{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "%reload_ext autoreload\n",
    "%autoreload 2\n",
    "%matplotlib inline\n",
    "\n",
    "from fastai import *\n",
    "from fastai.vision import *"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "tfms = get_transforms()\n",
    "bs = 64"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def show_image_batch_supercharged(dl:DataLoader, classes:Collection[str], rows:int=None, figsize:Tuple[int,int]=(12,15),\n",
    "                     denorm:Callable=None)->None:\n",
    "    b_idx = next(iter(dl.batch_sampler))\n",
    "    if rows is None: rows = int(math.sqrt(len(b_idx)))\n",
    "    fig, axs = plt.subplots(rows,rows,figsize=figsize)\n",
    "    for i, ax in zip(b_idx[:rows*rows], axs.flatten()):\n",
    "        x,y = dl.dataset[i]\n",
    "        x.show(ax=ax, y=y, classes=classes)\n",
    "    plt.tight_layout()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def show_batch_supercharged(data:DataBunch, rows:int=3, figsize:Tuple[int,int]=(9,10), is_train:bool=True)->None:\n",
    "        show_image_batch_supercharged(data.train_dl if is_train else data.valid_dl, data.classes,\n",
    "            denorm=getattr(data,'denorm',None), figsize=figsize, rows=rows)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Pets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "path = untar_data(URLs.PETS)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = (ImageFileList.from_folder(path/'images')\n",
    "        .label_from_re(r'^(.*)_\\d+.jpg$')\n",
    "        .random_split_by_pct(0.2)\n",
    "        .datasets(ImageClassificationDataset)\n",
    "        .transform(tfms, size=224)\n",
    "        .databunch(bs=bs, path=path))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = (ImageFileList.from_folder(path/'images')\n",
    "        .label_from_re(r'^(.*)_\\d+.jpg$')\n",
    "        .random_split_by_pct(0.2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.show_batch(rows=3, figsize=(7,7))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "show_batch_supercharged(data, rows=3, figsize=(7,7))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Planet"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "path = untar_data(URLs.PLANET_SAMPLE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = (InputList.from_folder(path)\n",
    "        .label_from_csv('labels.csv', sep=' ', suffix='.jpg', folder='train')\n",
    "        .random_split_by_pct(0.2)\n",
    "        .datasets(ImageMultiDataset)\n",
    "        .transform(tfms, size=128)\n",
    "        .databunch(bs=bs, path=path))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.show_batch(rows=3, figsize=(10,10))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "show_batch_supercharged(data, rows=3, figsize=(10,10))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Camvid"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "path = untar_data(URLs.CAMVID)\n",
    "path_lbl = path/'labels'\n",
    "path_img = path/'images'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "get_y_fn = lambda x: path_lbl/f'{x.stem}_P{x.suffix}'\n",
    "codes = np.loadtxt(path/'codes.txt', dtype=str); codes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = (ImageFileList.from_folder(path_img)\n",
    "        .label_from_func(get_y_fn)\n",
    "        .random_split_by_pct()\n",
    "        .datasets(SegmentationDataset, classes=codes)\n",
    "        .transform(get_transforms(), size=128, tfm_y=True)\n",
    "        .databunch(bs=64))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.show_batch(rows=2, figsize=(7,7))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "show_batch_supercharged(data, rows=2, figsize=(7,7))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Coco"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "path = untar_data(URLs.COCO_TINY)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "images, lbl_bbox = get_annotations(path/'train.json')\n",
    "img2bbox = {img:bb for img, bb in zip(images, lbl_bbox)}\n",
    "get_y_func = lambda o:img2bbox[o.name]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = (ImageFileList.from_folder(path)\n",
    "        .label_from_func(get_y_func)\n",
    "        .random_split_by_pct()\n",
    "        .datasets(ObjectDetectDataset)\n",
    "        .transform(tfms, size=128, tfm_y=True)\n",
    "        .databunch(bs=16, collate_fn=bb_pad_collate))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.show_batch(rows=3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}