03_SOM/03. SOM.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# SOM (Self Organizing Map, 자기조직화지도)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 개요\n",
    "* 비지도학습 중 하나로, 대뇌피질의 시각피질의 학습 과정을 모델화한 인공신경망으로써 자율 학습에 의한 클러스터링을 수행하는 알고리즘"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 역사\n",
    "* 1980년대 후반에 핀란드의 컴퓨터 과학자 Teuvo Kohonen에 의해 개발\n",
    "* SOM은 분류, 군집화, 데이터 시각화 등 다양한 분야에서 사용되며, 인공신경망 분야에서 가장 성공적인 알고리즘 중 하나로 평가됨\n",
    "* 초기에는 이미지 인식과 음성 인식 분야에서 주로 사용되었으나 현재는 데이터 분석 등의 다양한 분야에서도 사용됨"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 특징\n",
    "* 인간의 뇌에서 발견된 \"자기 조직화\" 원리를 모델링하여 입력 데이터의 유사성을 기반으로 뉴런들을 2D 혹은 3D 지도상에 배치함\n",
    "* Topological Ordering (위상적 순서)에 따라 지도상에서 인접한 뉴런들이 입력 데이터의 유사성을 반영하여 군집화(Clusturing)함\n",
    "입력 데이터에 대한 어떤 사전 정보나 라벨이 없이도, 데이터가 가지고 있는 패턴과 특성을 스스로 학습하고 이를 지도상으로 표현함\n",
    "* 주로 데이터의 시각화 및 차원 축소에 사용되며, 입력 데이터의 분포를 파악하여 그룹화하여 시각적으로 표현하므로, 복잡한 데이터를 시각적으로 이해하고 분석하는 데 도움을 줌"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 주요 용어\n",
    "* 클러스터링(clustering): 데이터의 유사성에 기초하여 데이터를 몇몇의 그룹으로 분류하는 기법\n",
    "* 입력층(input layer): 입력 벡터를 입력받는 층\n",
    "* 경쟁층(competitive layer): 입력 벡터의 특성에 따라 입력 벡터가 한 점으로 클러스터링 되는 층\n",
    "* 가중치(weight): 인공신경망에서 가중치는 각 입력 값에 대한 입력 값의 중요도를 값을 말함\n",
    "* 노드(node): 경쟁층에서 입력 벡터들이 서로의 유사성에 의해 모이는 하나의 영역"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![SOM Layer Structure](./images/SOM1.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 알고리즘\n",
    "1-1) 가중치 행렬 각 원소의 값을 임의의 0보다 크고 1보다 작은 값으로 초기화\n",
    "\n",
    "2-1) 입력 벡터와 경쟁층에 존재하는 j\n",
    "개의 노드에 대해 입력 벡터와 노드 간의 거리 Dij\n",
    "를 계산\n",
    "\n",
    "2-2) 현재 입력 벡터와 Dij\n",
    "값이 가장 작은 경쟁층의 노드를 선택\n",
    "\n",
    "2-3) 해당 노드의 가중치와 이웃 노드의 가중치를 수정\n",
    "\n",
    "5-1) 현재 입력 벡터가 마지막 입력 벡터라면 다음 과정으로 이동하고, 그렇지 않다면 과정 2로 돌아간다\n",
    "\n",
    "6-1) 현재 반복 횟수가 최대 반복 횟수라면 알고리즘을 종료\n",
    "\n",
    "6-2) 현재 반복 횟수가 최대 반복 횟수가 아니면 현재 입력 벡터를 처음 입력 벡터로 설정하고 과정 2로 돌아간다"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![SOM Algorithm](./images/SOM2.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 참고\n",
    "https://untitledtblog.tistory.com/5"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### SOM 예제 코드"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.76959259, 0.03105338],\n",
       "       [0.91933906, 0.8980492 ],\n",
       "       [0.74919767, 0.46775453],\n",
       "       [0.66010482, 0.70532455],\n",
       "       [0.07347433, 0.31437638],\n",
       "       [0.52162755, 0.75475037],\n",
       "       [0.07618824, 0.5595569 ],\n",
       "       [0.37694104, 0.88389   ],\n",
       "       [0.59261356, 0.49140373],\n",
       "       [0.11567152, 0.32084966],\n",
       "       [0.81627506, 0.7724125 ],\n",
       "       [0.7708446 , 0.98667885],\n",
       "       [0.57884267, 0.92658698],\n",
       "       [0.48815338, 0.04666865],\n",
       "       [0.07971729, 0.56926156],\n",
       "       [0.93371788, 0.20473364],\n",
       "       [0.71241481, 0.68994675],\n",
       "       [0.69855141, 0.55690579],\n",
       "       [0.87081581, 0.48218263],\n",
       "       [0.73239668, 0.47015934],\n",
       "       [0.48234206, 0.38374233],\n",
       "       [0.51407113, 0.3925697 ],\n",
       "       [0.97608349, 0.58044996],\n",
       "       [0.67676596, 0.51393561],\n",
       "       [0.37059459, 0.99168193],\n",
       "       [0.17015263, 0.17250751],\n",
       "       [0.6980668 , 0.47156525],\n",
       "       [0.9674704 , 0.29847506],\n",
       "       [0.85297608, 0.15389296],\n",
       "       [0.99135256, 0.95353731],\n",
       "       [0.75749781, 0.05744834],\n",
       "       [0.5850026 , 0.69425312],\n",
       "       [0.35572734, 0.77261987],\n",
       "       [0.96259553, 0.10738437],\n",
       "       [0.0178066 , 0.68856146],\n",
       "       [0.78219693, 0.6523648 ],\n",
       "       [0.18744908, 0.73438952],\n",
       "       [0.19713904, 0.56662014],\n",
       "       [0.11656314, 0.50708698],\n",
       "       [0.33800968, 0.03634445],\n",
       "       [0.64802194, 0.90416709],\n",
       "       [0.7668028 , 0.7619133 ],\n",
       "       [0.55338936, 0.78892134],\n",
       "       [0.20716839, 0.95280491],\n",
       "       [0.47688275, 0.30063862],\n",
       "       [0.22233577, 0.02612442],\n",
       "       [0.21021643, 0.66526042],\n",
       "       [0.87039248, 0.75634555],\n",
       "       [0.91421121, 0.60350262],\n",
       "       [0.32014712, 0.61909491],\n",
       "       [0.03320721, 0.69900356],\n",
       "       [0.0806021 , 0.68826953],\n",
       "       [0.52828811, 0.78636528],\n",
       "       [0.49048742, 0.60616199],\n",
       "       [0.07021355, 0.64982408],\n",
       "       [0.21773879, 0.55571391],\n",
       "       [0.02119119, 0.62099936],\n",
       "       [0.75589808, 0.88236869],\n",
       "       [0.99479197, 0.98599938],\n",
       "       [0.35444256, 0.84292181],\n",
       "       [0.90941177, 0.30758414],\n",
       "       [0.87616741, 0.18060184],\n",
       "       [0.22973575, 0.50353087],\n",
       "       [0.33569227, 0.82710827],\n",
       "       [0.03575231, 0.86915894],\n",
       "       [0.59710596, 0.91274408],\n",
       "       [0.61854338, 0.06007   ],\n",
       "       [0.69497861, 0.26505075],\n",
       "       [0.80347052, 0.93909288],\n",
       "       [0.8998602 , 0.34498986],\n",
       "       [0.96359265, 0.66780466],\n",
       "       [0.6180583 , 0.36145526],\n",
       "       [0.22507441, 0.24977   ],\n",
       "       [0.82067222, 0.65858647],\n",
       "       [0.26533719, 0.50457457],\n",
       "       [0.20615695, 0.52028333],\n",
       "       [0.78530409, 0.23287443],\n",
       "       [0.81894409, 0.32542015],\n",
       "       [0.66607148, 0.0717416 ],\n",
       "       [0.15435642, 0.57144805],\n",
       "       [0.60568206, 0.62824524],\n",
       "       [0.40981871, 0.51319754],\n",
       "       [0.14660118, 0.14759826],\n",
       "       [0.89091339, 0.12148139],\n",
       "       [0.48159329, 0.76667587],\n",
       "       [0.73766975, 0.76237422],\n",
       "       [0.32449576, 0.11193535],\n",
       "       [0.05351455, 0.37051478],\n",
       "       [0.15487227, 0.0884276 ],\n",
       "       [0.40902546, 0.73216362],\n",
       "       [0.6227993 , 0.54298886],\n",
       "       [0.70489165, 0.60729615],\n",
       "       [0.43506953, 0.26510733],\n",
       "       [0.17818975, 0.41551925],\n",
       "       [0.07820619, 0.31987945],\n",
       "       [0.27646196, 0.00825832],\n",
       "       [0.30999502, 0.67284126],\n",
       "       [0.6372436 , 0.48780002],\n",
       "       [0.66962877, 0.44225145],\n",
       "       [0.48185839, 0.17663244]])"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 입력 데이터 생성\n",
    "data = np.random.rand(100, 2)\n",
    "# data = np.array([[0.8, 0.0064],\n",
    "#         [0.12, 0.2848],\n",
    "#         [0.24, 0.7472],\n",
    "#         [0.68, 0.3488]])\n",
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [],
   "source": [
    "# SOM 파라미터 설정\n",
    "map_width = 10\n",
    "map_height = 10\n",
    "lr = 0.1\n",
    "num_iterations = 1000\n",
    "# map_width = 2\n",
    "# map_height = 2\n",
    "# lr = 0.1\n",
    "# num_iterations = 1000"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(10, 10, 2)\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# SOM 초기화\n",
    "som_map = np.random.rand(map_width, map_height, 2)\n",
    "print(som_map.shape)\n",
    "fig, (ax1, ax2) = plt.subplots(1, 2)\n",
    "ax1.imshow(som_map[:,:,0])\n",
    "ax2.imshow(som_map[:,:,1])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [],
   "source": [
    "# SOM 학습\n",
    "for i in range(num_iterations):\n",
    "    # 랜덤한 입력 데이터 선택\n",
    "    input_data = data[np.random.choice(data.shape[0])]\n",
    "    \n",
    "    # 가장 유사한 뉴런 찾기\n",
    "    # 3D som_map과 input_data의 차이^2를 구하여 z축으로 모두 합한 거리\n",
    "    distances = np.sum((som_map - input_data) ** 2, axis=2) \n",
    "    # 위에서 구한 z축의 거리들 중 가장 작은 뉴런\n",
    "    winner = np.argmin(distances)\n",
    "    # 3D z축에서 선택된 가장 작은 뉴런을 2D map으로 맵핑한 좌표\n",
    "    x, y = np.unravel_index(winner, (map_width, map_height))\n",
    "    \n",
    "    # 학습률 계산\n",
    "    learning_rate = lr * (1 - i/num_iterations)\n",
    "    \n",
    "    # 뉴런 가중치 업데이트\n",
    "    for j in range(map_width):\n",
    "        for k in range(map_height):\n",
    "            dist = np.sqrt((x-j)**2 + (y-k)**2)\n",
    "            if dist < 3:\n",
    "                som_map[j, k] += learning_rate * (input_data - som_map[j, k])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(10, 10, 2)\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# SOM 시각화\n",
    "print(som_map.shape)\n",
    "fig, (ax1, ax2) = plt.subplots(1, 2)\n",
    "ax1.imshow(som_map[:,:,0])\n",
    "ax2.imshow(som_map[:,:,1])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 참고\n",
    "ChatGPT"
   ]
  }
 ],
 "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": "3.10.6 (main, Mar 10 2023, 10:55:28) [GCC 11.3.0]"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "916dbcbb3f70747c44a77c7bcd40155683ae19c65e1c03b4aa3499c5328201f1"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}