Sotopia: an Open-ended Social Learning Environment

Installation

This package supports Python 3.11 and above. We recommend using a virtual environment to install this package, e.g. with anaconda3: conda create -n sotopia python=3.11; conda activate sotopia; conda install -c conda-forge pip. Then, install the requirements and this package.

python -m pip install -r requirements.txt # make sure the packages are installed in the specific conda environment
python -m pip install -e .

OpenAI key is required to run the code. Please set the environment variable OPENAI_API_KEY to your key. The recommend way is to add the key to the conda environment:

conda env config vars set OPENAI_API_KEY=your_key

For some experiments, TogetherAI key is required to run the code. Please set the environment variable TOGETHER_API_KEY to your key. The recommend way is to add the key to the conda environment:

conda env config vars set TOGETHER_API_KEY=your_key

A redis-stack server is required to run the code. Please follow the instruction to start a redis-stack server or use an existing server. The REDIS_OM_URL need to be set before loading and saving agents:

conda env config vars set REDIS_OM_URL="redis://user:password@host:port"

Easy Sample Server

You can view an episode demo with default parameters using the following command:

python -m sotopia_conf.server --gin_file="sotopia_conf/server_conf/server.gin" --gin_file="sotopia_conf/generation_utils_conf/generate.gin"

Contribution

Install dev options

Follow the installation instruction above and then, instead of running python -m pip install -e ., run the following commands:

python -m pip install -e ".[dev]"
mypy --install-types --non-interactive sotopia
python -m pip install pre-commit
pre-commit install

New branch for each feature

git checkout -b feature/feature-name and PR to main branch.

Before committing

Run pytest to make sure all tests pass (this will ensure dynamic typing passed with beartype) and mypy --strict . to check static typing. (You can also run pre-commit run --all-files to run all checks)

Check github action result

Check the github action result to make sure all tests pass. If not, fix the errors and push again.

Running Experiments

We use gin-config to configure the experiments. You don't need to be an expert to use it. The basic syntax is

python <code_file.py> --gin_file <gin_file1> --gin_file <gin_file2> '--gin.PARAM1=value1' '--gin.PARAM2=value2'

The --gin_file is used to load and compose the default configuration. The --gin.PARAM1=value1 is used to overwrite the default configuration. The later configuration will always overwrite the previous one.

Here is an example of running an experiment:

python examples/experiment_eval.py --gin_file sotopia_conf/generation_utils_conf/generate.gin --gin_file sotopia_conf/server_conf/server.gin --gin_file sotopia_conf/run_async_server_in_batch.gin '--gin.ENV_IDS=["01H7VFHPDZVVCDZR3AARA547CY"]' '--gin.AGENT1_MODEL="gpt-4"' '--gin.BATCH_SIZE=20' '--gin.PUSH_TO_DB=False' '--gin.TAG="test"'

For the complete set of parameters, please check the sotopia_conf folder.

To run a large batch of environments, you can change the ENV_IDS parameter in sotopia_conf/run_async_server_in_batch.gin to a list of environment ids. When gin.ENV_IDS==[], all environments on the DB will be used.

Check the generated episodes

Go to the examples/redis_stats.ipynb notebook to check the generated episodes (Episode Log section). As well as calculate the performance.

wwxFromTju/sotopia