examples/personalize_loggerwrapper.ipynb

{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "# Logger Wrapper personalization/configuration"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "In this notebook, we'll demonstrate how to customize the logger wrapper defined by Sinergym."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true,
    "pycharm": {
     "is_executing": true
    }
   },
   "outputs": [],
   "source": [
    "import gymnasium as gym\n",
    "import numpy as np\n",
    "import sinergym\n",
    "from sinergym.utils.wrappers import (LoggerWrapper, MultiObsWrapper,\n",
    "                                     NormalizeObservation)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Step 1 Inherit and modify the CSVloger"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "First, we need to modify the `create_row_contents` function in the CSV logger to change the values written into the file."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "\n",
    "from sinergym.utils.logger import CSVLogger\n",
    "from typing import Any, Dict, Optional, Sequence, Tuple, Union, List\n",
    "\n",
    "class CustomCSVLogger(CSVLogger):\n",
    "\n",
    "    def __init__(\n",
    "            self,\n",
    "            monitor_header: str,\n",
    "            progress_header: str,\n",
    "            log_progress_file: str,\n",
    "            log_file: Optional[str] = None,\n",
    "            flag: bool = True):\n",
    "        super(CustomCSVLogger, self).__init__(monitor_header,progress_header,log_progress_file,log_file,flag)\n",
    "        self.last_10_steps_reward = [0]*10\n",
    "\n",
    "    def _create_row_content(\n",
    "            self,\n",
    "            obs: List[Any],\n",
    "            action: Union[int, np.ndarray, List[Any]],\n",
    "            terminated: bool,\n",
    "            truncated: bool,\n",
    "            info: Optional[Dict[str, Any]]) -> List:\n",
    "            \n",
    "        if info.get('reward') is not None:\n",
    "            self.last_10_steps_reward.pop(0)\n",
    "            self.last_10_steps_reward.append(info['reward'])\n",
    "\n",
    "        \n",
    "        return [\n",
    "            info.get('timestep',0)] + list(obs) + list(action) + [\n",
    "            info.get('time_elapsed(hours)',0),\n",
    "            info.get('reward',None),\n",
    "            np.mean(self.last_10_steps_reward),\n",
    "            info.get('total_power_no_units'),\n",
    "            info.get('comfort_penalty'),\n",
    "            info.get('abs_comfort'),\n",
    "            terminated,\n",
    "            truncated]\n",
    "\n",
    "    "
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "## Step 2 Instantiate the LoggerWrapper\n"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "Next, we need to instantiate the `LoggerWrapper`, specifying the new headers for our file and the `CSVLogger` class we want to use."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "#==============================================================================================#\n",
      "\u001b[38;20m[ENVIRONMENT] (INFO) : Creating Gymnasium environment... [demo-v1]\u001b[0m\n",
      "#==============================================================================================#\n",
      "\u001b[38;20m[MODELING] (INFO) : Experiment working directory created [/workspaces/sinergym/examples/Eplus-env-demo-v1-res35042]\u001b[0m\n",
      "\u001b[38;20m[MODELING] (INFO) : runperiod established: {'start_day': 1, 'start_month': 1, 'start_year': 1991, 'end_day': 31, 'end_month': 12, 'end_year': 1991, 'start_weekday': 1, 'n_steps_per_hour': 4}\u001b[0m\n",
      "\u001b[38;20m[MODELING] (INFO) : Episode length (seconds): 31536000.0\u001b[0m\n",
      "\u001b[38;20m[MODELING] (INFO) : timestep size (seconds): 900.0\u001b[0m\n",
      "\u001b[38;20m[MODELING] (INFO) : timesteps per episode: 35040\u001b[0m\n",
      "\u001b[38;20m[MODELING] (INFO) : Model Config is correct.\u001b[0m\n",
      "\u001b[38;20m[REWARD] (INFO) : Reward function initialized.\u001b[0m\n",
      "\u001b[38;20m[ENVIRONMENT] (INFO) : Environment demo-v1 created successfully.\u001b[0m\n",
      "\u001b[38;20m[WRAPPER LoggerWrapper] (INFO) : Wrapper initialized.\u001b[0m\n"
     ]
    }
   ],
   "source": [
    "env=gym.make('Eplus-demo-v1')\n",
    "env=LoggerWrapper(env,logger_class=CustomCSVLogger,monitor_header = ['timestep'] + env.get_wrapper_attr('observation_variables') +\n",
    "                env.get_wrapper_attr('action_variables') + ['time (hours)', 'reward', '10-mean-reward',\n",
    "                'power_penalty', 'comfort_penalty', 'terminated', 'truncated'])"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, if you check the Sinergym output folder, you'll find `progress.csv` and `monitor.csv` files for each episode."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "#----------------------------------------------------------------------------------------------#\n",
      "\u001b[38;20m[ENVIRONMENT] (INFO) : Starting a new episode... [demo-v1] [Episode 1]\u001b[0m\n",
      "#----------------------------------------------------------------------------------------------#\n",
      "\u001b[38;20m[MODELING] (INFO) : Episode directory created [/workspaces/sinergym/examples/Eplus-env-demo-v1-res35042/Eplus-env-sub_run1]\u001b[0m\n",
      "\u001b[38;20m[MODELING] (INFO) : Weather file USA_PA_Pittsburgh-Allegheny.County.AP.725205_TMY3.epw used.\u001b[0m\n",
      "\u001b[38;20m[MODELING] (INFO) : Updated building model with whole Output:Variable available names\u001b[0m\n",
      "\u001b[38;20m[MODELING] (INFO) : Updated building model with whole Output:Meter available names\u001b[0m\n",
      "\u001b[38;20m[MODELING] (INFO) : Extra config: runperiod updated to {'apply_weekend_holiday_rule': 'No', 'begin_day_of_month': 1, 'begin_month': 1, 'begin_year': 1991, 'day_of_week_for_start_day': 'Tuesday', 'end_day_of_month': 1, 'end_month': 3, 'end_year': 1991, 'use_weather_file_daylight_saving_period': 'Yes', 'use_weather_file_holidays_and_special_days': 'Yes', 'use_weather_file_rain_indicators': 'Yes', 'use_weather_file_snow_indicators': 'Yes'}\u001b[0m\n",
      "\u001b[38;20m[MODELING] (INFO) : Updated episode length (seconds): 5184000.0\u001b[0m\n",
      "\u001b[38;20m[MODELING] (INFO) : Updated timestep size (seconds): 3600.0\u001b[0m\n",
      "\u001b[38;20m[MODELING] (INFO) : Updated timesteps per episode: 1440\u001b[0m\n",
      "\u001b[38;20m[MODELING] (INFO) : Adapting weather to building model. [USA_PA_Pittsburgh-Allegheny.County.AP.725205_TMY3.epw]\u001b[0m\n",
      "\u001b[38;20m[ENVIRONMENT] (INFO) : Saving episode output path... [/workspaces/sinergym/examples/Eplus-env-demo-v1-res35042/Eplus-env-sub_run1/output]\u001b[0m\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.10/dist-packages/opyplus/weather_data/weather_data.py:493: FutureWarning: the 'line_terminator'' keyword is deprecated, use 'lineterminator' instead.\n",
      "  epw_content = self._headers_to_epw(use_datetimes=use_datetimes) + df.to_csv(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[38;20m[SIMULATOR] (INFO) : Running EnergyPlus with args: ['-w', '/workspaces/sinergym/examples/Eplus-env-demo-v1-res35042/Eplus-env-sub_run1/USA_PA_Pittsburgh-Allegheny.County.AP.725205_TMY3.epw', '-d', '/workspaces/sinergym/examples/Eplus-env-demo-v1-res35042/Eplus-env-sub_run1/output', '/workspaces/sinergym/examples/Eplus-env-demo-v1-res35042/Eplus-env-sub_run1/5ZoneAutoDXVAV.epJSON']\u001b[0m\n",
      "\u001b[38;20m[ENVIRONMENT] (INFO) : Episode 1 started.\u001b[0m\n",
      "\u001b[38;20m[SIMULATOR] (INFO) : handlers initialized.\u001b[0m\n",
      "\u001b[38;20m[SIMULATOR] (INFO) : handlers are ready.\u001b[0m\n",
      "\u001b[38;20m[SIMULATOR] (INFO) : System is ready.\u001b[0m\n",
      "\u001b[38;20m[WRAPPER LoggerWrapper] (INFO) : Creating monitor.csv for current episode (episode 1) if logger is active\u001b[0m\n",
      "Reward:  -0.7359103431230598 {'time_elapsed(hours)': 2.5, 'month': 1, 'day': 1, 'hour': 1, 'is_raining': False, 'action': array([18.217491, 27.886911], dtype=float32), 'timestep': 2, 'reward': -0.7359103431230598, 'energy_term': -0.04774477368803297, 'comfort_term': -0.6881655694350268, 'reward_weight': 0.5, 'abs_energy': 954.8954737606593, 'abs_comfort': 1.3763311388700536, 'energy_values': [954.8954737606593], 'temp_values': [18.623668861129946]}\n",
      "Reward:  -183.89998888619655 {'time_elapsed(hours)': 745.3333333333334, 'month': 2, 'day': 1, 'hour': 0, 'is_raining': False, 'action': array([17.045027, 24.73458 ], dtype=float32), 'timestep': 745, 'reward': -0.04774477368803296, 'energy_term': -0.04774477368803296, 'comfort_term': -0.0, 'reward_weight': 0.5, 'abs_energy': 954.8954737606592, 'abs_comfort': 0.0, 'energy_values': [954.8954737606592], 'temp_values': [20.268929570656645]}\n",
      "Reward:  -301.35254046509186 {'time_elapsed(hours)': 1417.3333333333333, 'month': 3, 'day': 1, 'hour': 0, 'is_raining': False, 'action': array([15.840368, 25.000128], dtype=float32), 'timestep': 1417, 'reward': -0.10365036504194708, 'energy_term': -0.020063419083978743, 'comfort_term': -0.08358694595796834, 'reward_weight': 0.5, 'abs_energy': 401.2683816795749, 'abs_comfort': 0.16717389191593668, 'energy_values': [401.2683816795749], 'temp_values': [19.832826108084063]}\n",
      "\n",
      "Episode  0 Mean reward:  -0.21284685762769534 Cumulative reward:  -306.49947498388144\n",
      "\u001b[38;20m[WRAPPER LoggerWrapper] (INFO) : End of episode, recording summary (progress.csv) if logger is active\u001b[0m\n",
      "\u001b[38;20m[ENVIRONMENT] (INFO) : Environment closed. [demo-v1]\u001b[0m\n"
     ]
    }
   ],
   "source": [
    "for i in range(1):\n",
    "    obs, info = env.reset()\n",
    "    rewards = []\n",
    "    truncated = terminated = False\n",
    "    current_month = 0\n",
    "    while not (terminated or truncated):\n",
    "        a = env.action_space.sample()\n",
    "        obs, reward, terminated, truncated, info = env.step(a)\n",
    "        rewards.append(reward)\n",
    "        if info['month'] != current_month:  # display results every month\n",
    "            current_month = info['month']\n",
    "            print('Reward: ', sum(rewards), info)\n",
    "    print('Episode ', i, 'Mean reward: ', np.mean(\n",
    "        rewards), 'Cumulative reward: ', sum(rewards))\n",
    "env.close()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.10.4 64-bit",
   "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": "3.10.12"
  },
  "vscode": {
   "interpreter": {
    "hash": "e7370f93d1d0cde622a1f8e1c04877d8463912d04d973331ad4851f04de6915a"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}