ppo_kernel.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "f954cacd",
   "metadata": {},
   "outputs": [],
   "source": [
    "import argparse\n",
    "import os\n",
    "import random\n",
    "import time\n",
    "from distutils.util import strtobool\n",
    "\n",
    "import gym\n",
    "import numpy as np\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.optim as optim\n",
    "from torch.distributions.normal import Normal\n",
    "from torch.utils.tensorboard import SummaryWriter"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "2f799d48",
   "metadata": {},
   "outputs": [],
   "source": [
    "PERIOD = 7\n",
    "\n",
    "# Env\n",
    "import gym, json\n",
    "from gym import spaces\n",
    "from epipolicy.core.epidemic import construct_epidemic\n",
    "from epipolicy.obj.act import construct_act\n",
    "\n",
    "class EpiEnv(gym.Env):\n",
    "    \"\"\"Custom Environment that follows gym interface\"\"\"\n",
    "    metadata = {'render.modes': ['human']}\n",
    "\n",
    "    def __init__(self, session):\n",
    "        super(EpiEnv, self).__init__()\n",
    "        self.epi = construct_epidemic(session)\n",
    "        total_population = np.sum(self.epi.static.default_state.obs.current_comp)\n",
    "        obs_count = self.epi.static.compartment_count * self.epi.static.locale_count * self.epi.static.group_count\n",
    "        action_count = 0\n",
    "        action_param_count =  0\n",
    "        for itv in self.epi.static.interventions:\n",
    "            if not itv.is_cost:\n",
    "                action_count += 1\n",
    "                action_param_count += len(itv.cp_list)\n",
    "        self.act_domain = np.zeros((action_param_count, 2), dtype=np.float64)\n",
    "        index = 0\n",
    "        for itv in self.epi.static.interventions:\n",
    "            if not itv.is_cost:\n",
    "                for cp in itv.cp_list:\n",
    "                    self.act_domain[index, 0] = cp.min_value\n",
    "                    self.act_domain[index, 1] = cp.max_value\n",
    "                    index += 1\n",
    "        # Define action and observation space\n",
    "        # They must be gym.spaces objects\n",
    "        # Example when using discrete actions:\n",
    "        self.action_space = spaces.Box(low=0, high=1, shape=(action_count,), dtype=np.float64)\n",
    "        # Example for using image as input:\n",
    "        self.observation_space = spaces.Box(low=0, high=total_population, shape=(obs_count,), dtype=np.float64)\n",
    "\n",
    "    def step(self, action):\n",
    "        expanded_action = np.zeros(len(self.act_domain), dtype=np.float64)\n",
    "        index = 0\n",
    "        for i in range(len(self.act_domain)):\n",
    "            if self.act_domain[i, 0] == self.act_domain[i, 1]:\n",
    "                expanded_action[i] = self.act_domain[i, 0]\n",
    "            else:\n",
    "                expanded_action[i] = action[index]\n",
    "                index += 1\n",
    "\n",
    "        epi_action = []\n",
    "        index = 0\n",
    "        for itv_id, itv in enumerate(self.epi.static.interventions):\n",
    "            if not itv.is_cost:\n",
    "                epi_action.append(construct_act(itv_id, expanded_action[index:index+len(itv.cp_list)]))\n",
    "                index += len(itv.cp_list)\n",
    "\n",
    "        total_r = 0\n",
    "        for i in range(PERIOD):\n",
    "            state, r, done = self.epi.step(epi_action)\n",
    "            total_r += r\n",
    "            if done:\n",
    "                break\n",
    "        return state.obs.current_comp.flatten(), total_r, done, dict()\n",
    "\n",
    "    def reset(self):\n",
    "        state = self.epi.reset()\n",
    "        return state.obs.current_comp.flatten()  # reward, done, info can't be included\n",
    "    def render(self, mode='human'):\n",
    "        pass\n",
    "    def close(self):\n",
    "        pass"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "9e9d1fcc",
   "metadata": {},
   "outputs": [],
   "source": [
    "def parse_args(main_args = None):\n",
    "    # fmt: off\n",
    "    parser = argparse.ArgumentParser()\n",
    "    parser.add_argument(\"--exp-name\", type=str, default=\"PPO\",\n",
    "        help=\"the name of this experiment\")\n",
    "    parser.add_argument(\"--gym-id\", type=str, default=\"HalfCheetahBulletEnv-v0\",\n",
    "        help=\"the id of the gym environment\")\n",
    "    parser.add_argument(\"--learning-rate\", type=float, default=3e-4,\n",
    "        help=\"the learning rate of the optimizer\")\n",
    "    parser.add_argument(\"--seed\", type=int, default=1,\n",
    "        help=\"seed of the experiment\")\n",
    "    parser.add_argument(\"--total-timesteps\", type=int, default=700000,\n",
    "        help=\"total timesteps of the experiments\")\n",
    "    parser.add_argument(\"--torch-deterministic\", type=lambda x: bool(strtobool(x)), default=True, nargs=\"?\", const=True,\n",
    "        help=\"if toggled, `torch.backends.cudnn.deterministic=False`\")\n",
    "    parser.add_argument(\"--cuda\", type=lambda x: bool(strtobool(x)), default=True, nargs=\"?\", const=True,\n",
    "        help=\"if toggled, cuda will be enabled by default\")\n",
    "    parser.add_argument(\"--track\", type=lambda x: bool(strtobool(x)), default=False, nargs=\"?\", const=True,\n",
    "        help=\"if toggled, this experiment will be tracked with Weights and Biases\")\n",
    "    parser.add_argument(\"--wandb-project-name\", type=str, default=\"ppo-implementation-details\",\n",
    "        help=\"the wandb's project name\")\n",
    "    parser.add_argument(\"--wandb-entity\", type=str, default=None,\n",
    "        help=\"the entity (team) of wandb's project\")\n",
    "    parser.add_argument(\"--capture-video\", type=lambda x: bool(strtobool(x)), default=False, nargs=\"?\", const=True,\n",
    "        help=\"weather to capture videos of the agent performances (check out `videos` folder)\")\n",
    "\n",
    "    parser.add_argument(\"--policy_plot_interval\", type=int, default=1,\n",
    "        help=\"seed of the experiment\")\n",
    "\n",
    "    # Algorithm specific arguments\n",
    "    parser.add_argument(\"--num-envs\", type=int, default=1,\n",
    "        help=\"the number of parallel game environments\")\n",
    "    parser.add_argument(\"--num-steps\", type=int, default=2048,\n",
    "        help=\"the number of steps to run in each environment per policy rollout\")\n",
    "    parser.add_argument(\"--anneal-lr\", type=lambda x: bool(strtobool(x)), default=True, nargs=\"?\", const=True,\n",
    "        help=\"Toggle learning rate annealing for policy and value networks\")\n",
    "    parser.add_argument(\"--gae\", type=lambda x: bool(strtobool(x)), default=True, nargs=\"?\", const=True,\n",
    "        help=\"Use GAE for advantage computation\")\n",
    "    parser.add_argument(\"--gamma\", type=float, default=0.99,\n",
    "        help=\"the discount factor gamma\")\n",
    "    parser.add_argument(\"--gae-lambda\", type=float, default=0.95,\n",
    "        help=\"the lambda for the general advantage estimation\")\n",
    "    parser.add_argument(\"--num-minibatches\", type=int, default=32,\n",
    "        help=\"the number of mini-batches\")\n",
    "    parser.add_argument(\"--update-epochs\", type=int, default=10,\n",
    "        help=\"the K epochs to update the policy\")\n",
    "    parser.add_argument(\"--norm-adv\", type=lambda x: bool(strtobool(x)), default=True, nargs=\"?\", const=True,\n",
    "        help=\"Toggles advantages normalization\")\n",
    "    parser.add_argument(\"--clip-coef\", type=float, default=0.2,\n",
    "        help=\"the surrogate clipping coefficient\")\n",
    "    parser.add_argument(\"--clip-vloss\", type=lambda x: bool(strtobool(x)), default=True, nargs=\"?\", const=True,\n",
    "        help=\"Toggles whether or not to use a clipped loss for the value function, as per the paper.\")\n",
    "    parser.add_argument(\"--ent-coef\", type=float, default=0.0,\n",
    "        help=\"coefficient of the entropy\")\n",
    "    parser.add_argument(\"--vf-coef\", type=float, default=0.5,\n",
    "        help=\"coefficient of the value function\")\n",
    "    parser.add_argument(\"--max-grad-norm\", type=float, default=0.5,\n",
    "        help=\"the maximum norm for the gradient clipping\")\n",
    "    parser.add_argument(\"--target-kl\", type=float, default=None,\n",
    "        help=\"the target KL divergence threshold\")\n",
    "    if main_args is not None:\n",
    "        args = parser.parse_args(main_args.split())\n",
    "    else:\n",
    "        args = parser.parse_args()\n",
    "    args.num_steps //= PERIOD\n",
    "    args.total_timesteps //= PERIOD\n",
    "    args.batch_size = int(args.num_envs * args.num_steps)\n",
    "    args.minibatch_size = int(args.batch_size // args.num_minibatches)\n",
    "    # fmt: on\n",
    "    return args"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "91aa8265",
   "metadata": {},
   "outputs": [],
   "source": [
    "epi_ids = [\"SIR_A\", \"SIR_B\", \"SIRV_A\", \"SIRV_B\", \"COVID_A\", \"COVID_B\", \"COVID_C\"]\n",
    "\n",
    "def make_env(gym_id, seed, idx, capture_video, run_name):\n",
    "    def thunk():\n",
    "        if gym_id in epi_ids:\n",
    "            fp = open('jsons/{}.json'.format(gym_id), 'r')\n",
    "            session = json.load(fp)\n",
    "            env = EpiEnv(session)\n",
    "        else:\n",
    "            env = gym.make(gym_id)\n",
    "        env = gym.wrappers.RecordEpisodeStatistics(env)\n",
    "        if capture_video:\n",
    "            if idx == 0:\n",
    "                env = gym.wrappers.RecordVideo(env, f\"videos/{run_name}\")\n",
    "        env = gym.wrappers.ClipAction(env)\n",
    "        env = gym.wrappers.NormalizeObservation(env)\n",
    "        env = gym.wrappers.TransformObservation(env, lambda obs: np.clip(obs, -10, 10))\n",
    "        env = gym.wrappers.NormalizeReward(env)\n",
    "        env = gym.wrappers.TransformReward(env, lambda reward: np.clip(reward, -10, 10))\n",
    "        # Our env is deterministic\n",
    "        # env.seed(seed)\n",
    "        env.action_space.seed(seed)\n",
    "        env.observation_space.seed(seed)\n",
    "        return env\n",
    "\n",
    "    return thunk\n",
    "\n",
    "def make_primal_env(gym_id):\n",
    "    def thunk():\n",
    "        if gym_id in epi_ids:\n",
    "            fp = open('jsons/{}.json'.format(gym_id), 'r')\n",
    "            session = json.load(fp)\n",
    "            env = EpiEnv(session)\n",
    "        else:\n",
    "            env = gym.make(gym_id)\n",
    "        return env\n",
    "    return thunk"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "334de62b",
   "metadata": {},
   "outputs": [],
   "source": [
    "def layer_init(layer, std=np.sqrt(2), bias_const=0.0):\n",
    "    torch.nn.init.orthogonal_(layer.weight, std)\n",
    "    torch.nn.init.constant_(layer.bias, bias_const)\n",
    "    return layer\n",
    "\n",
    "\n",
    "class Agent(nn.Module):\n",
    "    def __init__(self, envs):\n",
    "        super(Agent, self).__init__()\n",
    "        self.critic = nn.Sequential(\n",
    "            layer_init(nn.Linear(np.array(envs.single_observation_space.shape).prod(), 64)),\n",
    "            nn.Tanh(),\n",
    "            layer_init(nn.Linear(64, 64)),\n",
    "            nn.Tanh(),\n",
    "            layer_init(nn.Linear(64, 1), std=1.0),\n",
    "        )\n",
    "        self.actor_mean = nn.Sequential(\n",
    "            layer_init(nn.Linear(np.array(envs.single_observation_space.shape).prod(), 64)),\n",
    "            nn.Tanh(),\n",
    "            layer_init(nn.Linear(64, 64)),\n",
    "            nn.Tanh(),\n",
    "            layer_init(nn.Linear(64, np.prod(envs.single_action_space.shape)), std=0.01),\n",
    "        )\n",
    "        self.actor_logstd = nn.Parameter(torch.zeros(1, np.prod(envs.single_action_space.shape)))\n",
    "\n",
    "    def get_value(self, x):\n",
    "        return self.critic(x)\n",
    "\n",
    "    def get_action_and_value(self, x, action=None):\n",
    "        action_mean = self.actor_mean(x)\n",
    "        action_logstd = self.actor_logstd.expand_as(action_mean)\n",
    "        action_std = torch.exp(action_logstd)\n",
    "        probs = Normal(action_mean, action_std)\n",
    "        if action is None:\n",
    "            action = probs.sample()\n",
    "        return action, probs.log_prob(action).sum(1), probs.entropy().sum(1), self.critic(x)\n",
    "\n",
    "    def get_action_mean(self, x):\n",
    "        return self.actor_mean(x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "9b3930b8",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running SIR_A__PPO__1__1659520286\n",
      "global_step=53, episodic_return=-133795024.0\n",
      "global_step=106, episodic_return=-249479552.0\n",
      "global_step=159, episodic_return=-330727008.0\n",
      "global_step=212, episodic_return=-109690456.0\n",
      "global_step=265, episodic_return=-127306376.0\n",
      "SPS: 35\n",
      "At global step 292, total_rewards=-2597174215.2497478\n",
      "global_step=345, episodic_return=-146012240.0\n",
      "global_step=398, episodic_return=-339600928.0\n",
      "global_step=451, episodic_return=-665188096.0\n",
      "global_step=504, episodic_return=-726868288.0\n",
      "global_step=557, episodic_return=-147837232.0\n",
      "SPS: 30\n",
      "At global step 584, total_rewards=-2738516316.4718237\n",
      "global_step=637, episodic_return=-307462176.0\n",
      "global_step=690, episodic_return=-99795904.0\n",
      "global_step=743, episodic_return=-158665904.0\n",
      "global_step=796, episodic_return=-114541552.0\n",
      "global_step=849, episodic_return=-901531520.0\n",
      "SPS: 28\n",
      "At global step 876, total_rewards=-2855688356.47755\n",
      "global_step=929, episodic_return=-445322368.0\n",
      "global_step=982, episodic_return=-83600248.0\n",
      "global_step=1035, episodic_return=-204369280.0\n",
      "global_step=1088, episodic_return=-86524952.0\n",
      "global_step=1141, episodic_return=-98369608.0\n",
      "SPS: 28\n",
      "At global step 1168, total_rewards=-74874336.45568526\n",
      "global_step=1221, episodic_return=-178432304.0\n",
      "global_step=1274, episodic_return=-87396688.0\n",
      "global_step=1327, episodic_return=-171928688.0\n",
      "global_step=1380, episodic_return=-104132392.0\n",
      "global_step=1433, episodic_return=-83349312.0\n",
      "SPS: 27\n",
      "At global step 1460, total_rewards=-77065033.63184081\n",
      "global_step=1513, episodic_return=-84811336.0\n",
      "global_step=1566, episodic_return=-88342184.0\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "Input \u001b[0;32mIn [12]\u001b[0m, in \u001b[0;36m<cell line: 49>\u001b[0;34m()\u001b[0m\n\u001b[1;32m     66\u001b[0m logprobs[step] \u001b[38;5;241m=\u001b[39m logprob\n\u001b[1;32m     68\u001b[0m \u001b[38;5;66;03m# TRY NOT TO MODIFY: execute the game and log data.\u001b[39;00m\n\u001b[0;32m---> 69\u001b[0m next_obs, reward, done, info \u001b[38;5;241m=\u001b[39m \u001b[43menvs\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mstep\u001b[49m\u001b[43m(\u001b[49m\u001b[43maction\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mcpu\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mnumpy\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m     70\u001b[0m rewards[step] \u001b[38;5;241m=\u001b[39m torch\u001b[38;5;241m.\u001b[39mtensor(reward)\u001b[38;5;241m.\u001b[39mto(device)\u001b[38;5;241m.\u001b[39mview(\u001b[38;5;241m-\u001b[39m\u001b[38;5;241m1\u001b[39m)\n\u001b[1;32m     71\u001b[0m next_obs, next_done \u001b[38;5;241m=\u001b[39m torch\u001b[38;5;241m.\u001b[39mTensor(next_obs)\u001b[38;5;241m.\u001b[39mto(device), torch\u001b[38;5;241m.\u001b[39mTensor(done)\u001b[38;5;241m.\u001b[39mto(device)\n",
      "File \u001b[0;32m~/.local/lib/python3.8/site-packages/gym/vector/vector_env.py:94\u001b[0m, in \u001b[0;36mVectorEnv.step\u001b[0;34m(self, actions)\u001b[0m\n\u001b[1;32m     71\u001b[0m \u001b[38;5;124mr\u001b[39m\u001b[38;5;124;03m\"\"\"Take an action for each sub-environments.\u001b[39;00m\n\u001b[1;32m     72\u001b[0m \n\u001b[1;32m     73\u001b[0m \u001b[38;5;124;03mParameters\u001b[39;00m\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m     90\u001b[0m \u001b[38;5;124;03m    A list of auxiliary diagnostic information dicts from sub-environments.\u001b[39;00m\n\u001b[1;32m     91\u001b[0m \u001b[38;5;124;03m\"\"\"\u001b[39;00m\n\u001b[1;32m     93\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mstep_async(actions)\n\u001b[0;32m---> 94\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mstep_wait\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[0;32m~/.local/lib/python3.8/site-packages/gym/vector/async_vector_env.py:288\u001b[0m, in \u001b[0;36mAsyncVectorEnv.step_wait\u001b[0;34m(self, timeout)\u001b[0m\n\u001b[1;32m    282\u001b[0m     \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_state \u001b[38;5;241m=\u001b[39m AsyncState\u001b[38;5;241m.\u001b[39mDEFAULT\n\u001b[1;32m    283\u001b[0m     \u001b[38;5;28;01mraise\u001b[39;00m mp\u001b[38;5;241m.\u001b[39mTimeoutError(\n\u001b[1;32m    284\u001b[0m         \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mThe call to `step_wait` has timed out after \u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m    285\u001b[0m         \u001b[38;5;124m\"\u001b[39m\u001b[38;5;132;01m{0}\u001b[39;00m\u001b[38;5;124m second\u001b[39m\u001b[38;5;132;01m{1}\u001b[39;00m\u001b[38;5;124m.\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;241m.\u001b[39mformat(timeout, \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124ms\u001b[39m\u001b[38;5;124m\"\u001b[39m \u001b[38;5;28;01mif\u001b[39;00m timeout \u001b[38;5;241m>\u001b[39m \u001b[38;5;241m1\u001b[39m \u001b[38;5;28;01melse\u001b[39;00m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[1;32m    286\u001b[0m     )\n\u001b[0;32m--> 288\u001b[0m results, successes \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mzip\u001b[39m(\u001b[38;5;241m*\u001b[39m[pipe\u001b[38;5;241m.\u001b[39mrecv() \u001b[38;5;28;01mfor\u001b[39;00m pipe \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mparent_pipes])\n\u001b[1;32m    289\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_raise_if_errors(successes)\n\u001b[1;32m    290\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_state \u001b[38;5;241m=\u001b[39m AsyncState\u001b[38;5;241m.\u001b[39mDEFAULT\n",
      "File \u001b[0;32m~/.local/lib/python3.8/site-packages/gym/vector/async_vector_env.py:288\u001b[0m, in \u001b[0;36m<listcomp>\u001b[0;34m(.0)\u001b[0m\n\u001b[1;32m    282\u001b[0m     \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_state \u001b[38;5;241m=\u001b[39m AsyncState\u001b[38;5;241m.\u001b[39mDEFAULT\n\u001b[1;32m    283\u001b[0m     \u001b[38;5;28;01mraise\u001b[39;00m mp\u001b[38;5;241m.\u001b[39mTimeoutError(\n\u001b[1;32m    284\u001b[0m         \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mThe call to `step_wait` has timed out after \u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m    285\u001b[0m         \u001b[38;5;124m\"\u001b[39m\u001b[38;5;132;01m{0}\u001b[39;00m\u001b[38;5;124m second\u001b[39m\u001b[38;5;132;01m{1}\u001b[39;00m\u001b[38;5;124m.\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;241m.\u001b[39mformat(timeout, \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124ms\u001b[39m\u001b[38;5;124m\"\u001b[39m \u001b[38;5;28;01mif\u001b[39;00m timeout \u001b[38;5;241m>\u001b[39m \u001b[38;5;241m1\u001b[39m \u001b[38;5;28;01melse\u001b[39;00m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[1;32m    286\u001b[0m     )\n\u001b[0;32m--> 288\u001b[0m results, successes \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mzip\u001b[39m(\u001b[38;5;241m*\u001b[39m[\u001b[43mpipe\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mrecv\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m \u001b[38;5;28;01mfor\u001b[39;00m pipe \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mparent_pipes])\n\u001b[1;32m    289\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_raise_if_errors(successes)\n\u001b[1;32m    290\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_state \u001b[38;5;241m=\u001b[39m AsyncState\u001b[38;5;241m.\u001b[39mDEFAULT\n",
      "File \u001b[0;32m/usr/lib/python3.8/multiprocessing/connection.py:250\u001b[0m, in \u001b[0;36m_ConnectionBase.recv\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    248\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_check_closed()\n\u001b[1;32m    249\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_check_readable()\n\u001b[0;32m--> 250\u001b[0m buf \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_recv_bytes\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    251\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m _ForkingPickler\u001b[38;5;241m.\u001b[39mloads(buf\u001b[38;5;241m.\u001b[39mgetbuffer())\n",
      "File \u001b[0;32m/usr/lib/python3.8/multiprocessing/connection.py:414\u001b[0m, in \u001b[0;36mConnection._recv_bytes\u001b[0;34m(self, maxsize)\u001b[0m\n\u001b[1;32m    413\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21m_recv_bytes\u001b[39m(\u001b[38;5;28mself\u001b[39m, maxsize\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mNone\u001b[39;00m):\n\u001b[0;32m--> 414\u001b[0m     buf \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_recv\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m4\u001b[39;49m\u001b[43m)\u001b[49m\n\u001b[1;32m    415\u001b[0m     size, \u001b[38;5;241m=\u001b[39m struct\u001b[38;5;241m.\u001b[39munpack(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m!i\u001b[39m\u001b[38;5;124m\"\u001b[39m, buf\u001b[38;5;241m.\u001b[39mgetvalue())\n\u001b[1;32m    416\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m size \u001b[38;5;241m==\u001b[39m \u001b[38;5;241m-\u001b[39m\u001b[38;5;241m1\u001b[39m:\n",
      "File \u001b[0;32m/usr/lib/python3.8/multiprocessing/connection.py:379\u001b[0m, in \u001b[0;36mConnection._recv\u001b[0;34m(self, size, read)\u001b[0m\n\u001b[1;32m    377\u001b[0m remaining \u001b[38;5;241m=\u001b[39m size\n\u001b[1;32m    378\u001b[0m \u001b[38;5;28;01mwhile\u001b[39;00m remaining \u001b[38;5;241m>\u001b[39m \u001b[38;5;241m0\u001b[39m:\n\u001b[0;32m--> 379\u001b[0m     chunk \u001b[38;5;241m=\u001b[39m \u001b[43mread\u001b[49m\u001b[43m(\u001b[49m\u001b[43mhandle\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mremaining\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    380\u001b[0m     n \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mlen\u001b[39m(chunk)\n\u001b[1;32m    381\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m n \u001b[38;5;241m==\u001b[39m \u001b[38;5;241m0\u001b[39m:\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "args = parse_args(\"--gym-id SIR_A --seed 1\")\n",
    "run_name = f\"{args.gym_id}__{args.exp_name}__{args.seed}__{int(time.time())}\"\n",
    "print(\"Running\", run_name)\n",
    "writer = SummaryWriter(f\"runs/{run_name}\")\n",
    "writer.add_text(\n",
    "    \"hyperparameters\",\n",
    "    \"|param|value|\\n|-|-|\\n%s\" % (\"\\n\".join([f\"|{key}|{value}|\" for key, value in vars(args).items()])),\n",
    ")\n",
    "\n",
    "# TRY NOT TO MODIFY: seeding\n",
    "random.seed(args.seed)\n",
    "np.random.seed(args.seed)\n",
    "torch.manual_seed(args.seed)\n",
    "torch.backends.cudnn.deterministic = args.torch_deterministic\n",
    "\n",
    "device = torch.device(\"cuda\" if torch.cuda.is_available() and args.cuda else \"cpu\")\n",
    "\n",
    "# env setup\n",
    "# envs = gym.vector.SyncVectorEnv(\n",
    "#     [make_env(args.gym_id, args.seed + i, i, args.capture_video, run_name) for i in range(args.num_envs)]\n",
    "# )\n",
    "envs = gym.vector.AsyncVectorEnv(\n",
    "    [make_env(args.gym_id, args.seed + i, i, args.capture_video, run_name) for i in range(args.num_envs)]\n",
    ")\n",
    "test_env = make_primal_env(args.gym_id)()\n",
    "# test_env = make_env(args.gym_id, args.seed, 0, args.capture_video, run_name)()\n",
    "assert isinstance(envs.single_action_space, gym.spaces.Box), \"only continuous action space is supported\"\n",
    "\n",
    "agent = Agent(envs).to(device)\n",
    "optimizer = optim.Adam(agent.parameters(), lr=args.learning_rate, eps=1e-5)\n",
    "\n",
    "# ALGO Logic: Storage setup\n",
    "obs = torch.zeros((args.num_steps, args.num_envs) + envs.single_observation_space.shape).to(device)\n",
    "actions = torch.zeros((args.num_steps, args.num_envs) + envs.single_action_space.shape).to(device)\n",
    "logprobs = torch.zeros((args.num_steps, args.num_envs)).to(device)\n",
    "rewards = torch.zeros((args.num_steps, args.num_envs)).to(device)\n",
    "dones = torch.zeros((args.num_steps, args.num_envs)).to(device)\n",
    "values = torch.zeros((args.num_steps, args.num_envs)).to(device)\n",
    "\n",
    "# TRY NOT TO MODIFY: start the game\n",
    "global_step = 0\n",
    "start_time = time.time()\n",
    "next_obs = torch.Tensor(envs.reset()).to(device)\n",
    "next_done = torch.zeros(args.num_envs).to(device)\n",
    "num_updates = args.total_timesteps // args.batch_size\n",
    "\n",
    "csv_file = open('runs/{}/records.csv'.format(run_name), 'w')\n",
    "\n",
    "for update in range(1, num_updates + 1):\n",
    "    # Annealing the rate if instructed to do so.\n",
    "    if args.anneal_lr:\n",
    "        frac = 1.0 - (update - 1.0) / num_updates\n",
    "        lrnow = frac * args.learning_rate\n",
    "        optimizer.param_groups[0][\"lr\"] = lrnow\n",
    "\n",
    "    for step in range(0, args.num_steps):\n",
    "        global_step += 1 * args.num_envs\n",
    "        obs[step] = next_obs\n",
    "        dones[step] = next_done\n",
    "\n",
    "        # ALGO LOGIC: action logic\n",
    "        with torch.no_grad():\n",
    "            action, logprob, _, value = agent.get_action_and_value(next_obs)\n",
    "            values[step] = value.flatten()\n",
    "        actions[step] = action\n",
    "        logprobs[step] = logprob\n",
    "\n",
    "        # TRY NOT TO MODIFY: execute the game and log data.\n",
    "        next_obs, reward, done, info = envs.step(action.cpu().numpy())\n",
    "        rewards[step] = torch.tensor(reward).to(device).view(-1)\n",
    "        next_obs, next_done = torch.Tensor(next_obs).to(device), torch.Tensor(done).to(device)\n",
    "\n",
    "        for item in info:\n",
    "            if \"episode\" in item.keys():\n",
    "                print(f\"global_step={global_step}, episodic_return={item['episode']['r']}\")\n",
    "                writer.add_scalar(\"charts/episodic_return\", item[\"episode\"][\"r\"], global_step)\n",
    "                writer.add_scalar(\"charts/episodic_length\", item[\"episode\"][\"l\"], global_step)\n",
    "                break\n",
    "\n",
    "    # bootstrap value if not done\n",
    "    with torch.no_grad():\n",
    "        next_value = agent.get_value(next_obs).reshape(1, -1)\n",
    "        if args.gae:\n",
    "            advantages = torch.zeros_like(rewards).to(device)\n",
    "            lastgaelam = 0\n",
    "            for t in reversed(range(args.num_steps)):\n",
    "                if t == args.num_steps - 1:\n",
    "                    nextnonterminal = 1.0 - next_done\n",
    "                    nextvalues = next_value\n",
    "                else:\n",
    "                    nextnonterminal = 1.0 - dones[t + 1]\n",
    "                    nextvalues = values[t + 1]\n",
    "                delta = rewards[t] + args.gamma * nextvalues * nextnonterminal - values[t]\n",
    "                advantages[t] = lastgaelam = delta + args.gamma * args.gae_lambda * nextnonterminal * lastgaelam\n",
    "            returns = advantages + values\n",
    "        else:\n",
    "            returns = torch.zeros_like(rewards).to(device)\n",
    "            for t in reversed(range(args.num_steps)):\n",
    "                if t == args.num_steps - 1:\n",
    "                    nextnonterminal = 1.0 - next_done\n",
    "                    next_return = next_value\n",
    "                else:\n",
    "                    nextnonterminal = 1.0 - dones[t + 1]\n",
    "                    next_return = returns[t + 1]\n",
    "                returns[t] = rewards[t] + args.gamma * nextnonterminal * next_return\n",
    "            advantages = returns - values\n",
    "\n",
    "    # flatten the batch\n",
    "    b_obs = obs.reshape((-1,) + envs.single_observation_space.shape)\n",
    "    b_logprobs = logprobs.reshape(-1)\n",
    "    b_actions = actions.reshape((-1,) + envs.single_action_space.shape)\n",
    "    b_advantages = advantages.reshape(-1)\n",
    "    b_returns = returns.reshape(-1)\n",
    "    b_values = values.reshape(-1)\n",
    "\n",
    "    # Optimizing the policy and value network\n",
    "    b_inds = np.arange(args.batch_size)\n",
    "    clipfracs = []\n",
    "    for epoch in range(args.update_epochs):\n",
    "        np.random.shuffle(b_inds)\n",
    "        for start in range(0, args.batch_size, args.minibatch_size):\n",
    "            end = start + args.minibatch_size\n",
    "            mb_inds = b_inds[start:end]\n",
    "\n",
    "            _, newlogprob, entropy, newvalue = agent.get_action_and_value(b_obs[mb_inds], b_actions[mb_inds])\n",
    "            logratio = newlogprob - b_logprobs[mb_inds]\n",
    "            ratio = logratio.exp()\n",
    "\n",
    "            with torch.no_grad():\n",
    "                # calculate approx_kl http://joschu.net/blog/kl-approx.html\n",
    "                old_approx_kl = (-logratio).mean()\n",
    "                approx_kl = ((ratio - 1) - logratio).mean()\n",
    "                clipfracs += [((ratio - 1.0).abs() > args.clip_coef).float().mean().item()]\n",
    "\n",
    "            mb_advantages = b_advantages[mb_inds]\n",
    "            if args.norm_adv:\n",
    "                mb_advantages = (mb_advantages - mb_advantages.mean()) / (mb_advantages.std() + 1e-8)\n",
    "\n",
    "            # Policy loss\n",
    "            pg_loss1 = -mb_advantages * ratio\n",
    "            pg_loss2 = -mb_advantages * torch.clamp(ratio, 1 - args.clip_coef, 1 + args.clip_coef)\n",
    "            pg_loss = torch.max(pg_loss1, pg_loss2).mean()\n",
    "\n",
    "            # Value loss\n",
    "            newvalue = newvalue.view(-1)\n",
    "            if args.clip_vloss:\n",
    "                v_loss_unclipped = (newvalue - b_returns[mb_inds]) ** 2\n",
    "                v_clipped = b_values[mb_inds] + torch.clamp(\n",
    "                    newvalue - b_values[mb_inds],\n",
    "                    -args.clip_coef,\n",
    "                    args.clip_coef,\n",
    "                )\n",
    "                v_loss_clipped = (v_clipped - b_returns[mb_inds]) ** 2\n",
    "                v_loss_max = torch.max(v_loss_unclipped, v_loss_clipped)\n",
    "                v_loss = 0.5 * v_loss_max.mean()\n",
    "            else:\n",
    "                v_loss = 0.5 * ((newvalue - b_returns[mb_inds]) ** 2).mean()\n",
    "\n",
    "            entropy_loss = entropy.mean()\n",
    "            loss = pg_loss - args.ent_coef * entropy_loss + v_loss * args.vf_coef\n",
    "\n",
    "            optimizer.zero_grad()\n",
    "            loss.backward()\n",
    "            nn.utils.clip_grad_norm_(agent.parameters(), args.max_grad_norm)\n",
    "            optimizer.step()\n",
    "\n",
    "        if args.target_kl is not None:\n",
    "            if approx_kl > args.target_kl:\n",
    "                break\n",
    "\n",
    "    y_pred, y_true = b_values.cpu().numpy(), b_returns.cpu().numpy()\n",
    "    var_y = np.var(y_true)\n",
    "    explained_var = np.nan if var_y == 0 else 1 - np.var(y_true - y_pred) / var_y\n",
    "\n",
    "    # TRY NOT TO MODIFY: record rewards for plotting purposes\n",
    "    writer.add_scalar(\"charts/learning_rate\", optimizer.param_groups[0][\"lr\"], global_step)\n",
    "    writer.add_scalar(\"losses/value_loss\", v_loss.item(), global_step)\n",
    "    writer.add_scalar(\"losses/policy_loss\", pg_loss.item(), global_step)\n",
    "    writer.add_scalar(\"losses/entropy\", entropy_loss.item(), global_step)\n",
    "    writer.add_scalar(\"losses/old_approx_kl\", old_approx_kl.item(), global_step)\n",
    "    writer.add_scalar(\"losses/approx_kl\", approx_kl.item(), global_step)\n",
    "    writer.add_scalar(\"losses/clipfrac\", np.mean(clipfracs), global_step)\n",
    "    writer.add_scalar(\"losses/explained_variance\", explained_var, global_step)\n",
    "    print(\"SPS:\", int(global_step / (time.time() - start_time)))\n",
    "    writer.add_scalar(\"charts/SPS\", int(global_step / (time.time() - start_time)), global_step)\n",
    "\n",
    "    # PLOT POLICY\n",
    "    if args.gym_id in epi_ids and (update - 1) % args.policy_plot_interval == 0:\n",
    "        test_obs = torch.Tensor(test_env.reset())\n",
    "        env_obs = torch.Tensor(envs.reset()).to(device)\n",
    "        timestep = 0\n",
    "        total_r = 0\n",
    "        done = False\n",
    "        itv_line = []\n",
    "        while not done:\n",
    "            with torch.no_grad():\n",
    "                action_mean = agent.get_action_mean(env_obs)\n",
    "                test_action_mean = torch.mean(action_mean, 0)\n",
    "                test_action_mean = torch.clamp(test_action_mean, 0, 1)\n",
    "             \n",
    "            test_obs, r, done, _ = test_env.step(test_action_mean.cpu().numpy())\n",
    "            test_obs = torch.Tensor(test_obs)\n",
    "            itv_index = 0\n",
    "            itv_array = []\n",
    "            for itv in test_env.epi.static.interventions:\n",
    "                if not itv.is_cost:\n",
    "                    v = float(test_action_mean[itv_index])\n",
    "                    writer.add_scalar('charts/policy_{}/{}'.format(global_step, itv.name), v, timestep)\n",
    "                    itv_array.append(v)\n",
    "                    itv_index += 1\n",
    "            itv_line.append(itv_array)\n",
    "            \n",
    "            env_obs, _, _, _ = envs.step(action_mean.cpu().numpy())\n",
    "            env_obs = torch.Tensor(env_obs).to(device)\n",
    "\n",
    "            total_r += r\n",
    "            timestep += PERIOD\n",
    "            \n",
    "        line = '|'.join([str(global_step), str(total_r), str(itv_line)]) + '\\n'\n",
    "        csv_file.write(line)\n",
    "        writer.add_scalar('charts/learning_curve', total_r, global_step)\n",
    "        print(\"At global step {}, total_rewards={}\".format(global_step, total_r))\n",
    "\n",
    "csv_file.close()\n",
    "envs.close()\n",
    "writer.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "19a5a8f7",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a31585c8",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "dca0cce3",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.8.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}