new feature&RF&GBDT&RNN.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from __future__ import print_function\n",
    "\n",
    "# For number crunching\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import scipy\n",
    "\n",
    "# For visualisation\n",
    "import matplotlib.pyplot as pl \n",
    "import seaborn as sns \n",
    "\n",
    "# For prediction \n",
    "import sklearn\n",
    "\n",
    "# Misc\n",
    "from itertools import cycle\n",
    "import json \n",
    "import os\n",
    "\n",
    "# Magic and formatting\n",
    "% matplotlib inline\n",
    "\n",
    "sns.set_context('poster')\n",
    "sns.set_style('darkgrid')\n",
    "\n",
    "current_palette = cycle(sns.color_palette())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Annotator 0\n",
      "  Annotation 1\n",
      "    Start time: 1774.195\n",
      "    End time:   1776.739\n",
      "    Duration:   2.544\n",
      "\n",
      "Annotator 1\n",
      "  Annotation 1\n",
      "    Start time: 1775.687\n",
      "    End time:   1776.348\n",
      "    Duration:   0.661\n",
      "\n",
      "  Annotation 2\n",
      "    Start time: 1776.901\n",
      "    End time:   1777.87\n",
      "    Duration:   0.969\n",
      "\n",
      "Annotator 2\n",
      "  Annotation 1\n",
      "    Start time: 1774.974\n",
      "    End time:   1778.844\n",
      "    Duration:   3.87\n",
      "\n",
      "(0, 1823.9170000000001)\n"
     ]
    }
   ],
   "source": [
    "from visualise_data import SequenceVisualisation\n",
    "\n",
    "plotter = SequenceVisualisation('public_data', 'public_data/train/00001')\n",
    "sequence_window = (plotter.meta['start'], plotter.meta['end'])\n",
    "\n",
    "# We can extract the time range of the activity 'jump' with the 'times_of_activity' function. \n",
    "# This function returns all of the times that jump was annotated, and it is indexed first \n",
    "# by the annotator, and then by the time at which it occurred. \n",
    "times_of_jump = plotter.times_of_activity('a_jump')\n",
    "\n",
    "# You can get the times at which these. \n",
    "for ai, annotator_jumps in enumerate(times_of_jump): \n",
    "    print ('Annotator {}'.format(ai))\n",
    "    \n",
    "    # The annotator_jumps is a list of tuples. The length of this list specifies the number of segments as\n",
    "    # annotated as 'a_jump' by this annotator. Each element of this list is a tuple that holds the start and \n",
    "    # end time of the t-th annotation in that order, ie: \n",
    "    for ti, (start, end) in enumerate(annotator_jumps, start=1): \n",
    "        print ('  Annotation {}'.format(ti))\n",
    "        print ('    Start time: {}'.format(start))\n",
    "        print ('    End time:   {}'.format(end))\n",
    "        print ('    Duration:   {}'.format(end - start))\n",
    "        print ()\n",
    "\n",
    "# The sequence object also holds metadata regarding the length of the sequence. \n",
    "sequence_window = (plotter.meta['start'], plotter.meta['end'])\n",
    "print (sequence_window)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABlAAAAiPCAYAAAC2I9hzAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xc41ccf8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "annotation_names = plotter.targets.columns.difference(['start', 'end'])\n",
    "\n",
    "# Select only the first minute of data \n",
    "sub_df = plotter.targets.ix[:60 * 2]\n",
    "\n",
    "# Select only the columns that are non-empty \n",
    "sub_df_cols = [col for col in annotation_names if sub_df[col].sum() > 0]\n",
    "\n",
    "# Plot a bar-plot w\n",
    "current_palette = cycle(sns.color_palette())\n",
    "sub_df[sub_df_cols].plot(\n",
    "    kind='bar', \n",
    "    subplots=True, \n",
    "    sharex=True, \n",
    "    sharey=True, \n",
    "    figsize=(20, 3 * len(sub_df_cols)), \n",
    "    width=1.0, \n",
    "    color=[next(current_palette) for _ in annotation_names]\n",
    ");"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Simple Feature extraction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from visualise_data import Sequence\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "\"\"\"\n",
    "For every data modality, we will extract some very simple features: the mean, min, max, median, and standard \n",
    "deviation of one second windows. We will put function pointers in a list called 'feature_functions' so that we \n",
    "can easily call these on our data for later\n",
    "\"\"\"\n",
    "feature_functions = [np.mean, np.std, np.min, np.median, np.max]\n",
    "feature_names = ['mean', 'std', 'min', 'median', 'max']\n",
    "\n",
    "# We will keep the number of extracted feature functions as a parameter \n",
    "num_ff = len(feature_functions)\n",
    "\n",
    "# We will want to keep track of the feature names for later, so we will collect these in the following list: \n",
    "column_names = []\n",
    "\n",
    "# These are the modalities that are available in the dataset, and the .iterate() function returns the data \n",
    "# in this order\n",
    "modality_names = ['acceleration', 'rssi', 'pir', 'video_living_room', 'video_kitchen', 'video_hallway']\n",
    "\n",
    "\"\"\"\n",
    "Iterate over all training directories\n",
    "\"\"\"\n",
    "for train_test in ('train', 'test', ): \n",
    "#     if train_test is 'train': \n",
    "#         print ('Extracting features from training data.\\n')\n",
    "#      else: \n",
    "#         print ('\\n\\n\\nExtracting features from testing data.\\n')\n",
    "        \n",
    "    for fi, file_id in enumerate(os.listdir('public_data/{}/'.format(train_test))):\n",
    "        stub_name = str(file_id).zfill(5)\n",
    "\n",
    "#         if train_test is 'train' or np.mod(fi, 50) == 0:\n",
    "#             print (\"Starting feature extraction for {}/{}\".format(train_test, stub_name))\n",
    "\n",
    "        # Use the sequence loader to load the data from the directory. \n",
    "        data = Sequence('public_data', 'public_data/{}/{}'.format(train_test, stub_name))\n",
    "        data.load()\n",
    "\n",
    "        \"\"\"\n",
    "        Populate the column_name list here. This needs to only be done on the first iteration\n",
    "        because the column names will be the same between datasets. \n",
    "        \"\"\"\n",
    "        if len(column_names) == 0:\n",
    "            for lu, modalities in data.iterate():\n",
    "                for modality, modality_name in zip(modalities, modality_names):\n",
    "                    for column_name, column_data in modality.transpose().iterrows():\n",
    "                        for feature_name in feature_names:\n",
    "                            column_names.append('{0}_{1}_{2}'.format(modality_name, column_name, feature_name))\n",
    "\n",
    "                # Break here \n",
    "                break \n",
    "\n",
    "        \"\"\"\n",
    "        Here, we will extract some features from the data. We will use the Sequence.iterate function. \n",
    "\n",
    "        This function partitions the data into one-second dataframes for the full data sequence. The first element \n",
    "        (which we call lu) represents the lower and upper times of the sliding window, and the second is a list of\n",
    "        dataframes, one for each modality. The dataframes may be empty (due to missing data), and feature extraction \n",
    "        has to be able to cope with this! \n",
    "\n",
    "        The list rows will store the features extracted for this dataset\n",
    "        \"\"\"\n",
    "        rows = []\n",
    "\n",
    "        for ri, (lu, modalities) in enumerate(data.iterate()):\n",
    "            row = []\n",
    "\n",
    "            \"\"\"\n",
    "            Iterate over the sensing modalities. The order is given in modality_names. \n",
    "\n",
    "            Note: If you want to treat each modality differently, you can do the following: \n",
    "\n",
    "            for ri, (lu, (accel, rssi, pir, vid_lr, vid_k, vid_h)) in enumerate(data.iterate()):\n",
    "                row.extend(extract_accel(accel))\n",
    "                row.extend(extract_rssi(rssi))\n",
    "                row.extend(extract_pir(pir))\n",
    "                row.extend(extract_video(vid_lr, vid_k, vid_h))\n",
    "\n",
    "            where the extract_accel/extract_rssi/extract_pir/extract_video functions are designed by you. \n",
    "            In the case here, we extract the same features from each modality, but optimal performance will \n",
    "            probably be achieved by considering different features for each modality. \n",
    "            \"\"\"\n",
    "\n",
    "            for modality in modalities:\n",
    "                \"\"\"\n",
    "                The accelerometer dataframe, for example, has three columns: x, y, and z. We want to extract features \n",
    "                from all of these, and so we iterate over the columns here. \n",
    "                \"\"\"\n",
    "                for name, column_data in modality.transpose().iterrows():\n",
    "                    if len(column_data) > 3:\n",
    "                        \"\"\"\n",
    "                        Extract the features stored in feature_functions on the column data if there is sufficient \n",
    "                        data in the dataframe. \n",
    "                        \"\"\"\n",
    "                        row.extend(map(lambda ff: ff(column_data), feature_functions))\n",
    "\n",
    "                    else:\n",
    "                        \"\"\"\n",
    "                        If no data is available, put nan placeholders to keep the column widths consistent\n",
    "                        \"\"\"\n",
    "                        row.extend([np.nan] * num_ff)\n",
    "\n",
    "            # Do a quick sanity check to ensure that the feature names and number of extracted features match\n",
    "            assert len(row) == len(column_names)\n",
    "\n",
    "            # Append the row to the full set of features\n",
    "            rows.append(row)\n",
    "\n",
    "            # Report progress \n",
    "#             if train_test is 'train':\n",
    "#                 if np.mod(ri + 1, 50) == 0:\n",
    "#                     print (\"{:5}\".format(str(ri + 1))),\n",
    "\n",
    "#                 if np.mod(ri + 1, 500) == 0:\n",
    "#                     print\n",
    "\n",
    "        \"\"\"\n",
    "        At this stage we have extracted a bunch of simple features from the data. In real implementation, \n",
    "        it would be advisable to look at more interesting features, eg\n",
    "\n",
    "          * acceleration: link\n",
    "          * environmental: link\n",
    "          * video: link\n",
    "\n",
    "        We will save these features to a new file called 'columns.csv' for use later. This file will be located \n",
    "        in the name of the training sequence. \n",
    "        \"\"\"\n",
    "        df = pd.DataFrame(rows)\n",
    "        df.columns = column_names\n",
    "        df.to_csv('public_data/{}/{}/simplefeatures.csv'.format(train_test, stub_name), index=False)\n",
    "\n",
    "#         if train_test is 'train' or np.mod(fi, 50) == 0:\n",
    "#             if train_test is 'train': print \n",
    "#             print (\"Finished feature extraction for {}/{}\\n\".format(train_test, stub_name))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Advanced Feature extraction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from visualise_data import Sequence\n",
    "import math\n",
    "from scipy.fftpack import fft, rfft\n",
    "from statsmodels import robust\n",
    "import scipy\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "feature_functions = [np.mean, np.std, np.min, np.median, np.max,scipy.stats.kurtosis,robust.mad, scipy.stats.entropy]\n",
    "feature_functions1 = [np.mean, np.std, np.min, np.median, np.max, scipy.stats.kurtosis,robust.mad]\n",
    "feature_functions2 = [np.max, np.mean]\n",
    "feature_functions3 = [np.mean, scipy.stats.kurtosis]\n",
    "feature_functions_video = [np.mean, np.std]\n",
    "feature_names0 = ['mean', 'std', 'min', 'median', 'max', 'kurtosis','mad','entropy','abs_max', 'abs_mean', 'fft_mean', 'fft_kurtosis']\n",
    "feature_names1 = ['mean', 'std', 'min', 'median', 'max', 'kurtosis','mad']\n",
    "feature_names_video = ['mean','std']\n",
    "feature_names_video1 = ['mean','std','0','-1']\n",
    "# We will keep the number of extracted feature functions as a parameter \n",
    "num_ff0 = len(feature_functions)+len(feature_functions2)+len(feature_functions3)\n",
    "num_ff1 = len(feature_functions1)\n",
    "num_video=len(feature_functions_video)\n",
    "num_video1=len(feature_functions_video)+2\n",
    "#num_ff1 = len(feature_functions)+len(feature_functions2)#+len(feature_functions2)+len(feature_functions3)\n",
    "\n",
    "# We will want to keep track of the feature names for later, so we will collect these in the following list: \n",
    "column_names0 = []\n",
    "column_names1 = []\n",
    "column_names2 = []\n",
    "column_names3 = []\n",
    "column_names4 = []\n",
    "column_names5 = []\n",
    "column_names_accel = []\n",
    "column_names_environment = []\n",
    "column_names_video = []\n",
    "column_names_all = []\n",
    "# These are the modalities that are available in the dataset, and the .iterate() function returns the data \n",
    "# in this order\n",
    "#modality_names = ['accel']\n",
    "for train_test in ('train', 'test', ): \n",
    "#     if train_test is 'train': \n",
    "#         print ('Extracting features from training data.\\n')\n",
    "#      else: \n",
    "#         print ('\\n\\n\\nExtracting features from testing data.\\n')\n",
    "        \n",
    "    for fi, file_id in enumerate(os.listdir('public_data/{}/'.format(train_test))):\n",
    "        stub_name = str(file_id).zfill(5)\n",
    "\n",
    "#         if train_test is 'train' or np.mod(fi, 50) == 0:\n",
    "#             print (\"Starting feature extraction for {}/{}\".format(train_test, stub_name))\n",
    "\n",
    "        # Use the sequence loader to load the data from the directory. \n",
    "        data = Sequence('public_data', 'public_data/{}/{}'.format(train_test, stub_name))\n",
    "        data.load()\n",
    "        \n",
    "        if len(column_names0) == 0:\n",
    "            for ri, (lu, (accel, rssi, pir, vid_lr, vid_k, vid_h)) in enumerate(data.iterate()):\n",
    "                #for modality, modality_name in zip(accel, modality_names):\n",
    "                for column_name0, column_data0 in accel.transpose().iterrows():\n",
    "                    for feature_name0 in feature_names0:\n",
    "                        column_names0.append('{0}_{1}_{2}'.format('accel', column_name0, feature_name0))\n",
    "                #column_names0.append('mean_jerk')\n",
    "                break\n",
    "                \n",
    "        if len(column_names1) == 0:\n",
    "            for ri, (lu, (accel, rssi, pir, vid_lr, vid_k, vid_h)) in enumerate(data.iterate()):    \n",
    "                for column_name1, column_data1 in rssi.transpose().iterrows():\n",
    "                    for feature_name1 in feature_names1:\n",
    "                        column_names1.append('{0}_{1}_{2}'.format('rssi', column_name1, feature_name1))\n",
    "                # Break here \n",
    "                break\n",
    "\n",
    "        if len(column_names2) == 0:\n",
    "            for ri, (lu, (accel, rssi, pir, vid_lr, vid_k, vid_h)) in enumerate(data.iterate()):    \n",
    "                for column_name2, column_data1 in pir.transpose().iterrows():\n",
    "                    for feature_name2 in feature_names1:\n",
    "                        column_names2.append('{0}_{1}_{2}'.format('pir', column_name2, feature_name2))\n",
    "                # Break here \n",
    "                break\n",
    "\n",
    "        if len(column_names3) == 0:\n",
    "            for ri, (lu, (accel, rssi, pir, vid_lr, vid_k, vid_h)) in enumerate(data.iterate()):    \n",
    "                for column_name3, column_data3 in vid_lr.transpose().iterrows():\n",
    "                    for feature_name3 in feature_names_video:\n",
    "                        column_names3.append('{0}_{1}_{2}'.format('vid_lr', column_name3, feature_name3))\n",
    "                column_names3.append('weight')\n",
    "                column_names3.append('height')\n",
    "                column_names3.append('mean_area')\n",
    "                column_names3.append('weight3d')\n",
    "                column_names3.append('height3d')\n",
    "                column_names3.append('length3d')\n",
    "                column_names3.append('volume')\n",
    "                column_names3.append('distance2d')\n",
    "                column_names3.append('distance3d')\n",
    "                # Break here \n",
    "                break\n",
    "                \n",
    "        if len(column_names4) == 0:\n",
    "            for ri, (lu, (accel, rssi, pir, vid_lr, vid_k, vid_h)) in enumerate(data.iterate()):    \n",
    "                for column_name4, column_data4 in vid_k.transpose().iterrows():\n",
    "                    for feature_name4 in feature_names_video:\n",
    "                        column_names4.append('{0}_{1}_{2}'.format('vid_k', column_name4, feature_name4))\n",
    "                column_names4.append('weight')\n",
    "                column_names4.append('height')\n",
    "                column_names4.append('mean_area')\n",
    "                column_names4.append('weight3d')\n",
    "                column_names4.append('height3d')\n",
    "                column_names4.append('length3d')\n",
    "                column_names4.append('volume')\n",
    "                column_names4.append('distance2d')\n",
    "                column_names4.append('distance3d')\n",
    "                # Break here \n",
    "                break\n",
    "                \n",
    "        if len(column_names5) == 0:\n",
    "            for ri, (lu, (accel, rssi, pir, vid_lr, vid_k, vid_h)) in enumerate(data.iterate()):    \n",
    "                for column_name5, column_data5 in vid_h.transpose().iterrows():\n",
    "                    for feature_name5 in feature_names_video:\n",
    "                        column_names5.append('{0}_{1}_{2}'.format('vid_h', column_name5, feature_name5))\n",
    "                column_names5.append('weight')\n",
    "                column_names5.append('height')\n",
    "                column_names5.append('mean_area')\n",
    "                column_names5.append('weight3d')\n",
    "                column_names5.append('height3d')\n",
    "                column_names5.append('length3d')\n",
    "                column_names5.append('volume')\n",
    "                column_names5.append('distance2d')\n",
    "                column_names5.append('distance3d')\n",
    "                # Break here \n",
    "                break\n",
    "                \n",
    "        column_names_all = column_names0+column_names1+column_names2+column_names3+column_names4+column_names5\n",
    "        column_names_acceleration = column_names0\n",
    "        column_names_environment = column_names1+column_names2\n",
    "        column_names_video = column_names2+column_names3+column_names4+column_names5\n",
    "        column_names_vid_accel = column_names0+column_names1+column_names3+column_names4+column_names5\n",
    "        \n",
    "        rows0=[]\n",
    "        rows1=[]\n",
    "        rows2=[]\n",
    "        rows3=[]\n",
    "        rows4=[]\n",
    "        rows5=[]\n",
    "        for ri, (lu, (accel, rssi, pir, vid_lr, vid_k, vid_h)) in enumerate(data.iterate()):\n",
    "            row0 = []\n",
    "            row1 = []\n",
    "            row2 = []\n",
    "            row3 = []\n",
    "            row4 = []\n",
    "            row5 = []\n",
    "            for name0, column_data0 in accel.transpose().iterrows():\n",
    "                if len(column_data0) > 3:\n",
    "                    A = np.abs(column_data0)\n",
    "                    B = column_data0\n",
    "                    C = rfft(B)\n",
    "                    row0.extend(map(lambda ff: ff(column_data0), feature_functions))\n",
    "                    row0.extend(map(lambda ff: ff(A), feature_functions2))\n",
    "                    row0.extend(map(lambda ff: ff(C), feature_functions3))\n",
    "                else:\n",
    "                    row0.extend([np.nan] * num_ff0)\n",
    "            #x=math.sqrt(pow(row0[0],2)+pow(row0[10],2)+pow(row0[20],2))\n",
    "            #row0.append(x)         \n",
    "            assert len(row0) == len(column_names0)\n",
    "            rows0.append(row0)\n",
    "            \n",
    "            for name1, column_data1 in rssi.transpose().iterrows():\n",
    "                if len(column_data1) > 3:\n",
    "                    #A1 = np.abs(column_data1)\n",
    "                    row1.extend(map(lambda ff: ff(column_data1), feature_functions1))\n",
    "                    #row1.extend(map(lambda ff: ff(A1), feature_functions2))\n",
    "                else:\n",
    "                    row1.extend([np.nan] * num_ff1)  \n",
    "            assert len(row1) == len(column_names1)\n",
    "            rows1.append(row1)\n",
    "\n",
    "            for name2, column_data2 in pir.transpose().iterrows():\n",
    "                if len(column_data2) > 3:\n",
    "                    #A1 = np.abs(column_data1)\n",
    "                    row2.extend(map(lambda ff: ff(column_data2), feature_functions1))\n",
    "                    #row1.extend(map(lambda ff: ff(A1), feature_functions2))\n",
    "                else:\n",
    "                    row2.extend([np.nan] * num_ff1)\n",
    "                    \n",
    "            assert len(row2) == len(column_names2)\n",
    "            rows2.append(row2)\n",
    "\n",
    "            for name3, column_data3 in vid_lr.transpose().iterrows():\n",
    "                if len(column_data3) > 3:\n",
    "                    #A3 = np.abs(column_data3)\n",
    "                    row3.extend(map(lambda ff: ff(column_data3), feature_functions_video))\n",
    "                    row3.append(np.array(column_data3)[0])\n",
    "                    row3.append(np.array(column_data3)[-1])\n",
    "                    #row3.extend(map(lambda ff: ff(A3), feature_functions2))\n",
    "                else:\n",
    "                    row3.extend([np.nan] * num_video1) \n",
    "            \n",
    "            weight=abs(row3[8]-row3[16])\n",
    "            height=abs(row3[12]-row3[20])\n",
    "            area=weight*height\n",
    "            weight3d=abs(row3[36]-row3[48])\n",
    "            height3d=abs(row3[40]-row3[52])\n",
    "            length3d=abs(row3[44]-row3[56])\n",
    "            volume=weight3d*height3d*length3d\n",
    "            distance2d=math.sqrt((pow((row3[2]-row3[3]),2))+(pow((row3[6]-row3[7]),2)))\n",
    "            distance3d=math.sqrt((pow((row3[26]-row3[27]),2))+(pow((row3[30]-row3[31]),2))+(pow((row3[34]-row3[35]),2)))\n",
    "            row3.append(weight)\n",
    "            row3.append(height)\n",
    "            row3.append(area)\n",
    "            row3.append(weight3d)\n",
    "            row3.append(height3d)\n",
    "            row3.append(length3d)\n",
    "            row3.append(volume)\n",
    "            row3.append (distance2d)\n",
    "            row3.append (distance3d)\n",
    "            a3=np.delete(row3,[2,3,6,7,10,11,14,15,18,19,22,23,26,27,30,31,34,35,38,39,42,43,46,47,50,51,54,55,58,59])\n",
    "            assert len(a3) == len(column_names3)\n",
    "            \n",
    "            rows3.append(a3)\n",
    "            \n",
    "            for name4, column_data4 in vid_k.transpose().iterrows():\n",
    "                if len(column_data4) > 3:\n",
    "                    #A4 = np.abs(column_data4)\n",
    "                    row4.extend(map(lambda ff: ff(column_data4), feature_functions_video))\n",
    "                    #row4.extend(map(lambda ff: ff(A4), feature_functions2))\n",
    "                    row4.append(np.array(column_data4)[0])\n",
    "                    row4.append(np.array(column_data4)[-1])\n",
    "                else:\n",
    "                    row4.extend([np.nan] * num_video1)  \n",
    "            weight=abs(row4[8]-row4[16])\n",
    "            height=abs(row4[12]-row4[20])\n",
    "            area=weight*height\n",
    "            weight3d=abs(row4[36]-row4[48])\n",
    "            height3d=abs(row4[40]-row4[52])\n",
    "            length3d=abs(row4[44]-row4[56])\n",
    "            volume=weight3d*height3d*length3d\n",
    "            distance2d=math.sqrt((pow((row4[2]-row4[3]),2))+(pow((row4[6]-row4[7]),2)))\n",
    "            distance3d=math.sqrt((pow((row4[26]-row4[27]),2))+(pow((row4[30]-row4[31]),2))+(pow((row4[34]-row4[35]),2)))\n",
    "            row4.append(weight)\n",
    "            row4.append(height)\n",
    "            row4.append(area)\n",
    "            row4.append(weight3d)\n",
    "            row4.append(height3d)\n",
    "            row4.append(length3d)\n",
    "            row4.append(volume)\n",
    "            row4.append (distance2d)\n",
    "            row4.append (distance3d)\n",
    "            a4=np.delete(row4,[2,3,6,7,10,11,14,15,18,19,22,23,26,27,30,31,34,35,38,39,42,43,46,47,50,51,54,55,58,59])\n",
    "            assert len(a4) == len(column_names4)\n",
    "            rows4.append(a4)\n",
    "\n",
    "            for name5, column_data5 in vid_h.transpose().iterrows():\n",
    "                if len(column_data5) > 3:\n",
    "                    #A5 = np.abs(column_data5)\n",
    "                    row5.extend(map(lambda ff: ff(column_data5), feature_functions_video))\n",
    "                    #row5.extend(map(lambda ff: ff(A5), feature_functions2))\n",
    "                    row5.append(np.array(column_data5)[0])\n",
    "                    row5.append(np.array(column_data5)[-1])\n",
    "                else:\n",
    "                    row5.extend([np.nan] * num_video1)  \n",
    "            weight=abs(row5[8]-row5[16])\n",
    "            height=abs(row5[12]-row5[20])\n",
    "            area=weight*height\n",
    "            weight3d=abs(row5[36]-row5[48])\n",
    "            height3d=abs(row5[40]-row5[52])\n",
    "            length3d=abs(row5[44]-row5[56])\n",
    "            volume=weight3d*height3d*length3d\n",
    "            distance2d=math.sqrt((pow((row3[2]-row3[3]),2))+(pow((row3[6]-row3[7]),2)))\n",
    "            distance3d=math.sqrt((pow((row3[26]-row3[27]),2))+(pow((row3[30]-row3[31]),2))+(pow((row3[34]-row3[35]),2)))\n",
    "            row5.append(weight)\n",
    "            row5.append(height)\n",
    "            row5.append(area)\n",
    "            row5.append(weight3d)\n",
    "            row5.append(height3d)\n",
    "            row5.append(length3d)\n",
    "            row5.append(volume)\n",
    "            row5.append (distance2d)\n",
    "            row5.append (distance3d)\n",
    "            a5=np.delete(row5,[2,3,6,7,10,11,14,15,18,19,22,23,26,27,30,31,34,35,38,39,42,43,46,47,50,51,54,55,58,59])\n",
    "            assert len(a5) == len(column_names5)\n",
    "            rows5.append(a5)\n",
    "            \n",
    "        df = pd.DataFrame(np.CC((rows0,rows1,rows2,rows3,rows4,rows5),axis=1))\n",
    "        #df = pd.DataFrame(rows0)\n",
    "        #df = pd.DataFrame(np.concatenate((rows0,rows1, rows3, rows4, rows5),axis=1))\n",
    "        df.columns = column_names_all\n",
    "        df.to_csv('public_data/{}/{}/advancedfeatures.csv'.format(train_test, stub_name), index=False)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data Imputation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Check whether the train/test features are all finite (before imputation)\n",
      "All training data finite: False\n",
      "All testing data finite: False\n",
      "Check whether the train/test features are all finite (after imputation)\n",
      "All training data finite: True\n",
      "All testing data finite: True\n",
      "Training data shapes:\n",
      "train_x.shape: (12520L, 241L)\n",
      "train_y.shape: (12520L, 20L)\n",
      "Testing data shapes\n",
      "test_x.shape: (3604L, 241L)\n",
      "test_y.shape: (3604L, 20L)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<function print>"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\"\"\"\n",
    "We will define two convenience function to load the extracted features and their \n",
    "\"\"\"\n",
    "\n",
    "\n",
    "def load_sequence(file_id):\n",
    "    filename = str(file_id).zfill(5)\n",
    "\n",
    "    df = pd.read_csv('public_data/train/{}/advancedfeatures.csv'.format(filename))\n",
    "    data = df.values\n",
    "    target = np.asarray(pd.read_csv('public_data/train/{}/targets.csv'.format(filename)))[:, 2:]\n",
    "\n",
    "    return data, target\n",
    "\n",
    "\n",
    "def load_sequences(file_ids):\n",
    "    x_es = []\n",
    "    y_es = []\n",
    "\n",
    "    for file_id in file_ids:\n",
    "        data, target = load_sequence(file_id)\n",
    "\n",
    "        x_es.append(data)\n",
    "        y_es.append(target)\n",
    "\n",
    "    return np.row_stack(x_es), np.row_stack(y_es)\n",
    "\n",
    "\n",
    "# Load the training and testing data \n",
    "train_x, train_y = load_sequences([1, 2, 3, 4, 5, 6, 7, 8])\n",
    "test_x, test_y = load_sequences([9, 10])\n",
    "\n",
    "print (\"Check whether the train/test features are all finite (before imputation)\")\n",
    "print ('All training data finite:', np.all(np.isfinite(train_x)))\n",
    "print ('All testing data finite:', np.all(np.isfinite(test_x)))\n",
    "print \n",
    "\n",
    "# We will want to impute the missing data \n",
    "from sklearn.preprocessing import Imputer\n",
    "imputer = Imputer()\n",
    "imputer.fit(train_x)\n",
    "\n",
    "train_x = imputer.transform(train_x)\n",
    "test_x = imputer.transform(test_x)\n",
    "\n",
    "print (\"Check whether the train/test features are all finite (after imputation)\")\n",
    "print ('All training data finite:', np.all(np.isfinite(train_x)))\n",
    "print ('All testing data finite:', np.all(np.isfinite(test_x)))\n",
    "print \n",
    "\n",
    "# Load the label names \n",
    "labels = json.load(open('public_data/annotations.json'))\n",
    "n_classes = len(labels)\n",
    "\n",
    "\"\"\"\n",
    "Note, not all data is annotated, so we select only the annotated rows\n",
    "\"\"\"\n",
    "train_y_has_annotation = np.isfinite(train_y.sum(1))\n",
    "train_x = train_x[train_y_has_annotation]\n",
    "train_y = train_y[train_y_has_annotation]\n",
    "\n",
    "test_y_has_annotation = np.isfinite(test_y.sum(1))\n",
    "test_x = test_x[test_y_has_annotation]\n",
    "test_y = test_y[test_y_has_annotation]\n",
    "\n",
    "\n",
    "\"\"\"\n",
    "Print simple statistics regarding the number of instances\n",
    "\"\"\"\n",
    "print (\"Training data shapes:\")\n",
    "print (\"train_x.shape: {}\".format(train_x.shape))\n",
    "print (\"train_y.shape: {}\".format(train_y.shape))\n",
    "print \n",
    "\n",
    "print (\"Testing data shapes\")\n",
    "print (\"test_x.shape: {}\".format(test_x.shape))\n",
    "print (\"test_y.shape: {}\".format(test_y.shape))\n",
    "print"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Gradient Boosting Decision Tree Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.multioutput import MultiOutputRegressor\n",
    "from sklearn.ensemble import GradientBoostingRegressor\n",
    "\n",
    "n_jobs = 4\n",
    "regr_multigbdt = MultiOutputRegressor(GradientBoostingRegressor(n_estimators=200, learning_rate=0.1,max_depth=3,max_features=110, random_state=0, loss='ls'),n_jobs)\n",
    "regr_multigbdt.fit(train_x, train_y)\n",
    "test_predicted_gbdt = regr_multigbdt.predict(test_x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "#for (x,y), value in np.ndenumerate(test_predicted_gbdt):\n",
    "    #if test_predicted_gbdt[x][y]< 0:\n",
    "        #test_predicted_gbdt[x][y] = 0\n",
    "    #if test_predicted_gbdt[x][y]< 0.001:\n",
    "        #test_predicted_gbdt[x][y] = 0\n",
    "g1=np.transpose(test_predicted_gbdt)\n",
    "g2=np.array(np.sum(test_predicted_gbdt, 1),dtype=float)\n",
    "test_predicted_gbdt=np.transpose(g1/g2) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Brier score on test set with the gbdt model\n",
      "0.0198393945762\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<function print>"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print (\"Brier score on test set with the gbdt model\")\n",
    "print (np.power(test_y-test_predicted_gbdt,2.0).mean())\n",
    "print "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "num_lines = 0\n",
    "\n",
    "se_cols = ['start', 'end']\n",
    "\n",
    "with open('submission_gbdt_AFs.csv', 'w') as fil: \n",
    "    fil.write(','.join(['index'] + annotation_names.tolist()))\n",
    "    fil.write('\\n')\n",
    "            \n",
    "    for te_ind_str in sorted(os.listdir(os.path.join('public_data', 'test'))):        \n",
    "        te_ind = int(te_ind_str)\n",
    "\n",
    "        meta = json.load(open(os.path.join('public_data', 'test', te_ind_str, 'meta.json')))\n",
    "        features = pd.read_csv(os.path.join('public_data', 'test', te_ind_str, 'advancedfeatures.csv')).values\n",
    "        features = imputer.transform(features)\n",
    "        #for (x,y), value in np.ndenumerate(test_predicted_gbdt):\n",
    "            #if test_predicted_gbdt[x][y]< 0:\n",
    "                #test_predicted_gbdt[x][y] = 0\n",
    "        probs = regr_multigbdt.predict(features)\n",
    "        #for (x,y), value in np.ndenumerate(probs):\n",
    "            #if probs[x][y]< 0:\n",
    "                #probs[x][y] = 0\n",
    "            #if probs[x][y]< 0.001:\n",
    "                #probs[x][y] = 0\n",
    "        g1=np.transpose(probs)\n",
    "        g2=np.array(np.sum(probs, 1))\n",
    "        probs_norm_gbdt=np.transpose(g1/g2) \n",
    "        pr=probs_norm_gbdt\n",
    "        for (x,y), value in np.ndenumerate(pr):\n",
    "            if pr[x][y]< 0:\n",
    "                pr[x][y] = 0\n",
    "            if pr[x][y]>1:\n",
    "                pr[x][y] = 1\n",
    "        starts = range(meta['end'])\n",
    "        ends = range(1, meta['end'] + 1)\n",
    "\n",
    "        for start, end, prob in zip(starts, ends, pr):\n",
    "            row = [str(te_ind) + '_' + str(start) + '_' + str(end)] + prob.tolist()\n",
    "\n",
    "            fil.write(','.join(map(str, row)))\n",
    "            fil.write('\\n')\n",
    "            \n",
    "            num_lines += 1\n",
    "            \n",
    "print (\"{} lines written.\".format(num_lines))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Random Forest Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.multioutput import MultiOutputRegressor\n",
    "from sklearn.ensemble import GradientBoostingRegressor\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "from sklearn.feature_selection import SelectFromModel\n",
    "\n",
    "regr_multirf = MultiOutputRegressor(RandomForestRegressor(max_depth= 30, random_state=0, n_estimators = 20))\n",
    "regr_multirf.fit(train_x, train_y)\n",
    "test_predicted_rf = regr_multirf.predict(test_x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "#for (x,y), value in np.ndenumerate(test_predicted_gbdt):\n",
    "    #if test_predicted_gbdt[x][y]< 0:\n",
    "        #test_predicted_gbdt[x][y] = 0\n",
    "    #if test_predicted_gbdt[x][y]< 0.001:\n",
    "        #test_predicted_gbdt[x][y] = 0\n",
    "g1=np.transpose(test_predicted_rf)\n",
    "g2=np.array(np.sum(test_predicted_rf, 1),dtype=float)\n",
    "test_predicted_gbdt=np.transpose(g1/g2) \n",
    "print (\"Brier score on test set with the rf model\")\n",
    "print (np.power(test_y-test_predicted_rf,2.0).mean())\n",
    "print "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "num_lines = 0\n",
    "\n",
    "se_cols = ['start', 'end']\n",
    "\n",
    "with open('submission_rf_AFs.csv', 'w') as fil: \n",
    "    fil.write(','.join(['index'] + annotation_names.tolist()))\n",
    "    fil.write('\\n')\n",
    "            \n",
    "    for te_ind_str in sorted(os.listdir(os.path.join('public_data', 'test'))):        \n",
    "        te_ind = int(te_ind_str)\n",
    "\n",
    "        meta = json.load(open(os.path.join('public_data', 'test', te_ind_str, 'meta.json')))\n",
    "        features = pd.read_csv(os.path.join('public_data', 'test', te_ind_str, 'advancedfeatures.csv')).values\n",
    "        features = imputer.transform(features)\n",
    "        #for (x,y), value in np.ndenumerate(test_predicted_gbdt):\n",
    "            #if test_predicted_gbdt[x][y]< 0:\n",
    "                #test_predicted_gbdt[x][y] = 0\n",
    "        probs = regr_multigbdt.predict(features)\n",
    "        #for (x,y), value in np.ndenumerate(probs):\n",
    "            #if probs[x][y]< 0:\n",
    "                #probs[x][y] = 0\n",
    "            #if probs[x][y]< 0.001:\n",
    "                #probs[x][y] = 0\n",
    "        g1=np.transpose(probs) \n",
    "        g2=np.array(np.sum(probs, 1))\n",
    "        probs_norm_rf =np.transpose(g1/g2) \n",
    "        pr=probs_norm_gbdt\n",
    "        for (x,y), value in np.ndenumerate(pr):\n",
    "            if pr[x][y]< 0:\n",
    "                pr[x][y] = 0\n",
    "            if pr[x][y]>1:\n",
    "                pr[x][y] = 1\n",
    "        starts = range(meta['end'])\n",
    "        ends = range(1, meta['end'] + 1)\n",
    "\n",
    "        for start, end, prob in zip(starts, ends, pr):\n",
    "            row = [str(te_ind) + '_' + str(start) + '_' + str(end)] + prob.tolist()\n",
    "\n",
    "            fil.write(','.join(map(str, row)))\n",
    "            fil.write('\\n')\n",
    "            \n",
    "            num_lines += 1\n",
    "            \n",
    "print (\"{} lines written.\".format(num_lines))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Recurrent Nueral Network\n",
    "   It cannot work..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from keras.models import Sequential  \n",
    "from keras.layers.core import Dense, Activation  \n",
    "from keras.layers.recurrent import LSTM\n",
    "from keras.layers import Embedding\n",
    "in_out_neurons = 2  \n",
    "hidden_neurons = 50\n",
    " \n",
    "input_size = train_x.shape[1]\n",
    "output_size = train_y.shape[1]\n",
    "hidden_size = 100\n",
    "hidden_size_2 = 200\n",
    "num_epochs = 10\n",
    "batch_size = 50\n",
    "\n",
    "nn = Sequential()\n",
    "nn.add(Embedding(input_size, hidden_size))\n",
    "nn.add(LSTM(hidden_size, return_sequences=False))\n",
    "nn.add(Dense(output_size))\n",
    "nn.add(Activation(\"linear\")) \n",
    "nn.compile(loss='mse', optimizer=\"rmsprop\")\n",
    "\n",
    "nn.fit(train_x, train_y, batch_size=batch_size, nb_epoch=num_epochs)\n",
    " \n",
    "predicted = model.predict(test_x)  \n",
    "rmse = np.sqrt(((predicted - test_y) ** 2).mean(axis=0))"
   ]
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python [default]",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}