votti/demonstrate_missmatch.ipynb

## demonstrate_missmatch.ipynb
{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import os\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import tifffile"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Demonstrate that the current CP behaviour of handling missing produces inconsistent output\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Aim:\n",
    "As discussed as issue 3582 in the CellProfiler Github (https://github.com/CellProfiler/CellProfiler/issues/3582),\n",
    "CellProfiler 3 currently does not handle segmentation masks with missing object. Explicitly this means that not all objects from 1:max(object_label) are present. These can be produced with valid modules such as IdentifySecondaryObjects and are an important case to consider in any case, e.g. also to maintain compatibility/consistency with other software that can generate such masks.\n",
    "\n",
    "The way it is implemented now, ObjectNumbers for masks with missing objects are just renumbered from 1:number_objects, implicitely without any notification.\n",
    "\n",
    "This breaks the relationship that ObjectNumbers should correspond 1:1 to the label number in segmentation masks.\n",
    "\n",
    "To demonstrate that this is broken in the current CellProfiler, I generated a test pipline that once just loads a mask with missing objects ('Original'), measures implicitly the X-Y positions and saves both these measurements as well as the segmetation mask.\n",
    "\n",
    "The same is done for the case where the mask is loaded and relabled from 1:number_of_objects ('Relabeled'). This is the control case where everything should work fine."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Set the paths"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "fol_cpout = './out'\n",
    "fn_image = 'MyExpt_Image.csv'\n",
    "fns_objects = ('MyExpt_CellRelabeled.csv', 'MyExpt_CellOriginal.csv')\n",
    "\n",
    "col_maskname_objects = ['FileName_'+c for c in ('CellRelabeledMask', 'CellOriginalMask')]\n",
    "\n",
    "COL_IMAGENUMBER = 'ImageNumber'\n",
    "COL_OBJECTNUMBER = 'ObjectNumber'\n",
    "\n",
    "IMG_NR = 1\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Define some helper functions to load masks and map values on masks:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_mask(dat_img, img_nr, base_folder, col_fn_mask):\n",
    "    \"\"\"\n",
    "    Retrieves a mask belonging to the image number\n",
    "    dat_img: an image data table from CP\n",
    "    img_nr: the current image_number\n",
    "    base_folder: the folder containing the masks\n",
    "    col_fn_mask: the column in the dat_img that indicates the mask filename\n",
    "    \n",
    "    Returns:\n",
    "        a segmentation mask array\n",
    "    \"\"\"\n",
    "    fn_mask = dat_img.loc[dat_img[COL_IMAGENUMBER] == img_nr, col_fn_mask].values[0]\n",
    "    fn = os.path.join(base_folder, fn_mask)\n",
    "    return tifffile.imread(fn)\n",
    "\n",
    "def get_object_meas(dat_obj, img_nr, meas_name):\n",
    "    \"\"\"\n",
    "    Retrieves the measurement for a certain image number from an object\n",
    "    measurement table\n",
    "    dat_obj: the object measuremetn table from CP\n",
    "    img_nr: the current image number\n",
    "    meas_name: the requested measurement\n",
    "    \n",
    "    Returns:\n",
    "        a pd.Series with the index being the OBJECTNUMBER and the value being the\n",
    "        measurement value\n",
    "    \"\"\"\n",
    "    map_dat = dat_obj.loc[dat_obj[COL_IMAGENUMBER] == img_nr, [COL_OBJECTNUMBER, meas_name]]\n",
    "    map_dat = map_dat.set_index(COL_OBJECTNUMBER)[meas_name]\n",
    "    return map_dat\n",
    "    \n",
    "def map_values_on_mask(mask, map_dat):\n",
    "    \"\"\"\n",
    "    Maps a pd.Series with index=OBJECTNUMBER on a segmentation mask\n",
    "    mask: a segmentation mask, pixels: 0=Background, >0= ObjectLabelNumber\n",
    "    map_dat: a pd.Series with the index being the OBJECTNUMBER and the value being the\n",
    "                measurement value to be mapped\n",
    "    \n",
    "    Returns:\n",
    "        an image where the object labels have been replaced with the value from map_dat or np.NAN\n",
    "    \"\"\"\n",
    "    labels = map_dat.index\n",
    "    values = map_dat.values.squeeze()\n",
    "    \n",
    "    val_vector = np.empty(mask.max()+1)\n",
    "    val_vector[:] = np.NAN\n",
    "    val_vector[0] = 0\n",
    "    val_vector[labels] = values\n",
    "    img_mask = np.take(val_vector, mask)\n",
    "    return img_mask\n",
    "\n",
    "def get_mapped_img(dat_obj, dat_img, img_nr, meas_name,\n",
    "                  base_folder, col_fn_mask):\n",
    "    \"\"\"\n",
    "    Convenience function to map a measurement from a cellprofiler object\n",
    "    table on a mask.\n",
    "    dat_obj: the object measuremetn table from CP\n",
    "    dat_img: an image data table from CP\n",
    "    img_nr: the current image number\n",
    "    meas_name: the requested measurement\n",
    "    base_folder: the folder containing the masks\n",
    "    col_fn_mask: the column in the dat_img that indicates the mask filename\n",
    "    \n",
    "    Returns:\n",
    "        an image where the object labels have been replaced with the value from the column meas_name \n",
    "        of the dat_obj or np.NAN\n",
    "    \"\"\"\n",
    "    \n",
    "    return map_values_on_mask(get_mask(dat_img, img_nr, base_folder, col_fn_mask),\n",
    "                   get_object_meas(dat_obj, img_nr, meas_name)\n",
    "                  )\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Load the image data data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "dat_image =pd.read_csv(os.path.join(fol_cpout, fn_image))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "dats_objects = [pd.read_csv(os.path.join(fol_cpout, fn)) for fn in fns_objects]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Mapping x and y position on the mask"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5,0.98,'Location_Center_X')"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1,len(dats_objects))\n",
    "\n",
    "measure_name = 'Location_Center_X'\n",
    "for a, dat, col_mask in zip(ax, dats_objects, col_maskname_objects):\n",
    "    img =get_mapped_img(dat, dat_image, IMG_NR, measure_name,\n",
    "                             fol_cpout, col_mask)\n",
    "    a.imshow(img)\n",
    "    a.set_title(col_mask)\n",
    "plt.suptitle(measure_name)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "-> This shows that there is indeed a shift in the the ObjectNumber and the labels in the corresponding masks happening in CellProfiller!\n",
    "Thus in such a case the labels in the masks do not longer correspond to the ObjectNumbers in CellProfiler."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5,0.98,'Location_Center_Y')"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1,len(dats_objects))\n",
    "\n",
    "measure_name = 'Location_Center_Y'\n",
    "for a, dat, col_mask in zip(ax, dats_objects, col_maskname_objects):\n",
    "    img =get_mapped_img(dat, dat_image, IMG_NR, measure_name,\n",
    "                             fol_cpout, col_mask)\n",
    "    a.imshow(img)\n",
    "    a.set_title(col_mask)\n",
    "plt.suptitle(measure_name)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "-> In the Y direction this is much less obvious"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Investigate a bit more why this is the case:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "FileName_CellRelabeledMask\n",
      "a) Number of objects in data:\n",
      "3718\n",
      "b) Number of objects in mask:\n",
      "3718\n",
      "c) Number of labels which are present both as labels and as ObjectNumber:\n",
      "3718\n",
      "d) Labels that are missing a corresponding ObjectNumber:\n",
      "[0]\n",
      "e) ObjectNumber that are missing a corresponding Label:\n",
      "[]\n",
      "\n",
      "\n",
      "FileName_CellOriginalMask\n",
      "a) Number of objects in data:\n",
      "3706\n",
      "b) Number of objects in mask:\n",
      "3706\n",
      "c) Number of labels which are present both as labels and as ObjectNumber:\n",
      "3701\n",
      "d) Labels that are missing a corresponding ObjectNumber:\n",
      "[   0 3707 3708 3709 3710 3711]\n",
      "e) ObjectNumber that are missing a corresponding Label:\n",
      "[2144 2186 2233 2393 2515]\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "masks = [get_mask(dat_img=dat_image, img_nr=IMG_NR, base_folder=fol_cpout, col_fn_mask=c) for c in col_maskname_objects]\n",
    "\n",
    "for d, name, mask in zip(dats_objects, col_maskname_objects, masks):\n",
    "    print(name)\n",
    "    print('a) Number of objects in data:')\n",
    "    print(d.shape[0])\n",
    "    print('b) Number of objects in mask:')\n",
    "    lab_mask =np.unique(mask)\n",
    "    n_obj_mask = len(lab_mask)-1\n",
    "    print(n_obj_mask)\n",
    "    print('c) Number of labels which are present both as labels and as ObjectNumber:')\n",
    "    lab_objectnumber =d[COL_OBJECTNUMBER].values\n",
    "    print(len(np.intersect1d(lab_mask, lab_objectnumber)))\n",
    "    print('d) Labels that are missing a corresponding ObjectNumber:')\n",
    "    print(lab_mask[~np.isin(lab_mask, lab_objectnumber)])\n",
    "    print('e) ObjectNumber that are missing a corresponding Label:')\n",
    "    print(lab_objectnumber[~np.isin(lab_objectnumber, lab_mask)])\n",
    "    print('\\n')\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "-> The above analysis shows that the discrepancy comes indeed from some internal relabeling in CellProfiler.\n",
    "\n",
    "Evidence:\n",
    "- Point a) and b): both mask and data seem to contain a consistent number of cells for both 'original' and 'relabeled' data. Interestingly the relabeling seems to produce additional labels, which maybe should be followed up seperately/likely comes from disjoint objects in the mask\n",
    "\n",
    "- Comparing b) and c) shows that there seem 5 objects not matching in the 'original' data\n",
    "\n",
    "- Point d) & e)  shows that with relabeling everything is matching and as expected. Without relabeling, masks with missing objects produce however NOT matching labels:ImageNumbers! From the pattern of the matching I would conclude that indeed CellProfiller internaly just renumbers the object as 1:number_unique_objects.\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "More formally test if the x and y ccoordinates are consistent between image label and ObjectNumber:\n",
    "(I think this will make for good unit-tests at one point)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "COL_CENTROID_X = 'Location_Center_X'\n",
    "COL_CENTROID_Y = 'Location_Center_Y'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "import skimage.measure as measure"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "def calc_centroid_x(mask):\n",
    "    return [(r.label, r.centroid[1]) for r in measure.regionprops(mask)]\n",
    "\n",
    "def calc_centroid_y(mask):\n",
    "    return [(r.label, r.centroid[0]) for r in measure.regionprops(mask)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "def test_centroid_x(dat_obj, dat_img, img_nr,\n",
    "                  base_folder, col_fn_mask):\n",
    "    label, x = zip(*calc_centroid_x(get_mask(dat_img, img_nr, base_folder, col_fn_mask)))\n",
    "    vals = get_object_meas(dat_obj, img_nr, COL_CENTROID_X)\n",
    "    return np.testing.assert_allclose(vals[list(label)], x)\n",
    "\n",
    "def test_centroid_y(dat_obj, dat_img, img_nr,\n",
    "                  base_folder, col_fn_mask):\n",
    "    label, x = zip(*calc_centroid_y(get_mask(dat_img, img_nr, base_folder, col_fn_mask)))\n",
    "    vals = get_object_meas(dat_obj, img_nr, COL_CENTROID_Y)\n",
    "    return np.testing.assert_allclose(vals[list(label)], x)\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Test first the 'Relabeld case' where we expect no differences"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "i=0 # This is the relabeld ones\n",
    "test_centroid_x(dats_objects[i], dat_image, IMG_NR,\n",
    "                         fol_cpout, col_maskname_objects[i])\n",
    "test_centroid_y(dats_objects[i], dat_image, IMG_NR,\n",
    "                         fol_cpout, col_maskname_objects[i])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "-> Indeed the tests pass!\n",
    "\n",
    "\n",
    "How does it look for the 'Original' masks with missing labels?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/mnt/bbvolume/server_homes/vitoz/.virtualenvs/python3_env/lib/python3.4/site-packages/pandas/core/series.py:696: FutureWarning: \n",
      "Passing list-likes to .loc or [] with any missing label will raise\n",
      "KeyError in the future, you can use .reindex() as an alternative.\n",
      "\n",
      "See the documentation here:\n",
      "http://pandas.pydata.org/pandas-docs/stable/indexing.html#deprecate-loc-reindex-listlike\n",
      "  return self.loc[key]\n"
     ]
    },
    {
     "ename": "AssertionError",
     "evalue": "\nNot equal to tolerance rtol=1e-07, atol=0\n\nx and y nan location mismatch:\n x: array([212.447368, 337.      , 672.157895, ...,        nan,        nan,\n              nan])\n y: array([212.447368, 337.      , 672.157895, ..., 317.681159, 291.916667,\n       697.830189])",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mAssertionError\u001b[0m                            Traceback (most recent call last)",
      "\u001b[0;32m/mnt/bbvolume/server_homes/vitoz/.virtualenvs/python3_env/lib/python3.4/site-packages/numpy/testing/nose_tools/utils.py\u001b[0m in \u001b[0;36mchk_same_position\u001b[0;34m(x_id, y_id, hasval)\u001b[0m\n\u001b[1;32m    699\u001b[0m         \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 700\u001b[0;31m             \u001b[0massert_array_equal\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx_id\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my_id\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    701\u001b[0m         \u001b[0;32mexcept\u001b[0m \u001b[0mAssertionError\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/mnt/bbvolume/server_homes/vitoz/.virtualenvs/python3_env/lib/python3.4/site-packages/numpy/testing/nose_tools/utils.py\u001b[0m in \u001b[0;36massert_array_equal\u001b[0;34m(x, y, err_msg, verbose)\u001b[0m\n\u001b[1;32m    854\u001b[0m     assert_array_compare(operator.__eq__, x, y, err_msg=err_msg,\n\u001b[0;32m--> 855\u001b[0;31m                          verbose=verbose, header='Arrays are not equal')\n\u001b[0m\u001b[1;32m    856\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/mnt/bbvolume/server_homes/vitoz/.virtualenvs/python3_env/lib/python3.4/site-packages/numpy/testing/nose_tools/utils.py\u001b[0m in \u001b[0;36massert_array_compare\u001b[0;34m(comparison, x, y, err_msg, verbose, header, precision, equal_nan, equal_inf)\u001b[0m\n\u001b[1;32m    778\u001b[0m                                 names=('x', 'y'), precision=precision)\n\u001b[0;32m--> 779\u001b[0;31m             \u001b[0;32mraise\u001b[0m \u001b[0mAssertionError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmsg\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    780\u001b[0m     \u001b[0;32mexcept\u001b[0m \u001b[0mValueError\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mAssertionError\u001b[0m: \nArrays are not equal\n\n(mismatch 0.1349163518618468%)\n x: array([False, False, False, ...,  True,  True,  True])\n y: array([False, False, False, ..., False, False, False])",
      "\nDuring handling of the above exception, another exception occurred:\n",
      "\u001b[0;31mAssertionError\u001b[0m                            Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-15-ebcce3722337>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[0mi\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      2\u001b[0m test_centroid_x(dats_objects[i], dat_image, IMG_NR,\n\u001b[0;32m----> 3\u001b[0;31m                          fol_cpout, col_maskname_objects[i])\n\u001b[0m",
      "\u001b[0;32m<ipython-input-13-98714eeeae8f>\u001b[0m in \u001b[0;36mtest_centroid_x\u001b[0;34m(dat_obj, dat_img, img_nr, base_folder, col_fn_mask)\u001b[0m\n\u001b[1;32m      3\u001b[0m     \u001b[0mlabel\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mzip\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mcalc_centroid_x\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mget_mask\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdat_img\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mimg_nr\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbase_folder\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcol_fn_mask\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      4\u001b[0m     \u001b[0mvals\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mget_object_meas\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdat_obj\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mimg_nr\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mCOL_CENTROID_X\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 5\u001b[0;31m     \u001b[0;32mreturn\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtesting\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0massert_allclose\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mvals\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mlist\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlabel\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      6\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      7\u001b[0m def test_centroid_y(dat_obj, dat_img, img_nr,\n",
      "\u001b[0;32m/mnt/bbvolume/server_homes/vitoz/.virtualenvs/python3_env/lib/python3.4/site-packages/numpy/testing/nose_tools/utils.py\u001b[0m in \u001b[0;36massert_allclose\u001b[0;34m(actual, desired, rtol, atol, equal_nan, err_msg, verbose)\u001b[0m\n\u001b[1;32m   1394\u001b[0m     \u001b[0mheader\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m'Not equal to tolerance rtol=%g, atol=%g'\u001b[0m \u001b[0;34m%\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mrtol\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0matol\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1395\u001b[0m     assert_array_compare(compare, actual, desired, err_msg=str(err_msg),\n\u001b[0;32m-> 1396\u001b[0;31m                          verbose=verbose, header=header, equal_nan=equal_nan)\n\u001b[0m\u001b[1;32m   1397\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1398\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/mnt/bbvolume/server_homes/vitoz/.virtualenvs/python3_env/lib/python3.4/site-packages/numpy/testing/nose_tools/utils.py\u001b[0m in \u001b[0;36massert_array_compare\u001b[0;34m(comparison, x, y, err_msg, verbose, header, precision, equal_nan, equal_inf)\u001b[0m\n\u001b[1;32m    724\u001b[0m                 \u001b[0mhas_nan\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0many\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx_isnan\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0many\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0my_isnan\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    725\u001b[0m                 \u001b[0;32mif\u001b[0m \u001b[0mhas_nan\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 726\u001b[0;31m                     \u001b[0mchk_same_position\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx_isnan\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my_isnan\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mhasval\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'nan'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    727\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    728\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mequal_inf\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/mnt/bbvolume/server_homes/vitoz/.virtualenvs/python3_env/lib/python3.4/site-packages/numpy/testing/nose_tools/utils.py\u001b[0m in \u001b[0;36mchk_same_position\u001b[0;34m(x_id, y_id, hasval)\u001b[0m\n\u001b[1;32m    704\u001b[0m                                 \u001b[0;34m%\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mhasval\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mverbose\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mverbose\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mheader\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mheader\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    705\u001b[0m                                 names=('x', 'y'), precision=precision)\n\u001b[0;32m--> 706\u001b[0;31m             \u001b[0;32mraise\u001b[0m \u001b[0mAssertionError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmsg\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    707\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    708\u001b[0m     \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mAssertionError\u001b[0m: \nNot equal to tolerance rtol=1e-07, atol=0\n\nx and y nan location mismatch:\n x: array([212.447368, 337.      , 672.157895, ...,        nan,        nan,\n              nan])\n y: array([212.447368, 337.      , 672.157895, ..., 317.681159, 291.916667,\n       697.830189])"
     ]
    }
   ],
   "source": [
    "i=1\n",
    "test_centroid_x(dats_objects[i], dat_image, IMG_NR,\n",
    "                         fol_cpout, col_maskname_objects[i])\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "-> Fails for X as visually expected, partially because there are NAN where things do not match"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/mnt/bbvolume/server_homes/vitoz/.virtualenvs/python3_env/lib/python3.4/site-packages/pandas/core/series.py:696: FutureWarning: \n",
      "Passing list-likes to .loc or [] with any missing label will raise\n",
      "KeyError in the future, you can use .reindex() as an alternative.\n",
      "\n",
      "See the documentation here:\n",
      "http://pandas.pydata.org/pandas-docs/stable/indexing.html#deprecate-loc-reindex-listlike\n",
      "  return self.loc[key]\n"
     ]
    },
    {
     "ename": "AssertionError",
     "evalue": "\nNot equal to tolerance rtol=1e-07, atol=0\n\nx and y nan location mismatch:\n x: array([3.210526, 5.244898, 4.289474, ...,      nan,      nan,      nan])\n y: array([  3.210526,   5.244898,   4.289474, ..., 796.434783, 797.5     ,\n       797.056604])",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mAssertionError\u001b[0m                            Traceback (most recent call last)",
      "\u001b[0;32m/mnt/bbvolume/server_homes/vitoz/.virtualenvs/python3_env/lib/python3.4/site-packages/numpy/testing/nose_tools/utils.py\u001b[0m in \u001b[0;36mchk_same_position\u001b[0;34m(x_id, y_id, hasval)\u001b[0m\n\u001b[1;32m    699\u001b[0m         \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 700\u001b[0;31m             \u001b[0massert_array_equal\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx_id\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my_id\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    701\u001b[0m         \u001b[0;32mexcept\u001b[0m \u001b[0mAssertionError\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/mnt/bbvolume/server_homes/vitoz/.virtualenvs/python3_env/lib/python3.4/site-packages/numpy/testing/nose_tools/utils.py\u001b[0m in \u001b[0;36massert_array_equal\u001b[0;34m(x, y, err_msg, verbose)\u001b[0m\n\u001b[1;32m    854\u001b[0m     assert_array_compare(operator.__eq__, x, y, err_msg=err_msg,\n\u001b[0;32m--> 855\u001b[0;31m                          verbose=verbose, header='Arrays are not equal')\n\u001b[0m\u001b[1;32m    856\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/mnt/bbvolume/server_homes/vitoz/.virtualenvs/python3_env/lib/python3.4/site-packages/numpy/testing/nose_tools/utils.py\u001b[0m in \u001b[0;36massert_array_compare\u001b[0;34m(comparison, x, y, err_msg, verbose, header, precision, equal_nan, equal_inf)\u001b[0m\n\u001b[1;32m    778\u001b[0m                                 names=('x', 'y'), precision=precision)\n\u001b[0;32m--> 779\u001b[0;31m             \u001b[0;32mraise\u001b[0m \u001b[0mAssertionError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmsg\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    780\u001b[0m     \u001b[0;32mexcept\u001b[0m \u001b[0mValueError\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mAssertionError\u001b[0m: \nArrays are not equal\n\n(mismatch 0.1349163518618468%)\n x: array([False, False, False, ...,  True,  True,  True])\n y: array([False, False, False, ..., False, False, False])",
      "\nDuring handling of the above exception, another exception occurred:\n",
      "\u001b[0;31mAssertionError\u001b[0m                            Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-16-8d152d54298e>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      1\u001b[0m test_centroid_y(dats_objects[i], dat_image, IMG_NR,\n\u001b[0;32m----> 2\u001b[0;31m                          fol_cpout, col_maskname_objects[i])\n\u001b[0m",
      "\u001b[0;32m<ipython-input-13-98714eeeae8f>\u001b[0m in \u001b[0;36mtest_centroid_y\u001b[0;34m(dat_obj, dat_img, img_nr, base_folder, col_fn_mask)\u001b[0m\n\u001b[1;32m      9\u001b[0m     \u001b[0mlabel\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mzip\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mcalc_centroid_y\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mget_mask\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdat_img\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mimg_nr\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbase_folder\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcol_fn_mask\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     10\u001b[0m     \u001b[0mvals\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mget_object_meas\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdat_obj\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mimg_nr\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mCOL_CENTROID_Y\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 11\u001b[0;31m     \u001b[0;32mreturn\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtesting\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0massert_allclose\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mvals\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mlist\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlabel\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     12\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/mnt/bbvolume/server_homes/vitoz/.virtualenvs/python3_env/lib/python3.4/site-packages/numpy/testing/nose_tools/utils.py\u001b[0m in \u001b[0;36massert_allclose\u001b[0;34m(actual, desired, rtol, atol, equal_nan, err_msg, verbose)\u001b[0m\n\u001b[1;32m   1394\u001b[0m     \u001b[0mheader\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m'Not equal to tolerance rtol=%g, atol=%g'\u001b[0m \u001b[0;34m%\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mrtol\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0matol\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1395\u001b[0m     assert_array_compare(compare, actual, desired, err_msg=str(err_msg),\n\u001b[0;32m-> 1396\u001b[0;31m                          verbose=verbose, header=header, equal_nan=equal_nan)\n\u001b[0m\u001b[1;32m   1397\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1398\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/mnt/bbvolume/server_homes/vitoz/.virtualenvs/python3_env/lib/python3.4/site-packages/numpy/testing/nose_tools/utils.py\u001b[0m in \u001b[0;36massert_array_compare\u001b[0;34m(comparison, x, y, err_msg, verbose, header, precision, equal_nan, equal_inf)\u001b[0m\n\u001b[1;32m    724\u001b[0m                 \u001b[0mhas_nan\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0many\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx_isnan\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0many\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0my_isnan\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    725\u001b[0m                 \u001b[0;32mif\u001b[0m \u001b[0mhas_nan\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 726\u001b[0;31m                     \u001b[0mchk_same_position\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx_isnan\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my_isnan\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mhasval\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'nan'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    727\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    728\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mequal_inf\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/mnt/bbvolume/server_homes/vitoz/.virtualenvs/python3_env/lib/python3.4/site-packages/numpy/testing/nose_tools/utils.py\u001b[0m in \u001b[0;36mchk_same_position\u001b[0;34m(x_id, y_id, hasval)\u001b[0m\n\u001b[1;32m    704\u001b[0m                                 \u001b[0;34m%\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mhasval\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mverbose\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mverbose\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mheader\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mheader\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    705\u001b[0m                                 names=('x', 'y'), precision=precision)\n\u001b[0;32m--> 706\u001b[0;31m             \u001b[0;32mraise\u001b[0m \u001b[0mAssertionError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmsg\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    707\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    708\u001b[0m     \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mAssertionError\u001b[0m: \nNot equal to tolerance rtol=1e-07, atol=0\n\nx and y nan location mismatch:\n x: array([3.210526, 5.244898, 4.289474, ...,      nan,      nan,      nan])\n y: array([  3.210526,   5.244898,   4.289474, ..., 796.434783, 797.5     ,\n       797.056604])"
     ]
    }
   ],
   "source": [
    "test_centroid_y(dats_objects[i], dat_image, IMG_NR,\n",
    "                         fol_cpout, col_maskname_objects[i])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Actually also fails for the Y coordinate, clearly as there are also NAN for the not matched cells"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "python3_env",
   "language": "python",
   "name": "python3_env"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.4.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}