{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "ad36b0ca",
   "metadata": {},
   "source": [
    "# Relabeling a label image\n",
    "See also [discussion on image.sc](https://forum.image.sc/t/making-np-where-faster/58641/2)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "e3a91561",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<NVIDIA GeForce RTX 3050 Ti Laptop GPU on Platform: NVIDIA CUDA (1 refs)>"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import numpy as np\n",
    "import numpy\n",
    "import cupy as cp\n",
    "import cupy\n",
    "import pyclesperanto_prototype as cle\n",
    "import timeit\n",
    "cle.select_device(\"RTX\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "1060e301",
   "metadata": {},
   "outputs": [],
   "source": [
    "sliceimage = (np.random.random((17, 2439, 2439)) * 4).astype(int)\n",
    "\n",
    "def relabel_numpy(sliceimage):\n",
    "    return numpy.take(numpy.array([0, 3, 3, 6, 4], numpy.uint8), sliceimage)\n",
    "\n",
    "def relabel_cupy(sliceimage):\n",
    "    return cupy.asnumpy(\n",
    "        cupy.take(\n",
    "            cupy.asarray([0, 3, 3, 6, 4]), \n",
    "            cupy.asarray(sliceimage)\n",
    "        )\n",
    "    )\n",
    "\n",
    "def relabel_cle(sliceimage):\n",
    "    return cle.replace_intensities(sliceimage, np.asarray([0, 3, 3, 6, 4]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "03a7d70d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "292 ms ± 24.8 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)\n"
     ]
    }
   ],
   "source": [
    "%timeit relabel_cupy(sliceimage)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "eaaad577",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "793 ms ± 12.7 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)\n"
     ]
    }
   ],
   "source": [
    "%timeit relabel_numpy(sliceimage)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "b9b3fc04",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "264 ms ± 5.14 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)\n"
     ]
    }
   ],
   "source": [
    "%timeit relabel_cle(sliceimage)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a1388823",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}