benchmark_usage_examples.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "c149747b",
   "metadata": {},
   "source": [
    "# Benchmark Usage"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "39c19bf9",
   "metadata": {},
   "source": [
    "\n",
    "1. [Create Object **BenchmarkHandler** to access the data](#first-bullet)\n",
    "2. [List the search spaces](#second-bullet)\n",
    "3. [Get the search space values](#third-bullet)\n",
    "4. [Get environments](#fourth-bullet)\n",
    "5. [Set the environment and search space you want to query](#fifth-bullet)\n",
    "6. [Query the already-set search space, environment and seed for a given budget and configuration](#sixth-bullet)\n",
    "7. [Query a subset of metrics](#seventh-bullet)\n",
    "8. [Create a BenchmarkHandler with customized settings](#eight-bullet)\n",
    "9. [Create a BenchmarkHandler with customized settings (dynamic)](#nineth-bullet)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "01154d05",
   "metadata": {},
   "source": [
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "151cfa82",
   "metadata": {},
   "source": [
    "<a class=\"anchor\" id=\"first-bullet\"></a>\n",
    "\n",
    "### 1) Create Object **BenchmarkHandler** to access the data:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "d95cc51d",
   "metadata": {},
   "outputs": [],
   "source": [
    "from benchmark_handler import BenchmarkHandler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "ee503ef0",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "benchmark = BenchmarkHandler()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e7375314",
   "metadata": {},
   "source": [
    "<a class=\"anchor\" id=\"second-bullet\"></a>\n",
    "\n",
    "### 2) List the search spaces:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "a7b6df70",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['PPO', 'DQN', 'A2C', 'DDPG', 'SAC', 'TD3']"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "benchmark.get_search_spaces_names(set=\"static\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "033a2ff4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['PPO']"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "benchmark.get_search_spaces_names(set=\"dynamic\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "09cc2cdd",
   "metadata": {},
   "source": [
    "<a class=\"anchor\" id=\"third-bullet\"></a>\n",
    "\n",
    "### 3) Get the search space values:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "2c4e1111",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'lr': [-6, -5, -4, -3, -2, -1],\n",
       " 'gamma': [0.8, 0.9, 0.95, 0.98, 0.99, 1.0],\n",
       " 'clip': [0.2, 0.3, 0.4]}"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "benchmark.get_search_space(\"PPO\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "996e60c4",
   "metadata": {},
   "source": [
    "<a class=\"anchor\" id=\"fourth-bullet\"></a>\n",
    "\n",
    "### 4) Get environments:\n",
    "\n",
    "The environments are contained in three groups. You can list the groups: "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "1a0fea92",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Atari', 'Mujoco', 'Control']"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "benchmark.get_environments_groups()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0e410b68",
   "metadata": {},
   "source": [
    "You can list the environments of a specific group"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "1d839e16",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Pong-v0',\n",
       " 'Alien-v0',\n",
       " 'BankHeist-v0',\n",
       " 'BeamRider-v0',\n",
       " 'Breakout-v0',\n",
       " 'Enduro-v0',\n",
       " 'Phoenix-v0',\n",
       " 'Seaquest-v0',\n",
       " 'SpaceInvaders-v0',\n",
       " 'Riverraid-v0',\n",
       " 'Tennis-v0',\n",
       " 'Skiing-v0',\n",
       " 'Boxing-v0',\n",
       " 'Bowling-v0',\n",
       " 'Asteroids-v0']"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "benchmark.get_environments_per_group(\"Atari\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cb5b3817",
   "metadata": {},
   "source": [
    "You can list all the environments:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "4f6a9d0e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Pong-v0',\n",
       " 'Ant-v2',\n",
       " 'Alien-v0',\n",
       " 'BankHeist-v0',\n",
       " 'BeamRider-v0',\n",
       " 'Breakout-v0',\n",
       " 'Enduro-v0',\n",
       " 'Phoenix-v0',\n",
       " 'Seaquest-v0',\n",
       " 'SpaceInvaders-v0',\n",
       " 'Riverraid-v0',\n",
       " 'Tennis-v0',\n",
       " 'Skiing-v0',\n",
       " 'Boxing-v0',\n",
       " 'Bowling-v0',\n",
       " 'Asteroids-v0',\n",
       " 'Hopper-v2',\n",
       " 'Humanoid-v2',\n",
       " 'CartPole-v1',\n",
       " 'MountainCar-v0',\n",
       " 'Acrobot-v1',\n",
       " 'Pendulum-v0']"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "benchmark.get_environments()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ec545a3f",
   "metadata": {},
   "source": [
    "<a class=\"anchor\" id=\"fifth-bullet\"></a>\n",
    "\n",
    "### 5) Set the environment, search space and seed you want to query:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "df10c9f7",
   "metadata": {},
   "outputs": [],
   "source": [
    "benchmark.set_env_space_seed(search_space=\"PPO\", environment=\"Pong-v0\", seed=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1f9c8bb7",
   "metadata": {},
   "source": [
    "<a class=\"anchor\" id=\"sixth-bullet\"></a>\n",
    "\n",
    "###  6) Query the already-set search space and environement for a given budget and configuration:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "f4aebfaa",
   "metadata": {},
   "outputs": [],
   "source": [
    "configuration = {\"lr\":-6, \"gamma\": 0.8, \"clip\": 0.2}\n",
    "budget = 50\n",
    "queried_data = benchmark.get_metrics(configuration, budget=budget)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "1e2ad5af",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys(['eval_avg_returns', 'eval_std_returns', 'eval_timestamps', 'eval_timesteps'])"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "queried_data.keys()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1b73f586",
   "metadata": {},
   "source": [
    "<a class=\"anchor\" id=\"seventh-bullet\"></a>\n",
    "\n",
    "### 7) Query a subset of metrics\n",
    "\n",
    "As you can see, the default functionality returns all the metrics. You can change the value of the attritbute \"return_names\" of a benchmark object and define the specific metrics you want."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "2423954d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys(['eval_avg_returns', 'eval_timesteps'])"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "benchmark.return_metrics = [\"eval_avg_returns\", \"eval_timesteps\"]\n",
    "configuration = {\"lr\":-4, \"gamma\": 0.98, \"clip\": 0.2}\n",
    "budget = 50\n",
    "queried_data = benchmark.get_metrics(configuration, budget=budget)\n",
    "queried_data.keys()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "701e5017",
   "metadata": {},
   "source": [
    "<a class=\"anchor\" id=\"eight-bullet\"></a>\n",
    "\n",
    "### 8) Create a BenchmarkHandler with customized settings (static)\n",
    "\n",
    "It is possible to define the environment, space and return metrics when creating the benchmark object:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "bfc4e2a5",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "benchmark = BenchmarkHandler(environment = \"Pong-v0\",\n",
    "                             search_space = \"PPO\",\n",
    "                             return_metrics = [\"eval_avg_returns\", \"eval_timesteps\"],\n",
    "                             seed = 0)\n",
    "queried_data = benchmark.get_metrics(configuration, budget=budget)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "17582764",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x1d8c782b1c0>]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(queried_data[\"eval_avg_returns\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "967c6557",
   "metadata": {},
   "source": [
    "<a class=\"anchor\" id=\"nineth-bullet\"></a>\n",
    "\n",
    "### 9) Create a BenchmarkHandler with customized settings (dynamic)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "bc9935c6",
   "metadata": {},
   "outputs": [],
   "source": [
    "benchmark.set_env_space_seed(search_space=\"PPO\", environment=\"Enduro-v0\", seed= 0)\n",
    "# Note: For the dynamic subset, you need to provide a list of max 3 values per hyperparameter.\n",
    "# Each of the 3 values signifies the value change at the switch.\n",
    "# If less than 3 values provided, the last value in the list is not changed until the end of the run.\n",
    "configuration = {\"lr\":[-5, -4], \"gamma\": [0.99,0.95, 0.99]}\n",
    "queried_data = benchmark.get_metrics(configuration, budget=budget, set=\"dynamic\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "34dbb0c8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x1d8c987bfd0>]"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(queried_data[\"eval_avg_returns\"])"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.19"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}