gymnasia 3.0.5

OpenAI Gymnasium environments in pure Rust
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125

use std::collections::VecDeque;

use crate::core::{Env, StepResult};

use super::Wrapper;

/// Tracks cumulative reward and episode length.
///
/// Access statistics via typed methods:
/// - [`episode_return`](RecordEpisodeStatistics::episode_return)
/// - [`episode_length`](RecordEpisodeStatistics::episode_length)
/// - [`episode_count`](RecordEpisodeStatistics::episode_count)
pub struct RecordEpisodeStatistics<E: Env> {
    env: E,
    episode_return: f64,
    episode_length: usize,
    episode_count: usize,
    return_history: VecDeque<f64>,
    length_history: VecDeque<usize>,
    buffer_length: usize,
}

impl<E: Env> RecordEpisodeStatistics<E> {
    /// Wrap `env` with episode statistics tracking.
    pub fn new(env: E) -> Self {
        Self::with_buffer(env, 100)
    }

    /// Wrap with a custom history buffer length.
    pub fn with_buffer(env: E, buffer_length: usize) -> Self {
        Self {
            env,
            episode_return: 0.0,
            episode_length: 0,
            episode_count: 0,
            return_history: VecDeque::with_capacity(buffer_length),
            length_history: VecDeque::with_capacity(buffer_length),
            buffer_length,
        }
    }

    /// Cumulative reward for the current episode.
    pub fn episode_return(&self) -> f64 {
        self.episode_return
    }

    /// Number of steps in the current episode.
    pub fn episode_length(&self) -> usize {
        self.episode_length
    }

    /// Total number of completed episodes.
    pub fn episode_count(&self) -> usize {
        self.episode_count
    }

    /// Returns from recent completed episodes.
    pub fn return_history(&self) -> &VecDeque<f64> {
        &self.return_history
    }

    /// Lengths of recent completed episodes.
    pub fn length_history(&self) -> &VecDeque<usize> {
        &self.length_history
    }

    fn record_episode(&mut self) {
        self.episode_count += 1;
        if self.return_history.len() >= self.buffer_length {
            self.return_history.pop_front();
            self.length_history.pop_front();
        }
        self.return_history.push_back(self.episode_return);
        self.length_history.push_back(self.episode_length);
    }
}

impl<E: Env> Env for RecordEpisodeStatistics<E> {
    type Action = E::Action;
    type Observation = E::Observation;
    type ActionSpace = E::ActionSpace;
    type ObservationSpace = E::ObservationSpace;
    type ResetOptions = E::ResetOptions;

    fn step(&mut self, action: Self::Action) -> StepResult<Self::Observation> {
        let result = self.env.step(action);
        self.episode_return += result.reward;
        self.episode_length += 1;
        if result.terminated || result.truncated {
            self.record_episode();
        }
        result
    }

    fn reset(&mut self, seed: Option<u64>, options: Self::ResetOptions) -> Self::Observation {
        self.episode_return = 0.0;
        self.episode_length = 0;
        self.env.reset(seed, options)
    }

    fn action_space(&self) -> &Self::ActionSpace {
        self.env.action_space()
    }

    fn observation_space(&self) -> &Self::ObservationSpace {
        self.env.observation_space()
    }

    fn close(&mut self) {
        self.env.close();
    }
}

impl<E: Env> Wrapper for RecordEpisodeStatistics<E> {
    type Inner = E;
    fn inner(&self) -> &E {
        &self.env
    }
    fn inner_mut(&mut self) -> &mut E {
        &mut self.env
    }
    fn into_inner(self) -> E {
        self.env
    }
}