gym-saturation 0.9.0

Gymnasium environments for saturation provers

gym-saturation

gym-saturation is a collection of Gymnasium environments for reinforcement learning (RL) agents striving to prove theorems. Currently, only theorems written in TPTP library formal language are supported.

There are two environments in gym-saturation: VampireEnv and IProverEnv. VampireEnv is a wrapper around a recent Vampire prover, and IProverEnv relies on an experimental version of iProver.

In contrast to monolithic architecture of a typical Automated Theorem Prover (ATP), gym-saturation gives different agents opportunities to select clauses themselves and train from their experience. Combined with a particular agent, gym-saturation can work as an ATP.

gym-saturation can be interesting for RL practitioners willing to apply their experience to theorem proving without coding all the logic-related stuff themselves. It also can be useful for automated deduction researchers who want to create an RL-empowered ATP.

How to Install

Attention!

If you want to use VampireEnv you should have a Vampire binary on your machine. For example, download the latest release.

To use IProverEnv, please build iProver from source using this branch.

The best way to install this package is to use pip:

pip install gym-saturation

Another option is to use conda:

conda install -c conda-forge gym-saturation

One can also run it in a Docker container (with a pre-packed vampire binary):

docker build -t gym-saturation https://github.com/inpefess/gym-saturation.git
docker run -it --rm -p 8888:8888 gym-saturation jupyter-lab --ip=0.0.0.0 --port=8888

How to use

One can use gym-saturation environments as any other Gymnasium environment:

import gym_saturation
import gymnasium

env = gymnasium.make("Vampire-v0")  # or "iProver-v0"
# skip this line to use the default problem
env.set_task("a-TPTP-problem-filename")
observation, info = env.reset()
terminated, truncated = False, False
while not (terminated or truncated):
    # apply policy (a valid random action here)
    action = env.action_space.sample(mask=observation["action_mask"])
    observation, reward, terminated, truncated, info = env.step(action)
env.close()

See the notebook for more information.

For more comprehensive experiments, please see the documentation page.

How to Contribute

Pull requests are welcome. To start:

git clone https://github.com/inpefess/gym-saturation
cd gym-saturation
# activate python virtual environment with Python 3.8+
pip install -U pip
pip install -U setuptools wheel poetry
poetry install
# recommended but not necessary
pre-commit install
# install vampire binary
wget https://github.com/vprover/vampire/releases/download/v4.7/vampire4.7.zip -O vampire.zip
unzip vampire.zip
# then use vampire_z3_rel_static_HEAD_6295 as an argument or add it to $PATH

All the tests in this package are doctests. One can run them with the following command:

pytest --doctest-modules gym-saturation

To check the code quality before creating a pull request, one might run the script local-build.sh. It locally does nearly the same as the CI pipeline after the PR is created.

Reporting issues or problems with the software

Questions and bug reports are welcome on the tracker.

More documentation

More documentation can be found here.