{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "##### 교재 페이지 (147-148) #####" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "6-1) Hypothesis using matrix" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "WARNING:tensorflow:From C:\\Users\\jsdata00010\\Anaconda3\\lib\\site-packages\\tensorflow\\python\\framework\\op_def_library.py:263: 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", "WARNING:tensorflow:From C:\\Users\\jsdata00010\\Anaconda3\\lib\\site-packages\\tensorflow\\python\\ops\\math_ops.py:3066: to_int32 (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.\n", "Instructions for updating:\n", "Use tf.cast instead.\n", "0 Cost: 348.10272 \n", "Prediction:\n", " [12.008278 21.277145 23.861647 30.65607 32.11578 55.47819 55.82252\n", " 45.47918 76.14809 ]\n", "10000 Cost: 10.863274 \n", "Prediction:\n", " [34.355064 42.821743 44.299553 49.76662 51.54307 70.2765 71.43204\n", " 62.962887 87.31517 ]\n", "20000 Cost: 9.20147 \n", "Prediction:\n", " [34.03128 43.539833 43.854923 49.818493 51.660923 70.393845 71.308205\n", " 62.93976 87.25981 ]\n", "30000 Cost: 7.884106 \n", "Prediction:\n", " [33.738117 44.17571 43.455902 49.862713 51.763973 70.50015 71.19954\n", " 62.91919 87.21547 ]\n", "40000 Cost: 6.839821 \n", "Prediction:\n", " [33.477085 44.74183 43.10065 49.9021 51.855736 70.594864 71.10286\n", " 62.90091 87.1761 ]\n", "50000 Cost: 6.0119457 \n", "Prediction:\n", " [33.244686 45.245903 42.784374 49.937195 51.93747 70.67925 71.01685\n", " 62.8847 87.141136]\n", "60000 Cost: 5.355751 \n", "Prediction:\n", " [33.037697 45.694523 42.50272 49.968327 52.010105 70.75424 70.94015\n", " 62.87013 87.10986 ]\n", "70000 Cost: 4.8354807 \n", "Prediction:\n", " [32.85362 46.09419 42.252094 49.996197 52.074947 70.82106 70.871895\n", " 62.857265 87.08196 ]\n", "80000 Cost: 4.4230633 \n", "Prediction:\n", " [32.68971 46.44997 42.02896 50.020985 52.13265 70.880554 70.81116\n", " 62.845802 87.05717 ]\n", "90000 Cost: 4.096322 \n", "Prediction:\n", " [32.5437 46.766384 41.83029 50.042934 52.18388 70.93344 70.75706\n", " 62.83553 87.035095]\n", "100000 Cost: 3.8372488 \n", "Prediction:\n", " [32.41334 47.047874 41.653217 50.062393 52.229424 70.98083 70.70916\n", " 62.826515 87.01607 ]\n", "110000 Cost: 3.6318321 \n", "Prediction:\n", " [32.29786 47.29892 41.495914 50.07994 52.270187 71.02275 70.666306\n", " 62.81845 86.99846 ]\n", "120000 Cost: 3.4686704 \n", "Prediction:\n", " [32.194767 47.52278 41.355534 50.095535 52.306496 71.060135 70.628044\n", " 62.811222 86.98277 ]\n", "130000 Cost: 3.3394291 \n", "Prediction:\n", " [32.103485 47.72227 41.23095 50.109615 52.338993 71.0933 70.593925\n", " 62.80484 86.96845 ]\n", "140000 Cost: 3.2370698 \n", "Prediction:\n", " [32.02142 47.89902 41.119556 50.12174 52.36751 71.12302 70.56379\n", " 62.7991 86.9565 ]\n", "150000 Cost: 3.1559892 \n", "Prediction:\n", " [31.948858 48.05656 41.020782 50.132732 52.393085 71.149376 70.53692\n", " 62.794056 86.94553 ]\n", "160000 Cost: 3.0917628 \n", "Prediction:\n", " [31.884005 48.19635 40.932697 50.14229 52.41561 71.17277 70.51298\n", " 62.78945 86.93589 ]\n", "170000 Cost: 3.0409749 \n", "Prediction:\n", " [31.827349 48.321095 40.855156 50.151234 52.436047 71.19342 70.49167\n", " 62.785526 86.92674 ]\n", "180000 Cost: 3.0006268 \n", "Prediction:\n", " [31.776636 48.432117 40.78591 50.1591 52.454166 71.21186 70.4727\n", " 62.782005 86.91875 ]\n", "190000 Cost: 2.9685714 \n", "Prediction:\n", " [31.73064 48.530464 40.72365 50.165756 52.46997 71.22854 70.45601\n", " 62.77882 86.91236 ]\n", "200000 Cost: 2.9430833 \n", "Prediction:\n", " [31.688828 48.617638 40.6676 50.17134 52.483734 71.24361 70.44127\n", " 62.775925 86.907295]\n" ] } ], "source": [ "import tensorflow as tf\n", "x1_data = [19., 23., 26., 29., 30., 43., 44., 38., 55.]\n", "x2_data = [16.6, 18.4, 17.1, 17.7, 18.2, 18.8, 19.2, 18.9, 19.5]\n", "x3_data = [13.,10., 12., 10., 10., 7., 8., 9., 5.]\n", "y_data = [33., 51., 40., 49., 50., 69., 70., 64., 89.]\n", "# placeholders for a tensor that will be always fed.\n", "x1 = tf.placeholder(tf.float32)\n", "x2 = tf.placeholder(tf.float32)\n", "x3 = tf.placeholder(tf.float32)\n", "Y = tf.placeholder(tf.float32)\n", "\n", "w1 = tf.Variable(tf.random_normal([1]), name='weight1')\n", "w2 = tf.Variable(tf.random_normal([1]), name='weight2')\n", "w3 = tf.Variable(tf.random_normal([1]), name='weight3')\n", "b = tf.Variable(tf.random_normal([1]), name='bias')\n", "hypothesis = x1 * w1 + x2 * w2 + x3 * w3 + b\n", "\n", "# cost/loss function\n", "cost = tf.reduce_mean(tf.square(hypothesis - Y))\n", "\n", "# Minimize. Need a very small learning rate for this data set\n", "optimizer = tf.train.GradientDescentOptimizer(learning_rate=1e-5)\n", "train = optimizer.minimize(cost)\n", "\n", "# Launch the graph in a session.\n", "sess = tf.Session()\n", "\n", "# Initializes global variables in the graph.\n", "sess.run(tf.global_variables_initializer())\n", "for step in range(200001):\n", " cost_val, hy_val, _ = sess.run([cost, hypothesis, train],\n", " feed_dict={x1: x1_data, x2: x2_data, x3: x3_data, Y: y_data})\n", " if step % 10000 == 0:\n", " print(step, \"Cost: \", cost_val, \"\\nPrediction:\\n\", hy_val)" ] } ], "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": "3.7.3" } }, "nbformat": 4, "nbformat_minor": 2 }