pulsar_frb_tutorial_2019_machine_learning.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h1><center>Pulsar-FRB Tutorial 2019</center></h1>\n",
    "<h1><center>Introduction To Machine-learning for Pulsar-Candidate Identification</center></h1>\n",
    "<h3><left>By Vishnu Balakrishnan, email: vishnu@mpifr-bonn.mpg.de</left></h3>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Credits\n",
    "These notes have been compiled from several different sources. Besides, plots from papers (that have been referenced within), some of the major slides I have borrowed are from MIT 6.S191 Lecture 1 & 2 Introduction to Deep Learning. \n",
    "\n",
    "http://introtodeeplearning.com/\n",
    "\n",
    "All gifs were borrowed from 3Blue1Brown\n",
    "\n",
    "https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw\n",
    "\n",
    "\n",
    "Hands On Session:\n",
    "\n",
    "Dependiences:\n",
    "\n",
    "Keras 2.1.5, Tensorflow (CPU/GPU), Sklearn, Numpy"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h2><center>Motivation</center></h2>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Pulsar Candidates have been steadily increasing over time. The rise is understood to be the result of expanding survey technical specifications (Stovall et al. 2013) occurring during the period depicted in Tables 2 & 3. Finer frequency resolution, longer dwell times, and acceleration searches (Eatough et al. 2013), have significantly increased the candidate yield (Lyon 2015)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "from IPython.display import Image\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 280,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<img src=\"images/candidate_numbers.png\" width=\"1000\" height=\"800\"/>"
      ],
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 280,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(url= \"images/candidate_numbers.png\", width=1000, height=800)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h2><center>What is a pulsar candidate?</center></h2>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 308,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<img src=\"images/pulsar.png\" width=\"800\" height=\"800\"/>"
      ],
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 308,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "Image(url= \"images/pulsar.png\", width=800, height=800)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 307,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<img src=\"images/non_pulsars.png\" width=\"800\" height=\"800\"/>"
      ],
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 307,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(url= \"images/non_pulsars.png\", width=800, height=800)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 306,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<img src=\"images/rfi.png\" width=\"800\" height=\"800\"/>"
      ],
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 306,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(url= \"images/rfi.png\", width=800, height=800)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h3><center>Literature Review</center></h3>\n",
    "Eatough et al. 2010: Used a single-layer perceptron (12:12:2) to train pulsar candidates using 259 pulsars and 1625 non-pulsars. Discovered one new pulsar!\n",
    "   \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 305,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<img src=\"images/ralph.png\" width=\"800\" height=\"800\"/>"
      ],
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 305,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Source Eatough et al. 2010\n",
    "Image(url= \"images/ralph.png\", width=800, height=800)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Bates et al. 2013: Single-Layer Perceptron with different features compared to Eatough et al. 2010. Rejected 99% of bad candidates and re-detected 85% of pulsars in HTRU Med-Lat"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 304,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<img src=\"images/bates.png\" width=\"800\" height=\"400\"/>"
      ],
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 304,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Source: Bates et al. 2013\n",
    "Image(url= \"images/bates.png\", width=800, height=400)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Lee et al. 2013: Pulsar Evaluation Algorithm for Candidate Extraction (PEACE). Hand Crafted features based on domain knowledge. Discovered atleast 47 pulsars (5 MSPs)\n",
    "\n",
    "Morello et al. 2014: Straightforward Pulsar Identification using Neural Networks (SPINN). Basically a perceptron built on scores from features chosen based on domain knowledge. Successfully discovered atleast 4 pulsars from HTRU Med Lat."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 303,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<img src=\"images/peace.png\" width=\"800\" height=\"400\"/>"
      ],
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 303,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Source: Lee et al. 2013\n",
    "Image(url= \"images/peace.png\", width=800, height=400)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Zhu et al. 2014: Pulsar Image Classification (PICS). Ensemble-Learning Algorithm. New model available called PICS-ResNet.\n",
    "\n",
    "https://github.com/zhuww/ubc_AI\n",
    "\n",
    "\n",
    "https://github.com/dzuwhf/PICS-ResNet"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 302,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<img src=\"images/pics.png\" width=\"800\" height=\"400\"/>"
      ],
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 302,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(url= \"images/pics.png\", width=800, height=400)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Lyon et. al 2016: Online Machine learning built to handle large data volumes in streaming mode. Uses an algorithm called Gaussian Hellinger Very Fast Decision Tree (GHVFDT)\n",
    "\n",
    "\n",
    "https://github.com/scienceguyrob/GHVFDT"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 301,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<img src=\"images/lyon.png\" width=\"800\" height=\"400\"/>"
      ],
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 301,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Source: Lyon et al. 2016\n",
    "Image(url= \"images/lyon.png\", width=800, height=400)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    " Agarwal et al. 2019: Deeper neural networks for Fast Radio Burst detection. Uses Transfer-Learning on Convolutional Neural Networks to identify FRB candidates. They have made an open-source tool available (FETCH) to check your candidates with their trained model.\n",
    " \n",
    "https://github.com/devanshkv/fetch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 300,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<img src=\"images/agarwal.png\" width=\"800\" height=\"400\"/>"
      ],
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 300,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Source: Agarwal et al. 2019\n",
    "Image(url= \"images/agarwal.png\", width=800, height=400)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    " <h2><center>What is Machine Learning?</center></h2>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 299,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<img src=\"images/definition.png\" width=\"800\" height=\"400\"/>"
      ],
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 299,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Source MIT Lecture 1 Introduction to Deep Learning\n",
    "Image(url= \"images/definition.png\", width=800, height=400)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 298,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<img src=\"images/ml_types.jpg\" width=\"800\" height=\"400\"/>"
      ],
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 298,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(url= \"images/ml_types.jpg\", width=800, height=400)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 297,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<img src=\"images/MLvsDL.png\" width=\"800\" height=\"400\"/>"
      ],
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 297,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(url= \"images/MLvsDL.png\", width=800, height=400)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 296,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<img src=\"images/perceptron.png\" width=\"800\" height=\"400\"/>"
      ],
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 296,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(url= \"images/perceptron.png\", width=800, height=400)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 295,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<img src=\"images/activation_fn.png\" width=\"800\" height=\"400\"/>"
      ],
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 295,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(url= \"images/activation_fn.png\", width=800, height=400)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 294,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<img src=\"images/single_layer_nn.png\" width=\"800\" height=\"400\"/>"
      ],
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 294,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(url= \"images/single_layer_nn.png\", width=800, height=400)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 293,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<img src=\"images/deep_layer_nn.png\" width=\"800\" height=\"400\"/>"
      ],
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 293,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(url= \"images/deep_layer_nn.png\", width=800, height=400)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h2><center>How do these algorithms learn?</center></h2>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 292,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<img src=\"images/loss_function.png\" width=\"800\" height=\"400\"/>"
      ],
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 292,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "# \n",
    "Image(url= \"images/loss_function.png\", width=800, height=400)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 290,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<img src=\"images/gradient_descent.png\" width=\"800\" height=\"400\"/>"
      ],
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 290,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(url= \"images/gradient_descent.png\", width=800, height=400)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 289,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<img src=\"https://thumbs.gfycat.com/GrandSlimyBadger-size_restricted.gif\" width=\"1000\" height=\"500\"/>"
      ],
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 289,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(url=\"https://thumbs.gfycat.com/GrandSlimyBadger-size_restricted.gif\", width=1000, height=500)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 288,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<img src=\"images/overfitting.png\" width=\"800\" height=\"400\"/>"
      ],
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 288,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(url= \"images/overfitting.png\", width=800, height=400)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 287,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<img src=\"images/dropout1.png\" width=\"800\" height=\"400\"/>"
      ],
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 287,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(url= \"images/dropout1.png\", width=800, height=400)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 286,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<img src=\"images/dropout2.png\" width=\"800\" height=\"400\"/>"
      ],
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 286,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(url= \"images/dropout2.png\", width=800, height=400)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 137,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Source: Gfycat\n",
    "#Image(filename=\"neural_net.gif\", width=800, height=400)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 285,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<img src=\"https://thumbs.gfycat.com/DeadlyDeafeningAtlanticblackgoby-size_restricted.gif\" width=\"1000\" height=\"500\"/>"
      ],
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 285,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(url=\"https://thumbs.gfycat.com/DeadlyDeafeningAtlanticblackgoby-size_restricted.gif\", width=1000, height=500)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 284,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<img src=\"https://thumbs.gfycat.com/ThirdTameBluebird-size_restricted.gif\" width=\"1000\" height=\"500\"/>"
      ],
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 284,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(url=\"https://thumbs.gfycat.com/ThirdTameBluebird-size_restricted.gif\", width=1000, height=500)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 282,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<img src=\"images/overview_nn.png\" width=\"1000\" height=\"400\"/>"
      ],
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 282,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(url= \"images/overview_nn.png\", width=1000, height=400)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Choosing the right algorithm for your problem is difficult!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 281,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<img src=\"images/sk_learn_algo_cheat_sheet.png\" width=\"1000\" height=\"400\"/>"
      ],
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 281,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(url= \"images/sk_learn_algo_cheat_sheet.png\", width=1000, height=400)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### What output to expect?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 309,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<img src=\"images/binary_classification.png\" width=\"1000\" height=\"400\"/>"
      ],
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 309,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(url= \"images/binary_classification.png\", width=1000, height=400)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 310,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<img src=\"images/prediction_matrix.png\" width=\"1000\" height=\"400\"/>"
      ],
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 310,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(url= \"images/prediction_matrix.png\", width=1000, height=400)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 311,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<img src=\"images/fscore.png\" width=\"1000\" height=\"400\"/>"
      ],
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 311,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(url= \"images/fscore.png\", width=1000, height=400)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 312,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<img src=\"images/imbalanced.png\" width=\"1000\" height=\"400\"/>"
      ],
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 312,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Source: Lecture: Washington State University\n",
    "Image(url= \"images/imbalanced.png\", width=1000, height=400)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 320,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<img src=\"images/yann_lecun.jpg\" width=\"400\" height=\"400\"/>"
      ],
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 320,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(url= \"images/yann_lecun.jpg\", width=400, height=400)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 318,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<img src=\"images/cnn_layout.png\" width=\"1000\" height=\"400\"/>"
      ],
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 318,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(url= \"images/cnn_layout.png\", width=1000, height=400)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# <h2><center>Practical Session</center></h2>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Let's train some models."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 422,
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Conv2D, MaxPooling2D\n",
    "from keras.layers import Activation, Dropout, Flatten, Dense\n",
    "from sklearn.metrics import confusion_matrix, f1_score, precision_score, recall_score\n",
    "from keras import backend as K\n",
    "from keras import optimizers\n",
    "import numpy as np\n",
    "import math, time\n",
    "import itertools\n",
    "from sklearn.model_selection import train_test_split\n",
    "from keras.callbacks import TensorBoard\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 423,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total Number of Pulsar Examples is 2847\n",
      "Total Number of Non-Pulsar Examples is 3100\n"
     ]
    }
   ],
   "source": [
    "# Load all data\n",
    "time_phase_pulsars = np.load('input_data/time_phase_data_pulsars.npy')\n",
    "time_phase_nonpulsars = np.load('input_data/time_phase_data_nonpulsars.npy')\n",
    "\n",
    "freq_phase_pulsars = np.load('input_data/freq_phase_data_pulsars.npy')\n",
    "freq_phase_nonpulsars = np.load('input_data/freq_phase_data_nonpulsars.npy')\n",
    "\n",
    "pulse_profile_pulsars = np.load('input_data/pulse_profile_data_pulsars.npy')\n",
    "pulse_profile_nonpulsars = np.load('input_data/pulse_profile_data_nonpulsars.npy')\n",
    "\n",
    "dm_curve_pulsars = np.load('input_data/dm_curve_data_pulsars.npy')\n",
    "dm_curve_nonpulsars = np.load('input_data/dm_curve_data_nonpulsars.npy')\n",
    "\n",
    "pulse_profile_gbncc_pulsars = np.load('input_data/pulse_profile_gbncc_test_data_pulsars.npy')\n",
    "pulse_profile_gbncc_nonpulsars = np.load('input_data/pulse_profile_gbncc_test_data_nonpulsars_part1.npy')\n",
    "\n",
    "reshaped_time_phase_pulsars = [np.reshape(f,(48,48,1)) for f in time_phase_pulsars]\n",
    "reshaped_time_phase_nonpulsars = [np.reshape(f,(48,48,1)) for f in time_phase_nonpulsars]\n",
    "\n",
    "reshaped_freq_phase_pulsars = [np.reshape(f,(48,48,1)) for f in freq_phase_pulsars]\n",
    "reshaped_freq_phase_nonpulsars = [np.reshape(f,(48,48,1)) for f in freq_phase_nonpulsars]\n",
    "\n",
    "print('Total Number of Pulsar Examples is %d' %len(dm_curve_pulsars))\n",
    "print('Total Number of Non-Pulsar Examples is %d' %len(dm_curve_nonpulsars))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 424,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Helper Functions \n",
    "def plot_confusion_matrix(cm, classes,\n",
    "                          normalize=False,\n",
    "                          title='Confusion matrix',\n",
    "                          cmap=plt.cm.Blues):\n",
    "    \"\"\"\n",
    "    This function prints and plots the confusion matrix.\n",
    "    Normalization can be applied by setting `normalize=True`.\n",
    "    \"\"\"\n",
    "    if normalize:\n",
    "        cm = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\n",
    "        print(\"Normalized confusion matrix\")\n",
    "    else:\n",
    "        print('Confusion matrix, without normalization')\n",
    "\n",
    "    print(cm)\n",
    "\n",
    "    plt.imshow(cm, interpolation='nearest', cmap=cmap)\n",
    "    plt.title(title)\n",
    "    plt.colorbar()\n",
    "    tick_marks = np.arange(len(classes))\n",
    "    plt.xticks(tick_marks, classes, rotation=45)\n",
    "    plt.yticks(tick_marks, classes)\n",
    "\n",
    "    fmt = '.5f' if normalize else 'd'\n",
    "    thresh = cm.max() / 2.\n",
    "    for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):\n",
    "        plt.text(j, i, format(cm[i, j], fmt),\n",
    "                 horizontalalignment=\"center\",\n",
    "                 color=\"white\" if cm[i, j] > thresh else \"black\")\n",
    "\n",
    "    plt.ylabel('True label')\n",
    "    plt.xlabel('Predicted label')\n",
    "    plt.tight_layout()\n",
    "    \n",
    "def plot_images(images, cls_true, cls_pred=None):\n",
    "    assert len(images) == len(cls_true) == 9\n",
    "    \n",
    "    # Create figure with 3x3 sub-plots.\n",
    "    fig, axes = plt.subplots(3, 3)\n",
    "    fig.subplots_adjust(hspace=0.3, wspace=0.3)\n",
    "    img_shape = 48, 48\n",
    "    for i, ax in enumerate(axes.flat):\n",
    "        # Plot image.\n",
    "        ax.imshow(images[i].reshape(img_shape), cmap='binary')\n",
    "\n",
    "        # Show true and predicted classes.\n",
    "        if cls_pred is None:\n",
    "            xlabel = \"True: {0}\".format(cls_true[i])\n",
    "        else:\n",
    "            xlabel = \"True: {0}, Pred: {1}\".format(cls_true[i], cls_pred[i])\n",
    "\n",
    "        # Show the classes as the label on the x-axis.\n",
    "        ax.set_xlabel(xlabel)\n",
    "        \n",
    "        # Remove ticks from the plot.\n",
    "        ax.set_xticks([])\n",
    "        ax.set_yticks([])\n",
    "    \n",
    "    # Ensure the plot is shown correctly with multiple plots\n",
    "    # in a single Notebook cell.\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h3><center>Let's Visualize our Input Data</center></h3>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 425,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Pulse Profile\n",
    "sample_pulse_profile_pulsar = pulse_profile_pulsars[0]\n",
    "sample_pulse_profile_nonpulsar = pulse_profile_nonpulsars[0]\n",
    "\n",
    "fig, axs = plt.subplots(nrows=1, ncols=2, sharex=True, figsize=(20,8))\n",
    "ax = axs[0]\n",
    "ax.plot(sample_pulse_profile_pulsar, color='black', linestyle='-.', linewidth=2)\n",
    "ax.set_title('Real Pulsar', size=18)\n",
    "ax.set_xlabel('Phase Bins')\n",
    "ax.set_ylabel('Normalised Intensity')\n",
    "ax = axs[1]\n",
    "ax.plot(sample_pulse_profile_nonpulsar, color='black', linestyle='-.', linewidth=2)\n",
    "ax.set_title('Non-Pulsar', size=18)\n",
    "ax.set_xlabel('Phase Bins')\n",
    "fig.suptitle('Normalised Pulse Profile', size=24)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 426,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Pulse Profile\n",
    "sample_pulse_profile_pulsar = pulse_profile_gbncc_pulsars[0]\n",
    "sample_pulse_profile_nonpulsar = pulse_profile_gbncc_nonpulsars[0]\n",
    "\n",
    "fig, axs = plt.subplots(nrows=1, ncols=2, sharex=True, figsize=(20,8))\n",
    "ax = axs[0]\n",
    "ax.plot(sample_pulse_profile_pulsar, color='black', linestyle='-.', linewidth=2)\n",
    "ax.set_title('Real Pulsar', size=18)\n",
    "ax.set_xlabel('Phase Bins')\n",
    "ax.set_ylabel('Normalised Intensity')\n",
    "ax = axs[1]\n",
    "ax.plot(sample_pulse_profile_nonpulsar, color='black', linestyle='-.', linewidth=2)\n",
    "ax.set_title('Non-Pulsar', size=18)\n",
    "ax.set_xlabel('Phase Bins')\n",
    "fig.suptitle('Real GBNCC data Normalised Pulse Profile', size=24)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 427,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#DM Curve\n",
    "sample_dm_curve_pulsar = dm_curve_pulsars[2]\n",
    "sample_dm_curve_nonpulsar = dm_curve_nonpulsars[22]\n",
    "\n",
    "fig, axs = plt.subplots(nrows=1, ncols=2, sharex=True, figsize=(20,8))\n",
    "ax = axs[0]\n",
    "ax.plot(sample_dm_curve_pulsar, color='black', linestyle='-.', linewidth=2)\n",
    "ax.set_title('Real Pulsar', size=18)\n",
    "ax.set_xlabel('Phase Bins')\n",
    "ax.set_ylabel('Normalised Intensity')\n",
    "ax = axs[1]\n",
    "ax.plot(sample_dm_curve_nonpulsar, color='black', linestyle='-.', linewidth=2)\n",
    "ax.set_title('Non-Pulsar', size=18)\n",
    "ax.set_xlabel('Phase Bins')\n",
    "fig.suptitle('Normalised and Centered DM Curve', size=24)\n",
    "\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 428,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Time vs Phase plot\n",
    "sample_time_phase_pulsar = np.reshape(time_phase_pulsars[0],(48,48))\n",
    "sample_time_phase_nonpulsar = np.reshape(time_phase_nonpulsars[0], (48,48))\n",
    "\n",
    "fig, axs = plt.subplots(nrows=1, ncols=2, sharex=True, figsize=(20,8))\n",
    "ax = axs[0]\n",
    "ax.imshow(sample_time_phase_pulsar, origin='lower')\n",
    "ax.set_title('Real Pulsar', size=18)\n",
    "ax.set_xlabel('Phase')\n",
    "ax.set_ylabel('Obs.Time')\n",
    "ax = axs[1]\n",
    "ax.imshow(sample_time_phase_nonpulsar, origin='lower')\n",
    "ax.set_title('Non-Pulsar', size=18)\n",
    "ax.set_xlabel('Phase')\n",
    "fig.suptitle('Time vs Phase Curve', size=24)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 429,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Freq vs Phase plot\n",
    "sample_freq_phase_pulsar = np.reshape(freq_phase_pulsars[34],(48,48))\n",
    "sample_freq_phase_nonpulsar = np.reshape(freq_phase_nonpulsars[34], (48,48))\n",
    "\n",
    "fig, axs = plt.subplots(nrows=1, ncols=2, sharex=True, figsize=(20,8))\n",
    "ax = axs[0]\n",
    "ax.imshow(sample_freq_phase_pulsar, origin='lower')\n",
    "ax.set_title('Real Pulsar', size=18)\n",
    "ax.set_xlabel('Phase')\n",
    "ax.set_ylabel('Obs.Frequency')\n",
    "ax = axs[1]\n",
    "ax.imshow(sample_freq_phase_nonpulsar, origin='lower')\n",
    "ax.set_title('Non-Pulsar', size=18)\n",
    "ax.set_xlabel('Phase')\n",
    "fig.suptitle('Frequency vs Phase Curve', size=24)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    " <h3><center>Single-Layer Perceptron (Similar to Eatough et al.2010 and Bates et al. 2012)</center></h3>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Build your model and compile"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 430,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "dense_71 (Dense)             (None, 64)                4160      \n",
      "_________________________________________________________________\n",
      "dropout_16 (Dropout)         (None, 64)                0         \n",
      "_________________________________________________________________\n",
      "dense_72 (Dense)             (None, 1)                 65        \n",
      "=================================================================\n",
      "Total params: 4,225\n",
      "Trainable params: 4,225\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "# Initialising the ANN\n",
    "classifier = Sequential()\n",
    "# Adding the Single Perceptron or Shallow network\n",
    "classifier.add(Dense(activation=\"relu\", input_dim=64, kernel_initializer=\"uniform\", units=64))\n",
    "# Adding dropout to prevent overfitting\n",
    "classifier.add(Dropout(rate=0.3))\n",
    "# Adding the output layer\n",
    "classifier.add(Dense(activation=\"sigmoid\", kernel_initializer=\"uniform\", units=1))\n",
    "# criterion loss and optimizer \n",
    "classifier.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])\n",
    "\n",
    "#Read about adam optimiser here: https://machinelearningmastery.com/adam-optimization-algorithm-for-deep-learning/\n",
    "classifier.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Lets test this network on Pulse Profile. Step 1 is to split your data into train and test datasets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 431,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total Samples in one epoch is 5947\n",
      "Total Number of Training Samples is 4757\n",
      "Total Number of Test Samples is 1190\n"
     ]
    }
   ],
   "source": [
    "# Pulsars are given label 1 and Non-Pulsars are given label 0\n",
    "label_pulsars = np.ones(len(dm_curve_pulsars))\n",
    "label_nonpulsars = np.zeros(len(dm_curve_nonpulsars))\n",
    "\n",
    "\n",
    "pulse_profile_data_combined = np.concatenate((pulse_profile_pulsars, pulse_profile_nonpulsars), axis = 0)\n",
    "pulse_profile_label_combined = np.concatenate((label_pulsars, label_nonpulsars), axis = 0)\n",
    "\n",
    "\n",
    "pulse_profile_train, pulse_profile_test, pulse_profile_label_train, \\\n",
    "pulse_profile_label_test = train_test_split(pulse_profile_data_combined, pulse_profile_label_combined, \\\n",
    "                                         test_size=0.2, random_state=42)\n",
    "\n",
    "print('Total Samples in one epoch is %d' %len(pulse_profile_data_combined))\n",
    "print('Total Number of Training Samples is %d' %len(pulse_profile_train))\n",
    "print('Total Number of Test Samples is %d' %len(pulse_profile_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Start Training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 432,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 3805 samples, validate on 952 samples\n",
      "Epoch 1/50\n",
      "3805/3805 [==============================] - 1s 201us/step - loss: 0.5615 - acc: 0.8460 - val_loss: 0.4017 - val_acc: 0.9454\n",
      "Epoch 2/50\n",
      "3805/3805 [==============================] - 0s 43us/step - loss: 0.2725 - acc: 0.9703 - val_loss: 0.1713 - val_acc: 0.9895\n",
      "Epoch 3/50\n",
      "3805/3805 [==============================] - 0s 36us/step - loss: 0.1155 - acc: 0.9908 - val_loss: 0.0761 - val_acc: 0.9958\n",
      "Epoch 4/50\n",
      "3805/3805 [==============================] - 0s 30us/step - loss: 0.0610 - acc: 0.9926 - val_loss: 0.0445 - val_acc: 0.9958\n",
      "Epoch 5/50\n",
      "3805/3805 [==============================] - 0s 37us/step - loss: 0.0435 - acc: 0.9929 - val_loss: 0.0314 - val_acc: 0.9968\n",
      "Epoch 6/50\n",
      "3805/3805 [==============================] - 0s 30us/step - loss: 0.0334 - acc: 0.9926 - val_loss: 0.0242 - val_acc: 0.9968\n",
      "Epoch 7/50\n",
      "3805/3805 [==============================] - 0s 26us/step - loss: 0.0301 - acc: 0.9934 - val_loss: 0.0202 - val_acc: 0.9968\n",
      "Epoch 8/50\n",
      "3805/3805 [==============================] - 0s 24us/step - loss: 0.0270 - acc: 0.9937 - val_loss: 0.0179 - val_acc: 0.9958\n",
      "Epoch 9/50\n",
      "3805/3805 [==============================] - 0s 26us/step - loss: 0.0234 - acc: 0.9942 - val_loss: 0.0159 - val_acc: 0.9958\n",
      "Epoch 10/50\n",
      "3805/3805 [==============================] - 0s 24us/step - loss: 0.0225 - acc: 0.9934 - val_loss: 0.0147 - val_acc: 0.9958\n",
      "Epoch 11/50\n",
      "3805/3805 [==============================] - 0s 26us/step - loss: 0.0211 - acc: 0.9932 - val_loss: 0.0137 - val_acc: 0.9979\n",
      "Epoch 12/50\n",
      "3805/3805 [==============================] - 0s 26us/step - loss: 0.0202 - acc: 0.9940 - val_loss: 0.0141 - val_acc: 0.9958\n",
      "Epoch 13/50\n",
      "3805/3805 [==============================] - 0s 26us/step - loss: 0.0208 - acc: 0.9950 - val_loss: 0.0138 - val_acc: 0.9979\n",
      "Epoch 14/50\n",
      "3805/3805 [==============================] - 0s 28us/step - loss: 0.0183 - acc: 0.9945 - val_loss: 0.0124 - val_acc: 0.9968\n",
      "Epoch 15/50\n",
      "3805/3805 [==============================] - 0s 26us/step - loss: 0.0183 - acc: 0.9942 - val_loss: 0.0116 - val_acc: 0.9968\n",
      "Epoch 16/50\n",
      "3805/3805 [==============================] - 0s 26us/step - loss: 0.0174 - acc: 0.9942 - val_loss: 0.0109 - val_acc: 0.9958\n",
      "Epoch 17/50\n",
      "3805/3805 [==============================] - 0s 28us/step - loss: 0.0187 - acc: 0.9937 - val_loss: 0.0115 - val_acc: 0.9968\n",
      "Epoch 18/50\n",
      "3805/3805 [==============================] - 0s 27us/step - loss: 0.0163 - acc: 0.9942 - val_loss: 0.0095 - val_acc: 0.9968\n",
      "Epoch 19/50\n",
      "3805/3805 [==============================] - 0s 26us/step - loss: 0.0161 - acc: 0.9945 - val_loss: 0.0095 - val_acc: 0.9968\n",
      "Epoch 20/50\n",
      "3805/3805 [==============================] - 0s 24us/step - loss: 0.0156 - acc: 0.9950 - val_loss: 0.0093 - val_acc: 0.9968\n",
      "Epoch 21/50\n",
      "3805/3805 [==============================] - 0s 27us/step - loss: 0.0139 - acc: 0.9958 - val_loss: 0.0092 - val_acc: 0.9968\n",
      "Epoch 22/50\n",
      "3805/3805 [==============================] - 0s 26us/step - loss: 0.0143 - acc: 0.9947 - val_loss: 0.0095 - val_acc: 0.9968\n",
      "Epoch 23/50\n",
      "3805/3805 [==============================] - 0s 25us/step - loss: 0.0124 - acc: 0.9961 - val_loss: 0.0089 - val_acc: 0.9968\n",
      "Epoch 24/50\n",
      "3805/3805 [==============================] - 0s 29us/step - loss: 0.0135 - acc: 0.9947 - val_loss: 0.0084 - val_acc: 0.9968\n",
      "Epoch 25/50\n",
      "3805/3805 [==============================] - 0s 29us/step - loss: 0.0133 - acc: 0.9958 - val_loss: 0.0083 - val_acc: 0.9968\n",
      "Epoch 26/50\n",
      "3805/3805 [==============================] - 0s 26us/step - loss: 0.0111 - acc: 0.9968 - val_loss: 0.0082 - val_acc: 0.9968\n",
      "Epoch 27/50\n",
      "3805/3805 [==============================] - 0s 27us/step - loss: 0.0135 - acc: 0.9958 - val_loss: 0.0091 - val_acc: 0.9968\n",
      "Epoch 28/50\n",
      "3805/3805 [==============================] - 0s 26us/step - loss: 0.0125 - acc: 0.9955 - val_loss: 0.0077 - val_acc: 0.9968\n",
      "Epoch 29/50\n",
      "3805/3805 [==============================] - 0s 24us/step - loss: 0.0116 - acc: 0.9966 - val_loss: 0.0079 - val_acc: 0.9968\n",
      "Epoch 30/50\n",
      "3805/3805 [==============================] - 0s 27us/step - loss: 0.0105 - acc: 0.9966 - val_loss: 0.0082 - val_acc: 0.9968\n",
      "Epoch 31/50\n",
      "3805/3805 [==============================] - 0s 25us/step - loss: 0.0098 - acc: 0.9966 - val_loss: 0.0076 - val_acc: 0.9968\n",
      "Epoch 32/50\n",
      "3805/3805 [==============================] - 0s 24us/step - loss: 0.0111 - acc: 0.9963 - val_loss: 0.0072 - val_acc: 0.9968\n",
      "Epoch 33/50\n",
      "3805/3805 [==============================] - 0s 26us/step - loss: 0.0118 - acc: 0.9963 - val_loss: 0.0068 - val_acc: 0.9979\n",
      "Epoch 34/50\n",
      "3805/3805 [==============================] - 0s 25us/step - loss: 0.0101 - acc: 0.9974 - val_loss: 0.0069 - val_acc: 0.9979\n",
      "Epoch 35/50\n",
      "3805/3805 [==============================] - 0s 25us/step - loss: 0.0102 - acc: 0.9963 - val_loss: 0.0071 - val_acc: 0.9979\n",
      "Epoch 36/50\n",
      "3805/3805 [==============================] - 0s 25us/step - loss: 0.0089 - acc: 0.9979 - val_loss: 0.0072 - val_acc: 0.9979\n",
      "Epoch 37/50\n",
      "3805/3805 [==============================] - 0s 28us/step - loss: 0.0077 - acc: 0.9976 - val_loss: 0.0068 - val_acc: 0.9979\n",
      "Epoch 38/50\n",
      "3805/3805 [==============================] - 0s 26us/step - loss: 0.0092 - acc: 0.9971 - val_loss: 0.0067 - val_acc: 0.9979\n",
      "Epoch 39/50\n",
      "3805/3805 [==============================] - 0s 23us/step - loss: 0.0098 - acc: 0.9963 - val_loss: 0.0069 - val_acc: 0.9979\n",
      "Epoch 40/50\n",
      "3805/3805 [==============================] - 0s 24us/step - loss: 0.0074 - acc: 0.9979 - val_loss: 0.0070 - val_acc: 0.9968\n",
      "Epoch 41/50\n",
      "3805/3805 [==============================] - 0s 23us/step - loss: 0.0083 - acc: 0.9974 - val_loss: 0.0067 - val_acc: 0.9979\n",
      "Epoch 42/50\n",
      "3805/3805 [==============================] - 0s 22us/step - loss: 0.0082 - acc: 0.9984 - val_loss: 0.0065 - val_acc: 0.9979\n",
      "Epoch 43/50\n",
      "3805/3805 [==============================] - 0s 24us/step - loss: 0.0088 - acc: 0.9976 - val_loss: 0.0064 - val_acc: 0.9979\n",
      "Epoch 44/50\n",
      "3805/3805 [==============================] - 0s 23us/step - loss: 0.0077 - acc: 0.9989 - val_loss: 0.0065 - val_acc: 0.9979\n",
      "Epoch 45/50\n",
      "3805/3805 [==============================] - 0s 24us/step - loss: 0.0073 - acc: 0.9982 - val_loss: 0.0064 - val_acc: 0.9979\n",
      "Epoch 46/50\n",
      "3805/3805 [==============================] - 0s 25us/step - loss: 0.0074 - acc: 0.9982 - val_loss: 0.0065 - val_acc: 0.9979\n",
      "Epoch 47/50\n",
      "3805/3805 [==============================] - 0s 23us/step - loss: 0.0060 - acc: 0.9984 - val_loss: 0.0066 - val_acc: 0.9979\n",
      "Epoch 48/50\n",
      "3805/3805 [==============================] - 0s 23us/step - loss: 0.0083 - acc: 0.9976 - val_loss: 0.0066 - val_acc: 0.9979\n",
      "Epoch 49/50\n",
      "3805/3805 [==============================] - 0s 23us/step - loss: 0.0075 - acc: 0.9982 - val_loss: 0.0064 - val_acc: 0.9979\n",
      "Epoch 50/50\n",
      "3805/3805 [==============================] - 0s 23us/step - loss: 0.0069 - acc: 0.9982 - val_loss: 0.0064 - val_acc: 0.9979\n"
     ]
    }
   ],
   "source": [
    "classifier.fit(pulse_profile_train, pulse_profile_label_train, validation_split=0.2, batch_size=100, epochs=50)\n",
    "classifier.save_weights('pulse_profile_perceptron.h5') # Save weights to avoid training in future"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Let's Test our Results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 433,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "F Score: 0.993127147766 Precision: 0.991423670669 Recall: 0.994836488812 Accuracy: 0.993277310924\n"
     ]
    }
   ],
   "source": [
    "y_pred = classifier.predict(pulse_profile_test)\n",
    "classified_results = np.rint(y_pred)\n",
    "classified_results = np.reshape(classified_results, len(classified_results))\n",
    "f_score = f1_score(pulse_profile_label_test, classified_results, average='binary')\n",
    "precision = precision_score(pulse_profile_label_test, classified_results, average='binary')\n",
    "recall = recall_score(pulse_profile_label_test, classified_results, average='binary')\n",
    "accuracy = (pulse_profile_label_test == classified_results).sum()/len(pulse_profile_label_test)\n",
    "print('F Score:', f_score, 'Precision:', precision, 'Recall:', recall, 'Accuracy:', accuracy)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 434,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion matrix, without normalization\n",
      "[[604   5]\n",
      " [  3 578]]\n",
      "Normalized confusion matrix\n",
      "[[ 0.99178982  0.00821018]\n",
      " [ 0.00516351  0.99483649]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "class_names = ['Non-Pulsar','Pulsar']\n",
    "\n",
    "cnf_matrix = confusion_matrix(pulse_profile_label_test, classified_results)\n",
    "plt.figure()\n",
    "plot_confusion_matrix(cnf_matrix, classes=class_names, title='Confusion matrix, without normalization')\n",
    "plt.figure()\n",
    "plot_confusion_matrix(cnf_matrix, classes=class_names, normalize=True, title='Confusion matrix, with normalization')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Test Network on real GBNCC Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 435,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "F Score: 0.0713128038898 Precision: 0.0391737891738 Recall: 0.397111913357 Accuracy: 0.873838566207\n",
      "Confusion matrix, without normalization\n",
      "[[19734  2698]\n",
      " [  167   110]]\n",
      "Normalized confusion matrix\n",
      "[[ 0.87972539  0.12027461]\n",
      " [ 0.60288809  0.39711191]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Pulsars are given label 1 and Non-Pulsars are given label 0\n",
    "label_gbncc_pulsars = np.ones(len(pulse_profile_gbncc_pulsars))\n",
    "label_gbncc_nonpulsars = np.zeros(len(pulse_profile_gbncc_nonpulsars))\n",
    "\n",
    "\n",
    "pulse_profile_data_combined = np.concatenate((pulse_profile_gbncc_pulsars, pulse_profile_gbncc_nonpulsars), axis = 0)\n",
    "pulse_profile_label_combined = np.concatenate((label_gbncc_pulsars, label_gbncc_nonpulsars), axis = 0)\n",
    "\n",
    "y_pred = classifier.predict(pulse_profile_data_combined)\n",
    "classified_results = np.rint(y_pred)\n",
    "classified_results = np.reshape(classified_results, len(classified_results))\n",
    "f_score = f1_score(pulse_profile_label_combined, classified_results, average='binary')\n",
    "precision = precision_score(pulse_profile_label_combined, classified_results, average='binary')\n",
    "recall = recall_score(pulse_profile_label_combined, classified_results, average='binary')\n",
    "accuracy = (pulse_profile_label_combined == classified_results).sum()/len(pulse_profile_data_combined)\n",
    "print('F Score:', f_score, 'Precision:', precision, 'Recall:', recall, 'Accuracy:', accuracy)\n",
    "\n",
    "class_names = ['Non-Pulsar','Pulsar']\n",
    "\n",
    "cnf_matrix = confusion_matrix(pulse_profile_label_combined, classified_results)\n",
    "plt.figure()\n",
    "plot_confusion_matrix(cnf_matrix, classes=class_names, title='Confusion matrix, without normalization')\n",
    "plt.figure()\n",
    "plot_confusion_matrix(cnf_matrix, classes=class_names, normalize=True, title='Confusion matrix, with normalization')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Let's use scikit-learn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 436,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Naive Bayes Classifier accuracy:  99.1596638655 %.\n",
      "Support Vector Machine Classifier Accuracy: 99.4117647059 %.\n"
     ]
    }
   ],
   "source": [
    "#Naive Bayes\n",
    "from sklearn.naive_bayes import GaussianNB\n",
    "\n",
    "classifier = GaussianNB()\n",
    "\n",
    "# First train the classifier with call to fit.\n",
    "classifier.fit(pulse_profile_train, pulse_profile_label_train)\n",
    "\n",
    "# Now obtain the classifiers 'score'\n",
    "accuracy = classifier.score(pulse_profile_test, pulse_profile_label_test)\n",
    "print (\"Naive Bayes Classifier accuracy: \", (100* accuracy), \"%.\")\n",
    "\n",
    "#Support Vector Machine\n",
    "from sklearn import svm\n",
    "clf = svm.SVC(gamma='scale')\n",
    "clf.fit(pulse_profile_train, pulse_profile_label_train)\n",
    "classified_results = clf.predict(pulse_profile_test)\n",
    "\n",
    "accuracy = (pulse_profile_label_test == classified_results).sum()/len(pulse_profile_label_test)\n",
    "print('Support Vector Machine Classifier Accuracy:', accuracy*100, \"%.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Convolutional Neural Network for 2D Images"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 437,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define Image Dimensions\n",
    "\n",
    "img_width, img_height = 48, 48\n",
    "input_shape = (48, 48, 1)\n",
    "batch_size = 200\n",
    "#tensor_board = TensorBoard(log_dir='./Graph', histogram_freq=0, write_graph=True, write_images=True)\n",
    "#NAME = \"Simple-CNN\"\n",
    "#tensorboard = TensorBoard(log_dir=\"logs/{}\".format(NAME))\n",
    "if K.image_data_format() == 'channels_first':\n",
    "    input_shape = (1, img_width, img_height)\n",
    "else:\n",
    "    input_shape = (img_width, img_height, 1)\n",
    "\n",
    "# Generator is used to avoid reading entire data into system memory.\n",
    "def generator(batch_size,from_list_x,from_list_y):\n",
    "\n",
    "    assert len(from_list_x) == len(from_list_y)\n",
    "    total_size = len(from_list_x)\n",
    "\n",
    "    while True:\n",
    "\n",
    "        for i in range(0,total_size,batch_size):\n",
    "            yield np.array(from_list_x[i:i+batch_size]), np.array(from_list_y[i:i+batch_size])\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 438,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 9 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "time_phase_data = time_phase_pulsars[0:9]\n",
    "time_phase_data = [np.reshape(f, (48,48)) for f in time_phase_data]\n",
    "plot_images(time_phase_data, np.ones(9))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 439,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 9 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "time_phase_data = time_phase_nonpulsars[0:9]\n",
    "time_phase_data = [np.reshape(f, (48,48)) for f in time_phase_data]\n",
    "plot_images(time_phase_data, np.zeros(9))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 440,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d_10 (Conv2D)           (None, 46, 46, 32)        320       \n",
      "_________________________________________________________________\n",
      "activation_16 (Activation)   (None, 46, 46, 32)        0         \n",
      "_________________________________________________________________\n",
      "max_pooling2d_10 (MaxPooling (None, 23, 23, 32)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_11 (Conv2D)           (None, 21, 21, 32)        9248      \n",
      "_________________________________________________________________\n",
      "activation_17 (Activation)   (None, 21, 21, 32)        0         \n",
      "_________________________________________________________________\n",
      "max_pooling2d_11 (MaxPooling (None, 10, 10, 32)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_12 (Conv2D)           (None, 8, 8, 64)          18496     \n",
      "_________________________________________________________________\n",
      "activation_18 (Activation)   (None, 8, 8, 64)          0         \n",
      "_________________________________________________________________\n",
      "max_pooling2d_12 (MaxPooling (None, 4, 4, 64)          0         \n",
      "_________________________________________________________________\n",
      "flatten_4 (Flatten)          (None, 1024)              0         \n",
      "_________________________________________________________________\n",
      "dense_73 (Dense)             (None, 64)                65600     \n",
      "_________________________________________________________________\n",
      "activation_19 (Activation)   (None, 64)                0         \n",
      "_________________________________________________________________\n",
      "dropout_17 (Dropout)         (None, 64)                0         \n",
      "_________________________________________________________________\n",
      "dense_74 (Dense)             (None, 1)                 65        \n",
      "_________________________________________________________________\n",
      "activation_20 (Activation)   (None, 1)                 0         \n",
      "=================================================================\n",
      "Total params: 93,729\n",
      "Trainable params: 93,729\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n",
      "None\n"
     ]
    }
   ],
   "source": [
    "model = Sequential()\n",
    "model.add(Conv2D(32, (3, 3), input_shape=input_shape))\n",
    "model.add(Activation('relu'))\n",
    "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
    "\n",
    "model.add(Conv2D(32, (3, 3)))\n",
    "model.add(Activation('relu'))\n",
    "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
    "\n",
    "model.add(Conv2D(64, (3, 3)))\n",
    "model.add(Activation('relu'))\n",
    "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
    "\n",
    "# After these layers, we convert our 3D feature maps to 1D feature vectors ith the help of 'flatten'. We use 'dropout' layer to prevent overfitting\n",
    "\n",
    "\n",
    "model.add(Flatten())\n",
    "model.add(Dense(64))\n",
    "model.add(Activation('relu'))\n",
    "model.add(Dropout(0.5))\n",
    "model.add(Dense(1))\n",
    "model.add(Activation('sigmoid'))\n",
    "\n",
    "model.compile(loss='binary_crossentropy',\n",
    "              optimizer='rmsprop',\n",
    "              metrics=['accuracy'])\n",
    "print(model.summary())\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 441,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n",
      "23/23 [==============================] - 7s 298ms/step - loss: 0.1304 - acc: 0.9457\n",
      "Epoch 2/20\n",
      "23/23 [==============================] - 6s 272ms/step - loss: 0.0312 - acc: 0.9928\n",
      "Epoch 3/20\n",
      "23/23 [==============================] - 7s 286ms/step - loss: 0.0255 - acc: 0.9939\n",
      "Epoch 4/20\n",
      "23/23 [==============================] - 7s 286ms/step - loss: 0.0186 - acc: 0.9952\n",
      "Epoch 5/20\n",
      "23/23 [==============================] - 7s 290ms/step - loss: 0.0161 - acc: 0.9950\n",
      "Epoch 6/20\n",
      "23/23 [==============================] - 7s 317ms/step - loss: 0.0161 - acc: 0.9952\n",
      "Epoch 7/20\n",
      "23/23 [==============================] - 8s 336ms/step - loss: 0.0119 - acc: 0.9954\n",
      "Epoch 8/20\n",
      "23/23 [==============================] - 7s 322ms/step - loss: 0.0100 - acc: 0.9957\n",
      "Epoch 9/20\n",
      "23/23 [==============================] - 8s 359ms/step - loss: 0.0104 - acc: 0.9959\n",
      "Epoch 10/20\n",
      "23/23 [==============================] - 8s 329ms/step - loss: 0.0127 - acc: 0.9967\n",
      "Epoch 11/20\n",
      "23/23 [==============================] - 7s 286ms/step - loss: 0.0083 - acc: 0.9967\n",
      "Epoch 12/20\n",
      "23/23 [==============================] - 7s 290ms/step - loss: 0.0094 - acc: 0.9976\n",
      "Epoch 13/20\n",
      "23/23 [==============================] - 7s 291ms/step - loss: 0.0096 - acc: 0.9976\n",
      "Epoch 14/20\n",
      "23/23 [==============================] - 7s 309ms/step - loss: 0.0061 - acc: 0.9989\n",
      "Epoch 15/20\n",
      "23/23 [==============================] - 7s 314ms/step - loss: 0.0089 - acc: 0.9987\n",
      "Epoch 16/20\n",
      "23/23 [==============================] - 8s 347ms/step - loss: 0.0056 - acc: 0.9991\n",
      "Epoch 17/20\n",
      "23/23 [==============================] - 7s 302ms/step - loss: 0.0048 - acc: 0.9989\n",
      "Epoch 18/20\n",
      "23/23 [==============================] - 7s 302ms/step - loss: 0.0060 - acc: 0.9993\n",
      "Epoch 19/20\n",
      "23/23 [==============================] - 8s 327ms/step - loss: 0.0049 - acc: 0.9991\n",
      "Epoch 20/20\n",
      "23/23 [==============================] - 8s 329ms/step - loss: 0.0038 - acc: 0.9996\n"
     ]
    }
   ],
   "source": [
    "label_reshaped_time_phase_pulsars = np.ones(len(reshaped_time_phase_pulsars))\n",
    "label_reshaped_time_phase_nonpulsars = np.zeros(len(reshaped_time_phase_nonpulsars))\n",
    "\n",
    "\n",
    "\n",
    "time_phase_data_combined = np.concatenate((reshaped_time_phase_pulsars, reshaped_time_phase_nonpulsars), axis = 0)\n",
    "time_phase_label_combined = np.concatenate((label_reshaped_time_phase_pulsars, label_reshaped_time_phase_nonpulsars), axis = 0)\n",
    "time_phase_train, time_phase_test, time_phase_label_train, time_phase_label_test = train_test_split(time_phase_data_combined, time_phase_label_combined, test_size=0.2, random_state=42)\n",
    "\n",
    "model.fit_generator(generator(batch_size, time_phase_train, time_phase_label_train), steps_per_epoch=len(time_phase_train)//batch_size, epochs=20, callbacks=[tensor_board])\n",
    "model.save_weights('cnn_time_phase.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 381,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "F Score: 0.994836488812 Precision: 0.994836488812 Recall: 0.994836488812\n",
      "Confusion matrix, without normalization\n",
      "[[606   3]\n",
      " [  3 578]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "predict = model.predict_generator(generator(batch_size, time_phase_test, time_phase_label_test), steps=math.ceil(len(time_phase_test)/batch_size))\n",
    "np.save('predictions.npy', predict)\n",
    "classified_results = np.rint(predict)\n",
    "f_score = f1_score(time_phase_label_test, classified_results, average='binary')\n",
    "precision = precision_score(time_phase_label_test, classified_results, average='binary')\n",
    "recall = recall_score(time_phase_label_test, classified_results, average='binary')\n",
    "print('F Score:', f_score, 'Precision:', precision, 'Recall:', recall)\n",
    "\n",
    "class_names = ['Non-Pulsar','Pulsar']\n",
    "\n",
    "cnf_matrix = confusion_matrix(time_phase_label_test, classified_results)\n",
    "plot_confusion_matrix(cnf_matrix, classes=class_names, title='Confusion matrix, without normalization')\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 382,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Normalized confusion matrix\n",
      "[[ 0.99507389  0.00492611]\n",
      " [ 0.00516351  0.99483649]]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUQAAAEYCAYAAAAkpo9KAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAIABJREFUeJzt3Xd8VFXex/HPl4QmICLYCCIoHaSDZVFRUVEQ1LVgWxXL6mNvu5ZVkV3Xtj72fXZ17eCC6CJVsKOw0kSpKqKgEFSKiIK0hN/zxz0Jk5BMhpBkZsjv7WteMveee+65M8kvp91zZWY455yDKskugHPOpQoPiM45F3hAdM65wAOic84FHhCdcy7wgOicc4EHxHIkqaakMZLWShqxE/mcK+nNsixbskg6QtIXqXI+SU0kmaTMiipTeZE0SNKQ8O/GktZJyijjc1To91fRPCACks6RNDP8AH0n6Q1JPcog69OBfYD6ZnZGaTMxs6FmdnwZlKdchcDSLF4aM/vQzFpWVJkKn0/SEkm9Kur8yWJm35pZbTPL3Zl8Cn+nFf39VbRKHxAl3QA8AvyVKHg1Bv4O9C+D7A8AFppZThnklfZ2hVpYWVGk0v/+pRwzq7QvoC6wDjgjTprqRAFzeXg9AlQP+3oCy4AbgRXAd8BFYd/dwGZgSzjHxcAgYEhM3k0AAzLD+wuBr4FfgMXAuTHbJ8ccdzgwA1gb/n94zL73gT8DU0I+bwINirm2vPL/Iab8pwAnAQuBH4HbYtJ3Bz4CfgppnwCqhX0fhGtZH673rJj8/wh8D7yUty0cc1A4R+fwviGwEuiZwHf3AnBj+HdWOPeVhfKtUuh8LwFbgQ2hjH+I+Q4uAL4FVgG3xznv88CTwLjw+U4DDtqB7+ae8N1sAJqFbX8B/hvKNAaoDwwFfg55NInJ41Fgadj3MXBEzL5BhJ+vmOvKBA4Leee9NgJLduY7jTln63ANPwHzgX6Jflap+Ep6AZJ68dAbyCEEpGLSDAamAnsDe4Uf3D+HfT3D8YOBqkSB5FegXuEf0GLex/7Q1go/5C3Dvv2AtuHfFxICIrAnsAY4Pxx3dnhfP+x/H/gKaAHUDO/vK+ba8sp/Zyj/pUQB6WWgDtCW6Be3aUjfBTg0nLcJ8BlwXUx+BjQrIv/7if6w1CziF+pSYAGwGzAR+FuC391AYEz49znhmofH7BsVU4bY8y0BehXxHTwdytcB2AS0Lua8zwOriQJJJlHgGrYD38234XPNDJ/5+8AioiBeN3wWC4FeIc2LwHMx5z+PKGBmEv0h/h6oUfjni0J/bGOOrwpMAu7die90WUxei4DbgGrAMUSBr2VJn1Wqvip7lb0+sMriN2nPBQab2QozW0lU8zs/Zv+WsH+LmY0n+kta2j6WrUA7STXN7Dszm19Emj7Al2b2kpnlmNm/gc+Bk2PSPGdmC81sA/AK0DHOObcA95jZFmAY0AB41Mx+CedfQBQkMLOPzWxqOO8S4J/AUQlc011mtimUpwAze5rol2oa0R+B20vIL88koEdodh4JPAD8Juw7KuzfEXeb2QYzmw3MJlxzMUaa2fTwczOUbZ9vIt/N82Y2P+zfErY9Z2Zfmdla4A3gKzN7O+Q/AuiUd7CZDTGz1eH4h4j+0OzIz9tjREHr9pBfab7TPIcCtYn+4G42s3eBsUR/CPIU91mlpMoeEFcDDUro22oIfBPz/puwLT+PQgH1V6Ifkh1iZuuJmiSXA99JGiepVQLlyStTVsz773egPKttW8d7XsD6IWb/hrzjJbWQNFbS95J+Jup3bRAnb4CVZraxhDRPA+2Ax81sUwlpATCzr4iach2BI4h+EZdLaknpAuKOfGbFpU3ku1laRH6FP+8iP38ASTdJ+izMXPiJqFZZ0neQd+zviWp455jZ1rCtNN9pnobA0ry8gp35WUy6yh4QPyJqHp0SJ81yosGRPI3DttJYT9Q0zLNv7E4zm2hmxxHVlD4nChQllSevTNmlLNOO+D+icjU3s92Jmkoq4Zi4yylJqk3UL/sMMEjSnjtQnklEI/nVzCw7vL8AqAd8Wpry7KREvptSn1/SEUT9nmcSdcvsQdRXWdJ3kHfsn4H+ZvZzzK7SfKd5lgP7FxocqqifxXJRqQNiaKLcCTwp6RRJu0mqKulESQ+EZP8G/iRpL0kNQvohpTzlp8CRYY5YXeDWvB2S9pHUX1ItoiC9jqi5Wdh4oEWYKpQp6SygDVENqbzVIernXBdqr1cU2v8DcOAO5vkoMNPMLiHqfP9H3o4wr+79OMdOAq4i6vyHqD/uKqL+1uKmm5SmjIkq7++mDlGf7EogU9KdwO4lHSRpf6Kuk9+Z2cIi8iztdzqNqNb3h/B705Ooe2BYYpeTeip1QAQI/TA3AH8i+kFbSvRL9XpI8hdgJjAHmAvMCttKc663gOEhr48p+ItSJZRjOdEI6VFs/8OJma0G+hJ1qK8mqjH0NbNVpSnTDrqJaADjF6La6/BC+wcBL0j6SdKZJWUmqT/RwFbedd4AdJZ0bni/P9GIbHEmEf1C5wXEyUQ18A+KPQLuJfoD95Okm0oq446ogO9mIjCBaNDlG6LR4qKa4IUdSzSl7NUw13adpLz+6VJ/p2a2mSgAnkg0Ov93oqD7eSmuLSUojAY5l3IkfQocGwKNc+XOA6JzzgWVvsnsnHN5PCA651zgAdE55wK/2b6MKLOmqVqdZBej0ujUunGyi1CpzJr18Soz26us8svY/QCznO1uXCrANqycaGa9y+qcifCAWEZUrQ7VW5Y408SVkSnTnkh2ESqVmlVV+A6cnWI5G0r8fdn46ZOJ3jFTZjwgOucqngRVynTt2jLhAdE5lxwpuBykB0TnXBJ4DdE557ZRomtIVBwPiM65iie8yeyccxFvMjvn3DbeZHbOOQB5k9k554CoD9GbzM45B15DdM65PAIyvIbonHMRH1RxzjnwJrNzzsXyQRXnnCNqLnuT2TnnAq8hOucceB+ic87F8iazc84RVsxOvfCTeiVyzlUOXkN0zrnAB1Wcc44w7cYHVZxzLuJNZuecC6t/VfEaonPOhWeqJLsQ2/OA6JxLAnkN0Tnn8sj7EJ1zLuIB0TnniIKhqnhAdM45wGuIzjmXLxUDYuoN8zjndn0CVVHcV0LZSL0lfSFpkaRbitjfWNJ7kj6RNEfSSfHy84DonEsKSXFfCRyfATwJnAi0Ac6W1KZQsj8Br5hZJ2AA8Pd4eXqT2TlX4VQ28xC7A4vM7GsAScOA/sCCmDQG7B7+XRdYHi9DD4jOueTY+S7ELGBpzPtlwCGF0gwC3pR0NVAL6BUvQ28yO+cqnhJqMjeQNDPmdVkpznQ28LyZNQJOAl6Sil9mx2uIzrmkSKDJvMrMusbZnw3sH/O+UdgW62KgN4CZfSSpBtAAWFFkmUoqkUs9xx3emtkj72DeqLu46aLjttvfeL96jP/H1UwffisTn76WrL33yN/3l2v6M3PEbcwccRunH985f/tTd5/HZ2MHMXXYLUwddgvtW2Tl73voD6czb9RdTB9+Kx1bNQLgyK7N89NOHXYLa6Y+zMk925fjVSfXmxMn0L5tS9q2asaDD9y33f5NmzZx3jln0bZVM444/BC+WbIkf9+D999L21bNaN+2JW+9ObHAcbm5uRzatROn9e+bv+39997lsG6d6dKxHZdcdAE5OTkAjBk9im6d2nNIl4785pCuTJk8uXwutgKI+LXDBKfkzACaS2oqqRrRoMnoQmm+BY4FkNQaqAGsLC5DryGmmSpVxCO3nEmfK54g+4efmDz0ZsZOmsvnX3+fn+be609l6LjpDB0zjaO6tWDw1f24+I4X6d2jLR1b788hA+6jetVM3vzXtUycsoBf1m8E4LZHXmfk258WON8JPdpwUOO9aNf/brof3ITHbhvAkb/7Gx/M/JJDB0SBod7uuzFv9F28PfWzivsgKlBubi7XXXMl4954i6xGjehxaDf69u1H6zbbBjSff/YZ6u1Rj/mfL+KV4cO4/bY/MuTl4Xy2YAEjhg9j1uz5fLd8OSf17sXcBQvJyIhWi37isUdp2bo1v/z8MwBbt27lkoEX8MbEd2jeogWDB93JkBdf4MKBF3P0McfS9+R+SGLunDmcd86ZzJ73eVI+k50Wpt3sDDPLkXQVMBHIAJ41s/mSBgMzzWw0cCPwtKTriQZYLjQzKy5PryGmmW7tmvDV0lUsyV7NlpxcRkycRd9CNbNWB+7HpOlfADBpxkL69jwYgNYH7svkWYvIzd3Krxs3M/fLbI4/vHXc8/U9qj0vj50OwPS5S6hbpyb7Nti9QJpTe3XizSkL2LBxS1ldZkqZMX06Bx3UjKYHHki1atU446wBjB0zqkCasWNGce75FwBw2m9P5/1338HMGDtmFGecNYDq1avTpGlTDjqoGTOmR5/nsmXLmPDGOC4aeEl+PqtXr6ZatWo0b9ECgGN6HcfrI18DoHbt2vk1p/Xr16fkxOYdUQY1RMxsvJm1MLODzOyesO3OEAwxswVm9hsz62BmHc3szXj5eUBMMw33rsuyH9bkv8/+YQ1Ze9UtkGbuwmz6H9MRgP7HdGD32jXZs24t5iyMAmDNGlWpv0ctjuragkb71ss/btCVJzN9+K08cONpVKuaGc63B8u+jz3fTzSMaYIDnHFCZ16Z8HGZX2uqWL48m0aNtnVVZWU1Ijs7e/s0+0dpMjMz2b1uXVavXk129vbHLl8eHXvzjddxz70PFOhLa9CgATk5OXw8cyYAI197lWVLtw2kjnp9JB3ateK0/n34x1PPlv3FVqCyCIhlrdwCoiST9FDM+5skDSqjvAdJypb0qaR5kvqVkL6JpHllce50cOvDIzmiSzM++vcfOaJLM7J/WENu7lbemfo5EyYv4L3nb+SFey9i2pzF5OZuBeDOx0fT4dQ/0+O8B6lXtxY3XhR3dkK+fRvsTtvmDXnrowUlJ3b5xo8by9577U3nLl0KbJfEi0OG8YebrqfHYd2pU6dOfvMaoP8ppzJ73ue88trrDB50R0UXu0yVxZ0qZa08a4ibgNMkNSin/B82s47AGcCz8YbSy0qYGZ9Uy1espdE+22p1WfvUI3vl2gJpvlu5lgE3/YvDzr6fu54YA8DadRsAeOCZiRw64D76XvEEkvjy22iw7ftVUR/W5i05vDhqKl3bNgnn+6lALTJrnz1YvuKn/Pe/Pa4zo9+dQ07O1rK/2BTRsGEWy5Ztq6VlZy8jKytr+zShJpeTk8PPa9dSv359srK2P7Zhwyw++u8Uxo4dTctmTfjduQN4/713ueh35wFw6GGH8c77HzL5o+n0OOJImoXmc6weRxzJ4sVfs2rVqvK45HJXUu1wl6shAjnAU8D1hXeEGtu74d7CdyQ1Dtufl/SYpP9K+lrS6SWdxMw+C+dqEI7PP0bSuiLO3VbS9FC7nCOpedj+uqSPJc2Pne8kaZ2khyTNBg4rxedQpmbO/4ZmjffigIb1qZqZwRkndGbc+3MKpKm/R638H6ibB57AC6OmAtGAzJ51awHQrnlD2jVvyNsfRZ3ysf2C/Y5uz4Kvogn94ybN5Zy+3QHofnATfl63IT94ApzZuwuvTJhZTlebGrp268aiRV+yZPFiNm/ezIjhw+jTt2CjpE/ffgx96QUA/vPaqxx19DFIok/ffowYPoxNmzaxZPFiFi36km7du/Pne+7lqyXL+GLREl4cOoyeRx/Dcy8OAWDFiuiP1KZNm3jowfu59LLLAfhq0SLyxgM+mTWLTZs2Ub9+/Yr6GMpclSpV4r6SobxHmZ8E5kh6oND2x4EXzOwFSQOBx4BTwr79gB5AK6Ih9FfjnUDSIcBW4gylF3I58KiZDQ1D9Xm1voFm9qOkmsAMSa+Z2Wqi2e3TzOzGIs59GRAFz6q1Ezz9zsnN3cr197/CmL9fSUYV8cKoqXz29ffccUUfZi34lnGT5nJk1+YMvrofZjB51iKuu/eVqIiZGbz97HUA/LJuIwNvfyG/yfzcPRfQoF4dJJjzxTKuvmcYABMmz+eEHm2ZP/ouft24hd8PGpJflsb77Umjfevx4ceLKuTakyUzM5OHH32Ck/ucQG5uLhdcOJA2bdsyeNCddO7Slb4n9+PCgRcz8MLzaduqGfXq7clLQ6PPr03btvz2jDPp1L4NmZmZPPLYkwWawEV5+KEHeWP8WLZu3cqll11Bz6OPAWDkyNd4eciLVM2sSo2aNXlp6PD0HlhJwaIrzgj0zmUsrTOz2mEIfAuwAahtZoMkrQL2M7MtkqoC35lZA0nPA2+Z2dCQxy9mVqeIvAcBlxIFwV+A28zsw3D8WDN7tVAZmoTt7SSdA9wOvAj8x8y+jMnz1HCKJsAJZjZVUg5Q3cxy411vld32tuotzyzdh+V22JoZTyS7CJVKzar6uIRJ0juk+j7NLevcR+OmWfxwnzI9ZyIqYh7iI8As4LkE02+K+bcAJN0D9AEI/YYQ9SH+rdCxOYRugNCnWK1w5mb2sqRpIb/xkn5PVMPsBRxmZr9Kep9oAifAxpKCoXNux0hRF06qKfeGupn9CLxCdAtNnv8SzSoHOBf4sIQ8bg9ziDrGSwcsAfKG7foBVQsnkHQg8LWZPQaMAtoTrYKxJgTDVsChJZzHObdTKt+gSqyHiO4fzHM1cJGkOcD5wLVldJ6ngaNiBkDWF5HmTGCepE+BdkRN5wlApqTPgPuAqWVUHudcMapUUdxXMpRbk9nMasf8+wdgt5j33wDHFHHMhcXlUWj7oGK2/0DB2t0fw/YlRMEPM7uPKOgVdmJJ1+GcKyOKms2pxu9lds5VOJGafYgeEJ1zSeEB0TnnwJvMzjmXR6TmY0g9IDrnkiB5I8nxeEB0ziWF1xCdc47UvVPFA6JzLilSsILoAdE5lxzeZHbOOQBvMjvnXCSadpPsUmzPA6JzLgl82o1zzuXzPkTnnAO/dc855/JEq92k3mPhPSA655LCa4jOOQfpN+1G0u7F7QMws5/j7XfOueKI5D03JZ54NcT5gFHw6al57w1oXI7lcs7t4lIwHhYfEM1s/4osiHOucskogyazpN7Ao0AG8K/wzKTCac4EBhFV5Gab2TnF5ZdQH6KkAcCBZvZXSY2Afczs41KU3znnkHZ+HqKkDOBJ4DhgGTBD0mgzWxCTpjlwK/AbM1sjae94eZY47i3pCeBooseFAvwK/KN0l+Ccc5GMKor7SkB3YJGZfW1mm4FhQP9CaS4FnjSzNQBmtiJeholMBDrczH4PbAwZ/ghUS6S0zjlXHCn+C2ggaWbM67JCWWQBS2PeLwvbYrUAWkiaImlqaGIXK5Em8xZJVYja30iqD2xN4DjnnCuSiEaaS7DKzLru5KkygeZAT6AR8IGkg83sp6ISJ1JDfBJ4DdhL0t3AZOD+nSykc64yU/zmcoJN5mwgdvC3UdgWaxkw2sy2mNliYCFRgCxSiTVEM3tR0sdAr7DpDDObl0hpnXOuOGUw7WYG0FxSU6JAOAAoPIL8OnA28JykBkRN6K+LyzDRO1UygC1EzebUuwHROZdWBFTZyYhoZjmSrgImEsWoZ81svqTBwEwzGx32HS9pAZAL3Gxmq4vLs8SAKOl2oqg7MlzHy5KGmtm9O3U1zrlKrSxu3TOz8cD4QtvujPm3ATeEV4kSqSH+DuhkZr8CSLoH+ATwgOicK5WYkeSUkkhA/K5QusywzTnnSi0jBSNivMUdHibqM/wRmC9pYnh/PFFnpnPOlVq6Le6QN5I8HxgXs31q+RXHOVcZRIMqyS7F9uIt7vBMRRbEOVeJKE0fMiXpIOAeoA1QI2+7mbUox3I553ZxqdhkTmRO4fPAc0S13BOBV4Dh5Vgm59wuTpTJ4g5lLpGAuJuZTQQws6/M7E9EgdE550pNJbySIZFpN5vC4g5fSbqc6BaZOuVbLOfcrkza+TtVykMiAfF6oBZwDVFfYl1gYHkWyjm360vLQRUzmxb++QvbFol1zrmdkoIVxLgTs0cS1kAsipmdVi4lcs7t8qTkDZzEE6+G+ESFlWIX0Kl1Y6ZM84+sotTrdlWyi+B2UipOu4k3MfudiiyIc65yScV1BBNdD9E558pM3jzEVOMB0TmXFCkYDxMPiJKqm9mm8iyMc65ykFKzhpjIc5m7S5oLfBned5D0eLmXzDm3S0vgMaQVLpF+zceAvsBqADObTfTgeuecK5W8Z6rEeyVDIk3mKmb2TaEh8txyKo9zrpLISL0Wc0IBcamk7oBJygCuJnq2qXPOlYqSWAuMJ5GAeAVRs7kx8APwdtjmnHOlloLxMKF7mVcQPQDaOefKhIDMFBxlTmTF7Kcp4p5mM7usXErknKsU0rKGSNREzlMDOBVYWj7Fcc5VCkqzx5DmMbMCjwuQ9BIwudxK5Jzb5aXdU/fiaArsU9YFcc5VLmkZECWtYVsfYhWiB9ffUp6Fcs7t2lJ1cYe4d6oomo3dAdgrvOqZ2YFm9kpFFM45t4sq4ba9RLsXJfWW9IWkRZKKrahJ+q0kk9Q1Xn5xA6KZGTDezHLDq9gVtJ1zLlF5027ivUrMI7pR5Emip4C2Ac6W1KaIdHWAa4FphfcVlsi9zJ9K6pRAOuecS1gZ1BC7A4vM7Gsz2wwMA/oXke7PwP3AxpIyLDYgSsrrX+wEzAjV0lmSPpE0K6HiOudckUSVEl5AA0kzY16F5z5nUXAK4LKwbdtZpM7A/mY2LpFSxRtUmQ50BvolkpFzziUqWg+xxGSrzCxun1/8c6gK8L/AhYkeEy8gCsDMviptgZxzrjhlsLhDNrB/zPtGYVueOkA74P2wWte+wGhJ/cxsZlEZxguIe0m6obidZva/iZbaOedildG0mxlAc0lNiQLhAOCcvJ1mthZokH9O6X3gpuKCIcQPiBlAbUJN0TnnytLOVhDNLEfSVcBEonj1rJnNlzQYmGlmo3c0z3gB8TszG1zKsjrnXLFE2TyG1MzGA+MLbbuzmLQ9S8qvxD5E55wrcyqTPsQyFy8gHlthpXDOVSp5z1RJNcUGRDP7sSIL4pyrXFLwVmZ/UL1zLhmE0qmG6Jxz5aWsBlXKmgdE51xSpFUfonPOlRvhTWbnnANvMjvnXAHeZHbOuSAF46EHROdcxRNp+hhS55wre0IpeHewB0TnXFKkYAXRA6JzruJJqdlkTsWRb5eANydOoH3blrRt1YwHH7hvu/2bNm3ivHPOom2rZhxx+CF8s2RJ/r4H77+Xtq2a0b5tS956c2L+9pbNmtC148Ec0qUjvzlk28rtr706gs4d2rJbtSp8PLPg2ppz58zhqB6H0blDW7p2PJiNG0t8jk9aOu7w1sweeQfzRt3FTRcdt93+xvvVY/w/rmb68FuZ+PS1ZO29R/6+v1zTn5kjbmPmiNs4/fjO2x370B9OZ+WUh/Lf779vPSY8dQ0f/fuPTB9+Kyf0KPgguf33rcfKKQ9x3fnpvf5KWTyGtKx5DTEN5ebmct01VzLujbfIatSIHod2o2/ffrRus+0X5/lnn6HeHvWY//kiXhk+jNtv+yNDXh7OZwsWMGL4MGbNns93y5dzUu9ezF2wkIyMDAAmvP0eDRo0KHC+tm3bMeyV/3DV//y+wPacnBwGXnAezzz/Eu07dGD16tVUrVq1/D+AClalinjkljPpc8UTZP/wE5OH3szYSXP5/Ovv89Pce/2pDB03naFjpnFUtxYMvrofF9/xIr17tKVj6/05ZMB9VK+ayZv/upaJUxbwy/roD0fnNo3Zo85uBc73x0t689pbs3h6xGRaHbgvrz9+Ba363JW///4bT+PNKfMr5uLLSaoOqngNMQ3NmD6dgw5qRtMDD6RatWqccdYAxo4ZVSDN2DGjOPf8CwA47ben8/6772BmjB0zijPOGkD16tVp0rQpBx3UjBnTp8c9X6vWrWnRsuV2299+603aHdye9h06AFC/fv38wLor6dauCV8tXcWS7NVsycllxMRZ9O3ZvkCaVgfux6TpXwAwacZC+vY8GIDWB+7L5FmLyM3dyq8bNzP3y2yOP7w1EAXav153Crc/+nqBvMyM3WvVAKBu7Zp8t3Jt/r6Te7ZnSfZqFnz1PelOJfyXDB4Q09Dy5dk0arTt2TpZWY3Izs7ePs3+UZrMzEx2r1uX1atXk529/bHLl0fHSuLkE4/n8O5deObpp0osx5cLF0bHnHQCh3XrzEN/e6AsLi/lNNy7Lst+WJP/PvuHNWTtVbdAmrkLs+l/TEcA+h/Tgd1r12TPurWYszAKgDVrVKX+HrU4qmsLGu1bD4ArzjqKcZPm8v2qnwvkdc8/xzPgpO4smvBnRj5+BTfcPwKAWjWrceNFx3HPPwssEJ22vMlchiTlAnOJruEz4AIz+zVO+ueBsWb2asWUMP288/5ksrKyWLFiBX17H0fLVq3occSRxabPyc3hv/+dzOSPZrDbbrtx4vHH0rlzF44+Jr37tkrj1odH8vAfz+C8focwZdYisn9YQ27uVt6Z+jld2h7Ae8/fyKo165g2ZzG5uVvZb6+6nHZcJ46/9NHt8jqzd1eGjJnKoy+9yyHtm/LMX35Hl9P/yp8u78PjQ95l/YbNSbjCspWqTea0DYjABjPrCCBpKHA50TNYy42kTDPLKc9zJKJhwyyWLdv2fO7s7GVkZWVtn2bpUho1akROTg4/r11L/fr1ycra/tiGDaNj8/LYe++96XfKqcyYMT1uQMzKakSPHkfm9zn2PvEkPvlk1i4XEJevWEujferlv8/apx7ZMc1YgO9WrmXATf8CoprcKcd2ZO26DQA88MxEHngmGrx6/q8X8uW3K+jQshEH7r8X80dHfYO71ajKvFF30a7/3VxwymH0v/JJAKbNWUyNalVpsEcturU7gFN7deSe606hbp2abN1qbNy8hX8M/6DcP4Oyl5rzEHeVJvOHQDNJTSTNy9so6SZJgwonlnSfpAWS5kj6W9h2sqRpkj6R9LakfcL2QZJekjQFeKmCrieurt26sWjRlyxZvJjNmzczYvgw+vTtVyBNn779GPrSCwD857VXOeroY5BEn779GDF8GJs2bWLJ4sUsWvQl3bp3Z/369fzyyy8ArF+/nrffepO2bdvFLcdxx5/A/Hlz+fXXX8nJyeHDDybRunU4Cv4fAAAQA0lEQVSbuMeko5nzv6FZ4704oGF9qmZmcMYJnRn3/pwCaervUSt/9ZabB57AC6OmAlE/4Z51awHQrnlD2jVvyNsffc6EyfNpetxttOpzF6363MWvG7fQrv/dACz9/kd6do/6bFs23Yca1auycs06el38SH76J4a+z4PPvJmmwZDwTJX4r2RI5xoiENXagBOBCQmmrw+cCrQyM5OUNz9iMnBo2HYJ8AfgxrCvDdDDzDYUyusy4DKA/Rs33ulrSVRmZiYPP/oEJ/c5gdzcXC64cCBt2rZl8KA76dylK31P7seFAy9m4IXn07ZVM+rV25OXhg6LLqRtW357xpl0at+GzMxMHnnsSTIyMljxww+cdfqpQNQUPmvAORx/Qm8ARr0+khuuu5pVK1dyWv8+tO/QkTHjJ1KvXj2uue4GehzWDUmc0PskTjypT4V9DhUlN3cr19//CmP+fiUZVcQLo6by2dffc8cVfZi14FvGTZrLkV2bM/jqfpjB5FmLuO7eVwCompnB289eB8Av6zYy8PYXyM3dGvd8t/zvSP5+x9lcfd7RmMGld6bE3+EylarPVJGZJbsMpRLThwhRDfFGoCFRP2G7kOYmoLaZDcrrQwReBz4Or7Eh/WZJBwMPAfsB1YDFZtY71DDNzO6OV54uXbralGnFPv/albF63a5KdhEqlY2fPvmxmXUtOWViWh/cyZ4b+V7cNIc1r1em50xEOjeZN5hZx/C62sw2AzkUvKYahQ8KfYDdgVeBvmyrWT4OPGFmBwO/L3Ts+vK4AOcqM0lxX8mQ9k3mQn4A9g7N4nUUDHgASKoN7GZm40O/4NdhV10gb+7KBRVUXucqrRRsMe9aAdHMtkgaDEwnCm6fF5GsDjBKUg2irowbwvZBwAhJa4B3gablX2LnKi8PiGXIzGoXs/0x4LEitl8Y87Z7EftHAaOK2D6o1IV0zhVJkJLTbtI2IDrn0lgS70aJJ50HVZxzaawsbt2T1FvSF5IWSbqliP03xMw5fkfSAfHy84DonEuCkpZ2KDkiSsoAniSah9wGOFtS4TsDPgG6mll7opklcW+494DonKtw0cTsnb5TpTuwyMy+DtPuhgH9YxOY2XsxaxxMBRrFy9ADonMuOVTCCxpImhnzuqxQDlnA0pj3y8K24lwMvBGvSD6o4pxLigSaxavK6k4VSecBXYGj4qXzgOicS4oyWMAhG9g/5n0jtt1ckU9SL+B24Cgz2xS3TDtdJOec21ElNZcTC5YzgOaSmkqqBgwARhc4jdQJ+CfQz8xWlJSh1xCdc0mxsxOzzSxH0lXARCADeNbM5oe71Waa2WjgQaA20V1oAN+aWb/i8vSA6JyrcHmjzDvLzMYD4wttuzPm3712JD8PiM655EjBO1U8IDrnkiIVF4j1gOicS4rUC4ceEJ1zyZKCEdEDonOuwkneZHbOuXypFw49IDrnkiJ5z02JxwOicy4pUjAeekB0zlW8xO/Oq1geEJ1zSeFNZuecC1IwHnpAdM4lQeKrYlcoD4jOuSRJvYjoAdE5V+GEN5mdcy6fN5mdcy7Y2QViy4MHROdcUniT2TnniIKhB0TnnAu8yeycc4HXEJ1zLvCA6JxzQNRgTr2I6AHROVfhfGK2c87F8IDonHMQFndIvYjoAdE5V+F8gVjnnIuVghHRA6JzLim8yeycc0HqhUMPiM65JEnFZ6rIzJJdhl2CpJXAN8kuRyk0AFYluxCVSLp+3geY2V5llZmkCUSfRTyrzKx3WZ0zER4QKzlJM82sa7LLUVn4553aqiS7AM45lyo8IDrnXOAB0T2V7AJUMv55pzDvQ3TOucBriM45F3hAdM65wAOic84FHhBdqSgVbzNwbid5QHQ7TJIsjMZJOlHS/skuU7qL/QMjyW+pTRIPiG6HxQTDo4DbgZ+TW6L0F/OZXgHckeTiVFr+l8iViqT+wL+Ay81sraTqZrYp2eVKZ5J6AqcBv01yUSotryG6hBTuMzSzUcAy4NrwfpOkjGSUbVcgqTlwKbB7zDbvp61gHhBdiQr1GbaW1B7AzDoBdSS9Gt7nelBMTBF/YL4EngeWAOdJqmdm5kGxYvmdKi5hkm4E+gAZwAzg9lAznAKsN7Pjk1rANFHoD8xFQC3gRzN7WdKJQF9gLjDCzFYnsaiVjtcQXUIknQf0MbNjgFnAOcC9kqqa2W+AKpIaJbWQaSImGF4PXACsBq6U9AAwARgFHAacIsl/RyuQf9iuSEU01RYAl0i6EmgGHElUW3xO0u5m1svMllV0OdNJoak1LYDOQC+gKbAOaAg8BLwFvAiMM7OtSShqpeWjzG47hZp05wDzzWyWpOrAIcBDZrZI0iigNf5zVKJCn2kDM1so6Q/AscBJwFFEo8sPALlmdnPySlt5eQ3RFaUKQKgN3kJUeyFMq5kPXC/pDqALcLWZ/ZisgqaDQsHw98BgSbXM7DugJvCumeUCtYkGVh5NWmErOf/L7vJJ6gTMDqPF7YALifoNl0rKe7bFbGANUXP5GjNbkpTCpglJmWaWE/59KXAJcIaZrQ93pMwDHpa0H9FnerR3PSSPjzI7AELn/XNE/VgnANWBB8P/Iern+gl41cyG+UTskkk6GOgATCQaOHkGeAn4nGgC9gCi2uAHQDtgsZl9nZzSOvAmsyO/SbcVuBpYCrxmZhuAscBi4FEz60VUO+weDtuclMKml4OB/kQDJxlEgfE14AmgBtEAynXAJjN7x4Nh8nmT2eVPAzGzn0Ozboik4cAAM5sAIOlcolrN2bHHuO3l9RmGeYUbgdOBDDMbImke8I2Z/SLpWKI/LDlJLbDL5zVEB4CksyW9Fjr3zyf6JX1dUhVJzYj6t843s8+TWtA0EDOA0g4YDbwMnBTmci4PwfBa4GHgWjNbl7zSuljeh1hJxY58hvfViEaQPzSzgaHD/xmgoZkdJ2k3M/s1WeVNN5IOBG4l6i98BDiRqHY9FniXaIR+sZl9lrRCuu14DbGSiqnFtJTUyMw2E3Xsd5f0chgZvQT4RlKWB8PiFXW/cegPnAg0IuqbfQP4N1FQPBJ4w4Nh6vEaYiUTptY0AL4FlhONeo4m+gX9LtQUlxLNjTs7eSVNH4Wm1vwWaGxmD4f3pwDHAfPM7P/CvcpzzCw7eSV2xfEaYiUi6QTgFeAsYDrQimik82jg2FAT3Aw8DhwiaR9fbSU+SccBL0m6JQS7ZcBFYQI2ZvY6sAq4VNLlZvaGB8PU5aPMlYSkw4FngXPMbJKkmUQ1wwOAvYGTgcaSahAFysPM7IekFTgNhMnqg4lq2XsDZwKPAf8DPCqpipn9H1E/4n7AyGSV1SXGA2IlEGp57YHJhInWZvYPSUcDdczsNUm/AG2AbsCdHgzjk7QnMB7ob2ZjJDUmmsjexMxGSroGeC18xp2J7vjxzzTFeR9iJSGpFnARcBDwJnAM0AI4M/aOk9j+MBefpD5EizEcFuZwDgU+MLN/hv2NgE7AXL/FMT14QKwE8qbYhKB4MVEnfxOga1jgtVroO3Q7KPQbPkY0otwQONfMNkjKCHM6XRrxgFhJhP6srSEongN0BMYQjSZ7MNwJknoR1br3NbMVkmqY2cZkl8vtOB9l3gUVHhkutOpybTN7mmiVldOJ1uJzO8HM3ia6k+c9SXt7MExfPqiyiym09l5tooUDtoT3hwP/kXQoMATIBT5KWmF3IWb2RpjDOUFS12iTN7/SjTeZdyGFguFNQA+iUeWBYdL1v4EhZjYumeXclUmq7fcmpy8PiLsgSccAdwKXEw2inEE09WNNGFwR+Io1zhXmTeZdjKSewFXAO2FlmptDAJwO9CS6k8KDoXNF8EGVNFfErXWLgZVAa0kdAMzsJqLHW06QP0jeuWJ5kzmNFeozPJloDcOfgI+Jlpz6kehh57NDmr3NbEWyyutcqvMa4i5A0v8AdxMNojxLtCz99cAewO/CQqUQ1Rydc8XwPsQ0FO6bXR2e3Ja3qMC5ZvaZpL8R1RCXA/cAfwS+B+83dK4kXkNMM5L2AW4ErghTPFYQLS+1GcDM1hDVEA8Oz/292cxWJa3AzqURD4jpZyUwg+i+2YvCoMoiYFhY9h+iJb0ahQEUX6jBuQT5oEqakNQcqGJmX4Qg2JfoOR2fmtlTkv6P6BnAc4BDiJrQC5JXYufSjwfENCCpPlHNcBXR4Eku8BTRIg3NgO/M7J+SDiF63u+3ZrY4WeV1Ll35oEoaMLPVYUWVt4m6OToAw4F1RH2HB4da43Oxaxs653aM1xDTSHh+x2NEAXEfokVeBwDdge+A35jZ2uSV0Ln05gExzYRVmh8GDjWzHyXVA6oCu/mqzM7tHG8ypxkzGydpKzBV0mFmtjrZZXJuV+EBMQ3FrL33tqQuZrY12WVyblfgTeY05mvvOVe2PCA651zgd6o451zgAdE55wIPiM45F3hAdM65wAOiS4ikXEmfSponaYSk3XYir56SxoZ/95N0S5y0e4QFcHf0HIPCkwcT2l4ozfOSTt+BczWRNG9Hy+hSjwdEl6gNZtbRzNoR3T99eexORXb458nMRpvZfXGS7AHscEB0rjQ8ILrS+BBoFmpGX0h6EZgH7C/peEkfSZoVapK1AST1lvS5pFnAaXkZSbpQ0hPh3/tIGilpdngdDtwHHBRqpw+GdDdLmiFpjqS7Y/K6XdJCSZOBliVdhKRLQz6zJb1WqNbbS9LMkF/fkD5D0oMx5/79zn6QLrV4QHQ7JCxCeyIwN2xqDvzdzNoC64E/Ab3MrDMwE7hBUg3gaeBkoAuwbzHZPwZMMrMORM+Rng/cAnwVaqc3Szo+nLM70BHoIulISV2IFrroCJwEdEvgcv5jZt3C+T4jeoZ1nibhHH2Af4RruBhYa2bdQv6XSmqawHlcmvBb91yiakr6NPz7Q+AZolW7vzGzqWH7oUAbYEp4Omo14COgFbDYzL4EkDQEuKyIcxwD/A7AzHKBtWHxiljHh9cn4X1togBZBxhpZr+Gc4xO4JraSfoLUbO8NjAxZt8r4ZbILyV9Ha7heKB9TP9i3XDuhQmcy6UBD4guURvMrGPshhD01sduAt4ys7MLpStw3E4ScK+Z/bPQOa4rRV7PA6eY2WxJFwI9Y/YVvoXLwrmvNrPYwImkJqU4t0tB3mR2ZWkq8BtJzQAk1ZLUAvgcaCLpoJDu7GKOfwe4IhybIaku8AtR7S/PRGBgTN9kVnjy4AfAKZJqSqpD1DwvSR3gO0lVgXML7TtDUpVQ5gOBL8K5rwjpkdRCUq0EzuPShNcQXZkxs5WhpvVvSdXD5j+Z2UJJlwHjJP1K1OSuU0QW1wJPSbqY6DEJV5jZR5KmhGktb4R+xNbAR6GGug44z8xmSRoOzAZWED2IqyR3ANOIHs8wrVCZvgWmA7sDl5vZRkn/IupbnBVWKF8JnJLYp+PSgS/u4JxzgTeZnXMu8IDonHOBB0TnnAs8IDrnXOAB0TnnAg+IzjkXeEB0zrng/wGTjMtui0XwlwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_confusion_matrix(cnf_matrix, classes=class_names, normalize=True, title='Confusion matrix, with normalization')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 384,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n",
      "23/23 [==============================] - 8s 361ms/step - loss: 0.0288 - acc: 0.9920\n",
      "Epoch 2/20\n",
      "23/23 [==============================] - 7s 322ms/step - loss: 0.0182 - acc: 0.9950\n",
      "Epoch 3/20\n",
      "23/23 [==============================] - 7s 321ms/step - loss: 0.0166 - acc: 0.9974\n",
      "Epoch 4/20\n",
      "23/23 [==============================] - 7s 310ms/step - loss: 0.0140 - acc: 0.9974\n",
      "Epoch 5/20\n",
      "23/23 [==============================] - 7s 324ms/step - loss: 0.0119 - acc: 0.9970\n",
      "Epoch 6/20\n",
      "23/23 [==============================] - 7s 306ms/step - loss: 0.0082 - acc: 0.9980\n",
      "Epoch 7/20\n",
      "23/23 [==============================] - 7s 297ms/step - loss: 0.0065 - acc: 0.9991\n",
      "Epoch 8/20\n",
      "23/23 [==============================] - 7s 297ms/step - loss: 0.0041 - acc: 0.9987\n",
      "Epoch 9/20\n",
      "23/23 [==============================] - 8s 330ms/step - loss: 0.0064 - acc: 0.9985\n",
      "Epoch 10/20\n",
      "23/23 [==============================] - 7s 316ms/step - loss: 0.0058 - acc: 0.9987\n",
      "Epoch 11/20\n",
      "23/23 [==============================] - 8s 327ms/step - loss: 0.0055 - acc: 0.9993\n",
      "Epoch 12/20\n",
      "23/23 [==============================] - 7s 325ms/step - loss: 0.0030 - acc: 0.9996\n",
      "Epoch 13/20\n",
      "23/23 [==============================] - 7s 318ms/step - loss: 0.0043 - acc: 0.9991\n",
      "Epoch 14/20\n",
      "23/23 [==============================] - 7s 311ms/step - loss: 0.0043 - acc: 0.9993\n",
      "Epoch 15/20\n",
      "23/23 [==============================] - 8s 329ms/step - loss: 0.0058 - acc: 0.9991\n",
      "Epoch 16/20\n",
      "23/23 [==============================] - 7s 318ms/step - loss: 0.0039 - acc: 0.9998\n",
      "Epoch 17/20\n",
      "23/23 [==============================] - 7s 299ms/step - loss: 0.0036 - acc: 0.9998\n",
      "Epoch 18/20\n",
      "23/23 [==============================] - 7s 289ms/step - loss: 0.0058 - acc: 0.9989\n",
      "Epoch 19/20\n",
      "23/23 [==============================] - 6s 282ms/step - loss: 0.0029 - acc: 0.9998\n",
      "Epoch 20/20\n",
      "23/23 [==============================] - 6s 274ms/step - loss: 0.0028 - acc: 0.9998\n"
     ]
    }
   ],
   "source": [
    "label_reshaped_freq_phase_pulsars = np.ones(len(reshaped_freq_phase_pulsars))\n",
    "label_reshaped_freq_phase_nonpulsars = np.zeros(len(reshaped_freq_phase_nonpulsars))\n",
    "\n",
    "freq_phase_data_combined = np.concatenate((reshaped_freq_phase_pulsars, reshaped_freq_phase_nonpulsars), axis = 0)\n",
    "freq_phase_label_combined = np.concatenate((label_reshaped_freq_phase_pulsars, label_reshaped_freq_phase_nonpulsars), axis = 0)\n",
    "freq_phase_train, freq_phase_test, freq_phase_label_train, freq_phase_label_test = train_test_split(freq_phase_data_combined, freq_phase_label_combined, test_size=0.2, random_state=42)\n",
    "\n",
    "model.fit_generator(generator(batch_size, freq_phase_train, freq_phase_label_train), steps_per_epoch=len(freq_phase_train)//batch_size, epochs=20, callbacks=[tensor_board])\n",
    "model.save_weights('cnn_freq_phase.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 385,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "F Score: 0.996557659208 Precision: 0.996557659208 Recall: 0.996557659208\n"
     ]
    }
   ],
   "source": [
    "predict = model.predict_generator(generator(batch_size, freq_phase_test, freq_phase_label_test), steps=math.ceil(len(freq_phase_test)/batch_size))\n",
    "np.save('predictions.npy', predict)\n",
    "classified_results = np.rint(predict)\n",
    "f_score = f1_score(freq_phase_label_test, classified_results, average='binary')\n",
    "precision = precision_score(freq_phase_label_test, classified_results, average='binary')\n",
    "recall = recall_score(freq_phase_label_test, classified_results, average='binary')\n",
    "print('F Score:', f_score, 'Precision:', precision, 'Recall:', recall)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Multiple Input Model. Use functional API of Keras"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 418,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "__________________________________________________________________________________________________\n",
      "Layer (type)                    Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      "input_12 (InputLayer)           (None, 60)           0                                            \n",
      "__________________________________________________________________________________________________\n",
      "input_11 (InputLayer)           (None, 64)           0                                            \n",
      "__________________________________________________________________________________________________\n",
      "dense_66 (Dense)                (None, 64)           3904        input_12[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "dense_64 (Dense)                (None, 8)            520         input_11[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "dense_67 (Dense)                (None, 32)           2080        dense_66[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "dense_65 (Dense)                (None, 4)            36          dense_64[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "dense_68 (Dense)                (None, 4)            132         dense_67[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "concatenate_5 (Concatenate)     (None, 8)            0           dense_65[0][0]                   \n",
      "                                                                 dense_68[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "dense_69 (Dense)                (None, 2)            18          concatenate_5[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "dense_70 (Dense)                (None, 1)            3           dense_69[0][0]                   \n",
      "==================================================================================================\n",
      "Total params: 6,693\n",
      "Trainable params: 6,693\n",
      "Non-trainable params: 0\n",
      "__________________________________________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "from keras.layers import Input, Dense, concatenate\n",
    "from keras.models import Model\n",
    "\n",
    "pulse_profile = Input(shape=(64,))\n",
    "dm_curve = Input(shape=(60,))\n",
    " \n",
    "# the first branch operates on the pulse profile\n",
    "x = Dense(8, activation=\"relu\")(pulse_profile)\n",
    "x = Dense(4, activation=\"relu\")(x)\n",
    "x = Model(inputs=pulse_profile, outputs=x)\n",
    " \n",
    "# the second branch opreates on the dm curve\n",
    "y = Dense(64, activation=\"relu\")(dm_curve)\n",
    "y = Dense(32, activation=\"relu\")(y)\n",
    "y = Dense(4, activation=\"relu\")(y)\n",
    "y = Model(inputs=dm_curve, outputs=y)\n",
    " \n",
    "# combine the output of the two branches\n",
    "combined = concatenate([x.output, y.output])\n",
    " \n",
    "# apply a FC layer and then a regression prediction on the\n",
    "# combined outputs\n",
    "z = Dense(2, activation=\"relu\")(combined)\n",
    "z = Dense(1, activation=\"linear\")(z)\n",
    " \n",
    "# our model will accept the inputs of the two branches and\n",
    "# then output a single value\n",
    "model = Model(inputs=[x.input, y.input], outputs=z)\n",
    "model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 419,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total Samples in one epoch is 5947\n",
      "Total Number of Training Samples is 4757\n",
      "Total Number of Test Samples is 1190\n"
     ]
    }
   ],
   "source": [
    "label_pulsars = np.ones(len(dm_curve_pulsars))\n",
    "label_nonpulsars = np.zeros(len(dm_curve_nonpulsars))\n",
    "\n",
    "\n",
    "dm_curve_data_combined = np.concatenate((dm_curve_pulsars, dm_curve_nonpulsars), axis = 0)\n",
    "dm_curve_label_combined = np.concatenate((label_pulsars, label_nonpulsars), axis = 0)\n",
    "\n",
    "dm_curve_train, dm_curve_test, dm_curve_label_train, \\\n",
    "dm_curve_label_test = train_test_split(dm_curve_data_combined, dm_curve_label_combined, \\\n",
    "                                         test_size=0.2, random_state=42)\n",
    "\n",
    "print('Total Samples in one epoch is %d' %len(dm_curve_data_combined))\n",
    "print('Total Number of Training Samples is %d' %len(dm_curve_train))\n",
    "print('Total Number of Test Samples is %d' %len(dm_curve_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 420,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 3805 samples, validate on 952 samples\n",
      "Epoch 1/50\n",
      "3805/3805 [==============================] - 1s 179us/step - loss: 6.9014 - acc: 0.4607 - val_loss: 5.8741 - val_acc: 0.5200\n",
      "Epoch 2/50\n",
      "3805/3805 [==============================] - 0s 29us/step - loss: 6.1822 - acc: 0.5243 - val_loss: 1.9480 - val_acc: 0.5567\n",
      "Epoch 3/50\n",
      "3805/3805 [==============================] - 0s 29us/step - loss: 0.2488 - acc: 0.8796 - val_loss: 0.0976 - val_acc: 0.9307\n",
      "Epoch 4/50\n",
      "3805/3805 [==============================] - 0s 31us/step - loss: 0.0808 - acc: 0.9346 - val_loss: 0.0825 - val_acc: 0.9317\n",
      "Epoch 5/50\n",
      "3805/3805 [==============================] - 0s 29us/step - loss: 0.0722 - acc: 0.9361 - val_loss: 0.0917 - val_acc: 0.9265\n",
      "Epoch 6/50\n",
      "3805/3805 [==============================] - 0s 28us/step - loss: 0.0691 - acc: 0.9359 - val_loss: 0.0899 - val_acc: 0.9265\n",
      "Epoch 7/50\n",
      "3805/3805 [==============================] - 0s 27us/step - loss: 0.0667 - acc: 0.9356 - val_loss: 0.0893 - val_acc: 0.9265\n",
      "Epoch 8/50\n",
      "3805/3805 [==============================] - 0s 29us/step - loss: 0.0692 - acc: 0.9335 - val_loss: 0.0788 - val_acc: 0.9254\n",
      "Epoch 9/50\n",
      "3805/3805 [==============================] - 0s 25us/step - loss: 0.0630 - acc: 0.9322 - val_loss: 0.0716 - val_acc: 0.9265\n",
      "Epoch 10/50\n",
      "3805/3805 [==============================] - 0s 30us/step - loss: 0.0600 - acc: 0.9343 - val_loss: 0.0721 - val_acc: 0.9265\n",
      "Epoch 11/50\n",
      "3805/3805 [==============================] - 0s 27us/step - loss: 0.0594 - acc: 0.9335 - val_loss: 0.0711 - val_acc: 0.9275\n",
      "Epoch 12/50\n",
      "3805/3805 [==============================] - 0s 29us/step - loss: 0.0587 - acc: 0.9330 - val_loss: 0.0704 - val_acc: 0.9275\n",
      "Epoch 13/50\n",
      "3805/3805 [==============================] - 0s 32us/step - loss: 0.0582 - acc: 0.9343 - val_loss: 0.0579 - val_acc: 0.9286\n",
      "Epoch 14/50\n",
      "3805/3805 [==============================] - 0s 30us/step - loss: 0.0574 - acc: 0.9353 - val_loss: 0.0592 - val_acc: 0.9286\n",
      "Epoch 15/50\n",
      "3805/3805 [==============================] - 0s 29us/step - loss: 0.0569 - acc: 0.9348 - val_loss: 0.0575 - val_acc: 0.9275\n",
      "Epoch 16/50\n",
      "3805/3805 [==============================] - 0s 30us/step - loss: 0.0565 - acc: 0.9330 - val_loss: 0.0564 - val_acc: 0.9275\n",
      "Epoch 17/50\n",
      "3805/3805 [==============================] - 0s 28us/step - loss: 0.0565 - acc: 0.9298 - val_loss: 0.0589 - val_acc: 0.9265\n",
      "Epoch 18/50\n",
      "3805/3805 [==============================] - 0s 29us/step - loss: 0.0564 - acc: 0.9325 - val_loss: 0.0551 - val_acc: 0.9296\n",
      "Epoch 19/50\n",
      "3805/3805 [==============================] - 0s 29us/step - loss: 0.0561 - acc: 0.9327 - val_loss: 0.0548 - val_acc: 0.9223\n",
      "Epoch 20/50\n",
      "3805/3805 [==============================] - 0s 32us/step - loss: 0.0554 - acc: 0.9290 - val_loss: 0.0677 - val_acc: 0.9244\n",
      "Epoch 21/50\n",
      "3805/3805 [==============================] - 0s 30us/step - loss: 0.0553 - acc: 0.9251 - val_loss: 0.0674 - val_acc: 0.9191\n",
      "Epoch 22/50\n",
      "3805/3805 [==============================] - 0s 29us/step - loss: 0.0556 - acc: 0.9251 - val_loss: 0.0572 - val_acc: 0.9202\n",
      "Epoch 23/50\n",
      "3805/3805 [==============================] - 0s 32us/step - loss: 0.0544 - acc: 0.9264 - val_loss: 0.0671 - val_acc: 0.9202\n",
      "Epoch 24/50\n",
      "3805/3805 [==============================] - 0s 31us/step - loss: 0.0540 - acc: 0.9225 - val_loss: 0.0667 - val_acc: 0.9170\n",
      "Epoch 25/50\n",
      "3805/3805 [==============================] - 0s 27us/step - loss: 0.0538 - acc: 0.9209 - val_loss: 0.0661 - val_acc: 0.9170\n",
      "Epoch 26/50\n",
      "3805/3805 [==============================] - 0s 29us/step - loss: 0.0533 - acc: 0.9198 - val_loss: 0.0659 - val_acc: 0.9128\n",
      "Epoch 27/50\n",
      "3805/3805 [==============================] - 0s 27us/step - loss: 0.0535 - acc: 0.9185 - val_loss: 0.0654 - val_acc: 0.9149\n",
      "Epoch 28/50\n",
      "3805/3805 [==============================] - 0s 31us/step - loss: 0.0530 - acc: 0.9146 - val_loss: 0.0793 - val_acc: 0.9128\n",
      "Epoch 29/50\n",
      "3805/3805 [==============================] - 0s 29us/step - loss: 0.0526 - acc: 0.9130 - val_loss: 0.0790 - val_acc: 0.9128\n",
      "Epoch 30/50\n",
      "3805/3805 [==============================] - 0s 28us/step - loss: 0.0530 - acc: 0.9096 - val_loss: 0.0670 - val_acc: 0.9107\n",
      "Epoch 31/50\n",
      "3805/3805 [==============================] - 0s 28us/step - loss: 0.0529 - acc: 0.9101 - val_loss: 0.0652 - val_acc: 0.9097\n",
      "Epoch 32/50\n",
      "3805/3805 [==============================] - 0s 28us/step - loss: 0.0521 - acc: 0.9088 - val_loss: 0.0788 - val_acc: 0.9034\n",
      "Epoch 33/50\n",
      "3805/3805 [==============================] - 0s 33us/step - loss: 0.0520 - acc: 0.9075 - val_loss: 0.0661 - val_acc: 0.9118\n",
      "Epoch 34/50\n",
      "3805/3805 [==============================] - 0s 30us/step - loss: 0.0517 - acc: 0.9049 - val_loss: 0.0647 - val_acc: 0.9065\n",
      "Epoch 35/50\n",
      "3805/3805 [==============================] - 0s 34us/step - loss: 0.0513 - acc: 0.9017 - val_loss: 0.0759 - val_acc: 0.9044\n",
      "Epoch 36/50\n",
      "3805/3805 [==============================] - 0s 28us/step - loss: 0.0510 - acc: 0.9028 - val_loss: 0.0755 - val_acc: 0.9076\n",
      "Epoch 37/50\n",
      "3805/3805 [==============================] - 0s 31us/step - loss: 0.0511 - acc: 0.9030 - val_loss: 0.0751 - val_acc: 0.9065\n",
      "Epoch 38/50\n",
      "3805/3805 [==============================] - 0s 28us/step - loss: 0.0510 - acc: 0.9038 - val_loss: 0.0747 - val_acc: 0.9055\n",
      "Epoch 39/50\n",
      "3805/3805 [==============================] - 0s 27us/step - loss: 0.0504 - acc: 0.9017 - val_loss: 0.0752 - val_acc: 0.9076\n",
      "Epoch 40/50\n",
      "3805/3805 [==============================] - 0s 27us/step - loss: 0.0506 - acc: 0.9014 - val_loss: 0.0762 - val_acc: 0.9044\n",
      "Epoch 41/50\n",
      "3805/3805 [==============================] - 0s 25us/step - loss: 0.0478 - acc: 0.9035 - val_loss: 0.0763 - val_acc: 0.9139\n",
      "Epoch 42/50\n",
      "3805/3805 [==============================] - 0s 32us/step - loss: 0.0466 - acc: 0.9125 - val_loss: 0.0746 - val_acc: 0.9149\n",
      "Epoch 43/50\n",
      "3805/3805 [==============================] - 0s 30us/step - loss: 0.0459 - acc: 0.9085 - val_loss: 0.0741 - val_acc: 0.9097\n",
      "Epoch 44/50\n",
      "3805/3805 [==============================] - 0s 27us/step - loss: 0.0457 - acc: 0.9075 - val_loss: 0.0605 - val_acc: 0.9107\n",
      "Epoch 45/50\n",
      "3805/3805 [==============================] - 0s 32us/step - loss: 0.3562 - acc: 0.8975 - val_loss: 0.3736 - val_acc: 0.8550\n",
      "Epoch 46/50\n",
      "3805/3805 [==============================] - 0s 29us/step - loss: 0.1447 - acc: 0.8302 - val_loss: 0.0725 - val_acc: 0.8403\n",
      "Epoch 47/50\n",
      "3805/3805 [==============================] - 0s 26us/step - loss: 0.0716 - acc: 0.8570 - val_loss: 0.0661 - val_acc: 0.8603\n",
      "Epoch 48/50\n",
      "3805/3805 [==============================] - 0s 26us/step - loss: 0.0627 - acc: 0.8762 - val_loss: 0.0638 - val_acc: 0.8708\n",
      "Epoch 49/50\n",
      "3805/3805 [==============================] - 0s 24us/step - loss: 0.0636 - acc: 0.8933 - val_loss: 0.0792 - val_acc: 0.8782\n",
      "Epoch 50/50\n",
      "3805/3805 [==============================] - 0s 28us/step - loss: 0.0605 - acc: 0.8773 - val_loss: 0.0689 - val_acc: 0.8918\n"
     ]
    }
   ],
   "source": [
    "model.fit([pulse_profile_train, dm_curve_train],pulse_profile_label_train, validation_split=0.2, batch_size=100, epochs=50)\n",
    "model.save_weights('pulse_profile_and_dm_curve_ann.h5')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Are you paying attention?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Mini test! (Fix the bug)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 421,
   "metadata": {},
   "outputs": [
    {
     "ename": "ValueError",
     "evalue": "Target is multiclass but average='binary'. Please choose another average setting, one of [None, 'micro', 'macro', 'weighted'].",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-421-0a50e32b1907>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m      3\u001b[0m \u001b[0mclassified_results\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0my_pred\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0mclassified_results\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mreshape\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mclassified_results\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mclassified_results\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 5\u001b[0;31m \u001b[0mf_score\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mf1_score\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpulse_profile_label_test\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mclassified_results\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0maverage\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'binary'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      6\u001b[0m \u001b[0mprecision\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mprecision_score\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpulse_profile_label_test\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mclassified_results\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0maverage\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'binary'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      7\u001b[0m \u001b[0mrecall\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mrecall_score\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpulse_profile_label_test\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mclassified_results\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0maverage\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'binary'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/usr/local/lib/python3.5/dist-packages/sklearn/metrics/classification.py\u001b[0m in \u001b[0;36mf1_score\u001b[0;34m(y_true, y_pred, labels, pos_label, average, sample_weight)\u001b[0m\n\u001b[1;32m   1057\u001b[0m     return fbeta_score(y_true, y_pred, 1, labels=labels,\n\u001b[1;32m   1058\u001b[0m                        \u001b[0mpos_label\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mpos_label\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0maverage\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0maverage\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1059\u001b[0;31m                        sample_weight=sample_weight)\n\u001b[0m\u001b[1;32m   1060\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1061\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/usr/local/lib/python3.5/dist-packages/sklearn/metrics/classification.py\u001b[0m in \u001b[0;36mfbeta_score\u001b[0;34m(y_true, y_pred, beta, labels, pos_label, average, sample_weight)\u001b[0m\n\u001b[1;32m   1180\u001b[0m                                                  \u001b[0maverage\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0maverage\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1181\u001b[0m                                                  \u001b[0mwarn_for\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'f-score'\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1182\u001b[0;31m                                                  sample_weight=sample_weight)\n\u001b[0m\u001b[1;32m   1183\u001b[0m     \u001b[0;32mreturn\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1184\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/usr/local/lib/python3.5/dist-packages/sklearn/metrics/classification.py\u001b[0m in \u001b[0;36mprecision_recall_fscore_support\u001b[0;34m(y_true, y_pred, beta, labels, pos_label, average, warn_for, sample_weight)\u001b[0m\n\u001b[1;32m   1413\u001b[0m         \u001b[0;32mraise\u001b[0m \u001b[0mValueError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"beta should be >0 in the F-beta score\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1414\u001b[0m     labels = _check_set_wise_labels(y_true, y_pred, average, labels,\n\u001b[0;32m-> 1415\u001b[0;31m                                     pos_label)\n\u001b[0m\u001b[1;32m   1416\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1417\u001b[0m     \u001b[0;31m# Calculate tp_sum, pred_sum, true_sum ###\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/usr/local/lib/python3.5/dist-packages/sklearn/metrics/classification.py\u001b[0m in \u001b[0;36m_check_set_wise_labels\u001b[0;34m(y_true, y_pred, average, labels, pos_label)\u001b[0m\n\u001b[1;32m   1252\u001b[0m             raise ValueError(\"Target is %s but average='binary'. Please \"\n\u001b[1;32m   1253\u001b[0m                              \u001b[0;34m\"choose another average setting, one of %r.\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1254\u001b[0;31m                              % (y_type, average_options))\n\u001b[0m\u001b[1;32m   1255\u001b[0m     \u001b[0;32melif\u001b[0m \u001b[0mpos_label\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32min\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1256\u001b[0m         warnings.warn(\"Note that pos_label (set to %r) is ignored when \"\n",
      "\u001b[0;31mValueError\u001b[0m: Target is multiclass but average='binary'. Please choose another average setting, one of [None, 'micro', 'macro', 'weighted']."
     ]
    }
   ],
   "source": [
    "\n",
    "y_pred = model.predict([pulse_profile_test, dm_curve_test])\n",
    "classified_results = np.rint(y_pred)\n",
    "classified_results = np.reshape(classified_results, len(classified_results))\n",
    "f_score = f1_score(pulse_profile_label_test, classified_results, average='binary')\n",
    "precision = precision_score(pulse_profile_label_test, classified_results, average='binary')\n",
    "recall = recall_score(pulse_profile_label_test, classified_results, average='binary')\n",
    "accuracy = (pulse_profile_label_test == classified_results).sum()/len(pulse_profile_label_test)\n",
    "print('F Score:', f_score, 'Precision:', precision, 'Recall:', recall, 'Accuracy:', accuracy)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Exercise: Combine Multiple inputs (2D images and 1D data) into a CNN and train"
   ]
  },
  {
   "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": "3.5.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}