Machine Learning - Generating Logscales for hyperparameter Arrays for tuning

Machine Learning - Generating Logscales for hyperparameter Arrays for tuning.ipynb

{
  "cells": [
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "# Generate a Logrithmic Distribution of Values"
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "import seaborn as sns\nimport pandas as pd",
      "execution_count": null,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "import numpy as np\nimport matplotlib.pyplot as plt\n%matplotlib inline",
      "execution_count": 83,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "from jupyterthemes import jtplot\njtplot.style(theme='grade3')",
      "execution_count": 84,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "low_limit = 0.0001\nhigh_limit = 10\n",
      "execution_count": 126,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "low, high = np.log10(low_limit), np.log10(high_limit)\nalpha= low*np.random.rand(25000)+high\nlearning_rate_values = pd.Series(np.power(10,alpha), name='alpha')",
      "execution_count": 138,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "plt.figure(figsize=(20,10))\nax = sns.distplot(learning_rate_values, bins=int(np.cbrt(len(learning_rate_values))), label='Learning Rate Distribution')\nax.set_xlim(-0.1,12)\nax.legend()",
      "execution_count": 139,
      "outputs": [
        {
          "output_type": "stream",
          "text": "/home/karl/anaconda2/envs/py36-test/lib/python3.6/site-packages/matplotlib/axes/_axes.py:6448: UserWarning: The 'normed' kwarg is deprecated, and has been replaced by the 'density' kwarg.\n  warnings.warn(\"The 'normed' kwarg is deprecated, and has been \"\n",
          "name": "stderr"
        },
        {
          "output_type": "execute_result",
          "execution_count": 139,
          "data": {
            "text/plain": "<matplotlib.legend.Legend at 0x7f29ea8b7320>"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": "<Figure size 1440x720 with 1 Axes>",
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "learning_rate_values.round(1).value_counts()",
      "execution_count": 136,
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 136,
          "data": {
            "text/plain": "0.0     10624\n0.1      2988\n0.2      1420\n0.3       921\n0.4       646\n0.5       535\n0.6       456\n0.7       388\n0.8       347\n0.9       295\n1.0       271\n1.1       248\n1.2       208\n1.3       202\n1.4       198\n1.7       175\n1.5       165\n1.6       162\n1.8       147\n1.9       142\n2.4       132\n2.1       129\n2.3       125\n2.2       123\n2.0       122\n2.6       112\n2.7       101\n2.5        99\n3.6        97\n2.8        96\n        ...  \n7.6        37\n8.5        36\n8.1        36\n7.0        36\n7.3        36\n9.7        35\n8.9        34\n8.0        34\n6.0        33\n5.4        33\n8.7        32\n6.8        32\n6.9        32\n7.1        32\n9.4        30\n8.6        30\n7.9        29\n9.8        29\n7.2        28\n9.0        28\n9.5        28\n8.4        28\n8.3        27\n9.6        26\n8.2        26\n9.2        26\n9.9        24\n9.1        23\n8.8        22\n10.0       10\nName: alpha, Length: 101, dtype: int64"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "np.random.logseries(0.6, 10000)",
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 26,
          "data": {
            "text/plain": "array([2, 1, 1, ..., 1, 1, 1])"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "np.log10(-4,10)",
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "error",
          "ename": "TypeError",
          "evalue": "return arrays must be of ArrayType",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-25-6a632206975c>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlog10\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m4\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;36m10\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
            "\u001b[0;31mTypeError\u001b[0m: return arrays must be of ArrayType"
          ]
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "np.power(10,r)",
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 8,
          "data": {
            "text/plain": "array([2.44571574e-13, 2.66347093e-02, 2.37082248e-07, 4.41591824e-04,\n       1.22820873e+08, 1.52132915e-02, 1.04780605e-02, 1.94214455e+02,\n       6.90705062e-09, 2.17351475e+05, 4.20589980e-05, 1.25748414e-01,\n       5.22412651e-05, 2.40040747e+02, 2.04950976e+00, 1.55532485e+00,\n       1.77052522e+05, 8.31740050e-02, 4.23639354e-04, 1.48540077e-01,\n       2.14272188e+07, 1.41907299e+05, 2.04689491e-02, 1.40553730e-09,\n       6.99522182e+03])"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "",
      "execution_count": null,
      "outputs": []
    }
  ],
  "metadata": {
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3",
      "language": "python"
    },
    "toc": {
      "nav_menu": {},
      "number_sections": true,
      "sideBar": true,
      "skip_h1_title": false,
      "toc_cell": false,
      "toc_position": {},
      "toc_section_display": "block",
      "toc_window_display": false
    },
    "language_info": {
      "name": "python",
      "version": "3.6.4",
      "mimetype": "text/x-python",
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "pygments_lexer": "ipython3",
      "nbconvert_exporter": "python",
      "file_extension": ".py"
    },
    "gist": {
      "id": "",
      "data": {
        "description": "Machine Learning - Generating Logscales for hyperparameter Arrays for tuning",
        "public": true
      }
    }
  },
  "nbformat": 4,
  "nbformat_minor": 2
}