results_curve.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "from IPython.display import display, Markdown, Latex\n",
    "import numpy as np\n",
    "import json\n",
    "import pandas as pd\n",
    "from scipy import stats\n",
    "import os\n",
    "\n",
    "#with open('transfer_experiment.json', 'r') as fp:\n",
    "#    data = json.load(fp)\n",
    "    \n",
    "def load_data(experiments):\n",
    "    data = { 'results' : {} }\n",
    "    for item in experiments:\n",
    "        experiment_title = item['id'] + '_' + item['source'] + '_' + item['target']\n",
    "        if os.path.isfile('experiments/' + experiment_title + '/' + experiment_title + '.json'):\n",
    "            with open('experiments/' + experiment_title + '/' + experiment_title + '.json', 'r') as fp:\n",
    "                results = json.load(fp)\n",
    "                data['results'][experiment_title] = results\n",
    "    return data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/markdown": [
       "# Results for AUC ROC"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Experiment</th>\n",
       "      <th>Transfer Learning</th>\n",
       "      <th>Transfer Learning with Revision Theory</th>\n",
       "      <th>Learning from scratch (RDN-B)</th>\n",
       "      <th>Learning from scratch (RDN)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1_imdb_uwcse_0.2</td>\n",
       "      <td>0.944 +/- 0.000</td>\n",
       "      <td>0.950 +/- 0.004</td>\n",
       "      <td>0.940 +/- 0.009</td>\n",
       "      <td>0.918 +/- 0.022</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1_imdb_uwcse_0.4</td>\n",
       "      <td>0.944 +/- 0.000</td>\n",
       "      <td>0.951 +/- 0.006</td>\n",
       "      <td>0.950 +/- 0.004</td>\n",
       "      <td>0.932 +/- 0.020</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1_imdb_uwcse_0.6</td>\n",
       "      <td>0.944 +/- 0.000</td>\n",
       "      <td>0.954 +/- 0.003</td>\n",
       "      <td>0.953 +/- 0.004</td>\n",
       "      <td>0.945 +/- 0.024</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1_imdb_uwcse_0.8</td>\n",
       "      <td>0.944 +/- 0.000</td>\n",
       "      <td>0.955 +/- 0.004</td>\n",
       "      <td>0.956 +/- 0.007</td>\n",
       "      <td>0.945 +/- 0.010</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1_imdb_uwcse_1.0</td>\n",
       "      <td>0.944 +/- 0.000</td>\n",
       "      <td>0.959 +/- 0.005</td>\n",
       "      <td>0.958 +/- 0.005</td>\n",
       "      <td>0.947 +/- 0.021</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2_uwcse_imdb_0.2</td>\n",
       "      <td>1.000 +/- 0.000</td>\n",
       "      <td>1.000 +/- 0.000</td>\n",
       "      <td>1.000 +/- 0.000</td>\n",
       "      <td>0.983 +/- 0.022</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2_uwcse_imdb_0.4</td>\n",
       "      <td>1.000 +/- 0.000</td>\n",
       "      <td>1.000 +/- 0.000</td>\n",
       "      <td>1.000 +/- 0.000</td>\n",
       "      <td>1.000 +/- 0.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>2_uwcse_imdb_0.6</td>\n",
       "      <td>1.000 +/- 0.000</td>\n",
       "      <td>1.000 +/- 0.000</td>\n",
       "      <td>1.000 +/- 0.000</td>\n",
       "      <td>0.998 +/- 0.003</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>2_uwcse_imdb_0.8</td>\n",
       "      <td>1.000 +/- 0.000</td>\n",
       "      <td>1.000 +/- 0.000</td>\n",
       "      <td>1.000 +/- 0.000</td>\n",
       "      <td>0.999 +/- 0.003</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>2_uwcse_imdb_1.0</td>\n",
       "      <td>1.000 +/- 0.000</td>\n",
       "      <td>1.000 +/- 0.000</td>\n",
       "      <td>1.000 +/- 0.000</td>\n",
       "      <td>1.000 +/- 0.000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Experiment Transfer Learning Transfer Learning with Revision Theory  \\\n",
       "0  1_imdb_uwcse_0.2   0.944 +/- 0.000                        0.950 +/- 0.004   \n",
       "1  1_imdb_uwcse_0.4   0.944 +/- 0.000                        0.951 +/- 0.006   \n",
       "2  1_imdb_uwcse_0.6   0.944 +/- 0.000                        0.954 +/- 0.003   \n",
       "3  1_imdb_uwcse_0.8   0.944 +/- 0.000                        0.955 +/- 0.004   \n",
       "4  1_imdb_uwcse_1.0   0.944 +/- 0.000                        0.959 +/- 0.005   \n",
       "5  2_uwcse_imdb_0.2   1.000 +/- 0.000                        1.000 +/- 0.000   \n",
       "6  2_uwcse_imdb_0.4   1.000 +/- 0.000                        1.000 +/- 0.000   \n",
       "7  2_uwcse_imdb_0.6   1.000 +/- 0.000                        1.000 +/- 0.000   \n",
       "8  2_uwcse_imdb_0.8   1.000 +/- 0.000                        1.000 +/- 0.000   \n",
       "9  2_uwcse_imdb_1.0   1.000 +/- 0.000                        1.000 +/- 0.000   \n",
       "\n",
       "  Learning from scratch (RDN-B) Learning from scratch (RDN)  \n",
       "0               0.940 +/- 0.009             0.918 +/- 0.022  \n",
       "1               0.950 +/- 0.004             0.932 +/- 0.020  \n",
       "2               0.953 +/- 0.004             0.945 +/- 0.024  \n",
       "3               0.956 +/- 0.007             0.945 +/- 0.010  \n",
       "4               0.958 +/- 0.005             0.947 +/- 0.021  \n",
       "5               1.000 +/- 0.000             0.983 +/- 0.022  \n",
       "6               1.000 +/- 0.000             1.000 +/- 0.000  \n",
       "7               1.000 +/- 0.000             0.998 +/- 0.003  \n",
       "8               1.000 +/- 0.000             0.999 +/- 0.003  \n",
       "9               1.000 +/- 0.000             1.000 +/- 0.000  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/markdown": [
       "# Results for AUC PR"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Experiment</th>\n",
       "      <th>Transfer Learning</th>\n",
       "      <th>Transfer Learning with Revision Theory</th>\n",
       "      <th>Learning from scratch (RDN-B)</th>\n",
       "      <th>Learning from scratch (RDN)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1_imdb_uwcse_0.2</td>\n",
       "      <td>0.459 +/- 0.000</td>\n",
       "      <td>0.421 +/- 0.047</td>\n",
       "      <td>0.367 +/- 0.093</td>\n",
       "      <td>0.320 +/- 0.063</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1_imdb_uwcse_0.4</td>\n",
       "      <td>0.459 +/- 0.000</td>\n",
       "      <td>0.450 +/- 0.048</td>\n",
       "      <td>0.425 +/- 0.045</td>\n",
       "      <td>0.351 +/- 0.055</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1_imdb_uwcse_0.6</td>\n",
       "      <td>0.459 +/- 0.000</td>\n",
       "      <td>0.463 +/- 0.016</td>\n",
       "      <td>0.468 +/- 0.029</td>\n",
       "      <td>0.451 +/- 0.043</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1_imdb_uwcse_0.8</td>\n",
       "      <td>0.459 +/- 0.000</td>\n",
       "      <td>0.464 +/- 0.042</td>\n",
       "      <td>0.489 +/- 0.070</td>\n",
       "      <td>0.417 +/- 0.092</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1_imdb_uwcse_1.0</td>\n",
       "      <td>0.459 +/- 0.000</td>\n",
       "      <td>0.501 +/- 0.047</td>\n",
       "      <td>0.490 +/- 0.037</td>\n",
       "      <td>0.463 +/- 0.027</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2_uwcse_imdb_0.2</td>\n",
       "      <td>1.000 +/- 0.000</td>\n",
       "      <td>1.000 +/- 0.000</td>\n",
       "      <td>1.000 +/- 0.000</td>\n",
       "      <td>0.866 +/- 0.085</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2_uwcse_imdb_0.4</td>\n",
       "      <td>1.000 +/- 0.000</td>\n",
       "      <td>1.000 +/- 0.000</td>\n",
       "      <td>1.000 +/- 0.000</td>\n",
       "      <td>1.000 +/- 0.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>2_uwcse_imdb_0.6</td>\n",
       "      <td>1.000 +/- 0.000</td>\n",
       "      <td>1.000 +/- 0.000</td>\n",
       "      <td>1.000 +/- 0.000</td>\n",
       "      <td>0.950 +/- 0.070</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>2_uwcse_imdb_0.8</td>\n",
       "      <td>1.000 +/- 0.000</td>\n",
       "      <td>1.000 +/- 0.000</td>\n",
       "      <td>1.000 +/- 0.000</td>\n",
       "      <td>0.975 +/- 0.070</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>2_uwcse_imdb_1.0</td>\n",
       "      <td>1.000 +/- 0.000</td>\n",
       "      <td>1.000 +/- 0.000</td>\n",
       "      <td>1.000 +/- 0.000</td>\n",
       "      <td>1.000 +/- 0.000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Experiment Transfer Learning Transfer Learning with Revision Theory  \\\n",
       "0  1_imdb_uwcse_0.2   0.459 +/- 0.000                        0.421 +/- 0.047   \n",
       "1  1_imdb_uwcse_0.4   0.459 +/- 0.000                        0.450 +/- 0.048   \n",
       "2  1_imdb_uwcse_0.6   0.459 +/- 0.000                        0.463 +/- 0.016   \n",
       "3  1_imdb_uwcse_0.8   0.459 +/- 0.000                        0.464 +/- 0.042   \n",
       "4  1_imdb_uwcse_1.0   0.459 +/- 0.000                        0.501 +/- 0.047   \n",
       "5  2_uwcse_imdb_0.2   1.000 +/- 0.000                        1.000 +/- 0.000   \n",
       "6  2_uwcse_imdb_0.4   1.000 +/- 0.000                        1.000 +/- 0.000   \n",
       "7  2_uwcse_imdb_0.6   1.000 +/- 0.000                        1.000 +/- 0.000   \n",
       "8  2_uwcse_imdb_0.8   1.000 +/- 0.000                        1.000 +/- 0.000   \n",
       "9  2_uwcse_imdb_1.0   1.000 +/- 0.000                        1.000 +/- 0.000   \n",
       "\n",
       "  Learning from scratch (RDN-B) Learning from scratch (RDN)  \n",
       "0               0.367 +/- 0.093             0.320 +/- 0.063  \n",
       "1               0.425 +/- 0.045             0.351 +/- 0.055  \n",
       "2               0.468 +/- 0.029             0.451 +/- 0.043  \n",
       "3               0.489 +/- 0.070             0.417 +/- 0.092  \n",
       "4               0.490 +/- 0.037             0.463 +/- 0.027  \n",
       "5               1.000 +/- 0.000             0.866 +/- 0.085  \n",
       "6               1.000 +/- 0.000             1.000 +/- 0.000  \n",
       "7               1.000 +/- 0.000             0.950 +/- 0.070  \n",
       "8               1.000 +/- 0.000             0.975 +/- 0.070  \n",
       "9               1.000 +/- 0.000             1.000 +/- 0.000  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/markdown": [
       "# Results for CLL"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Experiment</th>\n",
       "      <th>Transfer Learning</th>\n",
       "      <th>Transfer Learning with Revision Theory</th>\n",
       "      <th>Learning from scratch (RDN-B)</th>\n",
       "      <th>Learning from scratch (RDN)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1_imdb_uwcse_0.2</td>\n",
       "      <td>-0.227 +/- 0.017</td>\n",
       "      <td>-0.218 +/- 0.036</td>\n",
       "      <td>-0.224 +/- 0.030</td>\n",
       "      <td>-0.214 +/- 0.003</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1_imdb_uwcse_0.4</td>\n",
       "      <td>-0.220 +/- 0.010</td>\n",
       "      <td>-0.212 +/- 0.011</td>\n",
       "      <td>-0.211 +/- 0.012</td>\n",
       "      <td>-0.214 +/- 0.004</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1_imdb_uwcse_0.6</td>\n",
       "      <td>-0.219 +/- 0.003</td>\n",
       "      <td>-0.209 +/- 0.006</td>\n",
       "      <td>-0.212 +/- 0.016</td>\n",
       "      <td>-0.211 +/- 0.002</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1_imdb_uwcse_0.8</td>\n",
       "      <td>-0.220 +/- 0.008</td>\n",
       "      <td>-0.215 +/- 0.009</td>\n",
       "      <td>-0.214 +/- 0.022</td>\n",
       "      <td>-0.211 +/- 0.002</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1_imdb_uwcse_1.0</td>\n",
       "      <td>-0.220 +/- 0.007</td>\n",
       "      <td>-0.209 +/- 0.003</td>\n",
       "      <td>-0.212 +/- 0.002</td>\n",
       "      <td>-0.211 +/- 0.003</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2_uwcse_imdb_0.2</td>\n",
       "      <td>-0.067 +/- 0.003</td>\n",
       "      <td>-0.066 +/- 0.001</td>\n",
       "      <td>-0.072 +/- 0.004</td>\n",
       "      <td>-0.168 +/- 0.007</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2_uwcse_imdb_0.4</td>\n",
       "      <td>-0.067 +/- 0.002</td>\n",
       "      <td>-0.065 +/- 0.001</td>\n",
       "      <td>-0.066 +/- 0.003</td>\n",
       "      <td>-0.163 +/- 0.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>2_uwcse_imdb_0.6</td>\n",
       "      <td>-0.067 +/- 0.002</td>\n",
       "      <td>-0.066 +/- 0.001</td>\n",
       "      <td>-0.065 +/- 0.002</td>\n",
       "      <td>-0.163 +/- 0.001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>2_uwcse_imdb_0.8</td>\n",
       "      <td>-0.067 +/- 0.002</td>\n",
       "      <td>-0.065 +/- 0.000</td>\n",
       "      <td>-0.065 +/- 0.002</td>\n",
       "      <td>-0.163 +/- 0.001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>2_uwcse_imdb_1.0</td>\n",
       "      <td>-0.067 +/- 0.002</td>\n",
       "      <td>-0.065 +/- 0.001</td>\n",
       "      <td>-0.064 +/- 0.000</td>\n",
       "      <td>-0.163 +/- 0.000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Experiment Transfer Learning Transfer Learning with Revision Theory  \\\n",
       "0  1_imdb_uwcse_0.2  -0.227 +/- 0.017                       -0.218 +/- 0.036   \n",
       "1  1_imdb_uwcse_0.4  -0.220 +/- 0.010                       -0.212 +/- 0.011   \n",
       "2  1_imdb_uwcse_0.6  -0.219 +/- 0.003                       -0.209 +/- 0.006   \n",
       "3  1_imdb_uwcse_0.8  -0.220 +/- 0.008                       -0.215 +/- 0.009   \n",
       "4  1_imdb_uwcse_1.0  -0.220 +/- 0.007                       -0.209 +/- 0.003   \n",
       "5  2_uwcse_imdb_0.2  -0.067 +/- 0.003                       -0.066 +/- 0.001   \n",
       "6  2_uwcse_imdb_0.4  -0.067 +/- 0.002                       -0.065 +/- 0.001   \n",
       "7  2_uwcse_imdb_0.6  -0.067 +/- 0.002                       -0.066 +/- 0.001   \n",
       "8  2_uwcse_imdb_0.8  -0.067 +/- 0.002                       -0.065 +/- 0.000   \n",
       "9  2_uwcse_imdb_1.0  -0.067 +/- 0.002                       -0.065 +/- 0.001   \n",
       "\n",
       "  Learning from scratch (RDN-B) Learning from scratch (RDN)  \n",
       "0              -0.224 +/- 0.030            -0.214 +/- 0.003  \n",
       "1              -0.211 +/- 0.012            -0.214 +/- 0.004  \n",
       "2              -0.212 +/- 0.016            -0.211 +/- 0.002  \n",
       "3              -0.214 +/- 0.022            -0.211 +/- 0.002  \n",
       "4              -0.212 +/- 0.002            -0.211 +/- 0.003  \n",
       "5              -0.072 +/- 0.004            -0.168 +/- 0.007  \n",
       "6              -0.066 +/- 0.003            -0.163 +/- 0.000  \n",
       "7              -0.065 +/- 0.002            -0.163 +/- 0.001  \n",
       "8              -0.065 +/- 0.002            -0.163 +/- 0.001  \n",
       "9              -0.064 +/- 0.000            -0.163 +/- 0.000  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "experiments = [\n",
    "            {'id': '1', 'source':'imdb', 'target':'uwcse', 'predicate':'workedunder', 'to_predicate':'advisedby'},\n",
    "            {'id': '2', 'source':'uwcse', 'target':'imdb', 'predicate':'advisedby', 'to_predicate':'workedunder'},\n",
    "            {'id': '3', 'source':'imdb', 'target':'uwcse', 'predicate':'movie', 'to_predicate':'publication'},\n",
    "            {'id': '4', 'source':'uwcse', 'target':'imdb', 'predicate':'publication', 'to_predicate':'movie'},\n",
    "            {'id': '5', 'source':'imdb', 'target':'uwcse', 'predicate':'genre', 'to_predicate':'inphase'},\n",
    "            {'id': '6', 'source':'uwcse', 'target':'imdb', 'predicate':'inphase', 'to_predicate':'genre'},\n",
    "            {'id': '7', 'source':'imdb', 'target':'cora', 'predicate':'workedunder', 'to_predicate':'samevenue'},\n",
    "            {'id': '8', 'source':'imdb', 'target':'cora', 'predicate':'workedunder', 'to_predicate':'samebib'},\n",
    "            {'id': '9', 'source':'imdb', 'target':'cora', 'predicate':'workedunder', 'to_predicate':'sameauthor'},\n",
    "            {'id': '10', 'source':'imdb', 'target':'cora', 'predicate':'workedunder', 'to_predicate':'sametitle'},\n",
    "            {'id': '11', 'source':'uwcse', 'target':'cora', 'predicate':'advisedby', 'to_predicate':'samevenue'},\n",
    "            {'id': '12', 'source':'uwcse', 'target':'cora', 'predicate':'advisedby', 'to_predicate':'samebib'},\n",
    "            {'id': '13', 'source':'uwcse', 'target':'cora', 'predicate':'advisedby', 'to_predicate':'sameauthor'},\n",
    "            {'id': '14', 'source':'uwcse', 'target':'cora', 'predicate':'advisedby', 'to_predicate':'sametitle'},\n",
    "            {'id': '15', 'source':'yeast', 'target':'twitter', 'predicate':'proteinclass', 'to_predicate':'accounttype'},\n",
    "            {'id': '16', 'source':'yeast', 'target':'twitter', 'predicate':'interaction', 'to_predicate':'follows'},\n",
    "            {'id': '17', 'source':'yeast', 'target':'twitter', 'predicate':'location', 'to_predicate':'tweets'},\n",
    "            {'id': '18', 'source':'yeast', 'target':'twitter', 'predicate':'enzyme', 'to_predicate':'tweets'},\n",
    "            {'id': '19', 'source':'yeast', 'target':'twitter', 'predicate':'function', 'to_predicate':'tweets'},\n",
    "            {'id': '20', 'source':'yeast', 'target':'twitter', 'predicate':'phenotype', 'to_predicate':'tweets'},\n",
    "            {'id': '21', 'source':'yeast', 'target':'twitter', 'predicate':'complex', 'to_predicate':'tweets'},\n",
    "            {'id': '22', 'source':'twitter', 'target':'yeast', 'predicate':'accounttype', 'to_predicate':'proteinclass'},\n",
    "            {'id': '23', 'source':'twitter', 'target':'yeast', 'predicate':'follows', 'to_predicate':'interaction'},\n",
    "            {'id': '24', 'source':'twitter', 'target':'yeast', 'predicate':'tweets', 'to_predicate':'location'},\n",
    "            {'id': '25', 'source':'twitter', 'target':'yeast', 'predicate':'tweets', 'to_predicate':'enzyme'},\n",
    "            {'id': '26', 'source':'twitter', 'target':'yeast', 'predicate':'tweets', 'to_predicate':'function'},\n",
    "            {'id': '27', 'source':'twitter', 'target':'yeast', 'predicate':'tweets', 'to_predicate':'phenotype'},\n",
    "            {'id': '28', 'source':'twitter', 'target':'yeast', 'predicate':'tweets', 'to_predicate':'complex'},\n",
    "            {'id': '29', 'source':'nell_sports', 'target':'nell_finances', 'predicate':'teamalsoknownas', 'to_predicate':'companyalsoknownas'},\n",
    "            {'id': '30', 'source':'nell_sports', 'target':'nell_finances', 'predicate':'teamplaysagainstteam', 'to_predicate':'companyalsoknownas'},\n",
    "            {'id': '31', 'source':'nell_sports', 'target':'nell_finances', 'predicate':'teamplaysagainstteam', 'to_predicate':'acquired'},\n",
    "            {'id': '32', 'source':'nell_sports', 'target':'nell_finances', 'predicate':'teamplaysagainstteam', 'to_predicate':'bankboughtbank'},\n",
    "            {'id': '33', 'source':'nell_sports', 'target':'nell_finances', 'predicate':'athleteplayssport', 'to_predicate':'companyceo'},\n",
    "            {'id': '34', 'source':'nell_sports', 'target':'nell_finances', 'predicate':'athleteplayssport', 'to_predicate':'bankchiefexecutiveceo'},\n",
    "            {'id': '35', 'source':'nell_sports', 'target':'nell_finances', 'predicate':'athleteplaysforteam', 'to_predicate':'bankchiefexecutiveceo'},\n",
    "            {'id': '36', 'source':'nell_sports', 'target':'nell_finances', 'predicate':'athleteplaysforteam', 'to_predicate':'companyceo'},\n",
    "            {'id': '37', 'source':'nell_sports', 'target':'nell_finances', 'predicate':'teamplayssport', 'to_predicate':'companyeconomicsector'},\n",
    "            {'id': '38', 'source':'nell_finances', 'target':'nell_sports', 'predicate':'companyalsoknownas', 'to_predicate':'teamalsoknownas'},\n",
    "            {'id': '39', 'source':'nell_finances', 'target':'nell_sports', 'predicate':'companyalsoknownas', 'to_predicate':'teamplaysagainstteam'},\n",
    "            {'id': '40', 'source':'nell_finances', 'target':'nell_sports', 'predicate':'acquired', 'to_predicate':'teamplaysagainstteam'},\n",
    "            {'id': '41', 'source':'nell_finances', 'target':'nell_sports', 'predicate':'bankboughtbank', 'to_predicate':'teamplaysagainstteam'},\n",
    "            {'id': '42', 'source':'nell_finances', 'target':'nell_sports', 'predicate':'companyceo', 'to_predicate':'athleteplayssport'},\n",
    "            {'id': '43', 'source':'nell_finances', 'target':'nell_sports', 'predicate':'bankchiefexecutiveceo', 'to_predicate':'athleteplayssport'},\n",
    "            {'id': '44', 'source':'nell_finances', 'target':'nell_sports', 'predicate':'bankchiefexecutiveceo', 'to_predicate':'athleteplaysforteam'},\n",
    "            {'id': '45', 'source':'nell_finances', 'target':'nell_sports', 'predicate':'companyceo', 'to_predicate':'athleteplaysforteam'},\n",
    "            {'id': '46', 'source':'nell_finances', 'target':'nell_sports', 'predicate':'companyeconomicsector', 'to_predicate':'teamplayssport'},\n",
    "            {'id': '47', 'source':'yeast', 'target':'facebook', 'predicate':'interaction', 'to_predicate':'edge'},\n",
    "            {'id': '48', 'source':'twitter', 'target':'facebook', 'predicate':'follows', 'to_predicate':'edge'},\n",
    "            {'id': '49', 'source':'imdb', 'target':'facebook', 'predicate':'workedunder', 'to_predicate':'edge'},\n",
    "            {'id': '50', 'source':'uwcse', 'target':'facebook', 'predicate':'advisedby', 'to_predicate':'edge'},\n",
    "]\n",
    "\n",
    "data = load_data(experiments)\n",
    "\n",
    "for metric in ['AUC ROC', 'AUC PR', 'CLL']: #, 'Recall', 'F1', 'Precision', 'Learning and Revision time', 'Inference time']:\n",
    "    display(Markdown('# Results for ' + metric))\n",
    "    table = []\n",
    "    for j in range(len(experiments)):\n",
    "        dataset = experiments[j]['id'] + '_' + experiments[j]['source'] + '_' + experiments[j]['target']\n",
    "        if dataset in data['results']:\n",
    "            for amount in [0.2, 0.4, 0.6, 0.8, 1.0]:\n",
    "                #for metric in ['AUC ROC', 'AUC PR']:\n",
    "                    #display(Markdown('## ' + metric))\n",
    "                mapping_time = ''\n",
    "                if metric == 'Precision':\n",
    "                    scratch = np.array([(np.array([item['rdn_b_'+str(amount)]['Precision'][0] for item in sublist])).mean() for sublist in data['results'][str(dataset)]])\n",
    "                    scratch_rdn = np.array([(np.array([item['rdn_'+str(amount)]['Precision'][0] for item in sublist])).mean() for sublist in data['results'][str(dataset)]])\n",
    "                    parameter = np.array([(np.array([item['transfer_'+str(amount)]['parameter_'+str(amount)]['Precision'][0] for item in sublist])).mean() for sublist in data['results'][str(dataset)]])\n",
    "                    transfer = np.array([(np.array([item['transfer_'+str(amount)]['Precision'][0] for item in sublist])).mean() for sublist in data['results'][str(dataset)]])\n",
    "                elif metric == 'Learning and Revision time':\n",
    "                    scratch = np.array([(np.array([item['rdn_b_'+str(amount)]['Learning time'] for item in sublist])).mean() for sublist in data['results'][str(dataset)]])\n",
    "                    scratch_rdn = np.array([(np.array([item['rdn_'+str(amount)]['Learning time'] for item in sublist])).mean() for sublist in data['results'][str(dataset)]])\n",
    "                    parameter = np.array([(np.array([item['transfer_'+str(amount)]['parameter_'+str(amount)]['Learning time'] for item in sublist])).mean() for sublist in data['results'][str(dataset)]])\n",
    "                    transfer = np.array([(np.array([item['transfer_'+str(amount)]['Learning time'] for item in sublist])).mean() for sublist in data['results'][str(dataset)]])\n",
    "                    #mapping = np.array([(np.array([item['transfer']['Mapping results']['Total time'] for item in sublist])).mean() for sublist in data['results'][str(dataset)]])\n",
    "                    #mapping_time = ' (%.1f)' % mapping.mean()\n",
    "                elif metric == 'Inference time':\n",
    "                    scratch = np.array([(np.array([item['rdn_b_'+str(amount)]['Inference time'] for item in sublist])).mean() for sublist in data['results'][str(dataset)]])\n",
    "                    scratch_rdn = np.array([(np.array([item['rdn_'+str(amount)]['Inference time'] for item in sublist])).mean() for sublist in data['results'][str(dataset)]])\n",
    "                    parameter = np.array([(np.array([item['transfer_'+str(amount)]['parameter_'+str(amount)]['Inference time'] for item in sublist])).mean() for sublist in data['results'][str(dataset)]])\n",
    "                    transfer = np.array([(np.array([item['transfer_'+str(amount)]['Inference time'] for item in sublist])).mean() for sublist in data['results'][str(dataset)]])\n",
    "                else:\n",
    "                    scratch = np.array([(np.array([item['rdn_b_'+str(amount)][metric] for item in sublist])).mean() for sublist in data['results'][str(dataset)]])\n",
    "                    scratch_rdn = np.array([(np.array([item['rdn_'+str(amount)][metric] for item in sublist])).mean() for sublist in data['results'][str(dataset)]])\n",
    "                    parameter = np.array([(np.array([item['transfer_'+str(amount)]['parameter_'+str(amount)][metric] for item in sublist])).mean() for sublist in data['results'][str(dataset)]])\n",
    "                    transfer = np.array([(np.array([item['transfer_'+str(amount)][metric] for item in sublist])).mean() for sublist in data['results'][str(dataset)]])\n",
    "                table.append([dataset +'_' + str(amount), '%.3f +/- %.3f' % (parameter.mean(), 2 * parameter.std()),\n",
    "                '%.3f +/- %.3f' % (transfer.mean(), 2 * transfer.std()),\n",
    "                '%.3f +/- %.3f' % (scratch.mean(), 2 * scratch.std()),\n",
    "                '%.3f +/- %.3f' % (scratch_rdn.mean(), 2 * scratch_rdn.std())])\n",
    "                 #             ])\n",
    "    display(pd.DataFrame(table, columns=['Experiment', 'Transfer Learning', 'Transfer Learning with Revision Theory', 'Learning from scratch (RDN-B)', 'Learning from scratch (RDN)']))\n",
    "    #display(pd.DataFrame(table, columns=['Experiment', 'Transfer Learning', 'Transfer Learning with Revision Theory']))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/markdown": [
       "# Results for AUC ROC"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/markdown": [
       "# 1_imdb_uwcse"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "trRDN-B Rev [0.95040445, 0.95084325, 0.95394265, 0.9553802499999999, 0.95917405]\n",
      "trRDN-B [0.9443426, 0.9443426, 0.9443426, 0.9443426, 0.9443426]\n",
      "RDN-B [0.9395219500000002, 0.9496731499999999, 0.9529754500000001, 0.95565195, 0.95829365]\n",
      "RDN [0.9176458, 0.93207655, 0.9451099000000001, 0.94512575, 0.9470201999999999]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAikAAAGHCAYAAAB1bcIdAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fb882b3c668>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/markdown": [
       "# 2_uwcse_imdb"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "trRDN-B Rev [1.0, 1.0, 1.0, 1.0, 1.0]\n",
      "trRDN-B [1.0, 1.0, 1.0, 1.0, 1.0]\n",
      "RDN-B [1.0, 1.0, 1.0, 1.0, 1.0]\n",
      "RDN [0.9832157333333335, 1.0, 0.9980054666666667, 0.9990027333333332, 1.0]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjIAAAGHCAYAAACqI7gCAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRETi1v8D4299rQKSqdAAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fb882e1e4a8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/markdown": [
       "# Results for AUC PR"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/markdown": [
       "# 1_imdb_uwcse"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "trRDN-B Rev [0.4208026, 0.4497349, 0.46331285, 0.464372, 0.50056095]\n",
      "trRDN-B [0.4593376, 0.4593376, 0.4593376, 0.4593376, 0.4593376]\n",
      "RDN-B [0.36655055000000003, 0.42468, 0.46772145000000004, 0.4885398, 0.4902847]\n",
      "RDN [0.31981434999999997, 0.35118074999999993, 0.45111375, 0.4165454, 0.4631978]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAikAAAGHCAYAAAB1bcIdAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSimllApL/z8ZJpNrFSnq1AAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fb882cd5710>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/markdown": [
       "# 2_uwcse_imdb"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "trRDN-B Rev [1.0, 1.0, 1.0, 1.0, 1.0]\n",
      "trRDN-B [1.0, 1.0, 1.0, 1.0, 1.0]\n",
      "RDN-B [1.0, 1.0, 1.0, 1.0, 1.0]\n",
      "RDN [0.8660016666666666, 1.0, 0.9502348, 0.9751173999999999, 1.0]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAikAAAGHCAYAAAB1bcIdAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fb882ea37f0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/markdown": [
       "# Results for CLL"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/markdown": [
       "# 1_imdb_uwcse"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "trRDN-B Rev [-0.21825165000000002, -0.2121487, -0.20891790000000002, -0.21487255000000002, -0.2088936]\n",
      "trRDN-B [-0.22664069999999997, -0.22044814999999998, -0.21913115, -0.2196632, -0.2197029]\n",
      "RDN-B [-0.2243123, -0.2105414, -0.21193564999999998, -0.21394005000000002, -0.21165050000000002]\n",
      "RDN [-0.21402500000000002, -0.21391404999999997, -0.2114047, -0.21115625000000002, -0.21062240000000002]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAj4AAAGHCAYAAACwH1iMAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nCGEZ/x9btwtVGsym6gAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fb8827b1a20>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/markdown": [
       "# 2_uwcse_imdb"
      ],
      "text/plain": [
       "<IPython.core.display.Markdown object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "trRDN-B Rev [-0.0657136, -0.06535039999999999, -0.06559366666666666, -0.0650216, -0.06538519999999999]\n",
      "trRDN-B [-0.067332, -0.06684706666666668, -0.06685246666666667, -0.06690746666666667, -0.06687086666666667]\n",
      "RDN-B [-0.07243573333333332, -0.06637406666666666, -0.0650648, -0.0650636, -0.06414693333333332]\n",
      "RDN [-0.16824486666666663, -0.162615, -0.16335646666666667, -0.16298573333333333, -0.162615]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjUAAAGHCAYAAABI/6N7AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fb8831c1e80>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for metric in ['AUC ROC', 'AUC PR', 'CLL']:\n",
    "    display(Markdown('# Results for ' + metric))\n",
    "    for j in range(len(experiments)):\n",
    "        dataset = experiments[j]['id'] + '_' + experiments[j]['source'] + '_' + experiments[j]['target']\n",
    "        if dataset in data['results']:\n",
    "            display(Markdown('# ' + dataset))\n",
    "            transfer_p = []\n",
    "            parameter_p = []\n",
    "            rdn_p = []\n",
    "            rdn_b_p = []\n",
    "            for amount in [0.2, 0.4, 0.6, 0.8, 1.0]:\n",
    "                #for metric in ['AUC ROC', 'AUC PR']:\n",
    "                    #display(Markdown('## ' + metric))\n",
    "                mapping_time = ''\n",
    "                if metric == 'Precision':\n",
    "                    scratch = np.array([(np.array([item['rdn_b_'+str(amount)]['Precision'][0] for item in sublist])).mean() for sublist in data['results'][str(dataset)]])\n",
    "                    scratch_rdn = np.array([(np.array([item['rdn_'+str(amount)]['Precision'][0] for item in sublist])).mean() for sublist in data['results'][str(dataset)]])\n",
    "                    parameter = np.array([(np.array([item['transfer_'+str(amount)]['parameter_'+str(amount)]['Precision'][0] for item in sublist])).mean() for sublist in data['results'][str(dataset)]])\n",
    "                    transfer = np.array([(np.array([item['transfer_'+str(amount)]['Precision'][0] for item in sublist])).mean() for sublist in data['results'][str(dataset)]])\n",
    "                elif metric == 'Learning and Revision time':\n",
    "                    scratch = np.array([(np.array([item['rdn_b_'+str(amount)]['Learning time'] for item in sublist])).mean() for sublist in data['results'][str(dataset)]])\n",
    "                    scratch_rdn = np.array([(np.array([item['rdn_'+str(amount)]['Learning time'] for item in sublist])).mean() for sublist in data['results'][str(dataset)]])\n",
    "                    parameter = np.array([(np.array([item['transfer_'+str(amount)]['parameter_'+str(amount)]['Learning time'] for item in sublist])).mean() for sublist in data['results'][str(dataset)]])\n",
    "                    transfer = np.array([(np.array([item['transfer_'+str(amount)]['Learning time'] for item in sublist])).mean() for sublist in data['results'][str(dataset)]])\n",
    "                    #mapping = np.array([(np.array([item['transfer']['Mapping results']['Total time'] for item in sublist])).mean() for sublist in data['results'][str(dataset)]])\n",
    "                    #mapping_time = ' (%.1f)' % mapping.mean()\n",
    "                elif metric == 'Inference time':\n",
    "                    scratch = np.array([(np.array([item['rdn_b_'+str(amount)]['Inference time'] for item in sublist])).mean() for sublist in data['results'][str(dataset)]])\n",
    "                    scratch_rdn = np.array([(np.array([item['rdn_'+str(amount)]['Inference time'] for item in sublist])).mean() for sublist in data['results'][str(dataset)]])\n",
    "                    parameter = np.array([(np.array([item['transfer_'+str(amount)]['parameter_'+str(amount)]['Inference time'] for item in sublist])).mean() for sublist in data['results'][str(dataset)]])\n",
    "                    transfer = np.array([(np.array([item['transfer_'+str(amount)]['Inference time'] for item in sublist])).mean() for sublist in data['results'][str(dataset)]])\n",
    "                else:\n",
    "                    scratch = np.array([(np.array([item['rdn_b_'+str(amount)][metric] for item in sublist])).mean() for sublist in data['results'][str(dataset)]])\n",
    "                    scratch_rdn = np.array([(np.array([item['rdn_'+str(amount)][metric] for item in sublist])).mean() for sublist in data['results'][str(dataset)]])\n",
    "                    parameter = np.array([(np.array([item['transfer_'+str(amount)]['parameter_'+str(amount)][metric] for item in sublist])).mean() for sublist in data['results'][str(dataset)]])\n",
    "                    transfer = np.array([(np.array([item['transfer_'+str(amount)][metric] for item in sublist])).mean() for sublist in data['results'][str(dataset)]])\n",
    "                table.append([dataset, '%.3f +/- %.3f' % (parameter.mean(), 2 * parameter.std()),\n",
    "                '%.3f +/- %.3f' % (transfer.mean(), 2 * transfer.std()),\n",
    "                '%.3f +/- %.3f' % (scratch.mean(), 2 * scratch.std()),\n",
    "                '%.3f +/- %.3f' % (scratch_rdn.mean(), 2 * scratch_rdn.std())])\n",
    "                 #             ])\n",
    "                transfer_p.append(transfer.mean())\n",
    "                parameter_p.append(parameter.mean())\n",
    "                rdn_b_p.append(scratch.mean())\n",
    "                rdn_p.append(scratch_rdn.mean())\n",
    "\n",
    "            print('trRDN-B Rev %s' % transfer_p)\n",
    "            print('trRDN-B %s' % parameter_p)\n",
    "            print('RDN-B %s' % rdn_b_p)\n",
    "            print('RDN %s' % rdn_p)\n",
    "            t = [0.2, 0.4, 0.6, 0.8, 1.0]\n",
    "            fig, ax = plt.subplots()\n",
    "            ax.plot(t, rdn_p, '^-', linestyle=':', color='k', label='RDN')\n",
    "            ax.plot(t, rdn_b_p, 'D-', linestyle='-.', color='k', label='RDN-B')\n",
    "            ax.plot(t, parameter_p, 's-', linestyle='--', color='k', label='trRDN-B')\n",
    "            ax.plot(t, transfer_p, 'x-', linestyle='-', color='k', label='trRDN-B Rev')\n",
    "\n",
    "            ax.set(xlabel='time (s)', ylabel=metric,\n",
    "                   title='About as simple as it gets, folks')\n",
    "            ax.grid(linestyle = ':')\n",
    "            ax.legend(loc='lower right', fontsize='small', frameon=True)\n",
    "            plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python [default]",
   "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": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}