lab-02-1-Simple-Linear-Regression-eager.ipynb

{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"name":"lab-02-1-Simple-Liner-Regression-eager.ipynb","version":"0.3.2","provenance":[],"collapsed_sections":[]},"kernelspec":{"display_name":"Python 3","language":"python","name":"python3"}},"cells":[{"metadata":{"colab_type":"text","id":"T-dtOIu5FUV3","slideshow":{"slide_type":"slide"}},"cell_type":"markdown","source":["# 02 - Simple Liner Regression\n","\n","<img width=\"200\" src=\"https://i.imgur.com/hbPVe1T.png\">\n"]},{"metadata":{"colab_type":"code","id":"dk3ggjvmFUV5","slideshow":{"slide_type":"slide"},"colab":{}},"cell_type":"code","source":["import tensorflow as tf\n","import numpy as np\n","\n","tf.enable_eager_execution()"],"execution_count":0,"outputs":[]},{"metadata":{"colab_type":"text","collapsed":true,"id":"RFeXSskkFUV-","slideshow":{"slide_type":"slide"}},"cell_type":"markdown","source":["# Hypothesis and Cost\n","### Hypothesis \n","$$ H(x) = Wx + b $$\n","\n","\n","### Cost\n","$$ cost(W)=\\frac { 1 }{ m } \\sum _{i=1}^{m}{ { (W{ x }^{ i }-y^{ i } })^{ 2 } }  $$\n"]},{"metadata":{"colab_type":"code","id":"pvennmBIFUV_","slideshow":{"slide_type":"slide"},"colab":{}},"cell_type":"code","source":["x_data = [1, 2, 3, 4, 5]\n","y_data = [1, 2, 3, 4, 5]"],"execution_count":0,"outputs":[]},{"metadata":{"colab_type":"code","executionInfo":{"status":"ok","timestamp":1549783673077,"user_tz":-540,"elapsed":2167,"user":{"displayName":"Sean-June Lee","photoUrl":"https://lh3.googleusercontent.com/-j96Pvw9LbEw/AAAAAAAAAAI/AAAAAAACQxw/aY8i4PoItwc/s64/photo.jpg","userId":"12576390002043678893"}},"id":"0gJ136qSGiYm","outputId":"5eb3b45f-55df-4d8d-ded7-b3baf3f264b1","colab":{"base_uri":"https://localhost:8080/","height":369}},"cell_type":"code","source":["import matplotlib.pyplot as plt\n","plt.plot(x_data, y_data, 'o')\n","plt.ylim(0, 8)"],"execution_count":3,"outputs":[{"output_type":"execute_result","data":{"text/plain":["(0, 8)"]},"metadata":{"tags":[]},"execution_count":3},{"output_type":"display_data","data":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAAdkAAAFOCAYAAADKJGf5AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAElFTkSuQmCC\n","text/plain":["<Figure size 576x396 with 1 Axes>"]},"metadata":{"tags":[]}}]},{"metadata":{"colab_type":"text","id":"kJIo1YInVobJ"},"cell_type":"markdown","source":["## Hypothesis"]},{"metadata":{"id":"4F5aR9EEDaAs","colab_type":"code","colab":{"base_uri":"https://localhost:8080/","height":34},"outputId":"8bd52e07-f337-4b3d-c837-1a7bdc1a2760","executionInfo":{"status":"ok","timestamp":1549783673078,"user_tz":-540,"elapsed":2160,"user":{"displayName":"Sean-June Lee","photoUrl":"https://lh3.googleusercontent.com/-j96Pvw9LbEw/AAAAAAAAAAI/AAAAAAACQxw/aY8i4PoItwc/s64/photo.jpg","userId":"12576390002043678893"}}},"cell_type":"code","source":["v =[1., 2., 3., 4.]\n","tf.reduce_mean(v) # 2.5"],"execution_count":4,"outputs":[{"output_type":"execute_result","data":{"text/plain":["<tf.Tensor: id=6, shape=(), dtype=float32, numpy=2.5>"]},"metadata":{"tags":[]},"execution_count":4}]},{"metadata":{"id":"ci7fI9DdFcLh","colab_type":"code","colab":{"base_uri":"https://localhost:8080/","height":34},"outputId":"116b747d-9d10-4ae0-f1c3-069f0fd58186","executionInfo":{"status":"ok","timestamp":1549783673079,"user_tz":-540,"elapsed":2142,"user":{"displayName":"Sean-June Lee","photoUrl":"https://lh3.googleusercontent.com/-j96Pvw9LbEw/AAAAAAAAAAI/AAAAAAACQxw/aY8i4PoItwc/s64/photo.jpg","userId":"12576390002043678893"}}},"cell_type":"code","source":["tf.square(3) # 9"],"execution_count":5,"outputs":[{"output_type":"execute_result","data":{"text/plain":["<tf.Tensor: id=9, shape=(), dtype=int32, numpy=9>"]},"metadata":{"tags":[]},"execution_count":5}]},{"metadata":{"id":"vGg-WsdlAlNr","colab_type":"code","colab":{"base_uri":"https://localhost:8080/","height":90},"outputId":"c5eda8e8-9a4f-4713-a5c2-7c73998c6154","executionInfo":{"status":"ok","timestamp":1549783673080,"user_tz":-540,"elapsed":2135,"user":{"displayName":"Sean-June Lee","photoUrl":"https://lh3.googleusercontent.com/-j96Pvw9LbEw/AAAAAAAAAAI/AAAAAAACQxw/aY8i4PoItwc/s64/photo.jpg","userId":"12576390002043678893"}}},"cell_type":"code","source":["x_data = [1, 2, 3, 4, 5]\n","y_data = [1, 2, 3, 4, 5]\n","\n","W = tf.Variable(2.0)\n","b = tf.Variable(0.5)\n","\n","hypothesis = W * x_data + b"],"execution_count":6,"outputs":[{"output_type":"stream","text":["WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow/python/ops/resource_variable_ops.py:642: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.\n","Instructions for updating:\n","Colocations handled automatically by placer.\n"],"name":"stdout"}]},{"metadata":{"colab_type":"code","executionInfo":{"status":"ok","timestamp":1549783673081,"user_tz":-540,"elapsed":2131,"user":{"displayName":"Sean-June Lee","photoUrl":"https://lh3.googleusercontent.com/-j96Pvw9LbEw/AAAAAAAAAAI/AAAAAAACQxw/aY8i4PoItwc/s64/photo.jpg","userId":"12576390002043678893"}},"id":"f4AS7tpCHsXE","outputId":"919b261a-a193-476e-a742-04a3bf4646ce","colab":{"base_uri":"https://localhost:8080/","height":34}},"cell_type":"code","source":["W.numpy(), b.numpy()"],"execution_count":7,"outputs":[{"output_type":"execute_result","data":{"text/plain":["(2.0, 0.5)"]},"metadata":{"tags":[]},"execution_count":7}]},{"metadata":{"colab_type":"code","executionInfo":{"status":"ok","timestamp":1549783673082,"user_tz":-540,"elapsed":2125,"user":{"displayName":"Sean-June Lee","photoUrl":"https://lh3.googleusercontent.com/-j96Pvw9LbEw/AAAAAAAAAAI/AAAAAAACQxw/aY8i4PoItwc/s64/photo.jpg","userId":"12576390002043678893"}},"id":"5-9nQyH5IjJA","outputId":"36d8d098-a333-45c4-8d0a-e2b797837a96","colab":{"base_uri":"https://localhost:8080/","height":34}},"cell_type":"code","source":["hypothesis.numpy()"],"execution_count":8,"outputs":[{"output_type":"execute_result","data":{"text/plain":["array([ 2.5,  4.5,  6.5,  8.5, 10.5], dtype=float32)"]},"metadata":{"tags":[]},"execution_count":8}]},{"metadata":{"colab_type":"code","executionInfo":{"status":"ok","timestamp":1549783673347,"user_tz":-540,"elapsed":2375,"user":{"displayName":"Sean-June Lee","photoUrl":"https://lh3.googleusercontent.com/-j96Pvw9LbEw/AAAAAAAAAAI/AAAAAAACQxw/aY8i4PoItwc/s64/photo.jpg","userId":"12576390002043678893"}},"id":"5CK8K2HXXdB7","outputId":"c6ea512e-bbc6-4576-e0c6-af6b047f8a49","colab":{"base_uri":"https://localhost:8080/","height":351}},"cell_type":"code","source":["plt.plot(x_data, hypothesis.numpy(), 'r-')\n","plt.plot(x_data, y_data, 'o')\n","plt.ylim(0, 8)\n","plt.show()"],"execution_count":9,"outputs":[{"output_type":"display_data","data":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAAdkAAAFOCAYAAADKJGf5AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n3QAAAABJRU5ErkJggg==\n","text/plain":["<Figure size 576x396 with 1 Axes>"]},"metadata":{"tags":[]}}]},{"metadata":{"colab_type":"text","id":"wa9avNpfVdDD"},"cell_type":"markdown","source":["## Cost"]},{"metadata":{"colab_type":"code","id":"ZkOrmjTjKMJ7","colab":{}},"cell_type":"code","source":["cost = tf.reduce_mean(tf.square(hypothesis - y_data))"],"execution_count":0,"outputs":[]},{"metadata":{"colab_type":"code","executionInfo":{"status":"ok","timestamp":1549783673351,"user_tz":-540,"elapsed":2364,"user":{"displayName":"Sean-June Lee","photoUrl":"https://lh3.googleusercontent.com/-j96Pvw9LbEw/AAAAAAAAAAI/AAAAAAACQxw/aY8i4PoItwc/s64/photo.jpg","userId":"12576390002043678893"}},"id":"n00UWqSAKXHU","outputId":"a4191026-70c0-4e79-dff8-57ca14cd98f3","colab":{"base_uri":"https://localhost:8080/","height":34}},"cell_type":"code","source":["with tf.GradientTape() as tape:\n","    hypothesis = W * x_data + b\n","    cost = tf.reduce_mean(tf.square(hypothesis - y_data))\n","\n","W_grad, b_grad = tape.gradient(cost, [W, b])\n","W_grad.numpy(), b_grad.numpy()"],"execution_count":11,"outputs":[{"output_type":"execute_result","data":{"text/plain":["(25.0, 7.0)"]},"metadata":{"tags":[]},"execution_count":11}]},{"metadata":{"colab_type":"text","id":"1JhoCK8yLk_X"},"cell_type":"markdown","source":["## 파라미터 업데이트"]},{"metadata":{"colab_type":"code","executionInfo":{"status":"ok","timestamp":1549783673352,"user_tz":-540,"elapsed":2355,"user":{"displayName":"Sean-June Lee","photoUrl":"https://lh3.googleusercontent.com/-j96Pvw9LbEw/AAAAAAAAAAI/AAAAAAACQxw/aY8i4PoItwc/s64/photo.jpg","userId":"12576390002043678893"}},"id":"bEOlCDAoLkw0","outputId":"3e04ee39-a263-469e-87b9-53f4dd9980a1","colab":{"base_uri":"https://localhost:8080/","height":34}},"cell_type":"code","source":["learning_rate = 0.01\n","\n","W.assign_sub(learning_rate * W_grad)\n","b.assign_sub(learning_rate * b_grad)\n","\n","W.numpy(), b.numpy()"],"execution_count":12,"outputs":[{"output_type":"execute_result","data":{"text/plain":["(1.75, 0.43)"]},"metadata":{"tags":[]},"execution_count":12}]},{"metadata":{"colab_type":"code","executionInfo":{"status":"ok","timestamp":1549783673662,"user_tz":-540,"elapsed":2660,"user":{"displayName":"Sean-June Lee","photoUrl":"https://lh3.googleusercontent.com/-j96Pvw9LbEw/AAAAAAAAAAI/AAAAAAACQxw/aY8i4PoItwc/s64/photo.jpg","userId":"12576390002043678893"}},"id":"sYxk3o8VLUmJ","outputId":"c3f06239-6645-4b87-c600-267e89588d0c","colab":{"base_uri":"https://localhost:8080/","height":369}},"cell_type":"code","source":["plt.plot(x_data, hypothesis.numpy(), 'r-')\n","plt.plot(x_data, y_data, 'o')\n","plt.ylim(0, 8)"],"execution_count":13,"outputs":[{"output_type":"execute_result","data":{"text/plain":["(0, 8)"]},"metadata":{"tags":[]},"execution_count":13},{"output_type":"display_data","data":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAAdkAAAFOCAYAAADKJGf5AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n3QAAAABJRU5ErkJggg==\n","text/plain":["<Figure size 576x396 with 1 Axes>"]},"metadata":{"tags":[]}}]},{"metadata":{"colab_type":"text","id":"DIwrKgj-P-Id"},"cell_type":"markdown","source":["## 여러 번 반복"]},{"metadata":{"colab_type":"code","executionInfo":{"status":"ok","timestamp":1549783674437,"user_tz":-540,"elapsed":3427,"user":{"displayName":"Sean-June Lee","photoUrl":"https://lh3.googleusercontent.com/-j96Pvw9LbEw/AAAAAAAAAAI/AAAAAAACQxw/aY8i4PoItwc/s64/photo.jpg","userId":"12576390002043678893"}},"id":"iMia1ENFLUeJ","outputId":"0aa42d35-0cda-4e40-a40d-85cd6dd2547c","colab":{"base_uri":"https://localhost:8080/","height":545}},"cell_type":"code","source":["W = tf.Variable(2.9)\n","b = tf.Variable(0.5)\n","\n","for i in range(100):\n","    with tf.GradientTape() as tape:\n","        hypothesis = W * x_data + b\n","        cost = tf.reduce_mean(tf.square(hypothesis - y_data))\n","    W_grad, b_grad = tape.gradient(cost, [W, b])\n","    W.assign_sub(learning_rate * W_grad)\n","    b.assign_sub(learning_rate * b_grad)\n","    if i % 10 == 0:\n","      print(\"{:5}|{:10.4f}|{:10.4f}|{:10.6f}\".format(i, W.numpy(), b.numpy(), cost))\n","\n","plt.plot(x_data, y_data, 'o')\n","plt.plot(x_data, hypothesis.numpy(), 'r-')\n","plt.ylim(0, 8)"],"execution_count":14,"outputs":[{"output_type":"stream","text":["    0|    2.4520|    0.3760| 45.660004\n","   10|    1.1036|    0.0034|  0.206336\n","   20|    1.0128|   -0.0209|  0.001026\n","   30|    1.0065|   -0.0218|  0.000093\n","   40|    1.0059|   -0.0212|  0.000083\n","   50|    1.0057|   -0.0205|  0.000077\n","   60|    1.0055|   -0.0198|  0.000072\n","   70|    1.0053|   -0.0192|  0.000067\n","   80|    1.0051|   -0.0185|  0.000063\n","   90|    1.0050|   -0.0179|  0.000059\n"],"name":"stdout"},{"output_type":"execute_result","data":{"text/plain":["(0, 8)"]},"metadata":{"tags":[]},"execution_count":14},{"output_type":"display_data","data":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAAdkAAAFOCAYAAADKJGf5AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nBQDAIoQsAAAWIWQBALDI/wcb52eBRk/VFQAAAABJRU5ErkJggg==\n","text/plain":["<Figure size 576x396 with 1 Axes>"]},"metadata":{"tags":[]}}]},{"metadata":{"colab_type":"text","id":"6upnkrWZFUWF","slideshow":{"slide_type":"slide"}},"cell_type":"markdown","source":["# predict"]},{"metadata":{"colab_type":"code","executionInfo":{"status":"ok","timestamp":1549783674450,"user_tz":-540,"elapsed":3428,"user":{"displayName":"Sean-June Lee","photoUrl":"https://lh3.googleusercontent.com/-j96Pvw9LbEw/AAAAAAAAAAI/AAAAAAACQxw/aY8i4PoItwc/s64/photo.jpg","userId":"12576390002043678893"}},"id":"WnqG2Dn7LUNy","outputId":"577f4c5d-296d-497d-b444-651063be1d27","colab":{"base_uri":"https://localhost:8080/","height":52}},"cell_type":"code","source":["print(W * 5 + b)\n","print(W * 2.5 + b)"],"execution_count":15,"outputs":[{"output_type":"stream","text":["tf.Tensor(5.0066934, shape=(), dtype=float32)\n","tf.Tensor(2.4946523, shape=(), dtype=float32)\n"],"name":"stdout"}]},{"metadata":{"colab_type":"text","id":"MY4hGtbL7E5e"},"cell_type":"markdown","source":["## 전체 코드"]},{"metadata":{"colab_type":"code","executionInfo":{"status":"ok","timestamp":1549783674451,"user_tz":-540,"elapsed":3423,"user":{"displayName":"Sean-June Lee","photoUrl":"https://lh3.googleusercontent.com/-j96Pvw9LbEw/AAAAAAAAAAI/AAAAAAACQxw/aY8i4PoItwc/s64/photo.jpg","userId":"12576390002043678893"}},"id":"ZnPZBI0q7D_P","outputId":"fba7b647-97c6-402f-b22d-ef925014af6b","colab":{"base_uri":"https://localhost:8080/","height":246}},"cell_type":"code","source":["import tensorflow as tf\n","import numpy as np\n","tf.enable_eager_execution()\n","\n","# Data\n","x_data = [1, 2, 3, 4, 5]\n","y_data = [1, 2, 3, 4, 5]\n","\n","# W, b initialize\n","W = tf.Variable(2.9)\n","b = tf.Variable(0.5)\n","\n","# W, b update\n","for i in range(100):\n","    # Gradient descent\n","    with tf.GradientTape() as tape:\n","        hypothesis = W * x_data + b\n","        cost = tf.reduce_mean(tf.square(hypothesis - y_data))\n","    W_grad, b_grad = tape.gradient(cost, [W, b])\n","    W.assign_sub(learning_rate * W_grad)\n","    b.assign_sub(learning_rate * b_grad)\n","    if i % 10 == 0:\n","      print(\"{:5}|{:10.4f}|{:10.4f}|{:10.6f}\".format(i, W.numpy(), b.numpy(), cost))\n","\n","print()\n","\n","# predict\n","print(W * 5 + b)\n","print(W * 2.5 + b)"],"execution_count":16,"outputs":[{"output_type":"stream","text":["    0|    2.4520|    0.3760| 45.660004\n","   10|    1.1036|    0.0034|  0.206336\n","   20|    1.0128|   -0.0209|  0.001026\n","   30|    1.0065|   -0.0218|  0.000093\n","   40|    1.0059|   -0.0212|  0.000083\n","   50|    1.0057|   -0.0205|  0.000077\n","   60|    1.0055|   -0.0198|  0.000072\n","   70|    1.0053|   -0.0192|  0.000067\n","   80|    1.0051|   -0.0185|  0.000063\n","   90|    1.0050|   -0.0179|  0.000059\n","\n","tf.Tensor(5.0066934, shape=(), dtype=float32)\n","tf.Tensor(2.4946523, shape=(), dtype=float32)\n"],"name":"stdout"}]}]}