warenlg/training_stats.ipynb

## training_stats.ipynb
{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Populating the interactive namespace from numpy and matplotlib\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "%pylab inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "columns = [\"repository\", \"files_count\", \"lines_count\", \"features_shape\", \"rules_count\",\n",
    "           \"rules_av_length\", \"training_time (sec)\", \"model_size (kB)\"]\n",
    "df = pd.DataFrame(columns=columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.loc[len(df)] = {\"repository\":\"freeCodeCamp/freeCodeCamp\",\n",
    "                   \"files_count\":200,\n",
    "                   \"lines_count\":14209,\n",
    "                   \"features_shape\":(48788, 449),\n",
    "                   \"rules_count\":45,\n",
    "                   \"rules_av_length\":6.3,\n",
    "                   \"training_time (sec)\":1537,\n",
    "                   \"model_size (kB)\":5.1} "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.loc[len(df)] = {\"repository\":\"jquery/jquery\",\n",
    "                   \"files_count\":173,\n",
    "                   \"lines_count\":55384,\n",
    "                   \"features_shape\":(247249, 446),\n",
    "                   \"rules_count\":835,\n",
    "                   \"rules_av_length\":7.4,\n",
    "                   \"training_time (sec)\":1121,\n",
    "                   \"model_size (kB)\":15.3} "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.loc[len(df)] = {\"repository\":\"facebook/react\",\n",
    "                   \"files_count\":696,\n",
    "                   \"lines_count\":142090,\n",
    "                   \"features_shape\":(439952, 499),\n",
    "                   \"rules_count\":80,\n",
    "                   \"rules_av_length\":7.2,\n",
    "                   \"training_time (sec)\":1961,\n",
    "                   \"model_size (kB)\":6} "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.loc[len(df)] = {\"repository\":\"axios/axios\",\n",
    "                   \"files_count\":76,\n",
    "                   \"lines_count\":7338,\n",
    "                   \"features_shape\":(24811, 394),\n",
    "                   \"rules_count\":424,\n",
    "                   \"rules_av_length\":6,\n",
    "                   \"training_time (sec)\":81,\n",
    "                   \"model_size (kB)\":9.5} "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.loc[len(df)] = {\"repository\":\"zeit/next.js\",\n",
    "                   \"files_count\":952,\n",
    "                   \"lines_count\":33340,\n",
    "                   \"features_shape\":(115432, 478),\n",
    "                   \"rules_count\":61,\n",
    "                   \"rules_av_length\":6.7,\n",
    "                   \"training_time (sec)\":397,\n",
    "                   \"model_size (kB)\":5.5} "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.loc[len(df)] = {\"repository\":\"vuejs/vue\",\n",
    "                   \"files_count\":405,\n",
    "                   \"lines_count\":141674,\n",
    "                   \"features_shape\":(522046, 491),\n",
    "                   \"rules_count\":771,\n",
    "                   \"rules_av_length\":7.1,\n",
    "                   \"training_time (sec)\":3114,\n",
    "                   \"model_size (kB)\":13.1} "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.loc[len(df)] = {\"repository\":\"src-d/code-annotation\",\n",
    "                   \"files_count\":51,\n",
    "                   \"lines_count\":5493,\n",
    "                   \"features_shape\":(17058, 425),\n",
    "                   \"rules_count\":31,\n",
    "                   \"rules_av_length\":5.5,\n",
    "                   \"training_time (sec)\":63,\n",
    "                   \"model_size (kB)\":4.8} "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.loc[len(df)] = {\"repository\":\"facebook/react-native\",\n",
    "                   \"files_count\":904,\n",
    "                   \"lines_count\":167096,\n",
    "                   \"features_shape\":(482170, 498),\n",
    "                   \"rules_count\":90,\n",
    "                   \"rules_av_length\":7.1,\n",
    "                   \"training_time (sec)\":2284,\n",
    "                   \"model_size (kB)\":6.2} "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.loc[len(df)] = {\"repository\":\"lodash/lodash\",\n",
    "                   \"files_count\":248,\n",
    "                   \"lines_count\":8172,\n",
    "                   \"features_shape\":(13663, 389),\n",
    "                   \"rules_count\":294,\n",
    "                   \"rules_av_length\":6.6,\n",
    "                   \"training_time (sec)\":34.7,\n",
    "                   \"model_size (kB)\":9} "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.loc[len(df)] = {\"repository\":\"twbs/bootstraps\",\n",
    "                   \"files_count\":55,\n",
    "                   \"lines_count\":26483,\n",
    "                   \"features_shape\":(94553, 449),\n",
    "                   \"rules_count\":54,\n",
    "                   \"rules_av_length\":6.6,\n",
    "                   \"training_time (sec)\":337,\n",
    "                   \"model_size (kB)\":5.3} "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>repository</th>\n",
       "      <th>files_count</th>\n",
       "      <th>lines_count</th>\n",
       "      <th>features_shape</th>\n",
       "      <th>rules_count</th>\n",
       "      <th>rules_av_length</th>\n",
       "      <th>training_time (sec)</th>\n",
       "      <th>model_size (kB)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>facebook/react-native</td>\n",
       "      <td>904</td>\n",
       "      <td>167096</td>\n",
       "      <td>(482170, 498)</td>\n",
       "      <td>90</td>\n",
       "      <td>7.1</td>\n",
       "      <td>2284</td>\n",
       "      <td>6.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>facebook/react</td>\n",
       "      <td>696</td>\n",
       "      <td>142090</td>\n",
       "      <td>(439952, 499)</td>\n",
       "      <td>80</td>\n",
       "      <td>7.2</td>\n",
       "      <td>1961</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>vuejs/vue</td>\n",
       "      <td>405</td>\n",
       "      <td>141674</td>\n",
       "      <td>(522046, 491)</td>\n",
       "      <td>771</td>\n",
       "      <td>7.1</td>\n",
       "      <td>3114</td>\n",
       "      <td>13.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>jquery/jquery</td>\n",
       "      <td>173</td>\n",
       "      <td>55384</td>\n",
       "      <td>(247249, 446)</td>\n",
       "      <td>835</td>\n",
       "      <td>7.4</td>\n",
       "      <td>1121</td>\n",
       "      <td>15.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>zeit/next.js</td>\n",
       "      <td>952</td>\n",
       "      <td>33340</td>\n",
       "      <td>(115432, 478)</td>\n",
       "      <td>61</td>\n",
       "      <td>6.7</td>\n",
       "      <td>397</td>\n",
       "      <td>5.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>twbs/bootstraps</td>\n",
       "      <td>55</td>\n",
       "      <td>26483</td>\n",
       "      <td>(94553, 449)</td>\n",
       "      <td>54</td>\n",
       "      <td>6.6</td>\n",
       "      <td>337</td>\n",
       "      <td>5.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>freeCodeCamp/freeCodeCamp</td>\n",
       "      <td>200</td>\n",
       "      <td>14209</td>\n",
       "      <td>(48788, 449)</td>\n",
       "      <td>45</td>\n",
       "      <td>6.3</td>\n",
       "      <td>1537</td>\n",
       "      <td>5.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>lodash/lodash</td>\n",
       "      <td>248</td>\n",
       "      <td>8172</td>\n",
       "      <td>(13663, 389)</td>\n",
       "      <td>294</td>\n",
       "      <td>6.6</td>\n",
       "      <td>34.7</td>\n",
       "      <td>9.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>axios/axios</td>\n",
       "      <td>76</td>\n",
       "      <td>7338</td>\n",
       "      <td>(24811, 394)</td>\n",
       "      <td>424</td>\n",
       "      <td>6.0</td>\n",
       "      <td>81</td>\n",
       "      <td>9.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>src-d/code-annotation</td>\n",
       "      <td>51</td>\n",
       "      <td>5493</td>\n",
       "      <td>(17058, 425)</td>\n",
       "      <td>31</td>\n",
       "      <td>5.5</td>\n",
       "      <td>63</td>\n",
       "      <td>4.8</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  repository files_count lines_count features_shape  \\\n",
       "0      facebook/react-native         904      167096  (482170, 498)   \n",
       "1             facebook/react         696      142090  (439952, 499)   \n",
       "2                  vuejs/vue         405      141674  (522046, 491)   \n",
       "3              jquery/jquery         173       55384  (247249, 446)   \n",
       "4               zeit/next.js         952       33340  (115432, 478)   \n",
       "5            twbs/bootstraps          55       26483   (94553, 449)   \n",
       "6  freeCodeCamp/freeCodeCamp         200       14209   (48788, 449)   \n",
       "7              lodash/lodash         248        8172   (13663, 389)   \n",
       "8                axios/axios          76        7338   (24811, 394)   \n",
       "9      src-d/code-annotation          51        5493   (17058, 425)   \n",
       "\n",
       "  rules_count  rules_av_length training_time (sec)  model_size (kB)  \n",
       "0          90              7.1                2284              6.2  \n",
       "1          80              7.2                1961              6.0  \n",
       "2         771              7.1                3114             13.1  \n",
       "3         835              7.4                1121             15.3  \n",
       "4          61              6.7                 397              5.5  \n",
       "5          54              6.6                 337              5.3  \n",
       "6          45              6.3                1537              5.1  \n",
       "7         294              6.6                34.7              9.0  \n",
       "8         424              6.0                  81              9.5  \n",
       "9          31              5.5                  63              4.8  "
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = df.sort_values(\"lines_count\", ascending=False).reset_index(drop=True)\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[167096, 142090, 141674, 55384, 33340, 26483, 14209, 8172, 7338, 5493]"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[\"lines_count\"].tolist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "x = df[\"lines_count\"].tolist()\n",
    "y = df[\"training_time (sec)\"].tolist()\n",
    "\n",
    "figure(figsize=(15, 10))\n",
    "ax = subplot(111)\n",
    "\n",
    "ax.margins(0.05) # Optional, just adds 5% padding to the autoscaling\n",
    "ax.plot(x, y, marker='x', linestyle='--', label=r\"$\\alpha=0.6$,   $c=2$\")\n",
    "\n",
    "ax.set_ylabel(\"training time (sec)\", fontsize=17, labelpad=15)\n",
    "ax.set_xlabel(\"number of JS LOC\", fontsize=17, labelpad=15)\n",
    "ax.spines['right'].set_visible(False)\n",
    "ax.spines['top'].set_visible(False)\n",
    "\n",
    "#text(x = 520, y = 10, s=\"}\", size=100, family = 'Helvetica Neue UltraLight', alpha=0.5)\n",
    "\n",
    "#handles, labels = ax.get_legend_handles_labels()\n",
    "#legend(handles[::-1], labels[::-1], markerscale=1.5, loc=\"lower left\", fontsize=17, bbox_to_anchor=(1.02,0))\n",
    "tight_layout()\n",
    "#savefig(\"pooling_curves.png\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.6.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}