Go to file
Keerthana P G d4a6f4e023 add checkpoints 2022-12-13 14:59:48 -08:00
configs add robotics transformer 2022-12-09 11:58:47 -08:00
film_efficientnet add robotics transformer 2022-12-09 11:58:47 -08:00
tokenizers add robotics transformer 2022-12-09 11:58:47 -08:00
trained_checkpoints add checkpoints 2022-12-13 14:59:48 -08:00
.gitattributes add checkpoints 2022-12-13 14:59:48 -08:00
.gitignore add robotics transformer 2022-12-09 11:58:47 -08:00
.pylintrc add robotics transformer 2022-12-09 11:58:47 -08:00
CONTRIBUTING.md add robotics transformer 2022-12-09 11:58:47 -08:00
LICENSE add robotics transformer 2022-12-09 11:58:47 -08:00
README.md add checkpoints 2022-12-13 14:59:48 -08:00
__init__.py add robotics transformer 2022-12-09 11:58:47 -08:00
requirements.txt add robotics transformer 2022-12-09 11:58:47 -08:00
robotics_transformer.blueprint add robotics transformer 2022-12-09 11:58:47 -08:00
sequence_agent.py Update sequence_agent.py 2022-12-12 14:13:23 -08:00
sequence_agent_test.py add robotics transformer 2022-12-09 11:58:47 -08:00
sequence_agent_test_set_up.py add robotics transformer 2022-12-09 11:58:47 -08:00
transformer.py add robotics transformer 2022-12-09 11:58:47 -08:00
transformer_network.py Update transformer_network.py 2022-12-12 14:02:35 -08:00
transformer_network_test.py add robotics transformer 2022-12-09 11:58:47 -08:00
transformer_network_test_set_up.py add robotics transformer 2022-12-09 11:58:47 -08:00
transformer_test.py add robotics transformer 2022-12-09 11:58:47 -08:00

README.md

Robotics Transformer

This is not an officially supported Google product.

This repository is a collection code files and artifacts for running Robotics Transformer or RT-1.

Features

  • Film efficient net based image tokenizer backbone
  • Token learner based compression of input tokens
  • Transformer for end to end robotic control
  • Testing utilities

Getting Started

Installation

Clone the repo

git clone https://github.com/google-research/robotics_transformer.git
pip install -r robotics_transformer/requirements.txt
python -m robotics_transformer.tokenizers.action_tokenizer.test

Running Tests

To run RT-1 tests, you can clone the git repo and run bazel:

git clone https://github.com/google_research/robotics_transformer.git
cd robotics_transformer
bazel test ...

Using trained checkpoints

Checkpoints are included in trained_checkpoints/ folder for three models:

  1. RT-1 trained on 700 tasks
  2. RT-1 jointly trained on EDR and Kuka data
  3. RT-1 jointly trained on sim and real data They are tensorflow SavedModel files. Instructions on usage can be found (here)[https://www.tensorflow.org/guide/saved_model]

Future Releases

The current repository includes an initial set of libraries for early adoption. More components may come in future releases.

License

The Robotics Transformer library is licensed under the terms of the Apache license.