SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2002.06432 · 2020

PDDLGym: Gym Environments from PDDL Problems

arXiv · PDF · Open in the Atlas

Code that ran

3 of 11 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 3 have no contract verdict; 0 failures of our instrument; 8 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2002.06432.json · badge: /api/ran/2002.06432.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
tomsilver/pddlgym canonical (classifier label)
classifier basis: census
3 ran · 8 no recorded run
FunctionStatusWhere it lives
compute_necessary_literals Ran
ran on a synthesized input
tomsilver/pddlgym/pddlgym/downward_translate/axiom_rules.py
code served (permissive licence) · get_code("00af7df690907c53")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned set. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

# Build stand-in classes for the domain planning problem

class Literal:
    """Represents a literal (atom or its negation)"""
    def __init__(self, name, is_negated=False):
        self.name = name
        self._negated = is_negated
    
    def positive(self):
        """Return the positive version of this literal"""
        if self._negated:
            return Literal(self.name, False)
        return self
    
    def negated(self):
        """Check if this literal is negated"""
        return self._negated
    
    def negate(self):
        """Return the negation of this literal"""
        return Literal(self.name, not self._negated)
    
    def __eq__(self, other):
        return self.name == other.name and self._negated == other._negated
    
    def __hash__(self):
        return hash((self.name, self._negated))
    
    def __repr__(self):
        return f"~{self.name}" if self._negated else self.name


class Dependencies:
    """Represents dependencies between atoms"""
    def __init__(self, derived_variables, positive_deps, negative_deps):
        self.derived_variables = derived_variables
        self.positive_dependencies = positive_deps
        self.negative_dependencies = negative_deps


class Operator:
    """Represents a planning operator with preconditions and effects"""
    def __init__(self, name, precondition, add_effects, del_effects):
        self.name = name
        self.precondition = precondition
        self.add_effects = add_effects
        self.del_effects = del_effects


# Create test data
lit_a = Literal("a", False)
lit_b = Literal("b", False)
lit_c = Literal("c", False)

derived_vars = {"a", "b", "c"}

positive_deps = {
    "a": [lit_b],
    "b": [],
    "c": [lit_a],
}

negative_deps = {
    "a": [],
    "b": [lit_c],
    "c": [],
}

dependencies = Dependencies(derived_vars, positive_deps, negative_deps)

goals = [lit_a, lit_b]

op1 = Operator(
    "op1",
    precondition=[lit_b],
    add_effects=[([], lit_c)],
    del_effects=[]
)

op2 = Operator(
    "op2",
    precondition=[lit_a],
    add_effects=[],
    del_effects=[([], lit_b)]
)

operators = [op1, op2]

ARGS = {
    "dependencies": dependencies,
    "goals": goals,
    "operators": operators,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type set — a set output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a set output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
create_replanning_policy Ran
ran on a synthesized input
tomsilver/pddlgym/pddlgym/demo_planning.py
code served (permissive licence) · get_code("8db1674273fb34cc")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned function. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import copy
import torch
import numpy as np


class PlanningFailure(Exception):
    """Exception raised when planning fails."""
    pass


class SimpleState:
    """Minimal state representation."""
    def __init__(self, values):
        self.values = values
    
    def __repr__(self):
        return f"State({self.values})"


class SimplePlan:
    """Minimal plan that supports pop(0)."""
    def __init__(self, actions):
        self.actions = list(actions)
    
    def pop(self, index):
        return self.actions.pop(index)


class SimpleDomain:
    """Minimal domain with determinize method."""
    def __init__(self):
        self.is_deterministic = False
    
    def determinize(self):
        self.is_deterministic = True


class SimpleActionSpace:
    """Minimal action space with sample method."""
    def sample(self, state):
        return "random_action"


class SimpleEnv:
    """Minimal environment with domain and action_space."""
    def __init__(self):
        self.domain = SimpleDomain()
        self.action_space = SimpleActionSpace()


def simple_planner(domain, state):
    """Minimal planner that returns a plan."""
    # Return a plan with some actions
    return SimplePlan(["action_1", "action_2", "action_3"])


def failing_planner(domain, state):
    """Planner that raises PlanningFailure."""
    raise PlanningFailure("Planning failed")


ARGS = {
    "env": SimpleEnv(),
    "planner": simple_planner,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
create_single_plan_policy Ran
ran on a synthesized input
tomsilver/pddlgym/pddlgym/demo_planning.py
code served (permissive licence) · get_code("c833316e1abd284f")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned function. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

# Create a minimal environment class
class MinimalEnv:
    def __init__(self):
        self.domain = "test_domain"

# Create a minimal planner function
def minimal_planner(domain, state):
    # Returns a plan (list of actions)
    return ["action_1", "action_2", "action_3"]

# Create instances
env = MinimalEnv()
planner = minimal_planner

ARGS = {
    "env": env,
    "planner": planner,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
Anti Not yet run
unverified (harvested, no recorded run)
tomsilver/pddlgym/pddlgym/structs.py
code served (permissive licence) · get_code("24a7865bb85917cc")
How it ran

no invocation recorded.

Effect Not yet run
unverified (harvested, no recorded run)
tomsilver/pddlgym/pddlgym/structs.py
code served (permissive licence) · get_code("7b7d1ede4743c391")
How it ran

no invocation recorded.

Not Not yet run
unverified (harvested, no recorded run)
tomsilver/pddlgym/pddlgym/structs.py
code served (permissive licence) · get_code("c8facca750c9393a")
How it ran

no invocation recorded.

compute_model Not yet run
unverified (harvested, no recorded run)
tomsilver/pddlgym/pddlgym/downward_translate/build_model.py
code served (permissive licence) · get_code("75a6df62152e5791")
How it ran

no invocation recorded.

convert_rules Not yet run
unverified (harvested, no recorded run)
tomsilver/pddlgym/pddlgym/downward_translate/build_model.py
code served (permissive licence) · get_code("04701e1abd96a451")
How it ran

no invocation recorded.

get_strongly_connected_components Not yet run
unverified (harvested, no recorded run)
tomsilver/pddlgym/pddlgym/downward_translate/axiom_rules.py
code served (permissive licence) · get_code("62e5d19d8450127f")
How it ran

no invocation recorded.

handle_axioms Not yet run
unverified (harvested, no recorded run)
tomsilver/pddlgym/pddlgym/downward_translate/axiom_rules.py
code served (permissive licence) · get_code("2824e9e9cd15b30d")
How it ran

no invocation recorded.

variables_to_numbers Not yet run
unverified (harvested, no recorded run)
tomsilver/pddlgym/pddlgym/downward_translate/build_model.py
code served (permissive licence) · get_code("f9ee75ae536acc48")
How it ran

no invocation recorded.

Also present in other papers' code

Of 11 samples lifted from this paper's linked repositories, 11 carry a function identity and 0 do not. Of the 11 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 11 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

We present PDDLGym, a framework that automatically constructs OpenAI Gym environments from PDDL domains and problems. Observations and actions in PDDLGym are relational, making the framework particularly well-suited for research in relational reinforcement learning and relational sequential decision-making. PDDLGym is also useful as a generic framework for rapidly building numerous, diverse benchmarks from a concise and familiar specification language. We discuss design decisions and implementation details, and also illustrate empirical variations between the 20 built-in environments in terms of planning and model-learning difficulty. We hope that PDDLGym will facilitate bridge-building between the reinforcement learning community (from which Gym emerged) and the AI planning community (which produced PDDL). We look forward to gathering feedback from all those interested and expanding the set of available environments and features accordingly. Code: https://github.com/tomsilver/pddlgym

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2002.06432")
get_code_for_paper("2002.06432")
have("2002.06432")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2002.06432.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2002.06432.svg)](https://syntology.ai/paper/2002.06432)

Connect an agent — have() is free.