train val split

create_train_val_split.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [],
   "source": [
    "from glob import glob\n",
    "import os\n",
    "import shutil\n",
    "import random\n",
    "random.seed(42)\n",
    "from tqdm.auto import tqdm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset_dir = './combined/'\n",
    "ext = '*'\n",
    "minimum = 15000\n",
    "split = 0.8\n",
    "destination_dir = './bottom_type_dataset/'\n",
    "train_dir = 'train'\n",
    "val_dir = 'val'\n",
    "delete_old = True\n",
    "\n",
    "if delete_old:\n",
    "    shutil.rmtree(os.path.join(destination_dir, train_dir), ignore_errors=True)\n",
    "    shutil.rmtree(os.path.join(destination_dir, val_dir), ignore_errors=True)\n",
    "    \n",
    "os.makedirs(os.path.join(destination_dir, train_dir), exist_ok=True)\n",
    "os.makedirs(os.path.join(destination_dir, val_dir), exist_ok=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "classes = [os.path.dirname(d).split('/')[-1] for d in sorted(glob(dataset_dir+ '*/'))]\n",
    "if minimum == -1:\n",
    "    minimum = min([len(glob(os.path.join(dataset_dir, c)+ \"/*.\"+ ext)) for c in classes])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f36bf2b1c6824d919ce97d6b11ad7f11",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Pants', max=15000, style=ProgressStyle(description_width='ini…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "57f9461942b14e10805435d6cb9a9688",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Shorts', max=13264, style=ProgressStyle(description_width='in…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "bd83bfcebd8a477d8d43c7ccccec701c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Skirt', max=15000, style=ProgressStyle(description_width='ini…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "dff6b7f77c824af887b55e85a49578af",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Swimwear', max=8046, style=ProgressStyle(description_width='i…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "\n",
    "for cls in classes:\n",
    "    files = glob(os.path.join(dataset_dir, cls)+ \"/*.\"+ ext)\n",
    "    random.shuffle(files)\n",
    "    end_point = minimum if len(files)> minimum else len(files)\n",
    "    split_point = int(end_point * split)\n",
    "    \n",
    "    train_files = files[:split_point]\n",
    "    val_files = files[split_point:end_point]\n",
    "    os.makedirs(os.path.join(destination_dir, train_dir, cls), exist_ok=True)\n",
    "    with tqdm(total=end_point, desc=cls) as pbar:\n",
    "        for train_file  in train_files:\n",
    "            shutil.copyfile(train_file, os.path.join(destination_dir, train_dir, cls,  os.path.basename(train_file)))\n",
    "            pbar.update()\n",
    "\n",
    "        os.makedirs(os.path.join(destination_dir, val_dir, cls), exist_ok=True)    \n",
    "        for val_file  in val_files:\n",
    "            shutil.copyfile(val_file, os.path.join(destination_dir,val_dir, cls,  os.path.basename(val_file)))\n",
    "            pbar.update()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "python 3 tf2",
   "language": "python",
   "name": "venv"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}