current_step:self.current_step+5]\n", " self.state = self.discretize()\n", " self.done = False\n", " self.position = 0\n", " self.balance = 0\n", " self.old_price = 0\n", " self.current_price=0\n", " return self.state\n", "\n", " def step(self, action):\n", " self.current_step += 1\n", " if self.current_step > len(self.data) - 2:\n", " self.done = True\n", " self.current_price = self.data[self.current_step]\n", " reward = 0\n", " if action == 0: # Kaufen\n", " if self.position == 0: # Kaufe Aktie\n", " self.position = 1\n", " self.balance -= self.current_price\n", " self.old_price = self.current_price\n", " elif self.position == -1: # Verkaufe Short\n", " reward = 2 * (self.old_price - self.current_price)\n", " self.position = 0\n", " self.balance += self.old_price\n", " elif action == 1: # Verkaufen\n", " if self.position == 0: # Kaufe Short\n", " self.position = -1\n", " self.balance += self.current_price\n", " self.old_price = self.current_price\n", " elif self.position == 1: # Verkaufe Aktie\n", " reward = 2 * (self.current_price - self.old_price)\n", " self.position = 0\n", " self.balance -= self.old_price\n", " elif action == 2: # Halten\n", " pass\n", " self.state = self.discretize()\n", " return self.state, reward, self.done" ] }, { "cell_type": "markdown", "id": "c59d80c9-eb66-4f07-9b7d-df307f4e93fa", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "Als Beispiel kaufen wir Akten, halten sie für 6 Züge und verkaufen sie. Dann kaufen wir Shorts, um auf fallende Aktien zu wetten, halten sie wieder und verkaufen sie." ] }, { "cell_type": "code", "execution_count": 15, "id": "5d180ad5-ed3f-4192-9baf-efb4d3c12f64", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Startzustand: 0\n", "Nächster Zustand: 0 Belohnung: 0 Ziel erreicht: False\n", "Nächster Zustand: 0 Belohnung: 0 Ziel erreicht: False\n", "Nächster Zustand: 1 Belohnung: 0 Ziel erreicht: False\n", "Nächster Zustand: 1 Belohnung: 0 Ziel erreicht: False\n", "Nächster Zustand: 1 Belohnung: 0 Ziel erreicht: False\n", "Nächster Zustand: 2 Belohnung: 0 Ziel erreicht: False\n", "Nächster Zustand: 3 Belohnung: 14.098151214993003 Ziel erreicht: False\n", "Nächster Zustand: 3 Belohnung: 0 Ziel erreicht: False\n", "Nächster Zustand: 3 Belohnung: 0 Ziel erreicht: False\n", "Nächster Zustand: 3 Belohnung: 0 Ziel erreicht: False\n", "Nächster Zustand: 3 Belohnung: 0 Ziel erreicht: False\n", "Nächster Zustand: 3 Belohnung: 0 Ziel erreicht: False\n", "Nächster Zustand: 3 Belohnung: 0 Ziel erreicht: False\n", "Nächster Zustand: 2 Belohnung: 1.565646416803098 Ziel erreicht: False\n" ] } ], "source": [ "num_states=40\n", "\n", "# Beispiel für die Erstellung und Verwendung der StockTradingEnv-Umgebung\n", "env = StockTradingEnv(data, num_states)\n", "state = env.reset()\n", "print(\"Startzustand:\", state)\n", "\n", "# Beispiel für einen Schritt in der Umgebung\n", "for action in [0,2,2,2,2,2,1,1,2,2,2,2,2,0]:# 0 = Kaufen, 2 = Halten, 1 = Verkaufen\n", " next_state, reward, done = env.step(action) \n", " print(\"Nächster Zustand:\", next_state, \"Belohnung:\", reward, \"Ziel erreicht:\", done)" ] }, { "cell_type": "markdown", "id": "7c4b606e-e09c-4a9d-aad2-7bddf57e5dd1", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "Als nächstes implementieren wir wieder den Q-Learning Algorithmus. Der ist fast identisch zu dem Gridworld-Beispiel. Unterschiede gibt es nur, da wir nur eine zweidimensionale Q-Matrix haben, statt einer dreidimensionalen, da der Zustandsraum weniger Dimensionen hat." ] }, { "cell_type": "code", "execution_count": 16, "id": "93147524-1caa-4122-b98b-63b5dbb9c1c6", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Execution Time: 1.185906171798706\n" ] } ], "source": [ "import time\n", "\n", "# Beispiel für die Erstellung und Verwendung der StockTradingEnv-Umgebung\n", "tic=time.time()\n", "env = StockTradingEnv(data, num_states)\n", "\n", "# Initialisiere Q-Werte\n", "Q = np.zeros((num_states, 3))\n", "\n", "# Hyperparameter\n", "alpha = 0.1 # Lernrate\n", "gamma = 0.9 # Diskontierungsfaktor\n", "epsilon = 0.1 # Epsilon für die Epsilon-Greedy-Strategie\n", "\n", "# Training des Q-Learning-Agenten\n", "for episode in range(1000):\n", " state = env.reset()\n", " done = False\n", " actions = []\n", " rewards = []\n", " total_reward = 0\n", " \n", " while not done:\n", " if np.random.random() < epsilon:\n", " # Wähle Aktion entweder Zufällig aus\n", " action = np.random.randint(0, len(env.actions))\n", " else:\n", " # Wähle Aktion mit maximaler erwarteten Belohnung Q(s)\n", " action = np.argmax(Q[env.state])\n", " # Führe die Aktion aus und erhalte von der Umgebung den neuen Zustand und die Belohnung\n", " new_state, reward, done = env.step(action)\n", " new_state_idx = env.current_step\n", " # Aktualisiere Q-Werte auf Basis der erhaltenen Belohnung\n", " Q[state, action] += alpha * (reward + gamma * np.max(Q[new_state]) - Q[state, action])\n", " # Aktualisiere den Zustand\n", " state = new_state\n", " # Sammel Erfolgsstatistik\n", " actions.append(action)\n", " total_reward += reward\n", " rewards.append(total_reward)\n", "print(f\"Execution Time: {time.time()-tic}\")" ] }, { "cell_type": "markdown", "id": "543f7cc5-ce75-4501-94c9-9c4fd62489fc", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "Auch hier zeigt sich, dass der Q-Learning Algorithmus erfolgreich lernt, wann er Aktien kaufen, verkaufen oder shorten soll, so dass er am Ende Gewinn macht. " ] }, { "cell_type": "code", "execution_count": 17, "id": "ec5f7df7-bee4-4d8a-9dfa-9db6b9e7f74c", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "outputs": [ { "data": { "application/vnd.plotly.v1+json": { "config": { "plotlyServerURL": "https://plot.ly" }, "data": [ { "hovertemplate": "variable=0
index=%{x}
value=%{y}", "legendgroup": "0", "line": { "color": "#636efa", "dash": "solid" }, "marker": { "symbol": "circle" }, "mode": "lines", "name": "0", "orientation": "v", "showlegend": true, "type": "scatter", "x": [ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 191, 192, 193, 194, 195, 196, 197, 198 ], "xaxis": "x", "y": [ 0, 0, 0, 0, 0, 11.563281756687186, 11.563281756687186, 11.563281756687186, 11.563281756687186, 11.563281756687186, 11.563281756687186, 11.563281756687186, 11.563281756687186, 11.563281756687186, 11.563281756687186, 11.563281756687186, 11.563281756687186, 11.563281756687186, 11.563281756687186, 11.563281756687186, 11.563281756687186, 11.563281756687186, 11.563281756687186, 11.563281756687186, 11.563281756687186, 11.563281756687186, 11.563281756687186, 11.563281756687186, 11.563281756687186, 11.563281756687186, 11.563281756687186, 11.563281756687186, 11.563281756687186, 11.563281756687186, 11.563281756687186, 11.563281756687186, 11.563281756687186, 11.563281756687186, 11.563281756687186, 11.563281756687186, 11.563281756687186, 11.563281756687186, 11.563281756687186, 11.563281756687186, 11.563281756687186, 21.55460983870575, 21.55460983870575, 21.55460983870575, 21.55460983870575, 21.55460983870575, 21.55460983870575, 21.55460983870575, 21.55460983870575, 21.55460983870575, 21.55460983870575, 21.55460983870575, 21.55460983870575, 21.55460983870575, 21.55460983870575, 21.55460983870575, 21.55460983870575, 21.55460983870575, 21.55460983870575, 21.55460983870575, 21.55460983870575, 21.55460983870575, 21.55460983870575, 21.55460983870575, 21.55460983870575, 21.55460983870575, 21.55460983870575, 51.13066710024811, 51.13066710024811, 51.13066710024811, 51.13066710024811, 51.13066710024811, 51.13066710024811, 51.13066710024811, 51.13066710024811, 51.13066710024811, 51.13066710024811, 51.13066710024811, 51.13066710024811, 51.13066710024811, 51.13066710024811, 51.13066710024811, 51.13066710024811, 51.13066710024811, 51.13066710024811, 51.13066710024811, 51.13066710024811, 51.13066710024811, 51.13066710024811, 51.13066710024811, 51.13066710024811, 51.13066710024811, 51.13066710024811, 51.13066710024811, 51.13066710024811, 51.13066710024811, 51.13066710024811, 51.13066710024811, 66.40255014115607, 66.40255014115607, 66.40255014115607, 66.40255014115607, 66.40255014115607, 66.40255014115607, 66.40255014115607, 66.40255014115607, 66.40255014115607, 66.40255014115607, 66.40255014115607, 66.40255014115607, 66.40255014115607, 66.40255014115607, 66.40255014115607, 66.40255014115607, 66.40255014115607, 66.40255014115607, 66.40255014115607, 66.40255014115607, 66.40255014115607, 66.40255014115607, 66.40255014115607, 66.40255014115607, 99.66200034153601, 99.66200034153601, 99.66200034153601, 99.66200034153601, 99.66200034153601, 99.66200034153601, 99.66200034153601, 99.66200034153601, 99.66200034153601, 99.66200034153601, 99.66200034153601, 99.66200034153601, 99.66200034153601, 99.66200034153601, 99.66200034153601, 99.66200034153601, 99.66200034153601, 99.66200034153601, 112.04299134695576, 112.04299134695576, 112.04299134695576, 112.04299134695576, 112.04299134695576, 112.04299134695576, 112.04299134695576, 112.04299134695576, 112.04299134695576, 112.04299134695576, 112.04299134695576, 112.04299134695576, 112.04299134695576, 124.38935074452523, 124.38935074452523, 124.38935074452523, 124.38935074452523, 124.38935074452523, 124.38935074452523, 124.38935074452523, 124.38935074452523, 124.38935074452523, 124.38935074452523, 124.38935074452523, 124.38935074452523, 124.38935074452523, 140.4204688810774, 140.4204688810774, 140.4204688810774, 140.4204688810774, 140.4204688810774, 140.4204688810774, 140.4204688810774, 145.82422715267194, 145.82422715267194, 145.82422715267194, 145.82422715267194, 145.82422715267194, 145.82422715267194, 145.82422715267194, 145.82422715267194, 145.82422715267194, 145.82422715267194, 145.82422715267194, 145.82422715267194, 145.82422715267194, 145.82422715267194, 145.82422715267194, 145.82422715267194, 145.82422715267194, 145.82422715267194, 145.82422715267194, 145.82422715267194, 145.82422715267194, 145.82422715267194 ], "yaxis": "y" } ], "layout": { "autosize": true, "legend": { "title": { "text": "variable" }, "tracegroupgap": 0 }, "margin": { "t": 60 }, "template": { "data": { "bar": [ { "error_x": { "color": "#2a3f5f" }, "error_y": { "color": "#2a3f5f" }, "marker": { "line": { "color": "#E5ECF6", "width": 0.5 }, "pattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 } }, "type": "bar" } ], "barpolar": [ { "marker": { "line": { "color": "#E5ECF6", "width": 0.5 }, "pattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 } }, "type": "barpolar" } ], "carpet": [ { "aaxis": { "endlinecolor": "#2a3f5f", "gridcolor": "white", "linecolor": "white", "minorgridcolor": "white", "startlinecolor": "#2a3f5f" }, "baxis": { "endlinecolor": "#2a3f5f", "gridcolor": "white", "linecolor": "white", "minorgridcolor": "white", "startlinecolor": "#2a3f5f" }, "type": "carpet" } ], "choropleth": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "choropleth" } ], "contour": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "contour" } ], "contourcarpet": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "contourcarpet" } ], "heatmap": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "heatmap" } ], "heatmapgl": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "heatmapgl" } ], "histogram": [ { "marker": { "pattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 } }, "type": "histogram" } ], "histogram2d": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "histogram2d" } ], "histogram2dcontour": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "histogram2dcontour" } ], "mesh3d": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "mesh3d" } ], "parcoords": [ { "line": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "parcoords" } ], "pie": [ { "automargin": true, "type": "pie" } ], "scatter": [ { "fillpattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 }, "type": "scatter" } ], "scatter3d": [ { "line": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatter3d" } ], "scattercarpet": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattercarpet" } ], "scattergeo": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattergeo" } ], "scattergl": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattergl" } ], "scattermapbox": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattermapbox" } ], "scatterpolar": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterpolar" } ], "scatterpolargl": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterpolargl" } ], "scatterternary": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterternary" } ], "surface": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "surface" } ], "table": [ { "cells": { "fill": { "color": "#EBF0F8" }, "line": { "color": "white" } }, "header": { "fill": { "color": "#C8D4E3" }, "line": { "color": "white" } }, "type": "table" } ] }, "layout": { "annotationdefaults": { "arrowcolor": "#2a3f5f", "arrowhead": 0, "arrowwidth": 1 }, "autotypenumbers": "strict", "coloraxis": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "colorscale": { "diverging": [ [ 0, "#8e0152" ], [ 0.1, "#c51b7d" ], [ 0.2, "#de77ae" ], [ 0.3, "#f1b6da" ], [ 0.4, "#fde0ef" ], [ 0.5, "#f7f7f7" ], [ 0.6, "#e6f5d0" ], [ 0.7, "#b8e186" ], [ 0.8, "#7fbc41" ], [ 0.9, "#4d9221" ], [ 1, "#276419" ] ], "sequential": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "sequentialminus": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ] }, "colorway": [ "#636efa", "#EF553B", "#00cc96", "#ab63fa", "#FFA15A", "#19d3f3", "#FF6692", "#B6E880", "#FF97FF", "#FECB52" ], "font": { "color": "#2a3f5f" }, "geo": { "bgcolor": "white", "lakecolor": "white", "landcolor": "#E5ECF6", "showlakes": true, "showland": true, "subunitcolor": "white" }, "hoverlabel": { "align": "left" }, "hovermode": "closest", "mapbox": { "style": "light" }, "paper_bgcolor": "white", "plot_bgcolor": "#E5ECF6", "polar": { "angularaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "bgcolor": "#E5ECF6", "radialaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" } }, "scene": { "xaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" }, "yaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" }, "zaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" } }, "shapedefaults": { "line": { "color": "#2a3f5f" } }, "ternary": { "aaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "baxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "bgcolor": "#E5ECF6", "caxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" } }, "title": { "x": 0.05 }, "xaxis": { "automargin": true, "gridcolor": "white", "linecolor": "white", "ticks": "", "title": { "standoff": 15 }, "zerolinecolor": "white", "zerolinewidth": 2 }, "yaxis": { "automargin": true, "gridcolor": "white", "linecolor": "white", "ticks": "", "title": { "standoff": 15 }, "zerolinecolor": "white", "zerolinewidth": 2 } } }, "xaxis": { "anchor": "y", "autorange": true, "domain": [ 0, 1 ], "range": [ 0, 198 ], "title": { "text": "index" }, "type": "linear" }, "yaxis": { "anchor": "x", "autorange": true, "domain": [ 0, 1 ], "range": [ -8.10134595292622, 153.92557310559815 ], "title": { "text": "value" }, "type": "linear" } } }, "image/png": "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", "text/html": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "px.line(rewards)" ] }, { "cell_type": "markdown", "id": "8b1999db-808c-4d1a-b11f-db8f46637975", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "Hierbei macht der Algorithmus gar nicht so viele Transaktionen, wie er könnte, um das Ergebnis zu maximieren." ] }, { "cell_type": "code", "execution_count": 18, "id": "ecb73b27-0f63-4135-a9ba-feebf5ddde4b", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "outputs": [ { "data": { "application/vnd.plotly.v1+json": { "config": { "plotlyServerURL": "https://plot.ly" }, "data": [ { "hovertemplate": "variable=0
index=%{x}
value=%{y}", "legendgroup": "0", "line": { "color": "#636efa", "dash": "solid" }, "marker": { "symbol": "circle" }, "mode": "lines", "name": "0", "orientation": "v", "showlegend": true, "type": "scatter", "x": [ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 191, 192, 193, 194, 195, 196, 197, 198 ], "xaxis": "x", "y": [ 1, 1, 1, 1, 1, 0, 2, 2, 2, 2, 2, 2, 2, 2, 1, 1, 2, 1, 2, 1, 2, 1, 2, 1, 2, 1, 2, 1, 2, 1, 2, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 1, 2, 1, 0, 1, 2, 2, 1, 2, 1, 2, 1, 1, 1, 2, 1, 2, 2, 2, 2, 2, 1, 2, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1, 2, 1, 1, 1, 2, 2, 2, 2, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1, 2, 1, 1, 1, 1, 1, 1, 2, 1, 2, 2, 2, 2, 2, 2, 2, 2, 0, 2, 1, 2, 1, 2, 1, 2, 1, 1, 1, 1, 1, 1, 1, 1, 2, 1, 0, 2, 2, 1, 2, 1, 2, 1, 2, 1, 1, 1, 1, 0, 2, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 0, 2, 2, 2, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2 ], "yaxis": "y" } ], "layout": { "autosize": true, "legend": { "title": { "text": "variable" }, "tracegroupgap": 0 }, "margin": { "t": 60 }, "template": { "data": { "bar": [ { "error_x": { "color": "#2a3f5f" }, "error_y": { "color": "#2a3f5f" }, "marker": { "line": { "color": "#E5ECF6", "width": 0.5 }, "pattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 } }, "type": "bar" } ], "barpolar": [ { "marker": { "line": { "color": "#E5ECF6", "width": 0.5 }, "pattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 } }, "type": "barpolar" } ], "carpet": [ { "aaxis": { "endlinecolor": "#2a3f5f", "gridcolor": "white", "linecolor": "white", "minorgridcolor": "white", "startlinecolor": "#2a3f5f" }, "baxis": { "endlinecolor": "#2a3f5f", "gridcolor": "white", "linecolor": "white", "minorgridcolor": "white", "startlinecolor": "#2a3f5f" }, "type": "carpet" } ], "choropleth": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "choropleth" } ], "contour": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "contour" } ], "contourcarpet": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "contourcarpet" } ], "heatmap": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "heatmap" } ], "heatmapgl": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "heatmapgl" } ], "histogram": [ { "marker": { "pattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 } }, "type": "histogram" } ], "histogram2d": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "histogram2d" } ], "histogram2dcontour": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "histogram2dcontour" } ], "mesh3d": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "mesh3d" } ], "parcoords": [ { "line": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "parcoords" } ], "pie": [ { "automargin": true, "type": "pie" } ], "scatter": [ { "fillpattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 }, "type": "scatter" } ], "scatter3d": [ { "line": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatter3d" } ], "scattercarpet": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattercarpet" } ], "scattergeo": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattergeo" } ], "scattergl": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattergl" } ], "scattermapbox": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattermapbox" } ], "scatterpolar": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterpolar" } ], "scatterpolargl": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterpolargl" } ], "scatterternary": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterternary" } ], "surface": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "surface" } ], "table": [ { "cells": { "fill": { "color": "#EBF0F8" }, "line": { "color": "white" } }, "header": { "fill": { "color": "#C8D4E3" }, "line": { "color": "white" } }, "type": "table" } ] }, "layout": { "annotationdefaults": { "arrowcolor": "#2a3f5f", "arrowhead": 0, "arrowwidth": 1 }, "autotypenumbers": "strict", "coloraxis": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "colorscale": { "diverging": [ [ 0, "#8e0152" ], [ 0.1, "#c51b7d" ], [ 0.2, "#de77ae" ], [ 0.3, "#f1b6da" ], [ 0.4, "#fde0ef" ], [ 0.5, "#f7f7f7" ], [ 0.6, "#e6f5d0" ], [ 0.7, "#b8e186" ], [ 0.8, "#7fbc41" ], [ 0.9, "#4d9221" ], [ 1, "#276419" ] ], "sequential": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "sequentialminus": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ] }, "colorway": [ "#636efa", "#EF553B", "#00cc96", "#ab63fa", "#FFA15A", "#19d3f3", "#FF6692", "#B6E880", "#FF97FF", "#FECB52" ], "font": { "color": "#2a3f5f" }, "geo": { "bgcolor": "white", "lakecolor": "white", "landcolor": "#E5ECF6", "showlakes": true, "showland": true, "subunitcolor": "white" }, "hoverlabel": { "align": "left" }, "hovermode": "closest", "mapbox": { "style": "light" }, "paper_bgcolor": "white", "plot_bgcolor": "#E5ECF6", "polar": { "angularaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "bgcolor": "#E5ECF6", "radialaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" } }, "scene": { "xaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" }, "yaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" }, "zaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" } }, "shapedefaults": { "line": { "color": "#2a3f5f" } }, "ternary": { "aaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "baxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "bgcolor": "#E5ECF6", "caxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" } }, "title": { "x": 0.05 }, "xaxis": { "automargin": true, "gridcolor": "white", "linecolor": "white", "ticks": "", "title": { "standoff": 15 }, "zerolinecolor": "white", "zerolinewidth": 2 }, "yaxis": { "automargin": true, "gridcolor": "white", "linecolor": "white", "ticks": "", "title": { "standoff": 15 }, "zerolinecolor": "white", "zerolinewidth": 2 } } }, "xaxis": { "anchor": "y", "autorange": true, "domain": [ 0, 1 ], "range": [ 0, 198 ], "title": { "text": "index" }, "type": "linear" }, "yaxis": { "anchor": "x", "autorange": true, "domain": [ 0, 1 ], "range": [ -0.11111111111111112, 2.111111111111111 ], "title": { "text": "value" }, "type": "linear" } } }, "image/png": "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", "text/html": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "px.line(actions)" ] }, { "cell_type": "markdown", "id": "aaa03b11-688e-475e-bf73-b52ebf4b5d46", "metadata": { "editable": true, "slideshow": { "slide_type": "slide" }, "tags": [] }, "source": [ "## RL Frameworks" ] }, { "cell_type": "markdown", "id": "d986d52b-cd04-4d34-945f-7b8d4bc32bfb", "metadata": { "editable": true, "slideshow": { "slide_type": "skip" }, "tags": [] }, "source": [ "RL gehört nicht zu den traditionellen ML-Verfahren und ist deshalb auch nicht in SciKit-Learn oder Statsmodels enthalten. Das liegt unter anderem daran, dass man beim RL nicht einfach ein Modell auf einem Datensatz trainiert, wie es von der Standard-API von SciKit-Learn erwartet wird, sondern manuell ein Umgebungsmodell erstellen muss.\n", "\n", "Mit der wachsenden Popularität von RL in den letzten Jahren haben sich aber spezielle Bibliotheken für RL entwickelt, die dies vereinfachen und das Lösen komplexer Probleme vereinfachen. Eine Bibliothek dafür ist `Gymnasium` oder kurz `Gym` von OpenAI. Die `Gym`-Bibliothek ist eine Open-Source-Bibliothek, die speziell für die Entwicklung und das Testen von RL-Algorithmen entwickelt wurde. Sie bietet eine standardisierte API und eine Vielzahl von vordefinierten Umgebungen, die es ermöglichen RL-Algorithmen effizient zu implementieren und zu evaluieren. Hierbei spielt insbesondere die Definition der Umgebung eine wichtige Rolle." ] }, { "cell_type": "markdown", "id": "07868107-db9e-4af2-b74c-a7e1df9b758d", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "Das Standard-Interface in `Gym` für eine Umgebung für den Börsenfall ist mehr oder weniger identisch mit der Klasse, die wir oben definiert haben." ] }, { "cell_type": "code", "execution_count": 19, "id": "31cc25cd-fe00-4b15-a2a3-6f0fcaa13b15", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "outputs": [], "source": [ "import gymnasium as gym\n", "from gymnasium import spaces\n", "import numpy as np\n", "import pandas as pd\n", "\n", "class StockTradingEnvGym(gym.Env):\n", " def __init__(self, data):\n", " super(StockTradingEnvGym, self).__init__()\n", " self.data = data.copy()\n", " self.current_step = 0\n", " self.balance = 10000 # Startguthaben\n", " self.position = 0 # 0: neutral, 1: long, -1: short\n", " self.old_price = 0\n", " self.current_price = self.data[self.current_step]\n", " # Actions: 0 = Kaufen, 1 = Verkaufen, 2 = Halten\n", " self.action_space = spaces.Discrete(3)\n", " # Observation space: [current price, position, balance]\n", " self.observation_space = spaces.Box(\n", " low=np.array([-np.inf, -1, -np.inf]), \n", " high=np.array([np.inf, 1, np.inf]), \n", " dtype=np.float32\n", " )\n", "\n", " def reset(self, seed=0, options=None):\n", " self.current_step = 0\n", " self.balance = 10000\n", " self.position = 0\n", " self.current_price = self.data[self.current_step]\n", " return self._get_obs(), {}\n", "\n", " def _get_obs(self):\n", " return np.array([self.current_price, self.position, self.balance])\n", "\n", " def step(self, action):\n", " prev_price = self.current_price\n", " self.current_step += 1\n", " self.current_price = self.data[self.current_step]\n", " reward = 0\n", " if action == 0: # Kaufen\n", " if self.position == 0: # Kaufe Aktie\n", " self.position = 1\n", " self.balance -= self.current_price\n", " self.old_price = self.current_price\n", " elif self.position == -1: # Verkaufe Short\n", " reward = 2 * (self.old_price - self.current_price)\n", " self.position = 0\n", " self.balance += self.old_price\n", " elif action == 1: # Verkaufen\n", " if self.position == 0: # Kaufe Short\n", " self.position = -1\n", " self.balance += self.current_price\n", " self.old_price = self.current_price\n", " elif self.position == 1: # Verkaufe Aktie\n", " reward = 2 * (self.current_price - self.old_price)\n", " self.position = 0\n", " self.balance -= self.old_price\n", " elif action == 2: # Halten\n", " reward = 0\n", " done = self.current_step >= len(self.data) - 1\n", " return self._get_obs(), reward, done, False, {}\n", "\n", " def render(self, mode='human'):\n", " print(f'Step: {self.current_step}, Price: {self.current_price}, Position: {self.position}, Balance: {self.balance}')" ] }, { "cell_type": "markdown", "id": "e197790f-261e-42ba-ba5a-3b2b12e6bfdb", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "Ein Aspekt warum RL in den letzten Jahren so beliebt geworden ist, ist dass sich der Lernvorgang gut parallelisieren lässt. Da wir zum Lernen viele Experimente machen müssen, können wir diese natürlich auch parallel ausführen und dadurch gut in einem Rechenzentrum in der Cloud oder auf Grafikkarten mit ihren tausenden kleinen Prozessoren verteilen." ] }, { "cell_type": "markdown", "id": "f4bd7329-d31b-4d48-a330-4e108c93699c", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "source": [ "Eine Bibliothek, die hierbei viel genutzt wird, ist _Ray_ welche auch gerne zur Parallelisierung von ML-Aufgaben genutzt wird, da die Bibliothek das Verteilen der Lernaufgaben im Cluster und das Sammeln der Ergebnisse übernimmt." ] }, { "cell_type": "markdown", "id": "5af0a095-735f-403b-82f9-0cab9bf789c7", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "Wir nutzen hier den _Proximale Policy Optimization (PPO)_ Algorithmus. Er ist ein iteratives Verfahren, welches zur Optimierung von Richtlinien in Agenten verwendet wird und instabile Updates der Policy vermeidet und ermöglicht so die effiziente Handhabung komplexer Aufgaben mit kontinuierlichen Aktionsräumen. Dies wird durch die Einführung eines Clip-Parameters ε erreicht, der die maximale Änderung der Policy-Parameter begrenzt. PPO ist ein On-Policy-Verfahren, weil die Policy mit Daten aktualisiert wird, die von der aktuellen Policy selbst erzeugt wurden." ] }, { "cell_type": "code", "execution_count": 20, "id": "34fbb34f-f364-45fe-94b2-c283b5074588", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [ "scroll-output" ] }, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "2024-06-24 11:11:54,421\tWARNING algorithm_config.py:4078 -- You have specified 1 evaluation workers, but your `evaluation_interval` is 0 or None! Therefore, evaluation will not occur automatically with each call to `Algorithm.train()`. Instead, you will have to call `Algorithm.evaluate()` manually in order to trigger an evaluation run.\n", "/Users/jploennigs/miniconda3/envs/lehre4/lib/python3.11/site-packages/ray/rllib/algorithms/algorithm.py:525: RayDeprecationWarning:\n", "\n", "This API is deprecated and may be removed in future Ray releases. You could suppress this warning by setting env variable PYTHONWARNINGS=\"ignore::DeprecationWarning\"\n", "`UnifiedLogger` will be removed in Ray 2.7.\n", "\n", "/Users/jploennigs/miniconda3/envs/lehre4/lib/python3.11/site-packages/ray/tune/logger/unified.py:53: RayDeprecationWarning:\n", "\n", "This API is deprecated and may be removed in future Ray releases. You could suppress this warning by setting env variable PYTHONWARNINGS=\"ignore::DeprecationWarning\"\n", "The `JsonLogger interface is deprecated in favor of the `ray.tune.json.JsonLoggerCallback` interface and will be removed in Ray 2.7.\n", "\n", "/Users/jploennigs/miniconda3/envs/lehre4/lib/python3.11/site-packages/ray/tune/logger/unified.py:53: RayDeprecationWarning:\n", "\n", "This API is deprecated and may be removed in future Ray releases. You could suppress this warning by setting env variable PYTHONWARNINGS=\"ignore::DeprecationWarning\"\n", "The `CSVLogger interface is deprecated in favor of the `ray.tune.csv.CSVLoggerCallback` interface and will be removed in Ray 2.7.\n", "\n", "/Users/jploennigs/miniconda3/envs/lehre4/lib/python3.11/site-packages/ray/tune/logger/unified.py:53: RayDeprecationWarning:\n", "\n", "This API is deprecated and may be removed in future Ray releases. You could suppress this warning by setting env variable PYTHONWARNINGS=\"ignore::DeprecationWarning\"\n", "The `TBXLogger interface is deprecated in favor of the `ray.tune.tensorboardx.TBXLoggerCallback` interface and will be removed in Ray 2.7.\n", "\n", "2024-06-24 11:11:56,045\tINFO worker.py:1770 -- Started a local Ray instance.\n", "2024-06-24 11:11:59,353\tWARNING algorithm_config.py:4078 -- You have specified 1 evaluation workers, but your `evaluation_interval` is 0 or None! Therefore, evaluation will not occur automatically with each call to `Algorithm.train()`. Instead, you will have to call `Algorithm.evaluate()` manually in order to trigger an evaluation run.\n", "2024-06-24 11:12:01,524\tWARNING util.py:61 -- Install gputil for GPU system monitoring.\n", "2024-06-24 11:12:02,343\tWARNING deprecation.py:50 -- DeprecationWarning: `ray.rllib.execution.train_ops.multi_gpu_train_one_step` has been deprecated. This will raise an error in the future!\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Execution Time: 48.61755394935608\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "\u001b[36m(RolloutWorker pid=80435)\u001b[0m /Users/jploennigs/miniconda3/envs/lehre4/lib/python3.11/site-packages/gymnasium/core.py:311: UserWarning: \u001b[33mWARN: env.single_observation_space to get variables from other wrappers is deprecated and will be removed in v1.0, to get this variable you can do `env.unwrapped.single_observation_space` for environment variables or `env.get_wrapper_attr('single_observation_space')` that will search the reminding wrappers.\u001b[0m\n", "\u001b[36m(RolloutWorker pid=80435)\u001b[0m logger.warn(\n", "\u001b[36m(RolloutWorker pid=80435)\u001b[0m /Users/jploennigs/miniconda3/envs/lehre4/lib/python3.11/site-packages/gymnasium/core.py:311: UserWarning: \u001b[33mWARN: env.single_action_space to get variables from other wrappers is deprecated and will be removed in v1.0, to get this variable you can do `env.unwrapped.single_action_space` for environment variables or `env.get_wrapper_attr('single_action_space')` that will search the reminding wrappers.\u001b[0m\n", "\u001b[36m(RolloutWorker pid=80435)\u001b[0m logger.warn(\n", "\u001b[36m(RolloutWorker pid=80435)\u001b[0m 2024-06-24 11:12:46,958\tWARNING env_runner_v2.py:301 -- Could not import gymnasium.envs.classic_control.rendering! Try `pip install gymnasium[all]`.\n", "\u001b[36m(RolloutWorker pid=80656)\u001b[0m /Users/jploennigs/miniconda3/envs/lehre4/lib/python3.11/site-packages/gymnasium/core.py:311: UserWarning: \u001b[33mWARN: env.single_observation_space to get variables from other wrappers is deprecated and will be removed in v1.0, to get this variable you can do `env.unwrapped.single_observation_space` for environment variables or `env.get_wrapper_attr('single_observation_space')` that will search the reminding wrappers.\u001b[0m\n", "\u001b[36m(RolloutWorker pid=80656)\u001b[0m logger.warn(\n", "\u001b[36m(RolloutWorker pid=80656)\u001b[0m /Users/jploennigs/miniconda3/envs/lehre4/lib/python3.11/site-packages/gymnasium/core.py:311: UserWarning: \u001b[33mWARN: env.single_action_space to get variables from other wrappers is deprecated and will be removed in v1.0, to get this variable you can do `env.unwrapped.single_action_space` for environment variables or `env.get_wrapper_attr('single_action_space')` that will search the reminding wrappers.\u001b[0m\n", "\u001b[36m(RolloutWorker pid=80656)\u001b[0m logger.warn(\n" ] } ], "source": [ "import os\n", "os.environ[\"PYTHONWARNINGS\"]=\"ignore::DeprecationWarning\"\n", "os.environ[\"RAY_ENABLE_UV_RUN_RUNTIME_ENV\"]=\"0\"\n", "import ray\n", "from ray.rllib.algorithms.ppo import PPOConfig\n", "from ray.tune.registry import register_env\n", "\n", "tic=time.time()\n", "def env_creator(env_config):\n", " return StockTradingEnvGym(data) # return an env instance\n", "\n", "register_env(\"StockTradingEnvGym\", env_creator)\n", "\n", "if not ray.is_initialized():\n", " ray.init(ignore_reinit_error=True)\n", "\n", "config = (\n", " PPOConfig()\n", " .api_stack(\n", " enable_rl_module_and_learner=False,\n", " enable_env_runner_and_connector_v2=False\n", " )\n", " .environment(\"StockTradingEnvGym\")\n", " .env_runners(num_env_runners=2)\n", " .framework(\"torch\")\n", " .training()\n", " .evaluation(evaluation_num_env_runners=1)\n", ")\n", "\n", "algo = config.build_algo() # 2. build the algorithm,\n", "\n", "for _ in range(10):\n", " algo.train() # 3. train it,\n", "\n", "#algo.evaluate() # 4. and evaluate it.\n", "print(f\"Execution Time: {time.time()-tic}\")" ] }, { "cell_type": "code", "execution_count": 21, "id": "f2398e5f-c8de-4310-9fec-118cd2d3504e", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [ "scroll-output" ] }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Step: 1, Price: 1.358449851840048, Position: 0, Balance: 10000\n", "Step: 2, Price: 2.5061383899407406, Position: 0, Balance: 10000\n", "Step: 3, Price: 4.529168246348766, Position: 0, Balance: 10000\n", "Step: 4, Price: 4.79501487162543, Position: 1, Balance: 9995.204985128374\n", "Step: 5, Price: 5.060877914676249, Position: 1, Balance: 9995.204985128374\n", "Step: 6, Price: 7.1400907301836405, Position: 1, Balance: 9995.204985128374\n", "Step: 7, Price: 8.407525459336549, Position: 1, Balance: 9995.204985128374\n", "Step: 8, Price: 8.438051073401597, Position: 1, Balance: 9995.204985128374\n", "Step: 9, Price: 9.480611116987562, Position: 1, Balance: 9995.204985128374\n", "Step: 10, Price: 9.517193424175101, Position: 1, Balance: 9995.204985128374\n", "Step: 11, Price: 9.551463670604845, Position: 1, Balance: 9995.204985128374\n", "Step: 12, Price: 10.29342594217088, Position: 1, Balance: 9995.204985128374\n", "Step: 13, Price: 8.88014569751308, Position: 1, Balance: 9995.204985128374\n", "Step: 14, Price: 7.6552278650000485, Position: 1, Balance: 9995.204985128374\n", "Step: 15, Price: 7.592940335759076, Position: 1, Balance: 9995.204985128374\n", "Step: 16, Price: 7.0801092154246525, Position: 1, Balance: 9995.204985128374\n", "Step: 17, Price: 7.8943565480199265, Position: 1, Balance: 9995.204985128374\n", "Step: 18, Price: 7.486332472498716, Position: 1, Balance: 9995.204985128374\n", "Step: 19, Price: 6.574028771163424, Position: 1, Balance: 9995.204985128374\n", "Step: 20, Price: 8.539677540084979, Position: 1, Balance: 9995.204985128374\n", "Step: 21, Price: 8.813901239598444, Position: 1, Balance: 9995.204985128374\n", "Step: 22, Price: 9.381429444286368, Position: 1, Balance: 9995.204985128374\n", "Step: 23, Price: 8.456681258072912, Position: 1, Balance: 9995.204985128374\n", "Step: 24, Price: 8.41229853354773, Position: 1, Balance: 9995.204985128374\n", "Step: 25, Price: 9.023221123257596, Position: 1, Balance: 9995.204985128374\n", "Step: 26, Price: 8.372227545835292, Position: 1, Balance: 9995.204985128374\n", "Step: 27, Price: 9.247925564180964, Position: 1, Balance: 9995.204985128374\n", "Step: 28, Price: 9.147286874262159, Position: 1, Balance: 9995.204985128374\n", "Step: 29, Price: 9.355593124468882, Position: 1, Balance: 9995.204985128374\n", "Step: 30, Price: 9.253886512239486, Position: 1, Balance: 9995.204985128374\n", "Step: 31, Price: 11.606164696748424, Position: 1, Balance: 9995.204985128374\n", "Step: 32, Price: 12.09266747201049, Position: 1, Balance: 9995.204985128374\n", "Step: 33, Price: 11.53495654305459, Position: 1, Balance: 9995.204985128374\n", "Step: 34, Price: 12.85750145515778, Position: 1, Balance: 9995.204985128374\n", "Step: 35, Price: 12.136657805186758, Position: 0, Balance: 9990.409970256747\n", "Step: 36, Price: 12.845521400191513, Position: 0, Balance: 9990.409970256747\n", "Step: 37, Price: 11.385851276311739, Position: 1, Balance: 9979.024118980436\n", "Step: 38, Price: 10.557665227413308, Position: 1, Balance: 9979.024118980436\n", "Step: 39, Price: 11.254526463282431, Position: 1, Balance: 9979.024118980436\n", "Step: 40, Price: 12.492993043277842, Position: 1, Balance: 9979.024118980436\n", "Step: 41, Price: 13.164361324467812, Position: 1, Balance: 9979.024118980436\n", "Step: 42, Price: 13.548713042079571, Position: 1, Balance: 9979.024118980436\n", "Step: 43, Price: 13.747609346490282, Position: 1, Balance: 9979.024118980436\n", "Step: 44, Price: 12.769087356122855, Position: 1, Balance: 9979.024118980436\n", "Step: 45, Price: 12.549243147728147, Position: 1, Balance: 9979.024118980436\n", "Step: 46, Price: 12.588604376768359, Position: 1, Balance: 9979.024118980436\n", "Step: 47, Price: 14.145726602987274, Position: 1, Balance: 9979.024118980436\n", "Step: 48, Price: 14.989344892555735, Position: 1, Balance: 9979.024118980436\n", "Step: 49, Price: 13.726304737193, Position: 1, Balance: 9979.024118980436\n", "Step: 50, Price: 14.550388706587796, Position: 1, Balance: 9979.024118980436\n", "Step: 51, Price: 14.66530642617148, Position: 1, Balance: 9979.024118980436\n", "Step: 52, Price: 14.48838442586552, Position: 1, Balance: 9979.024118980436\n", "Step: 53, Price: 15.600060714706387, Position: 1, Balance: 9979.024118980436\n", "Step: 54, Price: 17.131060237202338, Position: 1, Balance: 9979.024118980436\n", "Step: 55, Price: 18.562340356318536, Position: 1, Balance: 9979.024118980436\n", "Step: 56, Price: 18.223122833095896, Position: 1, Balance: 9979.024118980436\n", "Step: 57, Price: 18.413910457244683, Position: 1, Balance: 9979.024118980436\n", "Step: 58, Price: 19.245173888648246, Position: 1, Balance: 9979.024118980436\n", "Step: 59, Price: 20.720719015770605, Position: 1, Balance: 9979.024118980436\n", "Step: 60, Price: 20.741544777925316, Position: 1, Balance: 9979.024118980436\n", "Step: 61, Price: 21.0558858012615, Position: 1, Balance: 9979.024118980436\n", "Step: 62, Price: 20.44955082725547, Position: 1, Balance: 9979.024118980436\n", "Step: 63, Price: 19.7533442031748, Position: 1, Balance: 9979.024118980436\n", "Step: 64, Price: 21.065870025568998, Position: 1, Balance: 9979.024118980436\n", "Step: 65, Price: 22.92211005413982, Position: 1, Balance: 9979.024118980436\n", "Step: 66, Price: 23.350099932559488, Position: 1, Balance: 9979.024118980436\n", "Step: 67, Price: 24.853632830451513, Position: 1, Balance: 9979.024118980436\n", "Step: 68, Price: 25.715268855499147, Position: 1, Balance: 9979.024118980436\n", "Step: 69, Price: 25.570149100894024, Position: 0, Balance: 9967.638267704126\n", "Step: 70, Price: 26.431544706402438, Position: 1, Balance: 9941.206722997724\n", "Step: 71, Price: 28.469581272868407, Position: 1, Balance: 9941.206722997724\n", "Step: 72, Price: 28.933755233758454, Position: 1, Balance: 9941.206722997724\n", "Step: 73, Price: 30.99839888957246, Position: 1, Balance: 9941.206722997724\n", "Step: 74, Price: 28.878653785482715, Position: 1, Balance: 9941.206722997724\n", "Step: 75, Price: 30.20055628985794, Position: 1, Balance: 9941.206722997724\n", "Step: 76, Price: 30.78760335809611, Position: 1, Balance: 9941.206722997724\n", "Step: 77, Price: 30.988596007630242, Position: 0, Balance: 9914.775178291322\n", "Step: 78, Price: 31.580356784165744, Position: 0, Balance: 9914.775178291322\n", "Step: 79, Price: 30.09278786956485, Position: 0, Balance: 9914.775178291322\n", "Step: 80, Price: 30.373115981727338, Position: 1, Balance: 9884.402062309595\n", "Step: 81, Price: 31.230228553239083, Position: 1, Balance: 9884.402062309595\n", "Step: 82, Price: 33.2081225979806, Position: 1, Balance: 9884.402062309595\n", "Step: 83, Price: 33.18985237970695, Position: 1, Balance: 9884.402062309595\n", "Step: 84, Price: 32.88135877681376, Position: 1, Balance: 9884.402062309595\n", "Step: 85, Price: 32.87960173322922, Position: 1, Balance: 9884.402062309595\n", "Step: 86, Price: 34.295003850931295, Position: 1, Balance: 9884.402062309595\n", "Step: 87, Price: 35.12375496059098, Position: 1, Balance: 9884.402062309595\n", "Step: 88, Price: 35.09399475682394, Position: 1, Balance: 9884.402062309595\n", "Step: 89, Price: 36.107262189937295, Position: 1, Balance: 9884.402062309595\n", "Step: 90, Price: 36.70433973928534, Position: 1, Balance: 9884.402062309595\n", "Step: 91, Price: 38.172984729818225, Position: 1, Balance: 9884.402062309595\n", "Step: 92, Price: 37.97093163594087, Position: 1, Balance: 9884.402062309595\n", "Step: 93, Price: 38.143269489343105, Position: 0, Balance: 9854.028946327868\n", "Step: 94, Price: 38.25116133621095, Position: -1, Balance: 9892.280107664079\n", "Step: 95, Price: 37.28764638807883, Position: 0, Balance: 9930.53126900029\n", "Step: 96, Price: 38.0837666651434, Position: 0, Balance: 9930.53126900029\n", "Step: 97, Price: 38.84482193732329, Position: 1, Balance: 9891.686447062966\n", "Step: 98, Price: 39.34993539396575, Position: 1, Balance: 9891.686447062966\n", "Step: 99, Price: 39.6153482605906, Position: 1, Balance: 9891.686447062966\n", "Step: 100, Price: 38.69997751854019, Position: 1, Balance: 9891.686447062966\n", "Step: 101, Price: 38.77933219577483, Position: 0, Balance: 9852.841625125642\n", "Step: 102, Price: 38.93661767924806, Position: 0, Balance: 9852.841625125642\n", "Step: 103, Price: 38.63434041002644, Position: 0, Balance: 9852.841625125642\n", "Step: 104, Price: 38.97305469836043, Position: 0, Balance: 9852.841625125642\n", "Step: 105, Price: 39.877105555174964, Position: 0, Balance: 9852.841625125642\n", "Step: 106, Price: 42.263291456385495, Position: 0, Balance: 9852.841625125642\n", "Step: 107, Price: 42.93786926921734, Position: 0, Balance: 9852.841625125642\n", "Step: 108, Price: 43.6954196599401, Position: 0, Balance: 9852.841625125642\n", "Step: 109, Price: 44.12097374417393, Position: 0, Balance: 9852.841625125642\n", "Step: 110, Price: 42.70220252887489, Position: 1, Balance: 9810.139422596767\n", "Step: 111, Price: 43.175688653425674, Position: 1, Balance: 9810.139422596767\n", "Step: 112, Price: 43.7359188633667, Position: 1, Balance: 9810.139422596767\n", "Step: 113, Price: 46.69916097585198, Position: 1, Balance: 9810.139422596767\n", "Step: 114, Price: 47.00680001107086, Position: 1, Balance: 9810.139422596767\n", "Step: 115, Price: 47.80834735340447, Position: 1, Balance: 9810.139422596767\n", "Step: 116, Price: 48.273635583699225, Position: 1, Balance: 9810.139422596767\n", "Step: 117, Price: 47.604957546079696, Position: 1, Balance: 9810.139422596767\n", "Step: 118, Price: 49.24778036059472, Position: 0, Balance: 9767.437220067892\n", "Step: 119, Price: 50.49971339328149, Position: 1, Balance: 9716.93750667461\n", "Step: 120, Price: 51.790745340324534, Position: 1, Balance: 9716.93750667461\n", "Step: 121, Price: 51.381357885529795, Position: 1, Balance: 9716.93750667461\n", "Step: 122, Price: 53.2841521964659, Position: 1, Balance: 9716.93750667461\n", "Step: 123, Price: 52.382301133673614, Position: 1, Balance: 9716.93750667461\n", "Step: 124, Price: 53.46915822747388, Position: 1, Balance: 9716.93750667461\n", "Step: 125, Price: 56.159613853283865, Position: 1, Balance: 9716.93750667461\n", "Step: 126, Price: 55.669077528153174, Position: 1, Balance: 9716.93750667461\n", "Step: 127, Price: 55.6027797985504, Position: 1, Balance: 9716.93750667461\n", "Step: 128, Price: 56.20243116363804, Position: 1, Balance: 9716.93750667461\n", "Step: 129, Price: 56.19895550952184, Position: 1, Balance: 9716.93750667461\n", "Step: 130, Price: 55.148292078455704, Position: 1, Balance: 9716.93750667461\n", "Step: 131, Price: 55.71685505326173, Position: 1, Balance: 9716.93750667461\n", "Step: 132, Price: 55.15455133953563, Position: 1, Balance: 9716.93750667461\n", "Step: 133, Price: 56.12814377017081, Position: 1, Balance: 9716.93750667461\n", "Step: 134, Price: 55.70871953593701, Position: 1, Balance: 9716.93750667461\n", "Step: 135, Price: 57.758653940954545, Position: 1, Balance: 9716.93750667461\n", "Step: 136, Price: 57.47540064861831, Position: 0, Balance: 9666.43779328133\n", "Step: 137, Price: 57.65333913241263, Position: 0, Balance: 9666.43779328133\n", "Step: 138, Price: 58.966856349782304, Position: 0, Balance: 9666.43779328133\n", "Step: 139, Price: 58.235992033348346, Position: -1, Balance: 9724.673785314679\n", "Step: 140, Price: 58.96345196795247, Position: -1, Balance: 9724.673785314679\n", "Step: 141, Price: 60.770594722234904, Position: -1, Balance: 9724.673785314679\n", "Step: 142, Price: 59.663111487673675, Position: -1, Balance: 9724.673785314679\n", "Step: 143, Price: 60.347745346205976, Position: -1, Balance: 9724.673785314679\n", "Step: 144, Price: 61.1076281404544, Position: -1, Balance: 9724.673785314679\n", "Step: 145, Price: 62.38945101223171, Position: -1, Balance: 9724.673785314679\n", "Step: 146, Price: 61.652500301353626, Position: 0, Balance: 9782.909777348028\n", "Step: 147, Price: 60.83204368826935, Position: 1, Balance: 9722.077733659758\n", "Step: 148, Price: 61.85398525388625, Position: 1, Balance: 9722.077733659758\n", "Step: 149, Price: 62.650969927119434, Position: 1, Balance: 9722.077733659758\n", "Step: 150, Price: 63.40146277746531, Position: 0, Balance: 9661.245689971489\n", "Step: 151, Price: 64.24791098696228, Position: 1, Balance: 9596.997778984527\n", "Step: 152, Price: 64.06788626538379, Position: 1, Balance: 9596.997778984527\n", "Step: 153, Price: 64.80013996254479, Position: 1, Balance: 9596.997778984527\n", "Step: 154, Price: 65.59321243584347, Position: 1, Balance: 9596.997778984527\n", "Step: 155, Price: 65.3788610178171, Position: 1, Balance: 9596.997778984527\n", "Step: 156, Price: 67.74463552896185, Position: 1, Balance: 9596.997778984527\n", "Step: 157, Price: 68.71846844987364, Position: 1, Balance: 9596.997778984527\n", "Step: 158, Price: 68.02716495267099, Position: 1, Balance: 9596.997778984527\n", "Step: 159, Price: 69.18371856130481, Position: 1, Balance: 9596.997778984527\n", "Step: 160, Price: 68.70903689107749, Position: 1, Balance: 9596.997778984527\n", "Step: 161, Price: 69.99612149481995, Position: 1, Balance: 9596.997778984527\n", "Step: 162, Price: 71.65471707382736, Position: 1, Balance: 9596.997778984527\n", "Step: 163, Price: 71.33403475547564, Position: 1, Balance: 9596.997778984527\n", "Step: 164, Price: 72.79741088471997, Position: 1, Balance: 9596.997778984527\n", "Step: 165, Price: 73.71019181165646, Position: 1, Balance: 9596.997778984527\n", "Step: 166, Price: 75.03225197165095, Position: 1, Balance: 9596.997778984527\n", "Step: 167, Price: 77.4290449543049, Position: 1, Balance: 9596.997778984527\n", "Step: 168, Price: 77.68365683830203, Position: 1, Balance: 9596.997778984527\n", "Step: 169, Price: 77.42992067394454, Position: 1, Balance: 9596.997778984527\n", "Step: 170, Price: 77.04040624431902, Position: 1, Balance: 9596.997778984527\n", "Step: 171, Price: 76.72459595935358, Position: 1, Balance: 9596.997778984527\n", "Step: 172, Price: 77.14749424993947, Position: 1, Balance: 9596.997778984527\n", "Step: 173, Price: 77.98864622475611, Position: 1, Balance: 9596.997778984527\n", "Step: 174, Price: 78.76533702408612, Position: 1, Balance: 9596.997778984527\n", "Step: 175, Price: 80.09252027312215, Position: 1, Balance: 9596.997778984527\n", "Step: 176, Price: 80.60552216500005, Position: 1, Balance: 9596.997778984527\n", "Step: 177, Price: 82.55905624215737, Position: 0, Balance: 9532.749867997565\n", "Step: 178, Price: 82.79439940891942, Position: 1, Balance: 9449.955468588645\n", "Step: 179, Price: 86.01456857550903, Position: 1, Balance: 9449.955468588645\n", "Step: 180, Price: 87.14023592327403, Position: 1, Balance: 9449.955468588645\n", "Step: 181, Price: 86.78307836685775, Position: 1, Balance: 9449.955468588645\n", "Step: 182, Price: 86.21218586879664, Position: 1, Balance: 9449.955468588645\n", "Step: 183, Price: 87.19465828403982, Position: 1, Balance: 9449.955468588645\n", "Step: 184, Price: 87.47119549871397, Position: 1, Balance: 9449.955468588645\n", "Step: 185, Price: 88.68519599280606, Position: 1, Balance: 9449.955468588645\n", "Step: 186, Price: 89.6584336173796, Position: 1, Balance: 9449.955468588645\n", "Step: 187, Price: 90.08560470472273, Position: 1, Balance: 9449.955468588645\n", "Step: 188, Price: 89.73881098665431, Position: 1, Balance: 9449.955468588645\n", "Step: 189, Price: 88.72396376196845, Position: 1, Balance: 9449.955468588645\n", "Step: 190, Price: 88.77744880990143, Position: 1, Balance: 9449.955468588645\n", "Step: 191, Price: 90.1338476042249, Position: 1, Balance: 9449.955468588645\n", "Step: 192, Price: 90.8479413483551, Position: 1, Balance: 9449.955468588645\n", "Step: 193, Price: 90.10220256964313, Position: 0, Balance: 9367.161069179725\n", "Step: 194, Price: 90.77538349549431, Position: 0, Balance: 9367.161069179725\n", "Step: 195, Price: 91.66070087522314, Position: 0, Balance: 9367.161069179725\n", "Step: 196, Price: 91.276843439022, Position: 0, Balance: 9367.161069179725\n", "Step: 197, Price: 91.93056854496753, Position: 1, Balance: 9275.230500634758\n", "Step: 198, Price: 92.48877726341352, Position: 1, Balance: 9275.230500634758\n", "Step: 199, Price: 91.8458069655829, Position: 1, Balance: 9275.230500634758\n", "Gesamtbelohnung: 146.08859019547393\n" ] } ], "source": [ "# Evaluierung des Agents\n", "env = StockTradingEnvGym(data)\n", "state = env.reset()\n", "done = False\n", "total_reward = 0\n", "\n", "actions = []\n", "rewards = []\n", "while not done:\n", " action = algo.compute_single_action(env._get_obs(), state)\n", " state, reward, done, _, _ = env.step(action[0])\n", " total_reward += reward\n", " actions.append(action)\n", " rewards.append(total_reward)\n", " env.render()\n", "\n", "print(\"Gesamtbelohnung:\", total_reward)" ] }, { "cell_type": "code", "execution_count": 22, "id": "189bb6de-e606-4cce-8300-81c93bb8b3bd", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "outputs": [ { "data": { "application/vnd.plotly.v1+json": { "config": { "plotlyServerURL": "https://plot.ly" }, "data": [ { "hovertemplate": "variable=0
index=%{x}
value=%{y}", "legendgroup": "0", "line": { "color": "#636efa", "dash": "solid" }, "marker": { "symbol": "circle" }, "mode": "lines", "name": "0", "orientation": "v", "showlegend": true, "type": "scatter", "x": [ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 191, 192, 193, 194, 195, 196, 197, 198 ], "xaxis": "x", "y": [ 2, 2, 2, 0, 2, 2, 2, 0, 0, 0, 2, 2, 2, 2, 0, 2, 0, 0, 0, 2, 0, 2, 0, 0, 2, 0, 2, 2, 2, 0, 0, 2, 2, 2, 1, 2, 0, 2, 2, 0, 2, 2, 2, 0, 0, 0, 2, 2, 2, 2, 2, 0, 2, 0, 0, 0, 0, 0, 2, 0, 2, 0, 2, 0, 0, 2, 2, 2, 1, 0, 2, 0, 0, 2, 0, 0, 1, 2, 2, 0, 2, 0, 0, 2, 2, 0, 0, 2, 2, 2, 2, 2, 1, 1, 0, 2, 0, 2, 2, 2, 1, 2, 2, 2, 2, 2, 2, 2, 2, 0, 0, 2, 0, 0, 2, 2, 0, 1, 0, 0, 0, 0, 0, 0, 0, 2, 2, 2, 2, 0, 0, 0, 2, 0, 2, 1, 2, 2, 1, 1, 2, 2, 2, 2, 2, 0, 0, 2, 2, 1, 0, 2, 2, 0, 0, 0, 2, 0, 0, 2, 2, 2, 2, 2, 2, 2, 0, 0, 0, 2, 0, 0, 0, 2, 2, 2, 1, 0, 0, 0, 2, 2, 2, 0, 2, 0, 0, 0, 0, 0, 2, 2, 1, 2, 2, 2, 0, 2, 2 ], "yaxis": "y" } ], "layout": { "autosize": true, "legend": { "title": { "text": "variable" }, "tracegroupgap": 0 }, "margin": { "t": 60 }, "template": { "data": { "bar": [ { "error_x": { "color": "#2a3f5f" }, "error_y": { "color": "#2a3f5f" }, "marker": { "line": { "color": "#E5ECF6", "width": 0.5 }, "pattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 } }, "type": "bar" } ], "barpolar": [ { "marker": { "line": { "color": "#E5ECF6", "width": 0.5 }, "pattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 } }, "type": "barpolar" } ], "carpet": [ { "aaxis": { "endlinecolor": "#2a3f5f", "gridcolor": "white", "linecolor": "white", "minorgridcolor": "white", "startlinecolor": "#2a3f5f" }, "baxis": { "endlinecolor": "#2a3f5f", "gridcolor": "white", "linecolor": "white", "minorgridcolor": "white", "startlinecolor": "#2a3f5f" }, "type": "carpet" } ], "choropleth": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "choropleth" } ], "contour": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "contour" } ], "contourcarpet": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "contourcarpet" } ], "heatmap": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "heatmap" } ], "heatmapgl": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "heatmapgl" } ], "histogram": [ { "marker": { "pattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 } }, "type": "histogram" } ], "histogram2d": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "histogram2d" } ], "histogram2dcontour": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "histogram2dcontour" } ], "mesh3d": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "mesh3d" } ], "parcoords": [ { "line": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "parcoords" } ], "pie": [ { "automargin": true, "type": "pie" } ], "scatter": [ { "fillpattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 }, "type": "scatter" } ], "scatter3d": [ { "line": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatter3d" } ], "scattercarpet": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattercarpet" } ], "scattergeo": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattergeo" } ], "scattergl": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattergl" } ], "scattermapbox": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattermapbox" } ], "scatterpolar": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterpolar" } ], "scatterpolargl": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterpolargl" } ], "scatterternary": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterternary" } ], "surface": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "surface" } ], "table": [ { "cells": { "fill": { "color": "#EBF0F8" }, "line": { "color": "white" } }, "header": { "fill": { "color": "#C8D4E3" }, "line": { "color": "white" } }, "type": "table" } ] }, "layout": { "annotationdefaults": { "arrowcolor": "#2a3f5f", "arrowhead": 0, "arrowwidth": 1 }, "autotypenumbers": "strict", "coloraxis": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "colorscale": { "diverging": [ [ 0, "#8e0152" ], [ 0.1, "#c51b7d" ], [ 0.2, "#de77ae" ], [ 0.3, "#f1b6da" ], [ 0.4, "#fde0ef" ], [ 0.5, "#f7f7f7" ], [ 0.6, "#e6f5d0" ], [ 0.7, "#b8e186" ], [ 0.8, "#7fbc41" ], [ 0.9, "#4d9221" ], [ 1, "#276419" ] ], "sequential": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "sequentialminus": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ] }, "colorway": [ "#636efa", "#EF553B", "#00cc96", "#ab63fa", "#FFA15A", "#19d3f3", "#FF6692", "#B6E880", "#FF97FF", "#FECB52" ], "font": { "color": "#2a3f5f" }, "geo": { "bgcolor": "white", "lakecolor": "white", "landcolor": "#E5ECF6", "showlakes": true, "showland": true, "subunitcolor": "white" }, "hoverlabel": { "align": "left" }, "hovermode": "closest", "mapbox": { "style": "light" }, "paper_bgcolor": "white", "plot_bgcolor": "#E5ECF6", "polar": { "angularaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "bgcolor": "#E5ECF6", "radialaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" } }, "scene": { "xaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" }, "yaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" }, "zaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" } }, "shapedefaults": { "line": { "color": "#2a3f5f" } }, "ternary": { "aaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "baxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "bgcolor": "#E5ECF6", "caxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" } }, "title": { "x": 0.05 }, "xaxis": { "automargin": true, "gridcolor": "white", "linecolor": "white", "ticks": "", "title": { "standoff": 15 }, "zerolinecolor": "white", "zerolinewidth": 2 }, "yaxis": { "automargin": true, "gridcolor": "white", "linecolor": "white", "ticks": "", "title": { "standoff": 15 }, "zerolinecolor": "white", "zerolinewidth": 2 } } }, "xaxis": { "anchor": "y", "autorange": true, "domain": [ 0, 1 ], "range": [ 0, 198 ], "title": { "text": "index" }, "type": "linear" }, "yaxis": { "anchor": "x", "autorange": true, "domain": [ 0, 1 ], "range": [ -0.11111111111111112, 2.111111111111111 ], "title": { "text": "value" }, "type": "linear" } } }, "image/png": "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", "text/html": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "px.line([a[0] for a in actions])" ] }, { "cell_type": "code", "execution_count": 23, "id": "e600205b-6acf-408f-ab1d-c37c347d3a0c", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "outputs": [ { "data": { "application/vnd.plotly.v1+json": { "config": { "plotlyServerURL": "https://plot.ly" }, "data": [ { "hovertemplate": "variable=0
index=%{x}
value=%{y}", "legendgroup": "0", "line": { "color": "#636efa", "dash": "solid" }, "marker": { "symbol": "circle" }, "mode": "lines", "name": "0", "orientation": "v", "showlegend": true, "type": "scatter", "x": [ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 191, 192, 193, 194, 195, 196, 197, 198 ], "xaxis": "x", "y": [ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 14.683285867122656, 14.683285867122656, 14.683285867122656, 14.683285867122656, 14.683285867122656, 14.683285867122656, 14.683285867122656, 14.683285867122656, 14.683285867122656, 14.683285867122656, 14.683285867122656, 14.683285867122656, 14.683285867122656, 14.683285867122656, 14.683285867122656, 14.683285867122656, 14.683285867122656, 14.683285867122656, 14.683285867122656, 14.683285867122656, 14.683285867122656, 14.683285867122656, 14.683285867122656, 14.683285867122656, 14.683285867122656, 14.683285867122656, 14.683285867122656, 14.683285867122656, 14.683285867122656, 14.683285867122656, 14.683285867122656, 14.683285867122656, 14.683285867122656, 14.683285867122656, 43.051881516287224, 43.051881516287224, 43.051881516287224, 43.051881516287224, 43.051881516287224, 43.051881516287224, 43.051881516287224, 43.051881516287224, 52.165984118742834, 52.165984118742834, 52.165984118742834, 52.165984118742834, 52.165984118742834, 52.165984118742834, 52.165984118742834, 52.165984118742834, 52.165984118742834, 52.165984118742834, 52.165984118742834, 52.165984118742834, 52.165984118742834, 52.165984118742834, 52.165984118742834, 52.165984118742834, 67.70629113397436, 67.70629113397436, 69.6333210302386, 69.6333210302386, 69.6333210302386, 69.6333210302386, 69.6333210302386, 69.6333210302386, 69.50234154714168, 69.50234154714168, 69.50234154714168, 69.50234154714168, 69.50234154714168, 69.50234154714168, 69.50234154714168, 69.50234154714168, 69.50234154714168, 69.50234154714168, 69.50234154714168, 69.50234154714168, 69.50234154714168, 69.50234154714168, 69.50234154714168, 69.50234154714168, 69.50234154714168, 82.59349721058133, 82.59349721058133, 82.59349721058133, 82.59349721058133, 82.59349721058133, 82.59349721058133, 82.59349721058133, 82.59349721058133, 82.59349721058133, 82.59349721058133, 82.59349721058133, 82.59349721058133, 82.59349721058133, 82.59349721058133, 82.59349721058133, 82.59349721058133, 82.59349721058133, 82.59349721058133, 96.54487172125498, 96.54487172125498, 96.54487172125498, 96.54487172125498, 96.54487172125498, 96.54487172125498, 96.54487172125498, 96.54487172125498, 96.54487172125498, 96.54487172125498, 89.71185518524442, 89.71185518524442, 89.71185518524442, 89.71185518524442, 94.85069336363634, 94.85069336363634, 94.85069336363634, 94.85069336363634, 94.85069336363634, 94.85069336363634, 94.85069336363634, 94.85069336363634, 94.85069336363634, 94.85069336363634, 94.85069336363634, 94.85069336363634, 94.85069336363634, 94.85069336363634, 94.85069336363634, 94.85069336363634, 94.85069336363634, 94.85069336363634, 94.85069336363634, 94.85069336363634, 94.85069336363634, 94.85069336363634, 94.85069336363634, 94.85069336363634, 94.85069336363634, 94.85069336363634, 94.85069336363634, 131.4729838740265, 131.4729838740265, 131.4729838740265, 131.4729838740265, 131.4729838740265, 131.4729838740265, 131.4729838740265, 131.4729838740265, 131.4729838740265, 131.4729838740265, 131.4729838740265, 131.4729838740265, 131.4729838740265, 131.4729838740265, 131.4729838740265, 131.4729838740265, 146.08859019547393, 146.08859019547393, 146.08859019547393, 146.08859019547393, 146.08859019547393, 146.08859019547393, 146.08859019547393 ], "yaxis": "y" } ], "layout": { "autosize": true, "legend": { "title": { "text": "variable" }, "tracegroupgap": 0 }, "margin": { "t": 60 }, "template": { "data": { "bar": [ { "error_x": { "color": "#2a3f5f" }, "error_y": { "color": "#2a3f5f" }, "marker": { "line": { "color": "#E5ECF6", "width": 0.5 }, "pattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 } }, "type": "bar" } ], "barpolar": [ { "marker": { "line": { "color": "#E5ECF6", "width": 0.5 }, "pattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 } }, "type": "barpolar" } ], "carpet": [ { "aaxis": { "endlinecolor": "#2a3f5f", "gridcolor": "white", "linecolor": "white", "minorgridcolor": "white", "startlinecolor": "#2a3f5f" }, "baxis": { "endlinecolor": "#2a3f5f", "gridcolor": "white", "linecolor": "white", "minorgridcolor": "white", "startlinecolor": "#2a3f5f" }, "type": "carpet" } ], "choropleth": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "choropleth" } ], "contour": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "contour" } ], "contourcarpet": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "contourcarpet" } ], "heatmap": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "heatmap" } ], "heatmapgl": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "heatmapgl" } ], "histogram": [ { "marker": { "pattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 } }, "type": "histogram" } ], "histogram2d": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "histogram2d" } ], "histogram2dcontour": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "histogram2dcontour" } ], "mesh3d": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "type": "mesh3d" } ], "parcoords": [ { "line": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "parcoords" } ], "pie": [ { "automargin": true, "type": "pie" } ], "scatter": [ { "fillpattern": { "fillmode": "overlay", "size": 10, "solidity": 0.2 }, "type": "scatter" } ], "scatter3d": [ { "line": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatter3d" } ], "scattercarpet": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattercarpet" } ], "scattergeo": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattergeo" } ], "scattergl": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattergl" } ], "scattermapbox": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scattermapbox" } ], "scatterpolar": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterpolar" } ], "scatterpolargl": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterpolargl" } ], "scatterternary": [ { "marker": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "type": "scatterternary" } ], "surface": [ { "colorbar": { "outlinewidth": 0, "ticks": "" }, "colorscale": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "type": "surface" } ], "table": [ { "cells": { "fill": { "color": "#EBF0F8" }, "line": { "color": "white" } }, "header": { "fill": { "color": "#C8D4E3" }, "line": { "color": "white" } }, "type": "table" } ] }, "layout": { "annotationdefaults": { "arrowcolor": "#2a3f5f", "arrowhead": 0, "arrowwidth": 1 }, "autotypenumbers": "strict", "coloraxis": { "colorbar": { "outlinewidth": 0, "ticks": "" } }, "colorscale": { "diverging": [ [ 0, "#8e0152" ], [ 0.1, "#c51b7d" ], [ 0.2, "#de77ae" ], [ 0.3, "#f1b6da" ], [ 0.4, "#fde0ef" ], [ 0.5, "#f7f7f7" ], [ 0.6, "#e6f5d0" ], [ 0.7, "#b8e186" ], [ 0.8, "#7fbc41" ], [ 0.9, "#4d9221" ], [ 1, "#276419" ] ], "sequential": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ], "sequentialminus": [ [ 0, "#0d0887" ], [ 0.1111111111111111, "#46039f" ], [ 0.2222222222222222, "#7201a8" ], [ 0.3333333333333333, "#9c179e" ], [ 0.4444444444444444, "#bd3786" ], [ 0.5555555555555556, "#d8576b" ], [ 0.6666666666666666, "#ed7953" ], [ 0.7777777777777778, "#fb9f3a" ], [ 0.8888888888888888, "#fdca26" ], [ 1, "#f0f921" ] ] }, "colorway": [ "#636efa", "#EF553B", "#00cc96", "#ab63fa", "#FFA15A", "#19d3f3", "#FF6692", "#B6E880", "#FF97FF", "#FECB52" ], "font": { "color": "#2a3f5f" }, "geo": { "bgcolor": "white", "lakecolor": "white", "landcolor": "#E5ECF6", "showlakes": true, "showland": true, "subunitcolor": "white" }, "hoverlabel": { "align": "left" }, "hovermode": "closest", "mapbox": { "style": "light" }, "paper_bgcolor": "white", "plot_bgcolor": "#E5ECF6", "polar": { "angularaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "bgcolor": "#E5ECF6", "radialaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" } }, "scene": { "xaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" }, "yaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" }, "zaxis": { "backgroundcolor": "#E5ECF6", "gridcolor": "white", "gridwidth": 2, "linecolor": "white", "showbackground": true, "ticks": "", "zerolinecolor": "white" } }, "shapedefaults": { "line": { "color": "#2a3f5f" } }, "ternary": { "aaxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "baxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" }, "bgcolor": "#E5ECF6", "caxis": { "gridcolor": "white", "linecolor": "white", "ticks": "" } }, "title": { "x": 0.05 }, "xaxis": { "automargin": true, "gridcolor": "white", "linecolor": "white", "ticks": "", "title": { "standoff": 15 }, "zerolinecolor": "white", "zerolinewidth": 2 }, "yaxis": { "automargin": true, "gridcolor": "white", "linecolor": "white", "ticks": "", "title": { "standoff": 15 }, "zerolinecolor": "white", "zerolinewidth": 2 } } }, "xaxis": { "anchor": "y", "autorange": true, "domain": [ 0, 1 ], "range": [ 0, 198 ], "title": { "text": "index" }, "type": "linear" }, "yaxis": { "anchor": "x", "autorange": true, "domain": [ 0, 1 ], "range": [ -8.11603278863744, 154.20462298411138 ], "title": { "text": "value" }, "type": "linear" } } }, "image/png": "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", "text/html": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "px.line(rewards)" ] }, { "cell_type": "markdown", "id": "1d6ef152", "metadata": { "editable": true, "slideshow": { "slide_type": "slide" }, "tags": [] }, "source": [ "## Robotik" ] }, { "cell_type": "markdown", "id": "f4115fa1-48a3-45a9-a515-3d83ae226058", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "DL ist insbesondere in der Robotik ein beliebtes Lernverfahren. Das liegt daran, das zum einen einfache Aufgaben, wie das Greifen von Objekten für Roboter hochkomplex sind und die Regelungen und Steuerungen sehr komplex zu entwickeln sind. Auch wir Menschen brauchen Wochen als Kleinkind um die Grob- und Feinmotorik dafür zu lernen. Trotzdem ist die Aufgabe und die Umgebung eines Roboters einfach zu simulieren. Deshalb setzt man vermehrt auf RL, um solche komplexen Steuerungsprobleme zu erlernen, statt selbstständisch Steuerungen zu entwickeln." ] }, { "cell_type": "markdown", "id": "52b730c0-c6b7-47c4-82fe-88d13a5aa5f6", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "### Robot Pusher" ] }, { "cell_type": "markdown", "id": "cb8b6f51-8a81-4c90-9e91-ae3783cdfaf7", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "Pusher-v5 ist eine Umgebung aus der Gymnasium-Bibliothek, die zur Simulation von Aufgaben im Bereich der Robotersteuerung verwendet wird. In dieser speziellen Umgebung wird ein Roboterarm simuliert, der darauf trainiert wird, ein Objekt zu einer bestimmten Zielposition zu schieben. " ] }, { "cell_type": "markdown", "id": "42e6ce4b-2f1d-45b0-8ffe-8d44511ab84d", "metadata": { "editable": true, "slideshow": { "slide_type": "skip" }, "tags": [] }, "source": [ "Das Ziel des Agenten (Roboterarms) in der Pusher-v5-Umgebung ist es, ein Objekt (in der Regel ein Block) zu einer festgelegten Zielposition zu schieben. Der Agent muss lernen, wie er seine Gelenke bewegen kann, um das Objekt erfolgreich zu schieben.\n", "\n", "Der Aktionsraum ist kontinuierlich und repräsentiert die Steuerung des Roboterarms. Typischerweise handelt es sich um einen Vektor von Gelenkbewegungen, die der Agent ausführen kann.\n", "\n", "Der Beobachtungsraum umfasst verschiedene Aspekte des Zustands der Umgebung, einschließlich der Position des Endeffektors des Roboterarms; die Position des zu schiebenden Objekts und die Zielposition, zu der das Objekt geschoben werden soll.\n", "\n", "Die Belohnung in der Pusher-v5-Umgebung basiert darauf, wie nah das Objekt an der Zielposition ist. Der Agent erhält eine höhere Belohnung, wenn das Objekt näher an der Zielposition ist, und eine geringere Belohnung, wenn es weiter entfernt ist." ] }, { "cell_type": "markdown", "id": "92d73f62-6315-4270-96f8-f77170068a9c", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "source": [ "Da das Modell in `Gym` enthalten ist, ist die Initialisierung einfach." ] }, { "cell_type": "code", "execution_count": 24, "id": "d09f40a7", "metadata": { "editable": true, "scrolled": true, "slideshow": { "slide_type": "" }, "tags": [ "scroll-output" ] }, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "2024-06-24 11:12:43,177\tWARNING algorithm_config.py:4078 -- You have specified 1 evaluation workers, but your `evaluation_interval` is 0 or None! Therefore, evaluation will not occur automatically with each call to `Algorithm.train()`. Instead, you will have to call `Algorithm.evaluate()` manually in order to trigger an evaluation run.\n", "/Users/jploennigs/miniconda3/envs/lehre4/lib/python3.11/site-packages/ray/rllib/algorithms/algorithm.py:525: RayDeprecationWarning:\n", "\n", "This API is deprecated and may be removed in future Ray releases. You could suppress this warning by setting env variable PYTHONWARNINGS=\"ignore::DeprecationWarning\"\n", "`UnifiedLogger` will be removed in Ray 2.7.\n", "\n", "/Users/jploennigs/miniconda3/envs/lehre4/lib/python3.11/site-packages/ray/tune/logger/unified.py:53: RayDeprecationWarning:\n", "\n", "This API is deprecated and may be removed in future Ray releases. You could suppress this warning by setting env variable PYTHONWARNINGS=\"ignore::DeprecationWarning\"\n", "The `JsonLogger interface is deprecated in favor of the `ray.tune.json.JsonLoggerCallback` interface and will be removed in Ray 2.7.\n", "\n", "/Users/jploennigs/miniconda3/envs/lehre4/lib/python3.11/site-packages/ray/tune/logger/unified.py:53: RayDeprecationWarning:\n", "\n", "This API is deprecated and may be removed in future Ray releases. You could suppress this warning by setting env variable PYTHONWARNINGS=\"ignore::DeprecationWarning\"\n", "The `CSVLogger interface is deprecated in favor of the `ray.tune.csv.CSVLoggerCallback` interface and will be removed in Ray 2.7.\n", "\n", "/Users/jploennigs/miniconda3/envs/lehre4/lib/python3.11/site-packages/ray/tune/logger/unified.py:53: RayDeprecationWarning:\n", "\n", "This API is deprecated and may be removed in future Ray releases. You could suppress this warning by setting env variable PYTHONWARNINGS=\"ignore::DeprecationWarning\"\n", "The `TBXLogger interface is deprecated in favor of the `ray.tune.tensorboardx.TBXLoggerCallback` interface and will be removed in Ray 2.7.\n", "\n", "2024-06-24 11:12:47,084\tWARNING algorithm_config.py:4078 -- You have specified 1 evaluation workers, but your `evaluation_interval` is 0 or None! Therefore, evaluation will not occur automatically with each call to `Algorithm.train()`. Instead, you will have to call `Algorithm.evaluate()` manually in order to trigger an evaluation run.\n", "2024-06-24 11:12:49,605\tWARNING util.py:61 -- Install gputil for GPU system monitoring.\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Execution Time: 79.14778399467468\n" ] } ], "source": [ "from ray.rllib.algorithms.ppo import PPOConfig\n", "\n", "if not ray.is_initialized():\n", " ray.init(ignore_reinit_error=True)\n", "\n", "config = ( # 1. Configure the algorithm,\n", " PPOConfig()\n", " .api_stack(\n", " enable_rl_module_and_learner=False,\n", " enable_env_runner_and_connector_v2=False\n", " )\n", " .environment(\"Pusher-v5\", render_env=True)\n", " .env_runners(num_env_runners=2)\n", " .framework(\"torch\")\n", " .training()\n", " .evaluation(evaluation_num_env_runners=1)\n", ")\n", "\n", "algo = config.build_algo() # 2. build the algorithm,\n", "\n", "for _ in range(5):\n", " algo.train() # 3. train it,\n", "\n", "#algo.evaluate() # 4. and evaluate it.\n", "print(f\"Execution Time: {time.time()-tic}\")" ] }, { "cell_type": "markdown", "id": "4dc7ea4f-ce9d-47c2-ac4c-442c16663cc4", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "Betrachten wir einmal das Ergebnis einer zufälligen Bewegung, so sehen wir wie der Arm vorerst orientierungslos agiert." ] }, { "cell_type": "markdown", "id": "e312cf30-fe86-430d-820f-63cd2031bc07", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "source": [ "![](images/15_Reinforcement_Learning/pusher.gif)" ] }, { "cell_type": "markdown", "id": "ddec7562-0e25-4628-8a73-8786a691449f", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "Nach mehreren Trainingsepisoden lernt der Algorithmus allerdings den Arm gut zu benutzen. Hier ein Vergleich unterschiedlicher RL-Ansätze. Zu beobachten ist, dass über die Trainings-Episoden die Modelle durch Zufall die Lösung entdecken und dann wiederholen können. Die finalen Lösungen sind dabei aber auch nach vielen Trainings-Episoden nicht perfekt." ] }, { "cell_type": "code", "execution_count": 2, "id": "c8aac465-d90a-437a-8b69-859f67bb2aa4", "metadata": { "editable": true, "slideshow": { "slide_type": "" } }, "outputs": [ { "data": { "text/html": [ "\n", " \n", " " ], "text/plain": [ "" ] }, "execution_count": 2, "metadata": {}, "output_type": "execute_result" } ], "source": [ "from IPython.display import IFrame\n", "IFrame(width=\"800\", height=\"413\", src=\"https://www.youtube.com/embed/_QmcH1TyNwg\", title=\"Gymnasium - Pusher-v4, Test with different algorithms\",\n", " frameborder=\"0\", allow=\"accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share\", referrerpolicy=\"strict-origin-when-cross-origin\", allowfullscreen=True)" ] }, { "cell_type": "markdown", "id": "acd89b35-dc69-4274-a27d-9f5f322e1f87", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "### Acrobot" ] }, { "cell_type": "markdown", "id": "4f50b4c6-211b-488a-b522-a0606050b743", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "`Acrobot-v1` ist eine klassische Gymnasium-Umgebung, in der ein zweigliedriges Pendelsystem durch geeignete Steuerimpulse in eine aufrechte Position gebracht werden soll." ] }, { "cell_type": "markdown", "id": "762fc42b-f1b7-4d53-a9ad-4fc1e9e948a5", "metadata": { "editable": true, "slideshow": { "slide_type": "skip" }, "tags": [] }, "source": [ "Das Hauptziel des Agenten in der `Acrobot-v1`-Umgebung ist es, das gekoppelte Pendelsystem so zu steuern, dass das freie Ende eine Zielhöhe erreicht. Der Agent muss lernen, wie die verfügbaren Drehmomente eingesetzt werden, um Schwung aufzubauen und das System kontrolliert aufzurichten.\n", "\n", "Der Aktionsraum ist diskret und beschreibt wenige mögliche Steuerimpulse am Gelenk. Der Beobachtungsraum enthält die Winkel- und Geschwindigkeitsinformationen des Systems in kompakter Form, sodass der Agent den aktuellen Zustand des Pendels einschätzen kann.\n", "\n", "Die Belohnungsstruktur ist einfach: Der Agent erhält in jedem Zeitschritt eine negative Rückmeldung, bis das Ziel erreicht ist. Dadurch wird er dazu motiviert, die Aufgabe mit möglichst wenigen Schritten zu lösen." ] }, { "cell_type": "markdown", "id": "62e5b008-5f71-4fa6-8923-9a236d2d1ae7", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "source": [ "Auch hier ist die Initialisierung einfach." ] }, { "cell_type": "code", "execution_count": 26, "id": "0356d6bf-f781-4c28-bb08-dff5e1ef5776", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [ "scroll-output" ] }, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "2024-06-24 11:13:13,573\tWARNING algorithm_config.py:4078 -- You have specified 1 evaluation workers, but your `evaluation_interval` is 0 or None! Therefore, evaluation will not occur automatically with each call to `Algorithm.train()`. Instead, you will have to call `Algorithm.evaluate()` manually in order to trigger an evaluation run.\n", "2024-06-24 11:13:17,302\tWARNING algorithm_config.py:4078 -- You have specified 1 evaluation workers, but your `evaluation_interval` is 0 or None! Therefore, evaluation will not occur automatically with each call to `Algorithm.train()`. Instead, you will have to call `Algorithm.evaluate()` manually in order to trigger an evaluation run.\n", "2024-06-24 11:13:20,204\tWARNING util.py:61 -- Install gputil for GPU system monitoring.\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Execution Time: 119.0861268043518\n" ] } ], "source": [ "from ray.rllib.algorithms.ppo import PPOConfig\n", "\n", "if not ray.is_initialized():\n", " ray.init(ignore_reinit_error=True)\n", "\n", "config = ( # 1. Configure the algorithm,\n", " PPOConfig()\n", " .api_stack(\n", " enable_rl_module_and_learner=False,\n", " enable_env_runner_and_connector_v2=False\n", " )\n", " .environment('Acrobot-v1')\n", " .env_runners(num_env_runners=2)\n", " .framework(\"torch\")\n", " .training()\n", " .evaluation(evaluation_num_env_runners=1)\n", ")\n", "\n", "algo = config.build_algo() # 2. build the algorithm,\n", "\n", "for _ in range(5):\n", " algo.train() # 3. train it,\n", "\n", "#algo.evaluate() # 4. and evaluate it.\n", "print(f\"Execution Time: {time.time()-tic}\")" ] }, { "cell_type": "markdown", "id": "8d650afb-22d1-49f2-9019-6bfbebb7a73e", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "Das RL-Modell lernt auch hier, ein dynamisches System schrittweise besser zu steuern." ] }, { "cell_type": "markdown", "id": "581a246e-c9fc-4525-a2f7-bcd461641f33", "metadata": { "editable": true, "slideshow": { "slide_type": "" }, "tags": [] }, "source": [ "" ] }, { "cell_type": "markdown", "id": "6de602b2", "metadata": { "editable": true, "slideshow": { "slide_type": "slide" }, "tags": [ "remove-cell" ] }, "source": [ "
\n", "
f  r  a  g  e  n  ?
\n", "
" ] } ], "metadata": { "jupytext": { "cell_metadata_filter": "-all", "main_language": "python", "notebook_metadata_filter": "-all" }, "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": "3.12.8" }, "rise": { "auto_select": "none", "center": false, "enable_chalkboard": true, "header": "", "scroll": true, "show_buttons_on_startup": false, "slideNumber": true, "theme": "white" } }, "nbformat": 4, "nbformat_minor": 5 }