terasakisatoshi/Playing_ChainerCV_in_advance.ipynb

## Playing_ChainerCV_in_advance.ipynb
{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Playing ChainerCV\n",
    "\n",
    "## Reference:\n",
    "- https://github.com/chainer/chainercv\n",
    "- http://www.iconsplace.com/black-icons/letter-f-icon\n",
    "- https://qiita.com/mitmul/items/1e35fba085eb07a92560"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## import "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from imageio import imread, imsave\n",
    "import skimage\n",
    "import numpy as np\n",
    "import matplotlib\n",
    "from matplotlib import pyplot as plt \n",
    "%matplotlib inline\n",
    "from IPython.display import Image\n",
    "import chainer\n",
    "from chainer.datasets import cifar\n",
    "import chainercv\n",
    "from chainercv import transforms"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## check version"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "chainercv 0.8.0\n",
      "matplotlib 2.2.0\n"
     ]
    }
   ],
   "source": [
    "print('chainercv',chainercv.__version__)\n",
    "print('matplotlib',matplotlib.__version__)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQAAAAEACAYAAABccqhmAAADSklEQVR4nO3csW3VABSG0ZvodUhUDJCGWSJGYgg2gT5VJmAFJCqaNBSUUcICBKXg5cZ850hurd+W9cluPAMAAAAAAAAAAAAAAAAAAAAAAAAAbLjYHvDKfZmZt9sjeNLHmfm6PeLIBODv7mbm3fYInvRhZm62RxzZ5fYAYI8AQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQNhpe0Dc+5n5vj3iwO63BxydAOy6Hw8xi3wCQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQNhpe0Dct5l52B5xANczc7s94n8kALsux1vYc7hHZ+LGQpgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJgAQJh/Au76PDO/tkccwI/tATTdzczjGY+rF7sS+AOfABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABAmABB22h7wyn2amTdnPP/PM54bAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACAf+43npEohsISdc8AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# image 'letterF.png' is taken from http://www.iconsplace.com/black-icons/letter-f-icon\n",
    "Image(filename='letterF.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(256, 256, 4)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "letter = imread('letterF.png')\n",
    "letter.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Flip letter 'F'  in horizontal direction"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## using numpy.flipud"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x11eaa9208>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "flip_np=np.flipud(letter)\n",
    "plt.imshow(flip_np)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# using chainercv.transforms.flip"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "uint8\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x120c112b0>"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "img=transforms.flip(letter, y_flip=False, x_flip=True, copy=True)\n",
    "plt.imshow(img)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Ops... we can't see any of shape of F.... "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Let's examine another example"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_dataset, test_dataset = cifar.get_cifar10(scale=255.)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## get an example"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "example=train_dataset[0][0].astype(np.uint8)\n",
    "example=example.transpose(2,1,0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## show the image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x12a411dd8>"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(example)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## using np.flipud"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x120c034a8>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(np.flipud(example))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Flip the example image in horizontal direction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x122f09400>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "img=transforms.flip(example, y_flip=False, x_flip=True, copy=True)\n",
    "plt.imshow(img)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Geez this result is not what I want"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Let's read ChainerCV Readme before creating new issues"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "the Readme.md of https://github.com/chainer/chainercv says...\n",
    "    \n",
    "# Data Conventions\n",
    "- Image\n",
    "  - The order of color channel is RGB.\n",
    "  -  Shape is CHW (i.e. (channel, height, width)).\n",
    "  -  The range of values is [0, 255]."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I see. We must convert an image data so that **the shape of the image is CHW**|"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "uint8\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x1232807f0>"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "img=transforms.flip(letter.transpose(2,1,0), y_flip=False, x_flip=True, copy=True)\n",
    "print(img.dtype)\n",
    "plt.imshow(img.transpose(2,1,0))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Yes, Got the result I wanted to see"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# try to apply an image of cifar10"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_dataset, test_dataset = cifar.get_cifar10(scale=255.)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(3, 32, 32)"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "example=train_dataset[0][0].astype(np.uint8)\n",
    "example.shape #<- the shape of example is originally CHW"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x1233eb6d8>"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "img=transforms.flip(example, y_flip=False, x_flip=True, copy=True)\n",
    "plt.imshow(img.transpose(2,1,0))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "example=train_dataset[0][0].astype(np.uint8)\n",
    "example=example.transpose(2,1,0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Conclusion \n",
    "- Sorry, I was quite mistaken.\n",
    "- image shape is very very important\n",
    "- I should have read the document in advance using new module"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.5.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}