lab-03-1-minimizing_cost_show_graph.ipynb

{
    "nbformat_minor": 2, 
    "cells": [
        {
            "source": "---\n* *** \uc6d0\ubcf8 \uc18c\uc2a4\ub294 \ud64d\ucf69\uacfc\uae30\ub300 \uae40\uc131\ud6c8 \uad50\uc218\ub2d8\uc758  [Github link](https://github.com/hunkim/DeepLearningZeroToAll)\ub97c \ucc38\uc870\ud558\uc138\uc694 ***     \n* *** Watson studio notebook \uc5d0\uc11c \uc791\uc5c5 \uac00\ub2a5\ud558\ub3c4\ub85d \uc218\uc815\ud558\uc600\uc73c\uba70***\n* *** \ud55c\uae00\ub85c \uc124\uba85\ub41c \ubd80\ubd84\uc740 \uc81c\uac00 \uc2a4\ud130\ub514 \ud558\uba74\uc11c \uc774\ud574\ud55c \ub0b4\uc6a9\uc744 \ucd94\uac00\ud55c \uac83\uc785\ub2c8\ub2e4. ***\n---", 
            "cell_type": "markdown", 
            "metadata": {}
        }, 
        {
            "source": "## Lab-03-1 Minimizing Cost show graph\n* #### \uc55e\uc11c Lab \uc5d0\uc11c \uc598\uae30\ud55c cost \uac12\uc744 \ucd5c\uc18c\ud654 \ud55c\ub2e4\ub294 \uc758\ubbf8\uac00 \ubb54\uac00\ub97c \uadf8\ub798\ud504\ub85c \uc124\uba85\ud569\ub2c8\ub2e4.\n* #### \ucd5c\uc885\uc801\uc73c\ub85c \uadf8\ub798\ud504\ub85c\ub97c \ubcf4\uc2dc\uba74 2\ucc28\uc6d0 \ud568\uc218 \uadf8\ub798\ud504\uac00 \ub418\ub294 \ub370 `(X-Y)**2` \uc640 \uac19\uc774 \uc81c\uacf1\uc744 \ud588\uae30 \ub54c\ubb38\uc774\uace0, \n* #### 0 (\ucd5c\uc18c\ud654) \uc73c\ub85c \uc218\ub834\ud558\uac8c learning \uc744 \uc218\ud589\ud558\ub294 \uac83\uc785\ub2c8\ub2e4.\n-------", 
            "cell_type": "markdown", 
            "metadata": {}
        }, 
        {
            "source": "* ***\uc55e\uc5d0 Lab \ub0b4\uc6a9\ud558\uace0 \ub3d9\uc77c\ud569\ub2c8\ub2e4.***", 
            "cell_type": "markdown", 
            "metadata": {}
        }, 
        {
            "execution_count": 1, 
            "cell_type": "code", 
            "metadata": {}, 
            "outputs": [], 
            "source": "import tensorflow as tf\nimport matplotlib.pyplot as plt\ntf.set_random_seed(777)  # for reproducibility  ==> \uc815\ud655\ud788 \uc5b4\ub5a4 \uc758\ubbf8 \uc778\uc9c0 ? "
        }, 
        {
            "execution_count": 2, 
            "cell_type": "code", 
            "metadata": {}, 
            "outputs": [], 
            "source": "X = [1, 2, 3]\nY = [1, 2, 3]"
        }, 
        {
            "execution_count": 3, 
            "cell_type": "code", 
            "metadata": {}, 
            "outputs": [], 
            "source": "W = tf.placeholder(tf.float32)"
        }, 
        {
            "execution_count": 4, 
            "cell_type": "code", 
            "metadata": {}, 
            "outputs": [], 
            "source": "# Our hypothesis for linear model X * W\nhypothesis = X * W\n\n# cost/loss function\ncost = tf.reduce_mean(tf.square(hypothesis - Y))"
        }, 
        {
            "execution_count": 5, 
            "cell_type": "code", 
            "metadata": {}, 
            "outputs": [], 
            "source": "# Launch the graph in a session.\nsess = tf.Session()\n# Initializes global variables in the graph.\nsess.run(tf.global_variables_initializer())"
        }, 
        {
            "source": "* ***W\uac12\uc758 \ubcc0\ud654\uc5d0 \ub530\ub77c  Cost \uac12\uc744 \uadf8\ub798\ud504\ub85c \ud45c\ud604\ud558\uae30 \uc704\ud574\uc11c history \uc5d0 Learning \ud560\ub54c \ub9c8\ub2e4\uc758 \ub370\uc774\ud130\ub97c \uc313\uc544\ub193\uc2b5\ub2c8\ub2e4.***", 
            "cell_type": "markdown", 
            "metadata": {}
        }, 
        {
            "execution_count": 6, 
            "cell_type": "code", 
            "metadata": {}, 
            "outputs": [], 
            "source": "# Variables for plotting cost function\nW_history = []\ncost_history = []\n\nfor i in range(-30, 50):\n    curr_W = i * 0.1\n    curr_cost = sess.run(cost, feed_dict={W: curr_W})\n    W_history.append(curr_W)\n    cost_history.append(curr_cost)"
        }, 
        {
            "source": "## Show graph", 
            "cell_type": "markdown", 
            "metadata": {}
        }, 
        {
            "execution_count": 7, 
            "cell_type": "code", 
            "metadata": {}, 
            "outputs": [
                {
                    "output_type": "display_data", 
                    "data": {
                        "image/png": "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\n", 
                        "text/plain": "<matplotlib.figure.Figure at 0x2af7fd280748>"
                    }, 
                    "metadata": {}
                }
            ], 
            "source": "# Show the cost function\nplt.plot(W_history, cost_history)\nplt.show()"
        }, 
        {
            "source": "* ***\uadf8\ub798\ud504\uc5d0 \ub300\ud55c \ud574\uc11d\uc744 \ud574\ubcf4\uba74***\n* ***cost \uac12\uc744 \ucd5c\uc18c\ud654 \ud558\ub294 W\uac12\uc740 \uacb0\uad6d 1 \uc774\ub77c\ub294 \uacb0\ub860 \uc785\ub2c8\ub2e4.***\n* ***\uc989 prediction \uc744 \uc704\ud55c \ud568\uc218 (hypothesis) \ub294 \uae30\uc6b8\uae30 1\uc778   1\ucc28 \ud568\uc218\uac00 \ub429\ub2c8\ub2e4. ***", 
            "cell_type": "markdown", 
            "metadata": {}
        }
    ], 
    "metadata": {
        "kernelspec": {
            "display_name": "Python 3.5", 
            "name": "python3", 
            "language": "python"
        }, 
        "language_info": {
            "mimetype": "text/x-python", 
            "nbconvert_exporter": "python", 
            "version": "3.5.4", 
            "name": "python", 
            "file_extension": ".py", 
            "pygments_lexer": "ipython3", 
            "codemirror_mode": {
                "version": 3, 
                "name": "ipython"
            }
        }
    }, 
    "nbformat": 4
}