bsuite
is a collection of carefully-designed experiments that investigate core
capabilities of a reinforcement learning (RL) agent with two main objectives.
- To collect clear, informative and scalable problems that capture key issues in the design of efficient and general learning algorithms.
- To study agent behavior through their performance on these shared benchmarks.
This library automates evaluation and analysis of any agent on these benchmarks. It serves to facilitate reproducible, and accessible, research on the core issues in RL, and ultimately the design of superior learning algorithms.
Going forward, we hope to incorporate more excellent experiments from the research community, and commit to a periodic review of the experiments from a committee of prominent researchers.
For a more comprehensive overview, see the accompanying paper.
bsuite
is a collection of experiments, defined in the experiments
subdirectory. Each subdirectory corresponds to one experiment and contains:
- A file defining an RL environment, which may be configurable to provide different levels of difficulty or different random seeds (for example).
- A sequence of keyword arguments for this environment, defined in the
SETTINGS
variable found in the experiment'ssweep.py
file. - A file
analysis.py
defining plots used in the provided Jupyter notebook.
bsuite
works by logging results from "within" each environment, when loading
environment via a
load_and_record*
function.
This means any experiment will automatically output data in the correct format
for analysis using the notebook, without any constraints on the structure of
agents or algorithms.
We collate all of the results and analysis in a pre-made jupyter notebook bit.ly/bsuite-colab.
If you are new to bsuite
you can get started in our
colab tutorial.
This Jupyter notebook is hosted with a free cloud server, so you can start
coding right away without installing anything on your machine. After this, you
can follow the instructions below to get bsuite
running on your local machine.
We have tested bsuite
on Python 3.6 & 3.7. To install the dependencies:
-
Optional: We recommend using a Python virtual environment to manage your dependencies, so as not to clobber your system installation:
python3 -m venv bsuite source bsuite/bin/activate pip install --upgrade pip setuptools
-
Install
bsuite
directly from PyPI:pip install bsuite
-
Optional: To also install dependencies for the
baselines
examples (excluding OpenAI and Dopamine examples), run:pip install bsuite[baselines]
Complete descriptions of each environment and their corresponding experiments
are found in the analysis/results.ipynb
Jupyter notebook.
These environments all have small observation sizes, allowing for reasonable performance with a small network on a CPU.
Environments are specified by a bsuite_id
string, for example "deep_sea/7"
.
This string denotes the experiment and the (index of the) environment settings
to use, as described in the technical overview section.
For a full description of each environment and its corresponding experiment settings, see the paper.
import bsuite
env = bsuite.load_from_id('catch/0')
The sequence of bsuite_id
s required to run all experiments can be accessed
programmatically via:
from bsuite import sweep
sweep.SWEEP
This module also contains bsuite_id
s for each experiment individually via
uppercase constants corresponding to the experiment name, for example:
sweep.DEEP_SEA
sweep.DISCOUNTING_CHAIN
In addition, sequences of bsuite_id
s with the same tag can be loaded via:
from bsuite import sweep
sweep.TAGS
The TAGS
variable groups bsuite
environments together by their underlying
tag, so all the basic
tasks or scale
tasks can be loaded with:
sweep.TAGS['basic']
sweep.TAGS['scale']
We include two implementations of automatic logging, available via:
bsuite.load_and_record_to_csv
. This outputs one CSV file perbsuite_id
, so is suitable for running a set of bsuite experiments split over multiple machines. The implementation is inlogging/csv_logging.py
bsuite.load_and_record_to_sqlite
. This outputs a single file, and is best suited when running a set of bsuite experiments via multiple processes on a single workstation. The implementation is inlogging/sqlite_logging.py
.
We also include a terminal logger in logging/terminal_logging.py
, exposed
via bsuite.load_and_record_to_terminal
.
It is easy to write your own logging mechanism, if you need to save results to a different storage system. See the CSV implementation for the simplest reference.
Our environments implement the Python interface defined in
dm_env
.
More specifically, all our environments accept a discrete, zero-based integer
action (or equivalently, a scalar numpy array with shape ()
).
To determine the number of actions for a specific environment, use
num_actions = env.action_spec().num_values
Each environment returns observations in the form of a numpy array.
We also expose a bsuite_num_episodes
property for each environment in bsuite.
This allows users to run exactly the number of episodes required for bsuite's
analysis, which may vary between environments used in different experiments.
Example run loop for a hypothetical agent with a step()
method.
for _ in range(env.bsuite_num_episodes):
timestep = env.reset()
while not timestep.last():
action = agent.step(timestep)
timestep = env.step(action)
agent.step(timestep)
To use bsuite
with a codebase that uses the
OpenAI Gym interface, use the GymFromDMEnv
class in utils/gym_wrapper.py
:
import bsuite
from bsuite.utils import gym_wrapper
env = bsuite.load_and_record_to_csv('catch/0', results_dir='/path/to/results')
gym_env = gym_wrapper.GymFromDMEnv(env)
Note that bsuite
does not include Gym in its default dependencies, so you may
need to pip install it separately.
We include implementations of several common agents in the [baselines/
]
subdirectory, along with a minimal run-loop.
See the installation section for how to include the required
dependencies at install time. These
dependencies are not installed by default, since bsuite
does not require users
to use any specific machine learning library.
Each of the agents in the baselines
folder contains a run
script which
serves as an example which can run against a single environment or against the
entire suite of experiments, by passing the --bsuite_id=SWEEP
flags; this will
start a pool of processes with which to run as many experiments in parallel as
the host machine allows. On a 12 core machine, this will complete overnight for
most agents. Alternatively, it is possible to run on Google Compute Platform
using run_on_gcp.sh
, steps of which are outlined below.
run_on_gcp.sh
does the following in order:
- Create an instance with specified specs (by default 64-core CPU optimized).
git clone
sbsuite
and installs it together with other dependencies.- Runs the specified agent (currently limited to
/baselines
) on a specified environment. - Copies the resulting SQLite file to
/tmp/bsuite.db
from the remote instance to you local machine. - Shuts down the created instance.
In order to run the script, you first need to create a billing account. Then
follow the instructions
here to setup and
initialize Cloud SDK. After completing gcloud init
, you are ready to run
bsuite
on Google Cloud.
For this make run_on_gcp.sh
executable and run it:
chmod x run_on_gcp.sh
./run_on_gcp.sh
After the instance is created, the instance name will be printed. Then you can
ssh into the instance by selecting Compute Engine -> Instances
and clicking
SSH
. Note that this is not necessary, as the result will be copied to your
local machine once it is ready. However, ssh
ing might be convenient if you
want to make local changes to agent and environments. In this case, after
ssh
ing, do
~/bsuite_env/bin/activate
to activate the virtual environment. Then you can run agents via
python ~/bsuite/bsuite/baselines/dqn/run.py --bsuite_id=SWEEP
for instance.
bsuite
comes with a ready-made analysis Jupyter notebook included in
analysis/results.ipynb
. This notebook loads and processes logged data, and
produces the scores and plots for each experiment. We recommend using this
notebook in conjunction with Colaboratory.
We provide an example of a such bsuite
report
here.
You can use bsuite
to generate an automated 1-page appendix, that summarizes
the core capabilities of your RL algorithm. This appendix is compatible with
most major ML conference formats. For example output run,
pdflatex bsuite/reports/neurips_2019/neurips_2019.tex
More examples of bsuite reports can be found in the reports/
subdirectory.
If you use bsuite
in your work, please cite the accompanying paper:
@inproceedings{osband2020bsuite,
title={Behaviour Suite for Reinforcement Learning},
author={Osband, Ian and
Doron, Yotam and
Hessel, Matteo and
Aslanides, John and
Sezener, Eren and
Saraiva, Andre and
McKinney, Katrina and
Lattimore, Tor and
{Sz}epesv{\'a}ri, Csaba and
Singh, Satinder and
Van Roy, Benjamin and
Sutton, Richard and
Silver, David and
van Hasselt, Hado},
booktitle={International Conference on Learning Representations},
year={2020},
url={https://openreview.net/forum?id=rygf-kSYwH}
}