pufferlib-core 3.0.10

PufferLib core functionality with vectorized environments and reinforcement learning

PufferLib Core

Minimal PufferLib core functionality with vectorized environments.

This package contains only the essential components:

• spaces: Observation/action space handling
• emulation: Environment compatibility layer for Gym/Gymnasium/PettingZoo
• vector: Vectorized environment implementations

For the full PufferLib with training capabilities and environments, see the main pufferlib package.

Installation

Basic Installation (Python-only)

pip install pufferlib-core

Installation with C++/CUDA Extensions

To enable the pufferlib._C extensions (required for advanced features like CUDA advantage computation):

# First install torch
pip install torch

# Then install with extensions
PUFFERLIB_BUILD_EXT=1 pip install pufferlib-core[ext]

Or you can install the dependencies and build in separate steps:

# Install with extra dependencies
pip install pufferlib-core[ext]

# Then rebuild with extensions
PUFFERLIB_BUILD_EXT=1 pip install --upgrade --force-reinstall --no-deps pufferlib-core

Usage

After installation with extensions, you should be able to import the C extensions:

import pufferlib
from pufferlib import _C  # This will only work if extensions were built

Development

To build extensions in development mode:

PUFFERLIB_BUILD_EXT=1 python setup.py build_ext --inplace