HyperOpt_result_and_run_all.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from hyperopt import Trials, space_eval\n",
    "import torch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "result = torch.load(\"search_result.pth\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'clipping_gradient': True,\n",
       " 'observation_normalization': False,\n",
       " 'orthogonal_initialization_and_layer_scaling': False,\n",
       " 'policy_noclip': False,\n",
       " 'reward_clipping': True,\n",
       " 'reward_standardization': True,\n",
       " 'value_clipping': False}"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result['best']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "tr = result['trial'].trials\n",
    "space = result['space']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-134.57761419999972"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tr[0]['result']['loss']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'clipping_gradient': [0],\n",
       " 'observation_normalization': [1],\n",
       " 'orthogonal_initialization_and_layer_scaling': [0],\n",
       " 'policy_noclip': [1],\n",
       " 'reward_clipping': [0],\n",
       " 'reward_standardization': [1],\n",
       " 'value_clipping': [0]}"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tr[0]['misc']['vals']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "d = {0: True, 1: False}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "tmp = []\n",
    "for i in range(len(tr)):\n",
    "    row_tmp = {}\n",
    "    row_tmp['score'] = -tr[i]['result']['loss']\n",
    "    row_tmp['clipping_gradient'] = d[tr[i]['misc']['vals']['clipping_gradient'][0]]\n",
    "    row_tmp['observation_normalization'] = d[tr[i]['misc']['vals']['observation_normalization'][0]]\n",
    "    row_tmp['orthogonal_initialization_and_layer_scaling'] = d[tr[i]['misc']['vals']['orthogonal_initialization_and_layer_scaling'][0]]\n",
    "    row_tmp['policy_noclip'] = d[tr[i]['misc']['vals']['policy_noclip'][0]]\n",
    "    row_tmp['reward_clipping'] = d[tr[i]['misc']['vals']['reward_clipping'][0]]\n",
    "    row_tmp['reward_standardization'] = d[tr[i]['misc']['vals']['reward_standardization'][0]]\n",
    "    row_tmp['value_clipping'] = d[tr[i]['misc']['vals']['value_clipping'][0]]\n",
    "    tmp.append(row_tmp)\n",
    "    \n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.DataFrame(tmp).sort_values(by='score', ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "df2 = df.drop_duplicates()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "options = df2.columns[1:].tolist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['clipping_gradient',\n",
       " 'observation_normalization',\n",
       " 'orthogonal_initialization_and_layer_scaling',\n",
       " 'policy_noclip',\n",
       " 'reward_clipping',\n",
       " 'reward_standardization',\n",
       " 'value_clipping']"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "options"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 8 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(4, 2, figsize=(10, 10), facecolor='white')\n",
    "for idx, opt in enumerate(options):\n",
    "    if idx == 0:\n",
    "        sns.distplot(df2.score, ax=ax[0][0])\n",
    "        ax[0][0].set_title(\"total score distribution\")\n",
    "    \n",
    "    row_n, col_n = (idx+1) % 4, (idx+1) // 4\n",
    "    ax[row_n][col_n].set_title(opt)\n",
    "    sns.distplot(df2[lambda x: x[opt] == True].score, label='ON', ax=ax[row_n][col_n])\n",
    "    sns.distplot(df2[lambda x: x[opt] == False].score, label='OFF', ax=ax[row_n][col_n])\n",
    "    ax[row_n][col_n].legend()\n",
    "    \n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "clipping_gradient\n",
      "False     80.246847\n",
      "True     114.132718\n",
      "Name: score, dtype: float64\n",
      "\n",
      "\n",
      "observation_normalization\n",
      "False    135.672125\n",
      "True     -25.034419\n",
      "Name: score, dtype: float64\n",
      "\n",
      "\n",
      "orthogonal_initialization_and_layer_scaling\n",
      "False     83.172379\n",
      "True     122.924345\n",
      "Name: score, dtype: float64\n",
      "\n",
      "\n",
      "policy_noclip\n",
      "False    136.639180\n",
      "True     -34.792868\n",
      "Name: score, dtype: float64\n",
      "\n",
      "\n",
      "reward_clipping\n",
      "False      0.870902\n",
      "True     122.184475\n",
      "Name: score, dtype: float64\n",
      "\n",
      "\n",
      "reward_standardization\n",
      "False     70.185904\n",
      "True     106.342454\n",
      "Name: score, dtype: float64\n",
      "\n",
      "\n",
      "value_clipping\n",
      "False    120.288015\n",
      "True      21.117910\n",
      "Name: score, dtype: float64\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# option on vs. off\n",
    "for opt in options:\n",
    "    print(df2.groupby(opt).score.mean())\n",
    "    print(\"\\n\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "df3 = df2.groupby(df2.columns[1:].tolist())[['score']].mean().reset_index().sort_values('score', ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-[1]-\n",
      "clipping_gradient                                True\n",
      "observation_normalization                       False\n",
      "orthogonal_initialization_and_layer_scaling      True\n",
      "policy_noclip                                   False\n",
      "reward_clipping                                  True\n",
      "reward_standardization                           True\n",
      "value_clipping                                  False\n",
      "score                                          246.65\n",
      "Name: 28, dtype: object\n",
      "\n",
      "\n",
      "-[2]-\n",
      "clipping_gradient                                 True\n",
      "observation_normalization                        False\n",
      "orthogonal_initialization_and_layer_scaling      False\n",
      "policy_noclip                                    False\n",
      "reward_clipping                                  False\n",
      "reward_standardization                            True\n",
      "value_clipping                                   False\n",
      "score                                          221.281\n",
      "Name: 22, dtype: object\n",
      "\n",
      "\n",
      "-[3]-\n",
      "clipping_gradient                                False\n",
      "observation_normalization                        False\n",
      "orthogonal_initialization_and_layer_scaling      False\n",
      "policy_noclip                                    False\n",
      "reward_clipping                                   True\n",
      "reward_standardization                            True\n",
      "value_clipping                                   False\n",
      "score                                          198.615\n",
      "Name: 4, dtype: object\n",
      "\n",
      "\n",
      "-[4]-\n",
      "clipping_gradient                                 True\n",
      "observation_normalization                        False\n",
      "orthogonal_initialization_and_layer_scaling      False\n",
      "policy_noclip                                    False\n",
      "reward_clipping                                   True\n",
      "reward_standardization                            True\n",
      "value_clipping                                   False\n",
      "score                                          192.678\n",
      "Name: 24, dtype: object\n",
      "\n",
      "\n",
      "-[5]-\n",
      "clipping_gradient                                 True\n",
      "observation_normalization                        False\n",
      "orthogonal_initialization_and_layer_scaling       True\n",
      "policy_noclip                                    False\n",
      "reward_clipping                                   True\n",
      "reward_standardization                           False\n",
      "value_clipping                                   False\n",
      "score                                          171.228\n",
      "Name: 26, dtype: object\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# top 5 results\n",
    "for i in range(5):\n",
    "    print(\"-[{}]-\".format(i+1))\n",
    "    print(df3.iloc[i])\n",
    "    print(\"\\n\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-[1]-\n",
      "clipping_gradient                                False\n",
      "observation_normalization                        False\n",
      "orthogonal_initialization_and_layer_scaling      False\n",
      "policy_noclip                                     True\n",
      "reward_clipping                                  False\n",
      "reward_standardization                            True\n",
      "value_clipping                                   False\n",
      "score                                         -116.133\n",
      "Name: 7, dtype: object\n",
      "\n",
      "\n",
      "-[2]-\n",
      "clipping_gradient                                False\n",
      "observation_normalization                         True\n",
      "orthogonal_initialization_and_layer_scaling      False\n",
      "policy_noclip                                     True\n",
      "reward_clipping                                  False\n",
      "reward_standardization                           False\n",
      "value_clipping                                    True\n",
      "score                                         -109.876\n",
      "Name: 16, dtype: object\n",
      "\n",
      "\n",
      "-[3]-\n",
      "clipping_gradient                               False\n",
      "observation_normalization                        True\n",
      "orthogonal_initialization_and_layer_scaling      True\n",
      "policy_noclip                                    True\n",
      "reward_clipping                                 False\n",
      "reward_standardization                          False\n",
      "value_clipping                                  False\n",
      "score                                         -96.486\n",
      "Name: 20, dtype: object\n",
      "\n",
      "\n",
      "-[4]-\n",
      "clipping_gradient                                False\n",
      "observation_normalization                         True\n",
      "orthogonal_initialization_and_layer_scaling      False\n",
      "policy_noclip                                     True\n",
      "reward_clipping                                   True\n",
      "reward_standardization                            True\n",
      "value_clipping                                    True\n",
      "score                                         -85.8142\n",
      "Name: 18, dtype: object\n",
      "\n",
      "\n",
      "-[5]-\n",
      "clipping_gradient                                 True\n",
      "observation_normalization                         True\n",
      "orthogonal_initialization_and_layer_scaling      False\n",
      "policy_noclip                                     True\n",
      "reward_clipping                                   True\n",
      "reward_standardization                            True\n",
      "value_clipping                                   False\n",
      "score                                         -83.2415\n",
      "Name: 33, dtype: object\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# bottom 5 results\n",
    "for i in range(5):\n",
    "    print(\"-[{}]-\".format(i+1))\n",
    "    print(df3.sort_values('score', ascending=True).iloc[i])\n",
    "    print(\"\\n\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import yaml"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open(\"config/base.yaml\") as file:\n",
    "    base_config = yaml.full_load(file)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# top 2\n",
    "# bottom 2\n",
    "# medium 2\n",
    "# PPO_M\n",
    "# PPO_Plus\n",
    "# PPO_Plus_NOCLIP"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def generate_config(base, expermiment_recipe, config_name):\n",
    "    copied_base = base.copy()\n",
    "    copied_base['exp_name'] = config_name\n",
    "    copied_base['experiment'].update(expermiment_recipe)\n",
    "    \n",
    "    file_path = \"config/{}.yaml\".format(config_name)\n",
    "    with open(file_path, 'w') as file:\n",
    "        yaml.dump(copied_base, file, default_flow_style=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "top1 = {k: bool(v) for k, v in df2.iloc[0].to_dict().items() if k in options}\n",
    "top2 = {k: bool(v) for k, v in df2.iloc[1].to_dict().items() if k in options}\n",
    "bot1 = {k: bool(v) for k, v in df2.iloc[-1].to_dict().items() if k in options}\n",
    "bot2 = {k: bool(v) for k, v in df2.iloc[-2].to_dict().items() if k in options}\n",
    "mid1 = {k: bool(v) for k, v in df2.iloc[int(len(df2) / 3)].to_dict().items() if k in options}\n",
    "mid2 = {k: bool(v) for k, v in df2.iloc[int(len(df2) / 3 * 2)].to_dict().items() if k in options}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "top1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# vanilla PPO\n",
    "PPO_M = {'clipping_gradient': False,\n",
    "         'observation_normalization': False,\n",
    "         'orthogonal_initialization_and_layer_scaling': False,\n",
    "         'reward_standardization': False,\n",
    "         'reward_clipping': False,\n",
    "         'value_clipping': False,\n",
    "         'policy_noclip': False}\n",
    "\n",
    "# PPO with all code-level optimization but policy_noclip\n",
    "PPO_Plus = {'clipping_gradient': True,\n",
    "         'observation_normalization': True,\n",
    "         'orthogonal_initialization_and_layer_scaling': True,\n",
    "         'reward_standardization': True,\n",
    "         'reward_clipping': True,\n",
    "         'value_clipping': True,\n",
    "         'policy_noclip': False}\n",
    "\n",
    "# PPO with all code-level optimization AND policy_noclip\n",
    "PPO_Plus_NOCLIP = {'clipping_gradient': True,\n",
    "         'observation_normalization': True,\n",
    "         'orthogonal_initialization_and_layer_scaling': True,\n",
    "         'reward_standardization': True,\n",
    "         'reward_clipping': True,\n",
    "         'value_clipping': True,\n",
    "         'policy_noclip': True}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "generate_config(base_config, top1, \"top1\")\n",
    "generate_config(base_config, top2, \"top2\")\n",
    "generate_config(base_config, bot1, \"bot1\")\n",
    "generate_config(base_config, bot2, \"bot2\")\n",
    "generate_config(base_config, mid1, \"mid1\")\n",
    "generate_config(base_config, mid2, \"mid2\")\n",
    "generate_config(base_config, PPO_M, \"PPO_M\")\n",
    "generate_config(base_config, PPO_Plus, \"PPO_Plus\")\n",
    "generate_config(base_config, PPO_Plus_NOCLIP, \"PPO_Plus_NOCLIP\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# run all!!!!! fuck yeah!\n",
    "import subprocess"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "recipes = ['top1', 'top2', 'bot1', 'bot2', 'mid1', 'mid2', 'PPO_M', 'PPO_Plus', 'PPO_Plus_NOCLIP']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# for recipe in recipes:\n",
    "#     subprocess.call(\"python main.py --config {}.yaml\".format(recipe), shell=True)"
   ]
  },
  {
   "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": "3.6.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}