markov-models/scratchpad.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "\n",
    "from constants import *\n",
    "from maths import gaussian\n",
    "from markov import MarkovModel\n",
    "from markovlog import LogMarkovModel"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Scratchpad\n",
    "===================="
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "state1.mean\n",
    "state_transition[1,1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "gaussian(490, 5, 500)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "x = list(range(0, 1000))\n",
    "y = [gaussian(i, 500, 50) for i in x]\n",
    "\n",
    "plt.plot(x, y)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "x = np.linspace(-500, 500, 120)\n",
    "y = [gaussian(i, 0, 100) for i in x]\n",
    "\n",
    "plt.plot(x, y)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "for i in state_transition:\n",
    "    print(i)\n",
    "    # for j in i:\n",
    "    #     print(j)\n",
    "\n",
    "print(state_transition[1][1])\n",
    "print(state_transition.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "a = np.array([[1, 2, 3], [4, 5, 6], [7, 8, 9]])\n",
    "print(a)\n",
    "print(a[:, 1] / 5)\n",
    "a[:, 2] / 2\n",
    "print(a)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "array1 = np.array([[1, 2], [3, 4]])\n",
    "array2 = np.array([[1, 2], [3, 4]])\n",
    "\n",
    "print(array1 * array2)\n",
    "print(array1[:, 1] * array2[:, 1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = MarkovModel(states=[state1, state2], observations=observations, state_transitions=state_transition)\n",
    "\n",
    "model.get_other_state_number(2)"
   ]
  },
  {
   "source": [
    "# Forward Probability Surface"
   ],
   "cell_type": "markdown",
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = MarkovModel(states=[state1, state2], observations=observations, state_transitions=state_transition)\n",
    "model.populate_forward()\n",
    "\n",
    "fig = plt.figure(figsize=(18,6))\n",
    "ax = fig.add_subplot(1, 1, 1, projection=\"3d\")\n",
    "\n",
    "y_width = 0.6\n",
    "\n",
    "X = np.arange(1, 10)\n",
    "Y = np.arange(1, 3) - 0.5*y_width\n",
    "X, Y = np.meshgrid(X, Y)\n",
    "Z = np.zeros(model.forward.size)\n",
    "\n",
    "dx = np.ones(model.forward.size)\n",
    "dy = y_width * np.ones(model.forward.size)\n",
    "\n",
    "# ax.plot_surface(X, Y, model.forward, cmap=cm.coolwarm)\n",
    "ax.bar3d(X.flatten(), Y.flatten(), Z, dx, dy, model.forward.flatten(), shade=True)\n",
    "\n",
    "ax.set_yticks([1, 2])\n",
    "\n",
    "ax.set_xlabel(\"Observation\")\n",
    "ax.set_ylabel(\"State\")\n",
    "ax.view_init(30, -80)\n",
    "# plt.zlabel(\"Forward Likelihood\")\n",
    "fig.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from numpy import log as ln\n",
    "\n",
    "ln(0.5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = LogMarkovModel(states=[state1, state2], observations=observations, state_transitions=state_transition)\n",
    "\n",
    "print(model.state_transitions)\n",
    "model.log_state_transitions()\n",
    "print(model.state_transitions)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = MarkovModel(states=[state1, state2], observations=observations, state_transitions=state_transition)\n",
    "model.populate()\n",
    "\n",
    "model.baum_welch_state_transitions()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.8.6-final"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}