Example_Algo_optimizations.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Algorithms Optimization"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Example of Package Usage"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Gradient Descent"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Declaring functions for optimization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "def f(x):\n",
    "    \"\"\"Apply function to point x\n",
    "    \n",
    "    Args:\n",
    "        x (float): point on x-axis\n",
    "        \n",
    "    Returns:\n",
    "        (float): f(x)\n",
    "    \"\"\"\n",
    "    return 0.1*x**2 + np.sin(0.1*x**2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "def df(x):\n",
    "    \"\"\"Apply function gradient to point x\n",
    "    \n",
    "    Args:\n",
    "        x (float): point on x-axis\n",
    "        \n",
    "    Returns:\n",
    "        (float): df(x)\n",
    "    \"\"\"\n",
    "    return 0.2*x + 0.2*x*np.cos(0.1*x**2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "from optimizationAlgorithms import GradientDescent"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Example of Inputs:\n",
    "- Function\n",
    "- Function Gradient\n",
    "- x_t: start point\n",
    "- learning_rate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "gd = GradientDescent(f, df, x_t=3, learning_rate = 0.1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Fitting optimizer\n",
    "- Function Global Minimum = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2.3961845786107356e-05"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gd.fit()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Total Number of Iterations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "289"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gd.get_n_iteration()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Plotting Function and Optimization Steps"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "gd.plot_optimization(x1 = -10, x2 = 10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Momentum"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Declaring functions for optimization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "def f(x):\n",
    "    \"\"\"Apply function to point x\n",
    "    \n",
    "    Args:\n",
    "        x (float): point on x-axis\n",
    "        \n",
    "    Returns:\n",
    "        (float): f(x)\n",
    "    \"\"\"\n",
    "    return 4*x**2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "def df(x):\n",
    "    \"\"\"Apply function gradient to point x\n",
    "    \n",
    "    Args:\n",
    "        x (float): point on x-axis\n",
    "        \n",
    "    Returns:\n",
    "        (float): df(x)\n",
    "    \"\"\"\n",
    "    return 8*x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "from optimizationAlgorithms import Momentum"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Example of Inputs:\n",
    "- Function\n",
    "- Function Gradient\n",
    "- x_t: start point\n",
    "- learning_rate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [],
   "source": [
    "mmt = Momentum(f, df, x_t=6, learning_rate = 0.1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Fitting optimizer\n",
    "- Function Global Minimum = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-3.265851595292336e-05"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mmt.fit()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Total Number of Iterations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "230"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mmt.get_n_iteration()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Plotting Function and Optimization Steps"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "mmt.plot_optimization(x1 = -10, x2 = 10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## RMSprop"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Declaring functions for optimization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [],
   "source": [
    "def f(x):\n",
    "    \"\"\"Apply function to point x\n",
    "    \n",
    "    Args:\n",
    "        x (float): point on x-axis\n",
    "        \n",
    "    Returns:\n",
    "        (float): f(x)\n",
    "    \"\"\"\n",
    "    return 5*x**6 - 36*x**5 + 165/2*x**4 - 60*x**3 + 36"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [],
   "source": [
    "def df(x):\n",
    "    \"\"\"Apply function gradient to point x\n",
    "    \n",
    "    Args:\n",
    "        x (float): point on x-axis\n",
    "        \n",
    "    Returns:\n",
    "        (float): df(x)\n",
    "    \"\"\"\n",
    "    return 30*x**5 -180*x**4 + 330*x**3 - 180*x**2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [],
   "source": [
    "from optimizationAlgorithms import RMSprop"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Example of Inputs:\n",
    "- Function\n",
    "- Function Gradient\n",
    "- x_t: start point\n",
    "- learning_rate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [],
   "source": [
    "rms = RMSprop(f, df, x_t=4, learning_rate = 0.01)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Fitting optimizer\n",
    "- Function Global Minimum = 3 "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3.0000004279434656"
      ]
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rms.fit()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Total Number of Iterations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "166"
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rms.get_n_iteration()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Plotting Function and Optimization Steps"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "rms.plot_optimization(x1 = -2, x2 = 6)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## NAG - Nesterov Accelerated Gradient"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [],
   "source": [
    "def f(x):\n",
    "    \"\"\"Apply function to point x\n",
    "    \n",
    "    Args:\n",
    "        x (float): point on x-axis\n",
    "        \n",
    "    Returns:\n",
    "        (float): f(x)\n",
    "    \"\"\"\n",
    "    return 0.1*x**2 + np.sin(0.1*x**2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [],
   "source": [
    "def df(x):\n",
    "    \"\"\"Apply function gradient to point x\n",
    "    \n",
    "    Args:\n",
    "        x (float): point on x-axis\n",
    "        \n",
    "    Returns:\n",
    "        (float): df(x)\n",
    "    \"\"\"\n",
    "    return 0.2*x + 0.2*x*np.cos(0.1*x**2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {},
   "outputs": [],
   "source": [
    "from optimizationAlgorithms import NAG"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Example of Inputs:\n",
    "- Function\n",
    "- Function Gradient\n",
    "- x_t: start point\n",
    "- learning_rate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {},
   "outputs": [],
   "source": [
    "nag = NAG(f, df, x_t=10, learning_rate = 0.01)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Fitting optimizer\n",
    "- Function Global Minimal = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.00017886759102741137"
      ]
     },
     "execution_count": 108,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nag.fit()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Total Number of Iterations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "894"
      ]
     },
     "execution_count": 109,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nag.get_n_iteration()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Plotting Function and Optimization Steps"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "nag.plot_optimization(x1 = -10, x2 = 10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## ADAM Optimizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {},
   "outputs": [],
   "source": [
    "def f(x):\n",
    "    \"\"\"Apply function to point x\n",
    "    \n",
    "    Args:\n",
    "        x (float): point on x-axis\n",
    "        \n",
    "    Returns:\n",
    "        (float): f(x)\n",
    "    \"\"\"\n",
    "    return 5*x**6 - 36*x**5 + 165/2*x**4 - 60*x**3 + 36"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {},
   "outputs": [],
   "source": [
    "def df(x):\n",
    "    \"\"\"Apply function gradient to point x\n",
    "    \n",
    "    Args:\n",
    "        x (float): point on x-axis\n",
    "        \n",
    "    Returns:\n",
    "        (float): df(x)\n",
    "    \"\"\"\n",
    "    return 30*x**5 -180*x**4 + 330*x**3 - 180*x**2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {},
   "outputs": [],
   "source": [
    "from optimizationAlgorithms import Adam"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Example of Inputs:\n",
    "- Function\n",
    "- Function Gradient\n",
    "- x_t: start point\n",
    "- learning_rate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 142,
   "metadata": {},
   "outputs": [],
   "source": [
    "adam = Adam(f, df, x_t=7, learning_rate = 0.01, max_iterations = 7000)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Fitting optimizer: \n",
    "- Function Global Minimal = 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3.0002648187117393"
      ]
     },
     "execution_count": 143,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "adam.fit()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Total Number of Iterations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "6866"
      ]
     },
     "execution_count": 144,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "adam.get_n_iteration()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Plotting Function and Optimization Steps"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 145,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "adam.plot_optimization(x1 = -10, x2 = 10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "optimization",
   "language": "python",
   "name": "optimization"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}