ppo_policy.py

import numpy as np
import torch
import torch.nn as nn
import torch.optim as optim
from torch.distributions.categorical import Categorical
from torch.distributions.normal import Normal
from torch.utils.tensorboard import SummaryWriter


def layer_init(layer, std=np.sqrt(2), bias_const=0.0):
    torch.nn.init.orthogonal_(layer.weight, std)
    torch.nn.init.constant_(layer.bias, bias_const)
    return layer


class Agent(nn.Module):
    def __init__(self, envs):
        super().__init__()
        self.critic = nn.Sequential(
            layer_init(nn.Linear(np.array(envs.single_observation_space.shape).prod(), 64)),
            nn.Tanh(),
            layer_init(nn.Linear(64, 64)),
            nn.Tanh(),
            layer_init(nn.Linear(64, 1), std=1.0),
        )
        self.actor_mean = nn.Sequential(
            layer_init(nn.Linear(np.array(envs.single_observation_space.shape).prod(), 64)),
            nn.Tanh(),
            layer_init(nn.Linear(64, 64)),
            nn.Tanh(),
            layer_init(nn.Linear(64, np.prod(envs.single_action_space.shape)), std=0.01),
        )
        self.actor_logstd = nn.Parameter(torch.zeros(1, np.prod(envs.single_action_space.shape)))


    def get_value(self, x):
        return self.critic(x)

    # Used in continuous spaces
    def get_action_and_value(self, x, action=None):
        action_mean = self.actor_mean(x)
        action_logstd = self.actor_logstd.expand_as(action_mean)
        action_std = torch.exp(action_logstd)
        probs = Normal(action_mean, action_std)
        if action is None:
            action = probs.sample()

        log_prob_sum = probs.log_prob(action).sum(1)
        entropy_sum = probs.entropy().sum(1)
        return action, log_prob_sum, entropy_sum, self.critic(x)

    # Only used in discrete action spaces.
    def get_discrete_action_and_value(self, x, action=None):
        logits = self.actor(x)
        probs = Categorical(logits=logits)
        if action is None:
            action = probs.sample()
        return action, probs.log_prob(action), probs.entropy(), self.critic(x)