{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true,
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "\n",
    "Quickstart\n",
    "==========\n",
    "\n",
    "1) Install ryomen with `pip install ryomen -U`\n",
    "\n",
    "2) Load a large microscopy image, create the slicer, and initialize an array to store the outputs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "outputs": [],
   "source": [
    "from skimage.io import imread\n",
    "from ryomen import Slicer\n",
    "import numpy as np\n",
    "\n",
    "huge_tilescan = imread('my_huge_tilescan.tif')   # shape is [C, X, Y] -> [3, 1000, 1000]\n",
    "output = np.zeros_like(huge_tilescan)\n",
    "\n",
    "slices = Slicer(\n",
    "    image=huge_tilescan,   # your image\n",
    "    crop_size=(512, 512),  # sliced up into 512x512 tiles\n",
    "    overlap=(64, 64),      # with 64px of overlap\n",
    "    pad=True,              # padded with reflection\n",
    "    batch_size=1           # batched one at a time\n",
    ")"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  },
  {
   "cell_type": "markdown",
   "source": [
    "3) Iterate over the slices\n",
    "4) Perform some expensive function on the slices\n",
    "5) slot the data from the expensive function into the output array"
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "outputs": [],
   "source": [
    "expensive_fn = lambda x: x + 1  # might be a ML model, huge filter, etc...\n",
    "\n",
    "for crop, source, destination in slices:\n",
    "    crop = expensive_fn(crop)\n",
    "    \n",
    "    # Take the non-overlapping data from crop and\n",
    "    # put it in the output array\n",
    "    output[destination] = crop[source]  "
   ],
   "metadata": {
    "collapsed": false
   }
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}