Model2_Seizure_Demo.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Model2_Seizure_Demo.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "authorship_tag": "ABX9TyPssO7dNelixcVbHQayyAyY",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/josephmhernandez/SeizureDetection/blob/master/Model2_Seizure_Demo.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QrSjadpfbUqY",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#To run this demo: \n",
        "# 1. Upload model gru-do-49-0.01.hdf5\n",
        "# 2. run the !pip install block so that the model can be extracted correctly from the file\n",
        "# 2. Restart kernel so changes are made \n",
        "# 3. Run the next block of code (openPose and using the model to predict.)\n",
        "#\n",
        "#"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "L3YcC4mO1zj6",
        "colab_type": "code",
        "outputId": "285c1679-2c29-4055-91a4-4a6254fec519",
        "colab": {
          "resources": {
            "http://localhost:8080/nbextensions/google.colab/files.js": {
              "data": "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",
              "ok": true,
              "headers": [
                [
                  "content-type",
                  "application/javascript"
                ]
              ],
              "status": 200,
              "status_text": ""
            }
          },
          "base_uri": "https://localhost:8080/",
          "height": 75
        }
      },
      "source": [
        "#  Upload model\n",
        "from google.colab import files\n",
        "src = list(files.upload().values())[0]"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              "\n",
              "     <input type=\"file\" id=\"files-fc779342-44f1-4ff5-a4b8-3eeee531a4c3\" name=\"files[]\" multiple disabled />\n",
              "     <output id=\"result-fc779342-44f1-4ff5-a4b8-3eeee531a4c3\">\n",
              "      Upload widget is only available when the cell has been executed in the\n",
              "      current browser session. Please rerun this cell to enable.\n",
              "      </output>\n",
              "      <script src=\"/nbextensions/google.colab/files.js\"></script> "
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Saving rnn2-improvement-29-100.hdf5 to rnn2-improvement-29-100.hdf5\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "My4Qoe34IyKZ",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "!pip install --upgrade pip setuptools wheel\n",
        "!pip install -I tensorflow\n",
        "!pip install -I keras"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "F6LQGRFTAZpb",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#Input Data. \n",
        "\n",
        "#These are public videos to whoever has the link. \n",
        "#Test-vid-3-hd :  https://youtu.be/d5ibbcjHMUo\n",
        "#Test-vid-4-hd : https://youtu.be/_Cybo4bEWPs\n",
        "#Test-vid-5-hd : https://youtu.be/Fz6vWTFux8o\n",
        "#Test-vid-6-hd : https://youtu.be/XQbGXOXaEl0\n",
        "#Test-vid-7-hd : https://youtu.be/pyHrRS7ND28\n",
        "#Test-vid-8-hd : https://youtu.be/1VH-f8kVy6k\n",
        "#Extensive-Test-Vid-2-hd : https://youtu.be/qVLRVXymyEY\n",
        "# Extensive-Test-Vid-1-hd : https://youtu.be/ZAqmHwV0Wtg\n",
        "# Living-Room-Seizure-Test-Vid-1 : https://youtu.be/GnEjfiW8xlw\n",
        "link_video = 'https://youtu.be/1VH-f8kVy6k'\n",
        " "
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gaEZTqSVQNz_",
        "colab_type": "code",
        "outputId": "cda0ebea-f61b-4ef1-cd39-69a9eb837ea2",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "!pip install youtube_dl\n",
        "!pip install pafy\n",
        "\n",
        "#Load Open Pose\n",
        "import os\n",
        "from os.path import exists, join, basename, splitext\n",
        "\n",
        "git_repo_url = 'https://github.com/CMU-Perceptual-Computing-Lab/openpose.git'\n",
        "project_name = splitext(basename(git_repo_url))[0]\n",
        "if not exists(project_name):\n",
        "  # see: https://github.com/CMU-Perceptual-Computing-Lab/openpose/issues/949\n",
        "  # install new CMake becaue of CUDA10\n",
        "  !wget -q https://cmake.org/files/v3.13/cmake-3.13.0-Linux-x86_64.tar.gz\n",
        "  !tar xfz cmake-3.13.0-Linux-x86_64.tar.gz --strip-components=1 -C /usr/local\n",
        "  # clone openpose\n",
        "  !git clone -q --depth 1 $git_repo_url\n",
        "  !sed -i 's/execute_process(COMMAND git checkout master WORKING_DIRECTORY ${CMAKE_SOURCE_DIR}\\/3rdparty\\/caffe)/execute_process(COMMAND git checkout f019d0dfe86f49d1140961f8c7dec22130c83154 WORKING_DIRECTORY ${CMAKE_SOURCE_DIR}\\/3rdparty\\/caffe)/g' openpose/CMakeLists.txt\n",
        "  # install system dependencies\n",
        "  !apt-get -qq install -y libatlas-base-dev libprotobuf-dev libleveldb-dev libsnappy-dev libhdf5-serial-dev protobuf-compiler libgflags-dev libgoogle-glog-dev liblmdb-dev opencl-headers ocl-icd-opencl-dev libviennacl-dev\n",
        "  # install python dependencies\n",
        "  !pip install -q youtube-dl\n",
        "  # build openpose\n",
        "  !cd openpose && rm -rf build || true && mkdir build && cd build && cmake .. && make -j`nproc`\n",
        "\n",
        "\n",
        "#Function: save video in colab memory \n",
        "!pip install youtube_dl\n",
        "!pip install pafy\n",
        "\n",
        "import youtube_dl\n",
        "import os\n",
        "import pafy\n",
        "import numpy as np\n",
        "\n",
        "def create_vid(link):\n",
        "    ydl = youtube_dl.YoutubeDL({'outtmpl': '%(id)s%(ext)s',\n",
        "            'noplaylist' : True})\n",
        "    link_video = link\n",
        "    with ydl:\n",
        "        result = ydl.extract_info(\n",
        "            link_video,\n",
        "            download=True\n",
        "        )\n",
        "\n",
        "    if 'entries' in result:\n",
        "        # Can be a playlist or a list of videos\n",
        "        video = result['entries'][0]\n",
        "    else:\n",
        "        # Just a video\n",
        "        video = result\n",
        "\n",
        "    # print(video)\n",
        "    title = video['id']\n",
        "\n",
        "    file_list = np.array(os.listdir(path=None))\n",
        "    video_name = None\n",
        "    # print(file_list)\n",
        "    for a in file_list:\n",
        "        if(title in a):\n",
        "            print('here')\n",
        "            video_name = a\n",
        "            break\n",
        "\n",
        "    nvideo = pafy.new(link_video)\n",
        "    video_duration=nvideo.length \n",
        "\n",
        "    os.rename(video_name, nvideo.title + '.' + video['ext'])\n",
        "    video_name = nvideo.title + '.' + video['ext']\n",
        "    title = nvideo.title\n",
        "    return video_name, title\n",
        "\n",
        "# Function\n",
        "# Iterate through JSON files of a video and save every 120 as an npy file. \n",
        "import json\n",
        "import numpy as np\n",
        "import os\n",
        "def create_120_npy_from_vid_json(read_json_folder, write_folder, save_name):\n",
        "    newJsonOutput = read_json_folder\n",
        "    # newJsonOutput = 'gdrive/My Drive/Seizure2/T10_ks3_hd/'\n",
        "    print('Change JSON files to npy files....')\n",
        "    jsonFiles = np.array(os.listdir(newJsonOutput))\n",
        "    jsonFiles.sort()\n",
        "\n",
        "    rtnList = np.zeros((len(jsonFiles),50)) \n",
        "\n",
        "    # print(jsonFiles[:7])\n",
        "    print('THIS ONE RIGHT HERE.>')\n",
        "    print(len(jsonFiles))\n",
        "    # print('total npy outputs: ' + str(len(jsonFiles) // 120))\n",
        "\n",
        "    for i,jsonFile in enumerate(jsonFiles): \n",
        "        if (i % 1000 == 0): print('\\tupdate:' + str(i))\n",
        "        with open(newJsonOutput + '/' + jsonFiles[i]) as jsonOutput:\n",
        "            data = json.load(jsonOutput)\n",
        "            try:\n",
        "                bodyLandmarks = data['people'][0]['pose_keypoints_2d']    \n",
        "                #Get rid of the confidence of points\n",
        "                bodyLandmarks = [xx for p, xx in enumerate(bodyLandmarks) if p%3 !=2]\n",
        "                # print(bodyLandmarks)\n",
        "                #Y - coordinates are opposite. Makes printing out coordinates easier to visualize. \n",
        "                for inc in range(len(bodyLandmarks)):\n",
        "                    if inc % 2 == 1:\n",
        "                        bodyLandmarks[inc] = 1 - bodyLandmarks[inc]\n",
        "                \n",
        "                rtnList[i,] = bodyLandmarks\n",
        "            except:\n",
        "                print('oof.' + str(i))\n",
        "                # print(data)\n",
        "                continue\n",
        "\n",
        "    # !mkdir {write_npy_files}\n",
        "    np.save(write_folder+save_name+'.npy', rtnList)\n",
        "    # print('write_npys')\n",
        "    # _count_id = 1\n",
        "\n",
        "    # for j in range(120, len(rtnList), 120):\n",
        "    #     _temp_data = rtnList[j-120:j]\n",
        "    #     np.save(write_folder + save_name + '_' + str(_count_id) + '.npy', _temp_data)\n",
        "    #     _count_id += 1\n",
        "    \n",
        "    print('finished')\n",
        "\n",
        "\n",
        "# Generate OpenPose Outputs \n",
        "\n",
        "_name, title = create_vid(link_video)\n",
        "json_output = '/content/temp'+title +'/'\n",
        "\n",
        "newvidInput = _name\n",
        "\n",
        "train_folder = '/content/'\n",
        "\n",
        "\n",
        "print(_name)\n",
        "newJsonOutput = json_output\n",
        "py_json_folder = json_output\n",
        "!./openpose/build/examples/openpose/openpose.bin --tracking 10 --render_pose 0 --video {newvidInput} --write_json {newJsonOutput} --model_folder openpose/models --disable_blending --display 0 --number_people_max 1 --scale_number 4 --scale_gap 0.25 --keypoint_scale 3\n",
        "\n",
        "write_npy_files = train_folder\n",
        "\n",
        "# !mkdir gdrive/My\\ Drive/Seizure2/all_data_hd_npy/\n",
        "\n",
        "create_120_npy_from_vid_json(read_json_folder=py_json_folder, write_folder=write_npy_files, save_name=title)\n",
        "\n",
        "\n",
        "#Delete json folder\n",
        "!rm -rf {newJsonOutput}\n",
        "#Delete video \n",
        "!rm -f {newvidInput}\n",
        "\n",
        "\n",
        "\n",
        "#Load model. \n",
        "from keras.models import load_model\n",
        "import numpy as np\n",
        "import os\n",
        "import json \n",
        "#Define function to give us list inputs from 120,50 data:\n",
        "def slice_data(opened_data):\n",
        "    _temp_r_arm_data = np.empty((120,6))\n",
        "    _temp_l_arm_data = np.empty((120,6))\n",
        "    _temp_r_leg_data = np.empty((120, 12))\n",
        "    _temp_l_leg_data = np.empty((120, 12))\n",
        "    _temp_torso_data = np.empty((120, 12))\n",
        "\n",
        "    for j,mat in enumerate(opened_data):\n",
        "        #iterate 120 timestamps\n",
        "        mult = 2\n",
        "        #(120, 6) pad with 6 zeros \n",
        "        _temp_r_arm_data[j,] = np.array([mat[2*mult], mat[mult*2+1], \n",
        "                                    mat[3*mult], mat[mult*3+1],\n",
        "                                    mat[4*mult], mat[mult*4+1]])\n",
        "        \n",
        "        #(120, 6) pad with 6 zeros\n",
        "        _temp_l_arm_data[j,] = np.array([mat[5*mult], mat[mult*5+1], \n",
        "                                    mat[6*mult], mat[mult*6+1],\n",
        "                                    mat[7*mult], mat[mult*7+1]])\n",
        "        \n",
        "        #(120, 12)\n",
        "        _temp_r_leg_data[j,] = np.array([mat[9*mult], mat[mult*9+1], \n",
        "                                    mat[10*mult], mat[mult*10+1],\n",
        "                                    mat[11*mult], mat[mult*11+1],\n",
        "                                    mat[22*mult], mat[mult*22+1], \n",
        "                                    mat[23*mult], mat[mult*23+1],\n",
        "                                    mat[24*mult], mat[mult*24+1]])\n",
        "        #(120, 12)\n",
        "        _temp_l_leg_data[j,] = np.array([mat[12*mult], mat[mult*12+1], \n",
        "                                    mat[13*mult], mat[mult*13+1],\n",
        "                                    mat[14*mult], mat[mult*14+1],\n",
        "                                    mat[19*mult], mat[mult*19+1], \n",
        "                                    mat[20*mult], mat[mult*20+1],\n",
        "                                    mat[21*mult], mat[mult*21+1]]) \n",
        "        \n",
        "        #(120, 12)\n",
        "        _temp_torso_data[j,] = np.array([mat[1*mult], mat[mult*1+1], \n",
        "                                    mat[2*mult], mat[mult*2+1],\n",
        "                                    mat[5*mult], mat[mult*5+1],\n",
        "                                    mat[8*mult], mat[mult*8+1], \n",
        "                                    mat[9*mult], mat[mult*9+1],\n",
        "                                    mat[12*mult], mat[mult*12+1]])\n",
        "    \n",
        "    return [_temp_torso_data.reshape(1,120,12), _temp_l_arm_data.reshape(1,120,6), _temp_r_arm_data.reshape(1,120,6), _temp_l_leg_data.reshape(1,120,12), _temp_r_leg_data.reshape(1,120,12)]\n",
        "\n",
        "model_path = 'gru-do-49-0.01.hdf5'\n",
        "my_model = load_model(model_path)\n",
        "\n",
        "\n",
        "data = np.load(write_npy_files + title + '.npy')\n",
        "print('Open this file ' + write_npy_files + title + '.npy')\n",
        "# data = np.load('Test-vid-1-hd.npy')\n",
        "print('Data: ' + title)\n",
        "# seconds = video_duration\n",
        "\n",
        "\n",
        "frames = len(data)\n",
        "print('frames?:', frames)\n",
        "fps = 30\n",
        "\n",
        "\n",
        "\n",
        "pred_data = [] #[ [timestamp (s) , p-value], ... ]\n",
        "time_vals = []\n",
        "pred_vals = []\n",
        "for t in range(0, frames//fps):\n",
        "    _curr_frame = round(t * fps)\n",
        "    if ((_curr_frame - 60) < 0) or (_curr_frame + 60 > frames - 1):\n",
        "        print('nope')\n",
        "        # pred_data.append([t,0])\n",
        "        # time_vals.append(t)\n",
        "        # pred_vals.append(0)\n",
        "    else:\n",
        "        _use = data[_curr_frame - 60 : _curr_frame + 60,]\n",
        "        \n",
        "\n",
        "        real_x = slice_data(_use)\n",
        "        \n",
        "\n",
        "        _pval = float(my_model.predict(x=real_x))\n",
        "        pred_data.append([t,_pval])\n",
        "        time_vals.append(t)\n",
        "\n",
        "        pred_vals.append(_pval)\n",
        "        if (_pval > .5): \n",
        "            print('Seizure Detected:', str(t), str(_pval))\n",
        "\n",
        "#Plot results.\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "plt.figure()\n",
        "plt.plot(time_vals, pred_vals, 'ro')\n",
        "plt.xlabel('Time (s)')\n",
        "plt.ylabel('P-Value')\n",
        "plt.plot(range(len(time_vals)), [.5 for x in range(len(time_vals))], 'b--')\n",
        "plt.savefig(title + '.png')\n",
        "\n",
        "plt.show()\n",
        "\n",
        "my_dict = {}\n",
        "my_dict['seizure'] = pred_data\n",
        "import json\n",
        "with open(title + '.json', 'w') as fp:\n",
        "    json.dump(my_dict, fp)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Requirement already satisfied: youtube_dl in /usr/local/lib/python3.6/dist-packages (2020.3.24)\n",
            "Requirement already satisfied: pafy in /usr/local/lib/python3.6/dist-packages (0.5.5)\n",
            "Requirement already satisfied: youtube_dl in /usr/local/lib/python3.6/dist-packages (2020.3.24)\n",
            "Requirement already satisfied: pafy in /usr/local/lib/python3.6/dist-packages (0.5.5)\n",
            "[youtube] 1VH-f8kVy6k: Downloading webpage\n",
            "[youtube] Downloading just video 1VH-f8kVy6k because of --no-playlist\n",
            "[youtube] 1VH-f8kVy6k: Downloading MPD manifest\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "WARNING: Requested formats are incompatible for merge and will be merged into mkv.\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "[dashsegments] Total fragments: 9\n",
            "[download] Destination: 1VH-f8kVy6kwebm.f247\n",
            "[download] 100% of 4.84MiB in 00:12\n",
            "[dashsegments] Total fragments: 6\n",
            "[download] Destination: 1VH-f8kVy6km4a.f140\n",
            "[download] 100% of 679.19KiB in 00:02\n",
            "[ffmpeg] Merging formats into \"1VH-f8kVy6kwebm.mkv\"\n",
            "Deleting original file 1VH-f8kVy6kwebm.f247 (pass -k to keep)\n",
            "Deleting original file 1VH-f8kVy6km4a.f140 (pass -k to keep)\n",
            "here\n",
            "Test-vid-8-hd.webm\n",
            "Starting OpenPose demo...\n",
            "Configuring OpenPose...\n",
            "Starting thread(s)...\n",
            "Auto-detecting all available GPUs... Detected 1 GPU(s), using 1 of them starting at GPU 0.\n",
            "Person tracking (`tracking` flag) is in experimental phase. Please, let us know if you find any bug on this alpha version.\n",
            "Empty frame detected, frame number 1286 of 1288. In /content/openpose/src/openpose/producer/producer.cpp:checkFrameIntegrity():290\n",
            "Empty frame detected, frame number 1286 of 1288. In /content/openpose/src/openpose/producer/producer.cpp:checkFrameIntegrity():290\n",
            "Empty frame detected, frame number 1286 of 1288. In /content/openpose/src/openpose/producer/producer.cpp:checkFrameIntegrity():290\n",
            "OpenPose demo successfully finished. Total time: 39.268441 seconds.\n",
            "Change JSON files to npy files....\n",
            "THIS ONE RIGHT HERE.>\n",
            "1286\n",
            "\tupdate:0\n",
            "\tupdate:1000\n",
            "finished\n",
            "Open this file /content/Test-vid-8-hd.npy\n",
            "Data: Test-vid-8-hd\n",
            "frames?: 1286\n",
            "nope\n",
            "nope\n",
            "Seizure Detected: 4 0.6366339325904846\n",
            "Seizure Detected: 5 0.6581339240074158\n",
            "Seizure Detected: 6 0.6968254446983337\n",
            "Seizure Detected: 7 0.6694623827934265\n",
            "Seizure Detected: 13 0.8431888818740845\n",
            "Seizure Detected: 14 0.9759030342102051\n",
            "Seizure Detected: 15 0.9767937660217285\n",
            "Seizure Detected: 16 0.9933316111564636\n",
            "Seizure Detected: 17 0.8476846218109131\n",
            "Seizure Detected: 26 0.999187171459198\n",
            "Seizure Detected: 27 0.9995853304862976\n",
            "Seizure Detected: 28 0.9998584985733032\n",
            "Seizure Detected: 29 0.9998031258583069\n",
            "Seizure Detected: 30 0.9985048770904541\n",
            "Seizure Detected: 31 0.9866874814033508\n",
            "Seizure Detected: 32 0.6700236797332764\n",
            "Seizure Detected: 34 0.9471073150634766\n",
            "Seizure Detected: 35 0.9217149615287781\n",
            "Seizure Detected: 36 0.9992713332176208\n",
            "Seizure Detected: 37 0.9992607235908508\n",
            "Seizure Detected: 38 0.9935653805732727\n",
            "nope\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EBjiS_QVgItJ",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "plt.figure()\n",
        "index = 0        \n",
        "# print(len(Seizure_Detected_Landmarks[0]))\n",
        "# print(Seizure_Detected_Landmarks[0][1].shape)\n",
        "\n",
        "# Seizure_Detected_Landmarks = np.array(Seizure_Detected_Landmarks)\n",
        "# print(Seizure_Detected_Landmarks.shape)\n",
        "_t10 = Seizure_Detected_Landmarks[index]\n",
        "for i in range(0, len(_t10), 1):\n",
        "    for j in range(0,len(_t10[0]),2):\n",
        "        plt.xlim(0, 1)\n",
        "        plt.ylim((0, 1))\n",
        "        plt.scatter(_t10[i][j], _t10[i][j+1])\n",
        "    \n",
        "\n",
        "    plt.pause(.0001)"
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}