SAC-discrete implementation

cleanrl/sac_atari.py

@@ -0,0 +1,374 @@
+# TODO: Must add header here
+import argparse
+import os
+import random
+import time
+from distutils.util import strtobool
+
+import gym
+import numpy as np
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+import torch.optim as optim
+from torch.distributions.categorical import Categorical
+from stable_baselines3.common.atari_wrappers import (
+    EpisodicLifeEnv,
+    FireResetEnv,
+    MaxAndSkipEnv,
+    NoopResetEnv,
+)
+from stable_baselines3.common.buffers import ReplayBuffer
+from torch.utils.tensorboard import SummaryWriter
+
+
+def parse_args():
+    # fmt: off
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--exp-name", type=str, default=os.path.basename(__file__).rstrip(".py"),
+        help="the name of this experiment")
+    parser.add_argument("--seed", type=int, default=1, 
+        help="seed of the experiment")
+    parser.add_argument("--torch-deterministic", type=lambda x: bool(strtobool(x)), default=True, nargs="?", const=True,
+        help="if toggled, `torch.backends.cudnn.deterministic=False`")
+    parser.add_argument("--cuda", type=lambda x: bool(strtobool(x)), default=True, nargs="?", const=True,
+        help="if toggled, cuda will be enabled by default")
+    parser.add_argument("--track", type=lambda x: bool(strtobool(x)), default=False, nargs="?", const=True,
+        help="if toggled, this experiment will be tracked with Weights and Biases")
+    # TODO: Must change this back to CLeanRL later
+    parser.add_argument("--wandb-project-name", type=str, default="SAC-discrete",
+        help="the wandb's project name")
+    parser.add_argument("--wandb-entity", type=str, default=None,
+        help="the entity (team) of wandb's project")
+    parser.add_argument("--capture-video", type=lambda x: bool(strtobool(x)), default=False, nargs="?", const=True,
+        help="weather to capture videos of the agent performances (check out `videos` folder)")
+
+    # Algorithm specific arguments
+    parser.add_argument("--env-id", type=str, default="Seaquest-v4",
+        help="the id of the environment")
+    parser.add_argument("--total-timesteps", type=int, default=120000,
+        help="total timesteps of the experiments")
+    parser.add_argument("--buffer-size", type=int, default=int(1.5e5),
+        help="the replay memory buffer size") # smaller than in original paper but evaluation is done only for 100k steps anyway
+    parser.add_argument("--gamma", type=float, default=0.99,
+        help="the discount factor gamma")
+    parser.add_argument("--tau", type=float, default=1.0,
+        help="target smoothing coefficient (default: 1)") # Default is 1 to perform replacement update
+    parser.add_argument("--batch-size", type=int, default=64,
+        help="the batch size of sample from the reply memory")
+    parser.add_argument("--learning-starts", type=int, default=2e4,
+        help="timestep to start learning")
+    parser.add_argument("--policy-lr", type=float, default=3e-4,
+        help="the learning rate of the policy network optimizer")
+    parser.add_argument("--q-lr", type=float, default=3e-4,
+        help="the learning rate of the Q network network optimizer")
+    # Update actor and policy each step and 4 minibatches each time
+    parser.add_argument("--update-frequency", type=int, default=1,
+        help="the frequency of learning updates")
+    parser.add_argument("--num-update-steps", type=int, default=4,
+        help="the number of ")
+    parser.add_argument("--target-network-frequency", type=int, default=8000, # Denis Yarats' implementation delays this by 2.
+        help="the frequency of updates for the target nerworks")
+    parser.add_argument("--alpha", type=float, default=0.2,
+        help="Entropy regularization coefficient.")
+    parser.add_argument("--autotune", type=lambda x:bool(strtobool(x)), default=True, nargs="?", const=True,
+        help="automatic tuning of the entropy coefficient")
+    args = parser.parse_args()
+    # fmt: on
+    return args
+
+def make_env(env_id, seed, idx, capture_video, run_name):
+    def thunk():
+        env = gym.make(env_id)
+        env = gym.wrappers.RecordEpisodeStatistics(env)
+        if capture_video:
+            if idx == 0:
+                env = gym.wrappers.RecordVideo(env, f"videos/{run_name}")
+        env = NoopResetEnv(env, noop_max=30)
+        env = MaxAndSkipEnv(env, skip=4)
+        env = EpisodicLifeEnv(env)
+        if "FIRE" in env.unwrapped.get_action_meanings():
+            env = FireResetEnv(env)
+        # Clip rewards: Clip rewards in a (-1, 1) range. SB3 reward_clip doesn't clip, it bins in [-1, 0, 1].
+        env = gym.wrappers.TransformReward(env, lambda reward: np.clip(reward, -1, 1))
+        env = gym.wrappers.ResizeObservation(env, (84, 84))
+        env = gym.wrappers.GrayScaleObservation(env)
+        env = gym.wrappers.FrameStack(env, 4)
+        env.seed(seed)
+        env.action_space.seed(seed)
+        env.observation_space.seed(seed)
+        return env
+
+    return thunk
+
+
+def network_init(net, bias_const=0.0):
+    # He initialization
+    for param in net.parameters():
+        if type(param) == nn.Linear or type(param) == nn.Conv2d:
+            nn.init.kaiming_normal_(param.weight)
+            torch.nn.init.constant_(param.bias, bias_const)
+
+    return net
+
+
+
+# ALGO LOGIC: initialize agent here:
+# NOTE: Sharing a CNN encoder between Actor and Critics is not recommended for SAC
+# See https://arxiv.org/abs/1910.01741 for more info
+# TL;DR The actor's gradients mess up the representation when using a joint encoder
+class SoftQNetwork(nn.Module):
+    """Soft Q network for pixel inputs."""
+
+    def __init__(self, envs):
+        super().__init__()
+        obs_shape = envs.single_observation_space.shape
+        self.conv = nn.Sequential(
+            nn.Conv2d(obs_shape[0], 32, kernel_size=8, stride=4),
+            nn.ReLU(),
+            nn.Conv2d(32, 64, kernel_size=4, stride=2),
+            nn.ReLU(),
+            nn.Conv2d(64, 64, kernel_size=3, stride=1),
+            nn.Flatten(),
+            nn.ReLU(),
+        )
+
+        with torch.inference_mode():
+            output_dim = self.conv(torch.zeros(1, *obs_shape)).shape[1]
+
+        self.Q_func = nn.Sequential(
+            nn.Linear(output_dim, 512),
+            nn.ReLU(),
+            nn.Linear(512, envs.single_action_space.n),
+        )
+
+    def forward(self, x):
+        x = self.conv(x / 255.0)
+        x = self.Q_func(x)
+        return x
+
+
+class Actor(nn.Module):
+    """Discrete SAC Actor for pixel inputs"""
+
+    def __init__(self, envs):
+        super().__init__()
+        obs_shape = envs.single_observation_space.shape
+        self.conv = nn.Sequential(
+            nn.Conv2d(obs_shape[0], 32, kernel_size=8, stride=4),
+            nn.ReLU(),
+            nn.Conv2d(32, 64, kernel_size=4, stride=2),
+            nn.ReLU(),
+            nn.Conv2d(64, 64, kernel_size=3, stride=1),
+            nn.Flatten(),
+            nn.ReLU(),
+        )
+
+        with torch.inference_mode():
+            output_dim = self.conv(torch.zeros(1, *obs_shape)).shape[1]
+
+        self.fc_logits = nn.Sequential(
+            nn.Linear(output_dim, 512),
+            nn.ReLU(),
+            nn.Linear(512, envs.single_action_space.n),
+        )
+
+        # He initialization
+        for param in self.parameters():
+            if type(param) == nn.Linear or type(param) == nn.Conv2d:
+                nn.init.kaiming_normal_(param.weight)
+
+    def forward(self, x):
+        x = self.conv(x)
+        logits = self.fc_logits(x)
+
+        return logits
+
+    def get_action(self, x):
+        logits = self(x / 255.0)
+        policy_dist = Categorical(logits=logits)
+        action = policy_dist.sample()
+        # Action probabilities for calculating the adapted soft-Q loss
+        action_probs = policy_dist.probs
+        # From https://github.com/p-christ/Deep-Reinforcement-Learning-Algorithms-with-PyTorch/blob/135d3e2e06bbde2868047d738e3fc2d73fd8cc93/agents/actor_critic_agents/SAC_Discrete.py#L58
+        # Add small numerical constant where action probability is 0 to avoid log(0)
+        z = action_probs == 0.0
+        z = z.float() * 1e-8
+        log_pi = torch.log(action_probs + z)
+        return action, log_pi, action_probs
+
+
+if __name__ == "__main__":
+    args = parse_args()
+    run_name = f"{args.env_id}__{args.exp_name}__{args.seed}__{int(time.time())}"
+    if args.track:
+        import wandb
+
+        wandb.init(
+            project=args.wandb_project_name,
+            entity=args.wandb_entity,
+            sync_tensorboard=True,
+            config=vars(args),
+            name=run_name,
+            monitor_gym=True,
+            save_code=True,
+        )
+    writer = SummaryWriter(f"runs/{run_name}")
+    writer.add_text(
+        "hyperparameters",
+        "|param|value|\n|-|-|\n%s" % ("\n".join([f"|{key}|{value}|" for key, value in vars(args).items()])),
+    )
+
+    # Don't utilize full CPU or else I might get warned
+    # TODO: Must change back before merge
+    torch.set_num_threads(1)
+
+    # TRY NOT TO MODIFY: seeding
+    random.seed(args.seed)
+    np.random.seed(args.seed)
+    torch.manual_seed(args.seed)
+    torch.backends.cudnn.deterministic = args.torch_deterministic
+
+    # TODO: Change back before merge
+    device = torch.device("cuda:0" if torch.cuda.is_available() and args.cuda else "cpu")
+
+    # env setup
+    envs = gym.vector.SyncVectorEnv([make_env(args.env_id, args.seed, 0, args.capture_video, run_name)])
+    assert isinstance(envs.single_action_space, gym.spaces.Discrete), "only discrete action space is supported"
+
+    actor = network_init(Actor(envs)).to(device)
+    qf1 = network_init(SoftQNetwork(envs)).to(device)
+    qf2 = network_init(SoftQNetwork(envs)).to(device)
+
+    qf1_target = SoftQNetwork(envs).to(device)
+    qf2_target = SoftQNetwork(envs).to(device)
+    qf1_target.load_state_dict(qf1.state_dict())
+    qf2_target.load_state_dict(qf2.state_dict())
+    q_optimizer = optim.Adam(list(qf1.parameters()) + list(qf2.parameters()), lr=args.q_lr, eps=1e-4)
+    actor_optimizer = optim.Adam(list(actor.parameters()), lr=args.policy_lr, eps=1e-4)
+
+    # Automatic entropy tuning
+    if args.autotune:
+        target_entropy = -0.98 * torch.log(torch.tensor(envs.single_action_space.n, dtype=float))
+        log_alpha = torch.zeros(1, requires_grad=True, device=device)
+        alpha = log_alpha.exp().item()
+        a_optimizer = optim.Adam([log_alpha], lr=args.policy_lr, eps=1e-4)
+    else:
+        alpha = args.alpha
+
+    rb = ReplayBuffer(
+        args.buffer_size,
+        envs.single_observation_space,
+        envs.single_action_space,
+        device,
+        # TODO: Can be changed back for cleanRL runs
+        optimize_memory_usage=False,
+        handle_timeout_termination=True,
+    )
+    start_time = time.time()
+
+    # TRY NOT TO MODIFY: start the game
+    obs = envs.reset()
+    for global_step in range(args.total_timesteps):
+        # ALGO LOGIC: put action logic here
+        if global_step < args.learning_starts:
+            actions = np.array([envs.single_action_space.sample() for _ in range(envs.num_envs)])
+        else:
+            actions, _, _ = actor.get_action(torch.Tensor(obs).to(device))
+            actions = actions.detach().cpu().numpy()
+
+        # TRY NOT TO MODIFY: execute the game and log data.
+        next_obs, rewards, dones, infos = envs.step(actions)
+
+        # TRY NOT TO MODIFY: record rewards for plotting purposes
+        for info in infos:
+            if "episode" in info.keys():
+                print(f"global_step={global_step}, episodic_return={info['episode']['r']}")
+                writer.add_scalar("charts/episodic_return", info["episode"]["r"], global_step)
+                writer.add_scalar("charts/episodic_length", info["episode"]["l"], global_step)
+                break
+
+        # TRY NOT TO MODIFY: save data to reply buffer; handle `terminal_observation`
+        real_next_obs = next_obs.copy()
+        for idx, d in enumerate(dones):
+            if d:
+                real_next_obs[idx] = infos[idx]["terminal_observation"]
+        rb.add(obs, real_next_obs, actions, rewards, dones, [info])
+
+        # TRY NOT TO MODIFY: CRUCIAL step easy to overlook
+        obs = next_obs
+
+        # ALGO LOGIC: training. Supports TD3 delayed update
+        if global_step > args.learning_starts and global_step % args.update_frequency == 0:
+            for _ in range(args.num_update_steps):
+                data = rb.sample(args.batch_size)
+                # CRITIC training
+                with torch.no_grad():
+                    _, next_state_log_pi, next_state_action_probs = actor.get_action(data.next_observations)
+                    qf1_next_target = qf1_target(data.next_observations)
+                    qf2_next_target = qf2_target(data.next_observations)
+                    # we can use the action probabilities instead of MC sampling to estimate the expectation
+                    min_qf_next_target = next_state_action_probs * (
+                        torch.min(qf1_next_target, qf2_next_target) - alpha * next_state_log_pi
+                    )
+                    # adapt Q-target for discrete Q-function
+                    min_qf_next_target = min_qf_next_target.sum(dim=1)[..., None]
+                    next_q_value = data.rewards.flatten() + (1 - data.dones.flatten()) * args.gamma * (min_qf_next_target).view(-1)
+
+                # Use Q-values only for the taken actions
+                qf1_a_values = qf1(data.observations).gather(1, data.actions.long())
+                qf2_a_values = qf2(data.observations).gather(1, data.actions.long())
+                qf1_loss = F.mse_loss(qf1_a_values, next_q_value[..., None], reduction="mean")
+                qf2_loss = F.mse_loss(qf2_a_values, next_q_value[..., None], reduction="mean")
+                qf_loss = qf1_loss + qf2_loss
+
+                q_optimizer.zero_grad()
+                qf_loss.backward()
+                q_optimizer.step()
+
+                # ACTOR training
+                _, log_pi, pi_probs = actor.get_action(data.observations)
+                with torch.no_grad():
+                    qf1_pi = qf1(data.observations)
+                    qf2_pi = qf2(data.observations)
+                    min_qf_pi = torch.min(qf1_pi, qf2_pi)
+                # no need for reparameterization, the expectation can be calculated for discrete actions
+                actor_loss = (pi_probs * ((alpha * log_pi) - min_qf_pi)).mean()
+
+                actor_optimizer.zero_grad()
+                actor_loss.backward()
+                actor_optimizer.step()
+
+                if args.autotune:
+                    with torch.no_grad():
+                        _, log_pi, action_probs = actor.get_action(data.observations)
+                    # use action probabilities for temperate loss
+                    alpha_loss = (action_probs * (-log_alpha * (log_pi + target_entropy))).mean()
+
+                    a_optimizer.zero_grad()
+                    alpha_loss.backward()
+                    a_optimizer.step()
+                    alpha = log_alpha.exp().item()
+
+            # update the target networks
+            if global_step % args.target_network_frequency == 0:
+                for param, target_param in zip(qf1.parameters(), qf1_target.parameters()):
+                    target_param.data.copy_(args.tau * param.data + (1 - args.tau) * target_param.data)
+                for param, target_param in zip(qf2.parameters(), qf2_target.parameters()):
+                    target_param.data.copy_(args.tau * param.data + (1 - args.tau) * target_param.data)
+
+            if global_step % 100 == 0:
+                writer.add_scalar("losses/qf1_values", qf1_a_values.mean().item(), global_step)
+                writer.add_scalar("losses/qf2_values", qf2_a_values.mean().item(), global_step)
+                writer.add_scalar("losses/qf1_loss", qf1_loss.item(), global_step)
+                writer.add_scalar("losses/qf2_loss", qf2_loss.item(), global_step)
+                writer.add_scalar("losses/qf_loss", qf_loss.item() / 2.0, global_step)
+                writer.add_scalar("losses/actor_loss", actor_loss.item(), global_step)
+                writer.add_scalar("losses/alpha", alpha, global_step)
+                print("SPS:", int(global_step / (time.time() - start_time)))
+                writer.add_scalar("charts/SPS", int(global_step / (time.time() - start_time)), global_step)
+                if args.autotune:
+                    writer.add_scalar("losses/alpha_loss", alpha_loss.item(), global_step)
+
+    envs.close()
+    writer.close()