Lab09/Lab09-1_ChanjaeYu.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Lab09-1 XOR - Logistic Regression - Eager Excuetion"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- XOR 문제를 Logistic Regression을 활용해 풀어보도록 하겠습니다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2.4.1\n"
     ]
    }
   ],
   "source": [
    "# 기본 Library 선언 및 Tensorflow 버전 확인\n",
    "\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "import tensorflow as tf\n",
    "\n",
    "tf.random.set_seed(777)  # for reproducibility\n",
    "\n",
    "print(tf.__version__)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 강의에 설명할 Data입니다.\n",
    "- x_data가 2차원 배열이기에 2차원 공간에 표현하여 x1과 x2를 기준으로 y_data 0과 1로 구분하는 예제입니다.\n",
    "- 붉은색과 푸른색으로 0과 1을 표시해 보도록 하겠습니다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "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": [
    "# 데이터 생성\n",
    "x_data = [[0, 0],\n",
    "          [0, 1],\n",
    "          [1, 0],\n",
    "          [1, 1]]\n",
    "y_data = [[0],\n",
    "          [1],\n",
    "          [1],\n",
    "          [0]]\n",
    "\n",
    "# 그래프에 각각의 x_data에 따른 y_data값 나타내기\n",
    "plt.scatter(x_data[0][0],x_data[0][1], c='red' , marker='^')\n",
    "plt.scatter(x_data[3][0],x_data[3][1], c='red' , marker='^')\n",
    "plt.scatter(x_data[1][0],x_data[1][1], c='blue' , marker='^')\n",
    "plt.scatter(x_data[2][0],x_data[2][1], c='blue' , marker='^')\n",
    "\n",
    "plt.xlabel(\"x1\")\n",
    "plt.ylabel(\"x2\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Tensorflow Eager\n",
    "#### 위 Data를 기준으로 XOR처리를 위한 모델을 만들도록 하겠습니다.\n",
    "- Tensorflow data API를 통해 학습시킬 값들을 담는다 (Batch Size는 한번에 학습시킬 Size로 정한다)\n",
    "- preprocess function으로 features,labels는 실재 학습에 쓰일 Data 연산을 위해 Type를 맞춰준다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# x_data, y_data 값을 tensorflow로 Dataset화 하기\n",
    "dataset = tf.data.Dataset.from_tensor_slices((x_data, y_data)).batch(len(x_data))\n",
    "\n",
    "# 위의 integer값을 연산을 위한 값으로 변환\n",
    "def preprocess_data(features, labels):\n",
    "    features = tf.cast(features, tf.float32)\n",
    "    labels = tf.cast(labels, tf.float32)\n",
    "    return features, labels"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1) Logistic Regression으로 XOR모델을 만들어 보겠습니다.\n",
    "##### W와 b은 학습을 통해 생성되는 모델에 쓰이는 Weight와 Bias (초기값을 variable : 0이나 Random값으로 가능 tf.random_normal([2, 1]) )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "W = [[0.]\n",
      " [0.]], B = [0.]\n"
     ]
    }
   ],
   "source": [
    "# Weight와 Bias의 초기값을 0으로 지정\n",
    "W = tf.Variable(tf.zeros((2,1)), name='weight')\n",
    "b = tf.Variable(tf.zeros((1,)), name='bias')\n",
    "print(\"W = {}, B = {}\".format(W.numpy(), b.numpy()))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Sigmoid 함수를 가설(예측값)로 선언합니다.\n",
    "- Sigmoid는 아래 그래프와 같이 0과 1의 값만을 리턴합니다 tf.sigmoid(tf.matmul(X, W) + b)와 같습니다.\n",
    "$$\n",
    "\\begin{align}\n",
    "sigmoid(x)  = \\frac{1}{1+e^{-x}}  \\\\\\\\\\\n",
    "\\end{align}\n",
    "$$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "def logistic_regression(features):\n",
    "    hypothesis  = tf.divide(1., 1. + tf.exp(tf.matmul(features, W) + b))\n",
    "    return hypothesis"
   ]
  },
  {
   "attachments": {
    "image-2.png": {
     "image/png": "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"
    },
    "image.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 가설을 검증할 Cost 함수를 정의합니다.\n",
    "![image.png](attachment:image.png)\n",
    "-  두 수식을 합치면 다음과 같습니다.\n",
    "![image-2.png](attachment:image-2.png)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "def loss_fn(hypothesis, features, labels):\n",
    "    cost = -tf.reduce_mean(labels * tf.math.log(logistic_regression(features)) + (1 - labels) * tf.math.log(1 - hypothesis))\n",
    "    return cost\n",
    "\n",
    "optimizer = tf.keras.optimizers.SGD(learning_rate=0.01)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 추론한 값은 0.5를 기준(Sigmoid 그래프 참조)로 0과 1의 값을 리턴합니다.\n",
    "- Sigmoid 함수를 통해 예측값이 0.5보다 크면 1을 반환하고 0.5보다 작으면 0으로 반환합니다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "def accuracy_fn(hypothesis, labels):\n",
    "    predicted = tf.cast(hypothesis > 0.5, dtype=tf.float32)\n",
    "    accuracy = tf.reduce_mean(tf.cast(tf.equal(predicted, labels), dtype=tf.float32))\n",
    "    return accuracy"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### GradientTape를 통해 경사값을 계산합니다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "def grad(hypothesis, features, labels):\n",
    "    with tf.GradientTape() as tape:\n",
    "        loss_value = loss_fn(logistic_regression(features),features,labels)\n",
    "    return tape.gradient(loss_value, [W,b])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Tensorflow를 통한 실행을 위해 Session를 선언합니다.\n",
    "- 위의 Data를 Cost함수를 통해 학습시킨 후 모델을 생성합니다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Iter: 0, Loss: 0.6931\n",
      "Iter: 100, Loss: 0.6931\n",
      "Iter: 200, Loss: 0.6931\n",
      "Iter: 300, Loss: 0.6931\n",
      "Iter: 400, Loss: 0.6931\n",
      "Iter: 500, Loss: 0.6931\n",
      "Iter: 600, Loss: 0.6931\n",
      "Iter: 700, Loss: 0.6931\n",
      "Iter: 800, Loss: 0.6931\n",
      "Iter: 900, Loss: 0.6931\n",
      "Iter: 1000, Loss: 0.6931\n",
      "W = [[0.]\n",
      " [0.]], B = [0.]\n",
      "Testset Accuracy: 0.5000\n"
     ]
    }
   ],
   "source": [
    "EPOCHS = 1001\n",
    "\n",
    "for step in range(EPOCHS):\n",
    "    for features, labels  in dataset:\n",
    "        features, labels = preprocess_data(features, labels)\n",
    "        grads = grad(logistic_regression(features), features, labels)\n",
    "        optimizer.apply_gradients(grads_and_vars=zip(grads,[W,b]))\n",
    "        if step % 100 == 0:\n",
    "            print(\"Iter: {}, Loss: {:.4f}\".format(step, loss_fn(logistic_regression(features),features,labels)))\n",
    "print(\"W = {}, B = {}\".format(W.numpy(), b.numpy()))\n",
    "x_data, y_data = preprocess_data(x_data, y_data)\n",
    "test_acc = accuracy_fn(logistic_regression(x_data),y_data)\n",
    "print(\"Testset Accuracy: {:.4f}\".format(test_acc))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "-> Iteration에 따라 Loss 감소하는 경향 보이지 않고 Accuracy 또한 50% 수준에 못 미친다."
   ]
  }
 ],
 "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": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}