Tensorflow(python3.6)で一次関数を学習させ一次関数を近似させる例

requirements.txt

numpy
matplotlib
pandas
tensorflow 

tensorflow_example1.ipynb

{
  "cells": [
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "import tensorflow as tf\nimport numpy as np\nimport time\nimport random\nimport math\nimport matplotlib.pyplot as plt\n\nprint(tf.__version__)\n",
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "text": "1.12.2\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "start = time.time()\n",
      "execution_count": 2,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "x_data = np.random.rand(100).astype(\"float32\")\ny_data = 0.1 * x_data + 0.3",
      "execution_count": 3,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "w = tf.Variable(tf.random_uniform([1], -1.0, 1.0))\nb = tf.Variable(tf.zeros([1]))\ny = w * x_data + b\n",
      "execution_count": 4,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "loss = tf.reduce_mean(tf.square(y_data - y))\noptimizer = tf.train.GradientDescentOptimizer(0.5)\ntrain = optimizer.minimize(loss)\n",
      "execution_count": 5,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "init = tf.global_variables_initializer()\nsess = tf.Session()\nsess.run(init)\n",
      "execution_count": 6,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "for step in range(1001):\n    sess.run(train)\n    if step % 100 == 0:\n        print(f\"{step} {sess.run(w)}, {sess.run(b)}\")",
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "text": "0 [-0.03927329], [0.4752938]\n100 [0.09979425], [0.3001]\n200 [0.09999977], [0.30000013]\n300 [0.0999999], [0.30000007]\n400 [0.0999999], [0.30000007]\n500 [0.0999999], [0.30000007]\n600 [0.0999999], [0.30000007]\n700 [0.0999999], [0.30000007]\n800 [0.0999999], [0.30000007]\n900 [0.0999999], [0.30000007]\n1000 [0.0999999], [0.30000007]\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "plot_x = []\nplot_y = []\nfor i in range(100):\n    result = sess.run(w) * i + sess.run(b)\n    plot_x.append(i)\n    plot_y.append(result)\n    print(f\"{i} {result}\")",
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "stream",
          "text": "0 [0.30000007]\n1 [0.39999998]\n2 [0.49999988]\n3 [0.5999998]\n4 [0.6999997]\n5 [0.7999996]\n6 [0.8999995]\n7 [0.9999994]\n8 [1.0999993]\n9 [1.1999992]\n10 [1.2999991]\n11 [1.399999]\n12 [1.4999989]\n13 [1.5999988]\n14 [1.6999987]\n15 [1.7999986]\n16 [1.8999985]\n17 [1.9999985]\n18 [2.0999985]\n19 [2.1999984]\n20 [2.2999983]\n21 [2.3999982]\n22 [2.499998]\n23 [2.599998]\n24 [2.699998]\n25 [2.7999978]\n26 [2.8999977]\n27 [2.9999976]\n28 [3.0999975]\n29 [3.1999974]\n30 [3.2999973]\n31 [3.3999972]\n32 [3.4999971]\n33 [3.599997]\n34 [3.699997]\n35 [3.7999969]\n36 [3.8999968]\n37 [3.9999967]\n38 [4.0999966]\n39 [4.1999965]\n40 [4.2999964]\n41 [4.3999963]\n42 [4.499996]\n43 [4.599996]\n44 [4.699996]\n45 [4.799996]\n46 [4.899996]\n47 [4.9999957]\n48 [5.0999956]\n49 [5.1999955]\n50 [5.2999954]\n51 [5.3999953]\n52 [5.499995]\n53 [5.599995]\n54 [5.699995]\n55 [5.799995]\n56 [5.899995]\n57 [5.9999948]\n58 [6.0999947]\n59 [6.1999946]\n60 [6.2999945]\n61 [6.3999944]\n62 [6.4999943]\n63 [6.599994]\n64 [6.699994]\n65 [6.799994]\n66 [6.899994]\n67 [6.999994]\n68 [7.0999937]\n69 [7.1999936]\n70 [7.2999935]\n71 [7.3999934]\n72 [7.4999933]\n73 [7.599993]\n74 [7.699993]\n75 [7.799993]\n76 [7.899993]\n77 [7.999993]\n78 [8.099993]\n79 [8.199992]\n80 [8.299993]\n81 [8.399993]\n82 [8.499992]\n83 [8.599992]\n84 [8.699992]\n85 [8.799993]\n86 [8.899992]\n87 [8.999991]\n88 [9.099992]\n89 [9.199992]\n90 [9.299992]\n91 [9.399991]\n92 [9.499991]\n93 [9.599992]\n94 [9.699991]\n95 [9.799991]\n96 [9.899991]\n97 [9.999991]\n98 [10.099991]\n99 [10.19999]\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "sess.close()",
      "execution_count": 9,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "plt.scatter(plot_x, plot_y)\nplt.show()",
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": "<Figure size 432x288 with 1 Axes>"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "timer = time.time() - start\nprint(f\"time:{timer}sec\")",
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "stream",
          "text": "time:6.7154457569122314sec\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "",
      "execution_count": null,
      "outputs": []
    }
  ],
  "metadata": {
    "kernelspec": {
      "name": "python36",
      "display_name": "Python 3.6",
      "language": "python"
    },
    "language_info": {
      "mimetype": "text/x-python",
      "nbconvert_exporter": "python",
      "name": "python",
      "pygments_lexer": "ipython3",
      "version": "3.6.6",
      "file_extension": ".py",
      "codemirror_mode": {
        "version": 3,
        "name": "ipython"
      }
    }
  },
  "nbformat": 4,
  "nbformat_minor": 2
}