models.py

# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
#
# This source code is licensed under the BSD-style license found in the
# LICENSE file in the root directory of this source tree.

"""
Data models for the Rlve Gym Environment.

The RLVE_Gym environment is a simple test environment that echoes back messages.
"""

from dataclasses import dataclass

from openenv_core.env_server.types import Action, Observation, State

from typing import Dict, Union


@dataclass(kw_only=True)
class RlveGymState(State):
    """State of the RLVE_Gym containing the seed."""
    seed: int
    problem_input: str = None

    num_samples: int = 0
    sum_accuracy: int = 0


@dataclass(kw_only=True)
class RlveGymAction(Action):
    """Action for the RLVE_Gym environment - just a model output."""

    output: str


@dataclass(kw_only=True)
class RlveGymObservation(Observation):
    """Observation from the RLVE_Gym environment."""

    problem_input: str
    verifier_result: Dict[str, Union[float, int]]

    success: bool
    message: str