queurunner vs dataset api

queuerunner vs api.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:root:Test\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "import numpy as np\n",
    "import time\n",
    "import logging\n",
    "import tensorflow as tf\n",
    "\n",
    "from hudl_beatrix.dataset import BrainWashDataset\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.patches as patches\n",
    "%matplotlib inline\n",
    "\n",
    "logging.info('Test')\n",
    "logger = logging.getLogger()\n",
    "logger.setLevel(logging.INFO)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset = BrainWashDataset(image_method='imageio')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "ERROR:root:Directory already exists, no data will be converted! Did you mean to set the overwrite flag to True?\n"
     ]
    }
   ],
   "source": [
    "dataset.convert(dataset_path='./brainwash_tf_records_imageio')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_dir = './brainwash_tf_records_imageio/train'\n",
    "records = [os.path.join(train_dir, record) for record in os.listdir(train_dir)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:root:Queue time : 13.206629753112793 sec\n",
      "INFO:root:API time : 13.020855903625488 sec\n"
     ]
    }
   ],
   "source": [
    "n_threads = 4\n",
    "batch_size = 1\n",
    "min_after_dequeue = 8\n",
    "\n",
    "input_fn_queue = dataset.input_fn(records, \n",
    "                                  batch_size=batch_size,\n",
    "                                  n_threads=n_threads,\n",
    "                                  min_after_dequeue=min_after_dequeue,\n",
    "                                  method='queuerunner')\n",
    "\n",
    "input_fn_api = dataset.input_fn(records, \n",
    "                                  batch_size=batch_size,\n",
    "                                  n_threads=n_threads,\n",
    "                                  min_after_dequeue=min_after_dequeue,\n",
    "                                  method='api')\n",
    "\n",
    "record_queue = input_fn_queue()\n",
    "record_api = input_fn_api()\n",
    "\n",
    "elapsed_queue = 0\n",
    "elapsed_api = 0\n",
    "\n",
    "with tf.Session() as sess:\n",
    "    sess.run(tf.global_variables_initializer())\n",
    "\n",
    "    coord = tf.train.Coordinator()\n",
    "    threads = tf.train.start_queue_runners(coord=coord)\n",
    "\n",
    "    for i in range(len(records)):\n",
    "    \n",
    "        start_time = time.time()\n",
    "        img, bx = sess.run(record_queue)\n",
    "        elapsed_queue += time.time() - start_time\n",
    "        \n",
    "        start_time = time.time()\n",
    "        img, bx = sess.run(record_api)\n",
    "        elapsed_api += time.time() - start_time\n",
    "\n",
    "    coord.request_stop()\n",
    "    coord.join(threads)\n",
    "\n",
    "logger.info('Queue time : {} sec'.format(elapsed_queue))\n",
    "logger.info('API time : {} sec'.format(elapsed_api))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.5.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}