GTSRB/06-Notebook-as-a-batch.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<img width=\"800px\" src=\"../fidle/img/00-Fidle-header-01.svg\"></img>\n",
    "\n",
    "# <!-- TITLE --> [GTSRB6] - Full convolutions as a batch\n",
    "<!-- DESC --> Episode 6 : To compute bigger, use your notebook in batch mode\n",
    "<!-- AUTHOR : Jean-Luc Parouty (CNRS/SIMaP) -->\n",
    "\n",
    "## Objectives :\n",
    "  - Run a notebook code as a **job**\n",
    "  - Follow up with Tensorboard\n",
    "  \n",
    "The German Traffic Sign Recognition Benchmark (GTSRB) is a dataset with more than 50,000 photos of road signs from about 40 classes.  \n",
    "The final aim is to recognise them !  \n",
    "Description is available there : http://benchmark.ini.rub.de/?section=gtsrb&subsection=dataset\n",
    "\n",
    "\n",
    "## What we're going to do :\n",
    "Our main steps:\n",
    " - Run Full-convolution.ipynb as a batch :\n",
    "    - Notebook mode\n",
    "    - Script mode \n",
    " - Tensorboard follow up"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Step 1 - How to run a notebook in a command line ?\n",
    "\n",
    "Two simple solutions are possible :-)\n",
    "\n",
    " - **Option 1 - As a notebook ! (a good choice)**\n",
    "\n",
    "  Very simple.  \n",
    "  The result is the executed notebook, so we can retrieve all the cell'soutputs of the notebook :  \n",
    "  ```jupyter nbconvert (...) --to notebook --execute <notebook>```   \n",
    "\n",
    "  Example :   \n",
    "  ```jupyter nbconvert --ExecutePreprocessor.timeout=-1 --to notebook --execute my_notebook.ipynb'```  \n",
    "  The result will be a notebook: 'my_notebook.nbconvert.ipynb'.\n",
    "  \n",
    "  See: [nbconvert documentation](https://nbconvert.readthedocs.io/en/latest/usage.html#convert-notebook)\n",
    "\n",
    "  **Note :** Do not forget the option: --ExecutePreprocessor.timeout=-1\n",
    "\n",
    " - **Option 2 - As a script**\n",
    "\n",
    "  Very simple too, but with some constraints on the notebook.  \n",
    "  We will convert the notebook to a Python script (IPython, to be precise) :  \n",
    "  ```jupyter nbconvert --to script <notebook>```  \n",
    "  \n",
    "  Then we can execute this script :  \n",
    "  ```ipython <script>```\n",
    "  \n",
    "  See: [nbconvert documentation](https://nbconvert.readthedocs.io/en/latest/usage.html#executable-script)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Step 2 - How to run a notebook in a batch ?\n",
    "\n",
    "Maybe not always the best solution, but this solution is very rustic !  \n",
    "\n",
    "### 2.1 - As a notebook ! (better choice)\n",
    "\n",
    "A direct execution with nbconvert (see option 1) is probably the best solution.  \n",
    "This allows to recover a complete notebook (graphics, traces, ...).\n",
    "\n",
    "### 2.2 - As a IPython script :\n",
    "**Important :** The generated python script must be executed with the iPython interpreter.\n",
    "\n",
    "Example :\n",
    "`$ ipython my_converted_notebook.py`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "! jupyter nbconvert --to script --output='05-full_convolutions' '05-Full-convolutions.ipynb'\n",
    "! ls -l *.py"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.3 - Batch submission\n",
    "\n",
    " See the two examples of bash launch script :\n",
    " - [batch_slurm.sh](batch_slurm.sh) using Slurm (like at IDRIS)\n",
    " - [batch_oar.sh](batch_oar.sh)   using OAR (like at GRICAD)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Example at IDRIS, using **slurm**\n",
    "\n",
    "On the frontal :\n",
    "```bash\n",
    "# hostname\n",
    "jean-zay2\n",
    "\n",
    "# sbatch $WORK/fidle/GTSRB/batch_slurm.sh \n",
    "Submitted batch job 249794\n",
    "\n",
    "#squeue -u $USER\n",
    "             JOBID PARTITION     NAME     USER ST       TIME  NODES NODELIST(REASON)\n",
    "            249794    gpu_p1 GTSRB Fu  jde45kb PD       0:00      1 (Resources)\n",
    "\n",
    "# ls -l _batch/\n",
    "total 32769\n",
    "-rw-r--r-- 1 jde45kb gensim07 13349 Sep 10 11:32 GTSRB_249794.err\n",
    "-rw-r--r-- 1 jde45kb gensim07   489 Sep 10 11:31 GTSRB_249794.out\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Example at GRICAD, using **oar**\n",
    "\n",
    "Have to be done on the frontal :\n",
    "```bash\n",
    "# hostname\n",
    "f-dahu\n",
    "\n",
    "# oarsub -S ~/fidle/GTSRB/batch_oar.sh\n",
    "[GPUNODE] Adding gpu node restriction\n",
    "[ADMISSION RULE] Modify resource description with type constraints\n",
    "\n",
    "#oarstat -u\n",
    "Job id    S User     Duration   System message\n",
    "--------- - -------- ---------- ------------------------------------------------\n",
    "5878410   R watsonb    0:19:56 R=8,W=1:0:0,J=I,P=fidle,T=gpu (Karma=0.005,quota_ok)\n",
    "5896266   W watsonb    0:00:00 R=8,W=1:0:0,J=B,N=Full convolutions,P=fidle,T=gpu\n",
    "\n",
    "# ls -l\n",
    "total 8\n",
    "-rw-r--r-- 1 watsonb l-simap    0 Feb 28 15:58 batch_oar_5896266.err\n",
    "-rw-r--r-- 1 watsonb l-simap 5703 Feb 28 15:58 batch_oar_5896266.out\n",
    "```\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "----\n",
    "**What you can do :**\n",
    "\n",
    "(If you have a calculation environment with a scheduler...)\n",
    "Your mission if you accept it: Run our full_convolution code in batch mode.<br>\n",
    " For that :\n",
    "   - Validate the full_convolution notebook on short tests</li>\n",
    "   - Submit it in batch mode for validation</li>\n",
    "   - Modify the notebook for a full run and submit it :-)</li>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---\n",
    "<img width=\"80px\" src=\"../fidle/img/00-Fidle-logo-01.svg\"></img>"
   ]
  }
 ],
 "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": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}