fastai CycleGAN

cyclegan.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Test CycleGAN with custom ImageTupleList\n",
    "\n",
    "Test fastai implementation of CycleGAN of a set of 1-channel TIFF brain MR images."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Setup notebook"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import sys\n",
    "\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import torch\n",
    "from torch import nn"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    " Support in-notebook plotting"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Report versions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "numpy version: 1.15.4\n",
      "matplotlib version: 3.0.1\n"
     ]
    }
   ],
   "source": [
    "print('numpy version: {}'.format(np.__version__))\n",
    "from matplotlib import __version__ as mplver\n",
    "print('matplotlib version: {}'.format(mplver))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "python version: 3.7.1\n"
     ]
    }
   ],
   "source": [
    "pv = sys.version_info\n",
    "print('python version: {}.{}.{}'.format(pv.major, pv.minor, pv.micro))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Reload packages where content for package development"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "%load_ext autoreload\n",
    "%autoreload 2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Define training/validation image location\n",
    "\n",
    "Images are all `.tif` located in `train_dir` (defined below). There are two types of images of interest located in `train_dir/t1` and `train_dir/flair`. In each of those directories there is a `train` and `valid` directory where the actual `.tif` files exist."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_dir = '/Users/jcreinhold/Research/data/nn_test/2d/small_dataset/'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "import fastai as fai\n",
    "import fastai.vision as faiv\n",
    "import torchvision"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "fastai version: 1.0.34\n",
      "pytorch version: 1.0.0.dev20181014\n",
      "torchvision version: 0.2.1\n"
     ]
    }
   ],
   "source": [
    "print(f'fastai version: {fai.__version__}')\n",
    "print(f'pytorch version: {torch.__version__}')\n",
    "print(f'torchvision version: {torchvision.__version__}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## CycleGAN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "import math\n",
    "from typing import Tuple\n",
    "from PIL import Image\n",
    "from pathlib import PosixPath\n",
    "\n",
    "\n",
    "def open_tiff(fn:fai.PathOrStr)->faiv.Image:\n",
    "    \"\"\" open a 1 channel tif image and transform it into a fastai image \"\"\"\n",
    "    return faiv.Image(torch.Tensor(np.asarray(Image.open(fn),dtype=np.float32)[None,...]))\n",
    "\n",
    "\n",
    "class TIFFImageList(faiv.ImageItemList):\n",
    "    \"\"\" custom item list for TIFF files \"\"\"\n",
    "    def open(self, fn:fai.PathOrStr)->faiv.Image: return open_tiff(fn)\n",
    "\n",
    "\n",
    "class ImageTuple(fai.ItemBase):\n",
    "    def __init__(self, img1, img2):\n",
    "        self.img1,self.img2 = img1,img2\n",
    "        self.obj,self.data = (img1,img2),[img1.data,img2.data]\n",
    "\n",
    "    def apply_tfms(self, tfms, **kwargs):\n",
    "        self.img1 = self.img1.apply_tfms(tfms, **kwargs)\n",
    "        self.img2 = self.img2.apply_tfms(tfms, **kwargs)\n",
    "        return self\n",
    "    \n",
    "    def to_one(self): return faiv.Image(torch.cat(self.data,2))\n",
    "    \n",
    "    def show_xys(self, xs, ys, figsize:Tuple[int,int]=(9,10), **kwargs):\n",
    "        \"Show the `xs` and `ys` on a figure of `figsize`. `kwargs` are passed to the show method.\"\n",
    "        rows = int(math.sqrt(len(xs)))\n",
    "        fig, axs = plt.subplots(rows,rows,figsize=figsize)\n",
    "        for i, ax in enumerate(axs.flatten() if rows > 1 else [axs]):\n",
    "            xs[i].show(ax=ax, y=ys[i], **kwargs)\n",
    "        plt.tight_layout()\n",
    "\n",
    "    def show_xyzs(self, xs, ys, zs, figsize:Tuple[int,int]=None, **kwargs):\n",
    "        \"\"\"Show `xs` (inputs), `ys` (targets) and `zs` (predictions) on a figure of `figsize`. \n",
    "        `kwargs` are passed to the show method.\"\"\"\n",
    "        figsize = fai.ifnone(figsize, (6,3*len(xs)))\n",
    "        fig,axs = plt.subplots(len(xs), 2, figsize=figsize)\n",
    "        fig.suptitle('Ground truth / Predictions', weight='bold', size=14)\n",
    "        for i,(x,y,z) in enumerate(zip(xs,ys,zs)):\n",
    "            x.show(ax=axs[i,0], y=y, **kwargs)\n",
    "            x.show(ax=axs[i,1], y=z, **kwargs)\n",
    "    \n",
    "    def __repr__(self): \n",
    "        return f'{self.__class__.__name__} - im1:{tuple(self.img1.shape)}, im2:{tuple(self.img2.shape)}'\n",
    "    \n",
    "\n",
    "class TargetTupleList(fai.ItemList):\n",
    "    def reconstruct(self, t:torch.Tensor):\n",
    "        if len(t.size()) == 0: return t\n",
    "        return ImageTuple(faiv.Image(t[0]),faiv.Image(t[1]))\n",
    "\n",
    "\n",
    "class TIFFTupleList(TIFFImageList):\n",
    "    _label_cls = TargetTupleList\n",
    "    def __init__(self, items, itemsB=None, **kwargs):\n",
    "        self.itemsB = itemsB\n",
    "        super().__init__(items, **kwargs)\n",
    "\n",
    "    def new(self, items, **kwargs):\n",
    "        return super().new(items, itemsB=self.itemsB, **kwargs)\n",
    "\n",
    "    def get(self, i):\n",
    "        img1 = super().get(i)\n",
    "        fn = self.itemsB[i]\n",
    "        return ImageTuple(img1, open_tiff(fn))\n",
    "\n",
    "    def reconstruct(self, t:torch.Tensor):\n",
    "        return ImageTuple(faiv.Image(t[0]),faiv.Image(t[1]))\n",
    "\n",
    "    @classmethod\n",
    "    def from_folders(cls, path, folderA, folderB, **kwargs):\n",
    "        path = PosixPath(path)\n",
    "        itemsB = TIFFImageList.from_folder(path/folderB).items\n",
    "        res = super().from_folder(path/folderA, itemsB=itemsB, **kwargs)\n",
    "        res.path = path\n",
    "        return res\n",
    "    \n",
    "    def show_xys(self, xs, ys, figsize:Tuple[int,int]=(12,6), **kwargs):\n",
    "        \"Show the `xs` and `ys` on a figure of `figsize`. `kwargs` are passed to the show method.\"\n",
    "        rows = int(math.sqrt(len(xs)))\n",
    "        fig, axs = plt.subplots(rows,rows,figsize=figsize)\n",
    "        for i, ax in enumerate(axs.flatten() if rows > 1 else [axs]):\n",
    "            xs[i].to_one().show(ax=ax, **kwargs)\n",
    "        plt.tight_layout()\n",
    "\n",
    "    def show_xyzs(self, xs, ys, zs, figsize:Tuple[int,int]=None, **kwargs):\n",
    "        \"\"\"Show `xs` (inputs), `ys` (targets) and `zs` (predictions) on a figure of `figsize`.\n",
    "        `kwargs` are passed to the show method.\"\"\"\n",
    "        figsize = fai.ifnone(figsize, (12,3*len(xs)))\n",
    "        fig,axs = plt.subplots(len(xs), 2, figsize=figsize)\n",
    "        fig.suptitle('Ground truth / Predictions', weight='bold', size=14)\n",
    "        for i,(x,z) in enumerate(zip(xs,zs)):\n",
    "            x.to_one().show(ax=axs[i,0], **kwargs)\n",
    "            z.to_one().show(ax=axs[i,1], **kwargs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = (TIFFTupleList.from_folders(train_dir, 't1', 'flair', extensions=('.tif'))\n",
    "                     .split_by_idx([])\n",
    "                     .label_const(0.)\n",
    "                     .transform([])\n",
    "                     .databunch(bs=4))\n",
    "data.valid_dl = None"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train\n",
      "(ImageTuple - im1:(1, 256, 256), im2:(1, 256, 256), 0.0)\n",
      "(ImageTuple - im1:(1, 256, 256), im2:(1, 256, 256), 0.0)\n",
      "(ImageTuple - im1:(1, 256, 256), im2:(1, 256, 256), 0.0)\n",
      "(ImageTuple - im1:(1, 256, 256), im2:(1, 256, 256), 0.0)\n",
      "(ImageTuple - im1:(1, 256, 256), im2:(1, 256, 256), 0.0)\n",
      "(ImageTuple - im1:(1, 256, 256), im2:(1, 256, 256), 0.0)\n",
      "(ImageTuple - im1:(1, 256, 256), im2:(1, 256, 256), 0.0)\n",
      "(ImageTuple - im1:(1, 256, 256), im2:(1, 256, 256), 0.0)\n",
      "(ImageTuple - im1:(1, 256, 256), im2:(1, 256, 256), 0.0)\n",
      "(ImageTuple - im1:(1, 256, 256), im2:(1, 256, 256), 0.0)\n",
      "(ImageTuple - im1:(1, 256, 256), im2:(1, 256, 256), 0.0)\n",
      "(ImageTuple - im1:(1, 256, 256), im2:(1, 256, 256), 0.0)\n",
      "(ImageTuple - im1:(1, 256, 256), im2:(1, 256, 256), 0.0)\n",
      "(ImageTuple - im1:(1, 256, 256), im2:(1, 256, 256), 0.0)\n",
      "(ImageTuple - im1:(1, 256, 256), im2:(1, 256, 256), 0.0)\n",
      "(ImageTuple - im1:(1, 256, 256), im2:(1, 256, 256), 0.0)\n"
     ]
    }
   ],
   "source": [
    "print('Train')\n",
    "for i in range(len(data.train_ds)): print(data.train_ds[i])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x432 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "data.show_batch(rows=2,cmap='gray')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "from functools import partial\n",
    "cycle_gan = faiv.models.gan.CycleGAN(1, 1, gen_blocks=1, disc_layers=1, n_features=2)\n",
    "learner = fai.Learner(data, cycle_gan, loss_func=faiv.models.gan.CycleGanLoss(cycle_gan), \n",
    "                      opt_func=partial(torch.optim.Adam, betas=(0.5,0.99)), \n",
    "                      callback_fns=[fai.callbacks.gan.CycleGANTrainer])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LR Finder is complete, type {learner_name}.recorder.plot() to see the graph.\n"
     ]
    }
   ],
   "source": [
    "learner.lr_find(num_it=20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "learner.recorder.plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "Total time: 00:14 <p><table style='width:525px; margin-bottom:10px'>\n",
       "  <tr>\n",
       "    <th>epoch</th>\n",
       "    <th>train_loss</th>\n",
       "    <th>idt_loss</th>\n",
       "    <th>gen_loss</th>\n",
       "    <th>cyc_loss</th>\n",
       "    <th>da_loss</th>\n",
       "    <th>db_loss</th>\n",
       "  </tr>\n",
       "  <tr>\n",
       "    <th>1</th>\n",
       "    <th>2080.854492</th>\n",
       "    <th>692.777954</th>\n",
       "    <th>2.341889</th>\n",
       "    <th>1385.734497</th>\n",
       "    <th>0.773312</th>\n",
       "    <th>0.692109</th>\n",
       "  </tr>\n",
       "  <tr>\n",
       "    <th>2</th>\n",
       "    <th>2082.499512</th>\n",
       "    <th>693.342163</th>\n",
       "    <th>2.300380</th>\n",
       "    <th>1386.856934</th>\n",
       "    <th>0.767366</th>\n",
       "    <th>0.680148</th>\n",
       "  </tr>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "learner.fit_one_cycle(2, 8e-4, moms=(0.5,0.5))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "Total time: 00:07 <p><table style='width:525px; margin-bottom:10px'>\n",
       "  <tr>\n",
       "    <th>epoch</th>\n",
       "    <th>train_loss</th>\n",
       "    <th>idt_loss</th>\n",
       "    <th>gen_loss</th>\n",
       "    <th>cyc_loss</th>\n",
       "    <th>da_loss</th>\n",
       "    <th>db_loss</th>\n",
       "  </tr>\n",
       "  <tr>\n",
       "    <th>1</th>\n",
       "    <th>2088.081787</th>\n",
       "    <th>695.234070</th>\n",
       "    <th>2.223558</th>\n",
       "    <th>1390.624268</th>\n",
       "    <th>0.753572</th>\n",
       "    <th>0.662040</th>\n",
       "  </tr>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "learner.fit(1, 8e-4)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "neuropp",
   "language": "python",
   "name": "neuropp"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}