lymph-node-extract-feature.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "\"\"\"\n",
    "Change the folders' name to make it easier to process. \n",
    "Basically this section is to switch the order of pos and neg.\n",
    "I want the pos cases go first, because there are fewer of\n",
    "them, and I'm going to sample the neg cases based on how \n",
    "many pos cases I have.\n",
    "\"\"\"\n",
    "\n",
    "import os\n",
    "path = '/Users/crl_imac/Documents/lymph_node_CT_data/ROI'\n",
    "\n",
    "for root, dirs, filenames in os.walk(path):\n",
    "    \n",
    "    if not '.DS_Store' in filenames:\n",
    "        continue\n",
    "    \n",
    "    if \"roi\" not in dirs[0]:\n",
    "        continue\n",
    "    \n",
    "    for one_dir in dirs:\n",
    "        \n",
    "        if \"neg\" in one_dir:\n",
    "            os.rename(os.path.join(root, one_dir),\n",
    "                     os.path.join(root, '2'+one_dir))\n",
    "        if \"pos\" in one_dir:\n",
    "            os.rename(os.path.join(root, one_dir),\n",
    "                     os.path.join(root, '1'+one_dir))\n",
    "            "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/Users/crl_imac/Documents/lymph_node_CT_data/ROI/MED_LYMPH_010\n",
      "/Users/crl_imac/Documents/lymph_node_CT_data/ROI/MED_LYMPH_030\n",
      "/Users/crl_imac/Documents/lymph_node_CT_data/ROI/MED_LYMPH_045\n",
      "/Users/crl_imac/Documents/lymph_node_CT_data/ROI/MED_LYMPH_062\n"
     ]
    }
   ],
   "source": [
    "\"\"\"\n",
    "This function is to point out the patients who only have \n",
    "negative roi.\n",
    "\"\"\"\n",
    "\n",
    "def find_only_neg(path):\n",
    "    \n",
    "    for root, dirs, filenames in os.walk(path, topdown=True):\n",
    "        \n",
    "        # \"len(dirs) -> return True if not empty \"\n",
    "        if ('.DS_Store' in filenames) and (len(dirs)):\n",
    "            continue\n",
    "        \n",
    "        if len(filenames):\n",
    "            continue\n",
    "            \n",
    "        if len(dirs) == 1:\n",
    "            print root\n",
    "\n",
    "find_only_neg('/Users/crl_imac/Documents/lymph_node_CT_data/ROI')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### From the output of checking function, we know that *MED10, MED30, MED45, MED62* just have negative cases. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 192,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "\"\"\"\n",
    "This function is used to delete input roi that has zero in their \n",
    "dimension. Since we're going to do 2.5D here, when we delete one\n",
    "patch, its corresponding sagittal/coronal/.. are going to be \n",
    "deleted as well. \n",
    "\"\"\"\n",
    "\n",
    "def remove_zero_dimension(path):\n",
    "    \n",
    "    directory = \"/Users/crl_imac/Documents/lymph_node_CT_data/Invalid_samples/Inconsistent_Dimension/invalid_info.txt\"\n",
    "    try:\n",
    "        os.remove(directory)\n",
    "    except OSError:\n",
    "        pass\n",
    "    \n",
    "    dustbin = \"/Users/crl_imac/Documents/lymph_node_CT_data/Invalid_samples/Inconsistent_Dimension/\"\n",
    "    for root, dirs, filenames in os.walk(path, topdown=True):\n",
    "        \n",
    "        if '.DS_Store' in filenames:\n",
    "            continue\n",
    "        \n",
    "        for filename in filenames:\n",
    "\n",
    "            image = np.load(os.path.join(root, filename))\n",
    "            \n",
    "            # if image.shape[0] == 0 or image.shape[1] == 0 or :\n",
    "            if image.shape[0] != image.shape[1]:\n",
    "                \n",
    "                if \"coronal\" in filename:\n",
    "                    other_name_one = filename.replace(\"coronal\", \"sagittal\")\n",
    "                    other_name_two = filename.replace(\"coronal\", \"axial\")\n",
    "                elif \"sagittal\" in filename:\n",
    "                    other_name_one = filename.replace(\"sagittal\", \"coronal\")\n",
    "                    other_name_two = filename.replace(\"sagittal\", \"axial\")\n",
    "                elif \"axial\" in filename:\n",
    "                    other_name_one = filename.replace(\"axial\", \"coronal\")\n",
    "                    other_name_two = filename.replace(\"axial\", \"sagittal\")\n",
    "                \n",
    "                filenames.remove(filename)\n",
    "                filenames.remove(other_name_one)\n",
    "                filenames.remove(other_name_two)\n",
    "                \n",
    "                if not os.path.exists(dustbin + root[-22:-9]):\n",
    "                    os.makedirs(dustbin + root[-22:-9])\n",
    "                \n",
    "                os.rename(os.path.join(root, filename), \n",
    "                          dustbin + root[-22:-9] +'/'+ root[-7:-4] + filename)\n",
    "                os.rename(os.path.join(root, other_name_one), \n",
    "                          dustbin + root[-22:-9] +'/'+ root[-7:-4] + other_name_one)\n",
    "                os.rename(os.path.join(root, other_name_two), \n",
    "                          dustbin + root[-22:-9] +'/'+ root[-7:-4] + other_name_two)\n",
    "                \n",
    "                base_info = os.path.join(root, filename)[-70:] + ' has a invalid shape of %s\\n' % (str(np.shape(image)))\n",
    "                other_info = \"%s and %s will be deleted together\\n\\n\" % (other_name_one, other_name_two)\n",
    "                \n",
    "                with open(directory, 'aw') as f:\n",
    "                    f.write(base_info)\n",
    "                    f.write(other_info)\n",
    "                \n",
    "                \n",
    "remove_zero_dimension('/Users/crl_imac/Documents/lymph_node_CT_data/ROI')\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### After we run the \"remove_zero_dimension\" function, the roi left in the \"ROI\" folder should be valid. Next, we're going to construct image from this folder, and feed it into CNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import sys\n",
    "\n",
    "caffe_root = '/Users/crl_imac/Documents/caffe/'\n",
    "sys.path.insert(0, caffe_root+'python')\n",
    "\n",
    "import caffe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "'''\n",
    "Set up CNN and do image preprocessing.\n",
    "'''\n",
    "import numpy as np\n",
    "caffe.set_mode_cpu()\n",
    "\n",
    "model_def = caffe_root + 'models/bvlc_reference_caffenet/deploy.prototxt'\n",
    "model_weight = caffe_root + 'models/bvlc_reference_caffenet/bvlc_reference_caffenet.caffemodel'\n",
    "\n",
    "net = caffe.Net(model_def, model_weight, caffe.TEST)\n",
    "\n",
    "net.blobs['data'].reshape(50,        # batch size \n",
    "                        3,          # 3-channel(BGR) images\n",
    "                        227, 227)   # image size\n",
    "\n",
    "mu = np.load(caffe_root+'python/caffe/imagenet/ilsvrc_2012_mean.npy') # mu:(3, 256, 256)\n",
    "mu = mu.mean(1).mean(1) # Obtain the mean (BGR) pixel values\n",
    "transformer = caffe.io.Transformer({'data':net.blobs['data'].data.shape})\n",
    "\n",
    "transformer.set_transpose('data', (2,0,1))  # move image channels to outermost dimension\n",
    "transformer.set_mean('data', mu)            # subtract the dataset-mean value in each channel\n",
    "transformer.set_raw_scale('data', 255)      # rescale from [0, 1] to [0, 255]\n",
    "transformer.set_channel_swap('data', (2,1,0))  # swap channels from RGB to BGR\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def preprocess(image):\n",
    "    \n",
    "    image = (image - np.mean(image))/np.mean(image)\n",
    "    transform_image = transformer.preprocess('data', image)\n",
    "    \n",
    "    return transform_image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 213,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "\"\"\"\n",
    "This function is to select three slices(axial, coronal, sagittal)\n",
    "and input it into caffe network. \n",
    "\n",
    "To avoid any intermediate saving, which might be huge, after we \n",
    "get one input image, we directly feed it into the network to \n",
    "extract its features. \n",
    "\"\"\"\n",
    "\n",
    "import os\n",
    "import csv\n",
    "from skimage.transform import resize\n",
    "\n",
    "def generate_CNNinput(path, blobnames):\n",
    "    \n",
    "    directory = \"/Users/crl_imac/Documents/DeepLearning/RESULT/CaffeNet/Off-the-shell/ABD_LYMPH/Input_info.txt\"\n",
    "    result_directory = \"/Users/crl_imac/Documents/DeepLearning/RESULT/CaffeNet/Off-the-shell/ABD_LYMPH\"\n",
    "    \n",
    "    blob_dict = {}\n",
    "    for blob in blob_names:\n",
    "        blob_dict[blob] = []\n",
    "        try:\n",
    "            os.remove(result_directory+'/'+blob+'.csv')\n",
    "        except OSError:\n",
    "            pass\n",
    "    \n",
    "    try:\n",
    "        os.remove(directory)                                              \n",
    "    except OSError:\n",
    "        pass\n",
    "    \n",
    "    for root, dirs, filenames in os.walk(path, topdown=True):\n",
    "        \n",
    "        if '.DS_Store' in filenames:\n",
    "            continue\n",
    "        \n",
    "        # Just do with ABD or MED once a time. \n",
    "        if \"MED\" in root:\n",
    "            continue      \n",
    "        \n",
    "        if \"pos\" in root:\n",
    "            \n",
    "            new_filenames = [x for x in filenames if \"ROT\" not in x]\n",
    "            newest_filenames = [y for y in new_filenames if \"SCALE_0\" in y]\n",
    "            \n",
    "            num_sample = len(newest_filenames)/3\n",
    "            \n",
    "            for filename in newest_filenames:\n",
    "                \n",
    "                image = np.load(os.path.join(root, filename))\n",
    "                \n",
    "                if \"axial\" in filename:\n",
    "                    image_3D = image\n",
    "                elif \"coronal\" in filename:\n",
    "                    image_3D = np.dstack((image_3D, image))\n",
    "                elif \"sagittal\" in filename:\n",
    "                    image_3D = np.dstack((image_3D, image))\n",
    "                \n",
    "                    CNN_input = preprocess(image_3D)\n",
    "                    net.blobs['data'].data[...] = CNN_input\n",
    "                    output = net.forward()\n",
    "\n",
    "                    for blob in blob_names:\n",
    "\n",
    "                        feature = net.blobs[blob].data[0,:]\n",
    "\n",
    "                        if \"fc\" in blob:\n",
    "                            blob_dict[blob] = list(feature)\n",
    "\n",
    "                        elif \"conv\" in blob:\n",
    "                            blob_dict[blob] = np.mean(np.mean(feature, axis=1),axis=1)\n",
    "                        \n",
    "                        \n",
    "                        with open(result_directory+'/'+blob+'.csv', 'aw') as fi:\n",
    "                            writer = csv.writer(fi)\n",
    "                            \n",
    "                            writer.writerow(blob_dict[blob])\n",
    "                \n",
    "        elif 'neg' in root:\n",
    "            \n",
    "            new_filenames = [x for x in filenames if \"ROT\" not in x]\n",
    "            newest_filenames = [y for y in new_filenames if \"SCALE_0\" in y]\n",
    "            \n",
    "            chosen_filenames = np.random.choice(newest_filenames, size=num_sample)\n",
    "            \n",
    "            for filename in chosen_filenames:\n",
    "                \n",
    "                image = np.load(os.path.join(root, filename))\n",
    "                \n",
    "                if \"axial\" in filename:\n",
    "                    \n",
    "                    coronal_filename = filename.replace('axial','coronal')\n",
    "                    sagittal_filename = filename.replace('axial', 'sagittal')\n",
    "                    coronal_image = np.load(os.path.join(root, coronal_filename))\n",
    "                    sagittal_image = np.load(os.path.join(root, sagittal_filename))\n",
    "                    image_3D = np.dstack((np.dstack((image, coronal_image)), sagittal_image))\n",
    "                \n",
    "                elif \"coronal\" in filename:\n",
    "                    \n",
    "                    axial_filename = filename.replace('coronal','axial')\n",
    "                    sagittal_filename = filename.replace('coronal', 'sagittal')\n",
    "                    axial_image = np.load(os.path.join(root, axial_filename))\n",
    "                    sagittal_image = np.load(os.path.join(root, sagittal_filename))\n",
    "                    image_3D = np.dstack((np.dstack((axial_image, image)), sagittal_image))\n",
    "                    \n",
    "                elif \"sagittal\" in filename:\n",
    "                    \n",
    "                    axial_filename = filename.replace('sagittal','axial')\n",
    "                    coronal_filename = filename.replace('sagittal','coronal')\n",
    "                    axial_image = np.load(os.path.join(root, axial_filename))\n",
    "                    coronal_image = np.load(os.path.join(root, coronal_filename))\n",
    "                    image_3D = np.dstack((np.dstack((axial_image, coronal_image)), image))\n",
    "                \n",
    "                CNN_input = preprocess(image_3D)\n",
    "                net.blobs['data'].data[...] = CNN_input\n",
    "                output = net.forward()\n",
    "                \n",
    "                for blob in blob_names:\n",
    "\n",
    "                        feature = net.blobs[blob].data[0,:]\n",
    "\n",
    "                        if \"fc\" in blob:\n",
    "                            blob_dict[blob] = list(feature)\n",
    "\n",
    "                        elif \"conv\" in blob:\n",
    "                            blob_dict[blob] = np.mean(np.mean(feature, axis=1),axis=1)\n",
    "                        \n",
    "                        \n",
    "                        with open(result_directory+'/'+blob+'.csv', 'aw') as fi:\n",
    "                            writer = csv.writer(fi)\n",
    "                            \n",
    "                            writer.writerow(blob_dict[blob])\n",
    "                \n",
    "                # pdb.set_trace()\n",
    "            \n",
    "            \n",
    "            print \"Processed %s\" % root[-23:-9]\n",
    "            info1 = \"%d pos samples from %s\\n\" % (num_sample, root[-23:-9])\n",
    "            info2 = \"%d neg samples from %s\\n\\n\" % (num_sample, root[-23:-9])\n",
    "\n",
    "            with open(directory, 'aw') as f:\n",
    "                f.write(info1)\n",
    "                f.write(info2)\n",
    "        \n",
    "   "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 214,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Processed /ABD_LYMPH_001\n",
      "Processed /ABD_LYMPH_002\n",
      "Processed /ABD_LYMPH_003\n",
      "Processed /ABD_LYMPH_004\n",
      "Processed /ABD_LYMPH_005\n",
      "Processed /ABD_LYMPH_006\n",
      "Processed /ABD_LYMPH_007\n",
      "Processed /ABD_LYMPH_008\n",
      "Processed /ABD_LYMPH_009\n",
      "Processed /ABD_LYMPH_010\n",
      "Processed /ABD_LYMPH_011\n",
      "Processed /ABD_LYMPH_012\n",
      "Processed /ABD_LYMPH_014\n",
      "Processed /ABD_LYMPH_015\n",
      "Processed /ABD_LYMPH_016\n",
      "Processed /ABD_LYMPH_017\n",
      "Processed /ABD_LYMPH_018\n",
      "Processed /ABD_LYMPH_019\n",
      "Processed /ABD_LYMPH_020\n",
      "Processed /ABD_LYMPH_021\n",
      "Processed /ABD_LYMPH_022\n",
      "Processed /ABD_LYMPH_023\n",
      "Processed /ABD_LYMPH_024\n",
      "Processed /ABD_LYMPH_025\n",
      "Processed /ABD_LYMPH_026\n",
      "Processed /ABD_LYMPH_027\n",
      "Processed /ABD_LYMPH_028\n",
      "Processed /ABD_LYMPH_029\n",
      "Processed /ABD_LYMPH_030\n",
      "Processed /ABD_LYMPH_031\n",
      "Processed /ABD_LYMPH_032\n",
      "Processed /ABD_LYMPH_033\n",
      "Processed /ABD_LYMPH_034\n",
      "Processed /ABD_LYMPH_035\n",
      "Processed /ABD_LYMPH_036\n",
      "Processed /ABD_LYMPH_037\n",
      "Processed /ABD_LYMPH_038\n",
      "Processed /ABD_LYMPH_039\n",
      "Processed /ABD_LYMPH_040\n",
      "Processed /ABD_LYMPH_041\n",
      "Processed /ABD_LYMPH_042\n",
      "Processed /ABD_LYMPH_043\n",
      "Processed /ABD_LYMPH_044\n",
      "Processed /ABD_LYMPH_045\n",
      "Processed /ABD_LYMPH_046\n",
      "Processed /ABD_LYMPH_047\n",
      "Processed /ABD_LYMPH_048\n",
      "Processed /ABD_LYMPH_049\n",
      "Processed /ABD_LYMPH_050\n",
      "Processed /ABD_LYMPH_051\n",
      "Processed /ABD_LYMPH_052\n",
      "Processed /ABD_LYMPH_053\n",
      "Processed /ABD_LYMPH_054\n",
      "Processed /ABD_LYMPH_055\n",
      "Processed /ABD_LYMPH_056\n",
      "Processed /ABD_LYMPH_057\n",
      "Processed /ABD_LYMPH_058\n",
      "Processed /ABD_LYMPH_059\n",
      "Processed /ABD_LYMPH_060\n",
      "Processed /ABD_LYMPH_061\n",
      "Processed /ABD_LYMPH_062\n",
      "Processed /ABD_LYMPH_063\n",
      "Processed /ABD_LYMPH_064\n",
      "Processed /ABD_LYMPH_065\n",
      "Processed /ABD_LYMPH_066\n",
      "Processed /ABD_LYMPH_067\n",
      "Processed /ABD_LYMPH_068\n",
      "Processed /ABD_LYMPH_069\n",
      "Processed /ABD_LYMPH_070\n",
      "Processed /ABD_LYMPH_071\n",
      "Processed /ABD_LYMPH_072\n",
      "Processed /ABD_LYMPH_073\n",
      "Processed /ABD_LYMPH_074\n",
      "Processed /ABD_LYMPH_075\n",
      "Processed /ABD_LYMPH_076\n",
      "Processed /ABD_LYMPH_077\n",
      "Processed /ABD_LYMPH_078\n",
      "Processed /ABD_LYMPH_079\n",
      "Processed /ABD_LYMPH_080\n",
      "Processed /ABD_LYMPH_081\n",
      "Processed /ABD_LYMPH_082\n",
      "Processed /ABD_LYMPH_083\n",
      "Processed /ABD_LYMPH_084\n",
      "Processed /ABD_LYMPH_085\n",
      "Processed /ABD_LYMPH_086\n"
     ]
    }
   ],
   "source": [
    "blob_names = ['fc8', 'fc7', 'fc6', 'conv5', 'conv4', 'conv3', 'conv2', 'conv1']\n",
    "data_path = '/Users/crl_imac/Documents/lymph_node_CT_data/ROI/'\n",
    "generate_CNNinput(data_path, blob_names)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[20, 20, 350, 350, 135, 135, 70, 70, 5, 5, 130, 130, 35, 35, 25, 25, 65, 65, 35, 35, 70, 70, 35, 35, 80, 80, 10, 10, 65, 65, 10, 10, 70, 70, 100, 100, 222, 222, 10, 10, 70, 70, 10, 10, 70, 70, 10, 10, 5, 5, 10, 10, 95, 95, 5, 5, 175, 175, 15, 15, 30, 30, 30, 30, 125, 125, 20, 20, 5, 5, 40, 40, 40, 40, 50, 50, 250, 250, 33, 33, 60, 60, 5, 5, 170, 170, 65, 65, 115, 115, 70, 70, 10, 10, 5, 5, 20, 20, 130, 130, 175, 175, 55, 55, 15, 15, 15, 15, 40, 40, 60, 60, 5, 5, 10, 10, 30, 30, 5, 5, 85, 85, 115, 115, 40, 40, 30, 30, 15, 15, 15, 15, 120, 120, 190, 190, 10, 10, 30, 30, 125, 125, 35, 35, 120, 120, 90, 90, 15, 15, 10, 10, 55, 55, 45, 45, 5, 5, 25, 25, 5, 5, 5, 5, 20, 20, 15, 15, 5, 5]\n"
     ]
    }
   ],
   "source": [
    "\"\"\"\n",
    "Through running the above functions, we've obtained the feature vectors from CNN. \n",
    "Next we're going to use its output file \"input_info.txt\" to generate the label\n",
    "message. \n",
    "\"\"\"\n",
    "\n",
    "label_path = \"/Users/crl_imac/Documents/DeepLearning/RESULT/CaffeNet/Off-the-shell/ABD_LYMPH/Input_info.txt\"\n",
    "import numpy as np\n",
    "import pdb\n",
    "\n",
    "def get_me_label(path):\n",
    "    \n",
    "    label = []\n",
    "    labels = []\n",
    "    with open(path, 'r') as file:\n",
    "        \n",
    "        for line in file:\n",
    "                \n",
    "            words = line.split()\n",
    "            if words:\n",
    "                labels.append(int(words[0]))\n",
    "            \n",
    "            if \"pos\" in words:\n",
    "                \n",
    "                label = np.append(label, np.zeros(int(words[0])))\n",
    "                \n",
    "            elif \"neg\" in words:\n",
    "                \n",
    "                label = np.append(label, np.ones(int(words[0])))\n",
    "        \n",
    "    \n",
    "    print labels\n",
    "    #np.savetxt('/Users/crl_imac/Documents/DeepLearning/RESULT/CaffeNet/Off-the-shell/ABD_LYMPH/label.txt',\n",
    "    #          label)\n",
    "        \n",
    "    return labels\n",
    "\n",
    "a = get_me_label(label_path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "first_fold = sum(a[0:26]); second_fold=sum(a[26:52]); third_fold=sum(a[52:78]);\n",
    "fourth_fold=sum(a[78:104]);fifth_fold=sum(a[104:130]);test_fold=sum(a[130:])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### The codes below are for debugging."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 147,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def check_stop_point(file1, file2, file3):\n",
    "    \n",
    "    image1 = np.load(file1); image2 = np.load(file2); image3 = np.load(file3);\n",
    "    image_3D = np.dstack((np.dstack((image1, image2)), image3))\n",
    "    CNN_input = preprocess(image_3D)\n",
    "    net.blobs['data'].data[...] = CNN_input\n",
    "    output = net.forward()\n",
    "    \n",
    "    feature = net.blobs[\"fc8\"].data[0,:]\n",
    "    np.savetxt(\"/Users/crl_imac/Documents/DeepLearning/RESULT/CaffeNet/Off-the-shell/ABD_LYMPH/test.csv\",\n",
    "              feature, delimiter=\",\")\n",
    "    \n",
    "    return None\n",
    "\n",
    "rdir = \"/Users/crl_imac/Documents/lymph_node_CT_data/ROI/ABD_LYMPH_003/1pos_roi/\"\n",
    "check_stop_point(rdir+\"ROI_20_SCALE_1_trans_3_ROT_4_axial.npy\",\n",
    "                rdir+\"ROI_20_SCALE_1_trans_3_ROT_4_coronal.npy\",\n",
    "                rdir+\"ROI_20_SCALE_1_trans_3_ROT_4_sagittal.npy\")\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 215,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "a = np.loadtxt(\"/Users/crl_imac/Documents/DeepLearning/RESULT/CaffeNet/Off-the-shell/ABD_LYMPH/conv3.csv\", delimiter=\",\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 179,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "a = np.load(\"/Users/crl_imac/Documents/lymph_node_CT_data/ROI/ABD_LYMPH_003/1pos_roi/ROI_2_SCALE_0_trans_1_ROT_4_axial.npy\")\n",
    "b = np.load(\"/Users/crl_imac/Documents/lymph_node_CT_data/ROI/ABD_LYMPH_003/1pos_roi/ROI_2_SCALE_0_trans_1_ROT_4_coronal.npy\")\n",
    "c = np.load(\"/Users/crl_imac/Documents/lymph_node_CT_data/ROI/ABD_LYMPH_003/1pos_roi/ROI_2_SCALE_0_trans_1_ROT_4_sagittal.npy\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 187,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x136445d10>"
      ]
     },
     "execution_count": 187,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAASAAAAD9CAYAAAAcVKrFAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n2cvOZA9Ae9nLXnYmewDay172sjP5/wAt5sb+Szq8VwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x134f66510>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "plt.imshow(c, cmap='gray')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 178,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ROI_2_SCALE_0_trans_1_ROT_4_axial.npy\n",
      "ROI_2_SCALE_0_trans_1_ROT_4_coronal.npy\n",
      "ROI_2_SCALE_0_trans_1_ROT_4_sagittal.npy\n"
     ]
    }
   ],
   "source": [
    "path1 = \"/Users/crl_imac/Documents/lymph_node_CT_data/ROI/ABD_LYMPH_003/1pos_roi/\"\n",
    "i = 0\n",
    "for root, dirs, filenames in os.walk(path1):\n",
    "    \n",
    "    for filename in filenames:\n",
    "        \n",
    "        i += 1\n",
    "        \n",
    "        if i == 7228 or i == 7229 or i==7230:\n",
    "            print filename"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "now we're going to 7 sample from the negative set\n",
      "> /Users/crl_imac/anaconda/lib/python2.7/site-packages/IPython/core/interactiveshell.py(1803)showtraceback()\n",
      "-> try:\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m----------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m              Traceback (most recent call last)",
      "\u001b[0;32m/Users/crl_imac/anaconda/lib/python2.7/site-packages/IPython/core/interactiveshell.pyc\u001b[0m in \u001b[0;36mrun_code\u001b[0;34m(self, code_obj, result)\u001b[0m\n\u001b[1;32m   2900\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mresult\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2901\u001b[0m                 \u001b[0mresult\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0merror_in_exec\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0msys\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mexc_info\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2902\u001b[0;31m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshowtraceback\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   2903\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2904\u001b[0m             \u001b[0moutflag\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/Users/crl_imac/anaconda/lib/python2.7/site-packages/IPython/core/interactiveshell.pyc\u001b[0m in \u001b[0;36mshowtraceback\u001b[0;34m(self, exc_tuple, filename, tb_offset, exception_only)\u001b[0m\n\u001b[1;32m   1801\u001b[0m         simply call this method.\"\"\"\n\u001b[1;32m   1802\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1803\u001b[0;31m         \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1804\u001b[0m             \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1805\u001b[0m                 \u001b[0metype\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mvalue\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtb\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_get_exc_info\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mexc_tuple\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/Users/crl_imac/anaconda/lib/python2.7/site-packages/IPython/core/interactiveshell.pyc\u001b[0m in \u001b[0;36mshowtraceback\u001b[0;34m(self, exc_tuple, filename, tb_offset, exception_only)\u001b[0m\n\u001b[1;32m   1801\u001b[0m         simply call this method.\"\"\"\n\u001b[1;32m   1802\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1803\u001b[0;31m         \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1804\u001b[0m             \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1805\u001b[0m                 \u001b[0metype\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mvalue\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtb\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_get_exc_info\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mexc_tuple\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/Users/crl_imac/anaconda/lib/python2.7/bdb.pyc\u001b[0m in \u001b[0;36mtrace_dispatch\u001b[0;34m(self, frame, event, arg)\u001b[0m\n\u001b[1;32m     47\u001b[0m             \u001b[0;32mreturn\u001b[0m \u001b[0;31m# None\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     48\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mevent\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;34m'line'\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 49\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdispatch_line\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mframe\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     50\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mevent\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;34m'call'\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     51\u001b[0m             \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdispatch_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mframe\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0marg\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/Users/crl_imac/anaconda/lib/python2.7/bdb.pyc\u001b[0m in \u001b[0;36mdispatch_line\u001b[0;34m(self, frame)\u001b[0m\n\u001b[1;32m     65\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mdispatch_line\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mframe\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     66\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstop_here\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mframe\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbreak_here\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mframe\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 67\u001b[0;31m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0muser_line\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mframe\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     68\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mquitting\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0;32mraise\u001b[0m \u001b[0mBdbQuit\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     69\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrace_dispatch\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/Users/crl_imac/anaconda/lib/python2.7/pdb.pyc\u001b[0m in \u001b[0;36muser_line\u001b[0;34m(self, frame)\u001b[0m\n\u001b[1;32m    156\u001b[0m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_wait_for_mainpyfile\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    157\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbp_commands\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mframe\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 158\u001b[0;31m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0minteraction\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mframe\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mNone\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    159\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    160\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mbp_commands\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mframe\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/Users/crl_imac/anaconda/lib/python2.7/pdb.pyc\u001b[0m in \u001b[0;36minteraction\u001b[0;34m(self, frame, traceback)\u001b[0m\n\u001b[1;32m    208\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msetup\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mframe\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtraceback\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    209\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mprint_stack_entry\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstack\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcurindex\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 210\u001b[0;31m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcmdloop\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    211\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mforget\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    212\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/Users/crl_imac/anaconda/lib/python2.7/cmd.pyc\u001b[0m in \u001b[0;36mcmdloop\u001b[0;34m(self, intro)\u001b[0m\n\u001b[1;32m    128\u001b[0m                     \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0muse_rawinput\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    129\u001b[0m                         \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 130\u001b[0;31m                             \u001b[0mline\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mraw_input\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mprompt\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    131\u001b[0m                         \u001b[0;32mexcept\u001b[0m \u001b[0mEOFError\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    132\u001b[0m                             \u001b[0mline\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m'EOF'\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/Users/crl_imac/anaconda/lib/python2.7/site-packages/ipykernel/kernelbase.pyc\u001b[0m in \u001b[0;36mraw_input\u001b[0;34m(self, prompt)\u001b[0m\n\u001b[1;32m    687\u001b[0m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_parent_ident\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    688\u001b[0m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_parent_header\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 689\u001b[0;31m             \u001b[0mpassword\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mFalse\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    690\u001b[0m         )\n\u001b[1;32m    691\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/Users/crl_imac/anaconda/lib/python2.7/site-packages/ipykernel/kernelbase.pyc\u001b[0m in \u001b[0;36m_input_request\u001b[0;34m(self, prompt, ident, parent, password)\u001b[0m\n\u001b[1;32m    717\u001b[0m             \u001b[0;32mexcept\u001b[0m \u001b[0mKeyboardInterrupt\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    718\u001b[0m                 \u001b[0;31m# re-raise KeyboardInterrupt, to truncate traceback\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 719\u001b[0;31m                 \u001b[0;32mraise\u001b[0m \u001b[0mKeyboardInterrupt\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    720\u001b[0m             \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    721\u001b[0m                 \u001b[0;32mbreak\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "import pdb\n",
    "\n",
    "pdb.run(generate_CNNinput(data_path, blob_names))"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [Root]",
   "language": "python",
   "name": "Python [Root]"
  },
  "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": 0
}