linear_regression_tensorflow.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "source": [
    "# Linear Regression\n",
    "\n",
    "- Generate dummy dataset as [x,y]\n",
    "- Sampling: split as [x_train, y_train] & [x_test, y_test]\n",
    "- Model by **tensorflow** + *keras*\n",
    "- Performance check [optimization: SGD, loss: MSE]\n",
    "- Prediction\n"
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-08-09 20:04:27.215496: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  SSE4.1 SSE4.2 AVX AVX2 FMA\n",
      "To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "import matplotlib.pyplot as plt"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-08-09T17:04:35.601812Z",
     "start_time": "2023-08-09T17:04:26.995602Z"
    }
   }
  },
  {
   "cell_type": "markdown",
   "source": [
    "## preparation - dataset"
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "x [0.         1.02040816 2.04081633 3.06122449 4.08163265]\n",
      "y [0.         1.02040816 2.04081633 3.06122449 4.08163265]\n"
     ]
    }
   ],
   "source": [
    "# generate 50 size dataset(x, y=h(x))\n",
    "# add noise to dataset(x, y)\n",
    "# graph dataset(x, y)\n",
    "\n",
    "x = np.linspace(0, 50, 50)   #0 to 50, 50 sample, float\n",
    "y = np.linspace(0, 50, 50)\n",
    "\n",
    "print('x', x[0:5])\n",
    "print('y', y[0:5])"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-08-09T17:04:35.619931Z",
     "start_time": "2023-08-09T17:04:35.605455Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "x [0.51182162 1.97087186 2.18497594 4.00987394 4.39346411]\n",
      "y [ 0.11821625  5.52504513 -1.51758755  7.54771896  2.19994717]\n",
      "shape (50,) (50,)\n"
     ]
    }
   ],
   "source": [
    "# add noise\n",
    "\n",
    "x += np.random.default_rng(1).random(50)\n",
    "y += np.random.default_rng(1).uniform(-5, 5, 50)\n",
    "\n",
    "print('x', x[0:5])\n",
    "print('y', y[0:5])\n",
    "\n",
    "print('shape', x.shape, y.shape)"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-08-09T17:04:35.680748Z",
     "start_time": "2023-08-09T17:04:35.622054Z"
    }
   }
  },
  {
   "cell_type": "markdown",
   "source": [
    "## split - train vs test"
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "outputs": [
    {
     "data": {
      "text/plain": "<Figure size 300x300 with 1 Axes>",
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# sampling\n",
    "# split data as train and test set\n",
    "\n",
    "x_train = x[0:40]\n",
    "y_train = y[0:40]\n",
    "\n",
    "x_test = x[40:]\n",
    "y_test = y[40:]\n",
    "\n",
    "plt.figure(figsize=(3,3))\n",
    "plt.title(\"train vs test\")\n",
    "plt.scatter(x_train, y_train, label=\"train\")\n",
    "plt.scatter(x_test, y_test, label=\"test\" )\n",
    "plt.legend()\n",
    "plt.show()"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-08-09T17:04:57.592727Z",
     "start_time": "2023-08-09T17:04:56.511306Z"
    }
   }
  },
  {
   "cell_type": "markdown",
   "source": [
    "## model"
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "outputs": [
    {
     "data": {
      "text/plain": "(array([[ 0.5118216],\n        [ 1.9708718],\n        [ 2.1849759],\n        [ 4.009874 ],\n        [ 4.393464 ],\n        [ 5.5253673],\n        [ 6.9501514],\n        [ 7.5520563],\n        [ 8.712859 ],\n        [ 9.211232 ],\n        [10.957595 ],\n        [11.762633 ],\n        [12.57463  ],\n        [14.053735 ],\n        [14.588909 ],\n        [15.759621 ],\n        [16.460573 ],\n        [17.750051 ],\n        [18.570803 ],\n        [19.650068 ],\n        [21.158527 ],\n        [21.70898  ],\n        [22.93417  ],\n        [24.450125 ],\n        [25.451452 ],\n        [26.234995 ],\n        [27.071838 ],\n        [27.827911 ],\n        [28.73208  ],\n        [30.561762 ],\n        [31.128313 ],\n        [31.748518 ],\n        [33.27655  ],\n        [34.450153 ],\n        [35.30688  ],\n        [36.631584 ],\n        [36.77429  ],\n        [38.28369  ],\n        [39.234844 ],\n        [39.85827  ]], dtype=float32),\n (40, 1))"
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# reshape numpy arrays using numpy\n",
    "\n",
    "x_train = np.reshape(x_train, (-1,1))   # (40,) -> (40, 1)\n",
    "# y_train                               # (40,)\n",
    "\n",
    "x_train, x_train.shape"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-08-09T17:07:43.752608Z",
     "start_time": "2023-08-09T17:07:43.711312Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "outputs": [
    {
     "data": {
      "text/plain": "(<tf.Tensor: shape=(40, 1), dtype=float32, numpy=\n array([[ 0.5118216],\n        [ 1.9708718],\n        [ 2.1849759],\n        [ 4.009874 ],\n        [ 4.393464 ],\n        [ 5.5253673],\n        [ 6.9501514],\n        [ 7.5520563],\n        [ 8.712859 ],\n        [ 9.211232 ],\n        [10.957595 ],\n        [11.762633 ],\n        [12.57463  ],\n        [14.053735 ],\n        [14.588909 ],\n        [15.759621 ],\n        [16.460573 ],\n        [17.750051 ],\n        [18.570803 ],\n        [19.650068 ],\n        [21.158527 ],\n        [21.70898  ],\n        [22.93417  ],\n        [24.450125 ],\n        [25.451452 ],\n        [26.234995 ],\n        [27.071838 ],\n        [27.827911 ],\n        [28.73208  ],\n        [30.561762 ],\n        [31.128313 ],\n        [31.748518 ],\n        [33.27655  ],\n        [34.450153 ],\n        [35.30688  ],\n        [36.631584 ],\n        [36.77429  ],\n        [38.28369  ],\n        [39.234844 ],\n        [39.85827  ]], dtype=float32)>,\n TensorShape([40, 1]))"
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# reshape numpy arrays using tensorflow !\n",
    "\n",
    "x_train = tf.constant(x_train, dtype=tf.float32, shape=(40,1)) # (40,) -> (40, 1)\n",
    "# y_train\n",
    "\n",
    "x_train, x_train.shape"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-08-09T17:08:29.763629Z",
     "start_time": "2023-08-09T17:08:29.725504Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/50\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-07-09 20:26:49.398618: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  SSE4.1 SSE4.2 AVX AVX2 FMA\n",
      "To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2/2 [==============================] - 0s 16ms/step - loss: 3494.6772\n",
      "Epoch 2/50\n",
      "2/2 [==============================] - 0s 5ms/step - loss: 2272.4067\n",
      "Epoch 3/50\n",
      "2/2 [==============================] - 0s 5ms/step - loss: 1448.7974\n",
      "Epoch 4/50\n",
      "2/2 [==============================] - 0s 6ms/step - loss: 950.7256\n",
      "Epoch 5/50\n",
      "2/2 [==============================] - 0s 5ms/step - loss: 628.7716\n",
      "Epoch 6/50\n",
      "2/2 [==============================] - 0s 7ms/step - loss: 405.1072\n",
      "Epoch 7/50\n",
      "2/2 [==============================] - 0s 6ms/step - loss: 247.7197\n",
      "Epoch 8/50\n",
      "2/2 [==============================] - 0s 6ms/step - loss: 151.4881\n",
      "Epoch 9/50\n",
      "2/2 [==============================] - 0s 6ms/step - loss: 98.9445\n",
      "Epoch 10/50\n",
      "2/2 [==============================] - 0s 7ms/step - loss: 63.7313\n",
      "Epoch 11/50\n",
      "2/2 [==============================] - 0s 6ms/step - loss: 43.8396\n",
      "Epoch 12/50\n",
      "2/2 [==============================] - 0s 7ms/step - loss: 30.0763\n",
      "Epoch 13/50\n",
      "2/2 [==============================] - 0s 6ms/step - loss: 21.1467\n",
      "Epoch 14/50\n",
      "2/2 [==============================] - 0s 6ms/step - loss: 15.6189\n",
      "Epoch 15/50\n",
      "2/2 [==============================] - 0s 7ms/step - loss: 12.2410\n",
      "Epoch 16/50\n",
      "2/2 [==============================] - 0s 7ms/step - loss: 9.5892\n",
      "Epoch 17/50\n",
      "2/2 [==============================] - 0s 5ms/step - loss: 8.0343\n",
      "Epoch 18/50\n",
      "2/2 [==============================] - 0s 6ms/step - loss: 7.5282\n",
      "Epoch 19/50\n",
      "2/2 [==============================] - 0s 5ms/step - loss: 7.0380\n",
      "Epoch 20/50\n",
      "2/2 [==============================] - 0s 7ms/step - loss: 6.7830\n",
      "Epoch 21/50\n",
      "2/2 [==============================] - 0s 6ms/step - loss: 6.7446\n",
      "Epoch 22/50\n",
      "2/2 [==============================] - 0s 6ms/step - loss: 6.6968\n",
      "Epoch 23/50\n",
      "2/2 [==============================] - 0s 6ms/step - loss: 6.6333\n",
      "Epoch 24/50\n",
      "2/2 [==============================] - 0s 7ms/step - loss: 6.6370\n",
      "Epoch 25/50\n",
      "2/2 [==============================] - 0s 6ms/step - loss: 6.6217\n",
      "Epoch 26/50\n",
      "2/2 [==============================] - 0s 5ms/step - loss: 6.6338\n",
      "Epoch 27/50\n",
      "2/2 [==============================] - 0s 6ms/step - loss: 6.6226\n",
      "Epoch 28/50\n",
      "2/2 [==============================] - 0s 6ms/step - loss: 6.6225\n",
      "Epoch 29/50\n",
      "2/2 [==============================] - 0s 6ms/step - loss: 6.6221\n",
      "Epoch 30/50\n",
      "2/2 [==============================] - 0s 5ms/step - loss: 6.6246\n",
      "Epoch 31/50\n",
      "2/2 [==============================] - 0s 5ms/step - loss: 6.6313\n",
      "Epoch 32/50\n",
      "2/2 [==============================] - 0s 5ms/step - loss: 6.6334\n",
      "Epoch 33/50\n",
      "2/2 [==============================] - 0s 7ms/step - loss: 6.6337\n",
      "Epoch 34/50\n",
      "2/2 [==============================] - 0s 5ms/step - loss: 6.6920\n",
      "Epoch 35/50\n",
      "2/2 [==============================] - 0s 5ms/step - loss: 6.6904\n",
      "Epoch 36/50\n",
      "2/2 [==============================] - 0s 6ms/step - loss: 6.6651\n",
      "Epoch 37/50\n",
      "2/2 [==============================] - 0s 7ms/step - loss: 6.6793\n",
      "Epoch 38/50\n",
      "2/2 [==============================] - 0s 7ms/step - loss: 6.6757\n",
      "Epoch 39/50\n",
      "2/2 [==============================] - 0s 7ms/step - loss: 6.6441\n",
      "Epoch 40/50\n",
      "2/2 [==============================] - 0s 5ms/step - loss: 6.6393\n",
      "Epoch 41/50\n",
      "2/2 [==============================] - 0s 5ms/step - loss: 6.6301\n",
      "Epoch 42/50\n",
      "2/2 [==============================] - 0s 7ms/step - loss: 6.6375\n",
      "Epoch 43/50\n",
      "2/2 [==============================] - 0s 5ms/step - loss: 6.6456\n",
      "Epoch 44/50\n",
      "2/2 [==============================] - 0s 4ms/step - loss: 6.6503\n",
      "Epoch 45/50\n",
      "2/2 [==============================] - 0s 7ms/step - loss: 6.6285\n",
      "Epoch 46/50\n",
      "2/2 [==============================] - 0s 4ms/step - loss: 6.6297\n",
      "Epoch 47/50\n",
      "2/2 [==============================] - 0s 6ms/step - loss: 6.6389\n",
      "Epoch 48/50\n",
      "2/2 [==============================] - 0s 6ms/step - loss: 6.6466\n",
      "Epoch 49/50\n",
      "2/2 [==============================] - 0s 8ms/step - loss: 6.6217\n",
      "Epoch 50/50\n",
      "2/2 [==============================] - 0s 7ms/step - loss: 6.6251\n"
     ]
    }
   ],
   "source": [
    "# create a sequential model\n",
    "# ..keras.model\n",
    "# ..keras.layer\n",
    "# ..keras.optimizers\n",
    "# ..keras.losses\n",
    "\n",
    "# hypothesis\n",
    "model = tf.keras.models.Sequential()\n",
    "model.add(tf.keras.layers.Dense(1)) # output (none, 1)\n",
    "\n",
    "# optimizer # loss\n",
    "model.compile(optimizer=tf.keras.optimizers.SGD(learning_rate=0.0001),\n",
    "              loss=tf.keras.losses.mse)\n",
    "\n",
    "# train\n",
    "train = model.fit(x_train, y_train, epochs=50)"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-07-09T17:26:50.674716Z",
     "start_time": "2023-07-09T17:26:49.393278Z"
    }
   }
  },
  {
   "cell_type": "markdown",
   "source": [
    "## model - performance"
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "outputs": [
    {
     "data": {
      "text/plain": "<Figure size 300x300 with 1 Axes>",
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# performance analysis\n",
    "\n",
    "plt.figure(figsize=(3,3))\n",
    "plt.plot(train.history['loss'], label=\"loss\")\n",
    "plt.legend()\n",
    "plt.show()"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-07-09T17:26:54.758132Z",
     "start_time": "2023-07-09T17:26:54.534332Z"
    }
   }
  },
  {
   "cell_type": "markdown",
   "source": [
    "## model - prediction"
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 86ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": "<Figure size 500x300 with 1 Axes>",
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# prediction\n",
    "\n",
    "y_pred = model.predict(np.reshape(x_test, newshape=(-1,1)))  #reshape (10,) -> (10,1)\n",
    "\n",
    "plt.figure(figsize=(5,3))\n",
    "plt.plot(x_test, y_pred, color=\"red\", label=\"best fit\")\n",
    "\n",
    "plt.scatter(x_test, y_test, label= \"test data\")\n",
    "plt.legend()\n",
    "plt.show()"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-07-09T17:26:59.171988Z",
     "start_time": "2023-07-09T17:26:58.776292Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "outputs": [],
   "source": [],
   "metadata": {
    "collapsed": false
   }
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}