iclbench.environments package

Subpackages

• iclbench.environments.baba_is_ai package
  • Submodules
  • iclbench.environments.baba_is_ai.base module
    • BabaIsAIWrapper
      • BabaIsAIWrapper.__init__()
      • BabaIsAIWrapper.default_action
      • BabaIsAIWrapper.get_ruleset()
      • BabaIsAIWrapper.get_stats()
      • BabaIsAIWrapper.get_text_action()
      • BabaIsAIWrapper.get_text_observation()
      • BabaIsAIWrapper.reset()
      • BabaIsAIWrapper.step()
      • BabaIsAIWrapper.textworld_process_obsv()
  • Module contents
    • get_instruction_prompt()
• iclbench.environments.babyai_text package
  • Submodules
  • iclbench.environments.babyai_text.clean_lang_wrapper module
    • BabyAITextCleanLangWrapper
      • BabyAITextCleanLangWrapper.__init__()
      • BabyAITextCleanLangWrapper.default_action
      • BabyAITextCleanLangWrapper.get_prompt()
      • BabyAITextCleanLangWrapper.get_stats()
      • BabyAITextCleanLangWrapper.get_text_action()
      • BabyAITextCleanLangWrapper.interleaving_token
      • BabyAITextCleanLangWrapper.reset()
      • BabyAITextCleanLangWrapper.step()
  • Module contents
    • get_instruction_prompt()
• iclbench.environments.craftax package
  • Submodules
  • iclbench.environments.craftax.env module
    • CraftaxLanguageWrapper
      • CraftaxLanguageWrapper.__init__()
      • CraftaxLanguageWrapper.default_action
      • CraftaxLanguageWrapper.get_stats()
      • CraftaxLanguageWrapper.get_text_action()
      • CraftaxLanguageWrapper.render()
      • CraftaxLanguageWrapper.reset()
      • CraftaxLanguageWrapper.step()
  • Module contents
    • get_instruction_prompt()
• iclbench.environments.crafter package
  • Submodules
  • iclbench.environments.crafter.env module
    • CrafterLanguageWrapper
      • CrafterLanguageWrapper.__init__()
      • CrafterLanguageWrapper.default_iter
      • CrafterLanguageWrapper.default_steps
      • CrafterLanguageWrapper.get_stats()
      • CrafterLanguageWrapper.get_text_action()
      • CrafterLanguageWrapper.process_obs()
      • CrafterLanguageWrapper.reset()
      • CrafterLanguageWrapper.step()
      • CrafterLanguageWrapper.update_progress()
    • describe_act()
    • describe_env()
    • describe_frame()
    • describe_inventory()
    • describe_loc()
    • describe_status()
    • get_player_idx()
    • rotation_matrix()
  • Module contents
    • get_instruction_prompt()
• iclbench.environments.minihack package
  • Module contents
    • get_available_actions()
    • get_instruction_prompt()
• iclbench.environments.nle package
  • Submodules
  • iclbench.environments.nle.base module
  • iclbench.environments.nle.progress module
    • BaseProgress
      • BaseProgress.end_reason
      • BaseProgress.episode_return
      • BaseProgress.progression
      • BaseProgress.update()
    • Progress
      • Progress.__init__()
      • Progress.depth
      • Progress.dlvl_list
      • Progress.end_reason
      • Progress.episode_return
      • Progress.experience_level
      • Progress.gold
      • Progress.highest_achievement
      • Progress.progression
      • Progress.score
      • Progress.time
      • Progress.update()
      • Progress.xplvl_list
    • get_progress_system()
  • iclbench.environments.nle.render module
    • create_texture_map()
    • make_atlas()
    • tty_render_image()
    • tty_render_image_action_history()
  • iclbench.environments.nle.render_rgb module
    • load_atlas()
    • rgb_render_image()
  • Module contents
    • Role
      • Role.ARCHEOLOGIST
      • Role.BARBARIAN
      • Role.CAVEMAN
      • Role.HEALER
      • Role.KNIGHT
      • Role.MONK
      • Role.PRIEST
      • Role.RANGER
      • Role.ROGUE
      • Role.SAMURAI
      • Role.TOURIST
      • Role.VALKYRIE
      • Role.WIZARD
    • get_instruction_prompt()
• iclbench.environments.textworld package
  • Submodules
  • iclbench.environments.textworld.base module
    • AlwaysTrue
    • TextWorldFactory
      • TextWorldFactory.get_textworld_env()
      • TextWorldFactory.initialize()
    • TextWorldWrapper
      • TextWorldWrapper.__init__()
      • TextWorldWrapper.default_action
      • TextWorldWrapper.filter_objective()
      • TextWorldWrapper.get_stats()
      • TextWorldWrapper.get_text_action()
      • TextWorldWrapper.reset()
      • TextWorldWrapper.step()
      • TextWorldWrapper.textworld_process_obsv()
  • Module contents
    • get_instruction_prompt()
    • global_textworld_context()

Submodules

iclbench.environments.env_wrapper module

class iclbench.environments.env_wrapper.EnvWrapper(env, env_name, task_name)

Bases: Wrapper

__init__(env, env_name, task_name)

Wraps an environment to allow a modular transformation of the step() and reset() methods.

Parameters:

env – The environment to wrap

property actions

check_action_validity(candidate_action)

get_instruction_prompt(instructions=None)

get_stats()

get_text_action(action)

property max_steps

reset()

Resets the environment with kwargs.

step(action)

Steps through the environment with action.

Module contents

class iclbench.environments.Strings(values, seed=None)

Bases: Space

__init__(values, seed=None)

Constructor of Space.

Parameters:

• shape (Optional[Sequence[int]]) – If elements of the space are numpy arrays, this should specify their shape.

• dtype (Optional[Type | str]) – If elements of the space are numpy arrays, this should specify their dtype.

• seed – Optionally, you can use this argument to seed the RNG that is used to sample from the space

contains(value)

Return boolean specifying if x is a valid member of this space.

map(action)

sample()

Randomly sample an element of this space.

Can be uniform or non-uniform sampling based on boundedness of space.

Parameters:

mask – A mask used for sampling, expected dtype=np.int8 and see sample implementation for expected shape.

Returns:

A sampled actions from the space

iclbench.environments.make_env(env_name, task, config)

Creates and initializes an environment based on the specified environment name and task.

This function supports multiple environment types, each with its own configuration requirements. The function will return a wrapped environment suitable for use with agents in the context of the ICLBench framework.

Parameters:

• env_name (str) – The name of the environment to create. Supported values include: - “nle” - “minihack” - “babyai” - “crafter” - “craftax” - “textworld” - “babaisai”

• task (str) – The specific task to be performed within the environment.

• config (Config) – An object containing configuration settings, which must include environment-specific keys such as: - envs.nle_kwargs (dict): Arguments specific to the NLE environment. - envs.minihack_kwargs (dict): Arguments specific to the MiniHack environment. - envs.babyai_kwargs (dict): Arguments specific to the BabyAI environment. - envs.crafter_kwargs (dict): Arguments specific to the Crafter environment. - envs.craftax_kwargs (dict): Arguments specific to the Craftax environment. - envs.textworld_kwargs (dict): Arguments specific to the TextWorld environment. - envs.babaisai_kwargs (dict): Arguments specific to the Baba Is AI environment.

Returns:

A wrapped environment instance that includes language processing capabilities

and task-specific functionality.

Return type:

EnvWrapper

Raises:

ValueError – If the provided environment name is not recognized.