Metadata-Version: 2.3
Name: pi-zero-pytorch
Version: 0.0.4
Summary: π0 in Pytorch
Project-URL: Homepage, https://pypi.org/project/pi-zero-pytorch/
Project-URL: Repository, https://github.com/lucidrains/pi-zero-pytorch
Author-email: Phil Wang <lucidrains@gmail.com>
License: MIT License
        
        Copyright (c) 2024 Phil Wang
        
        Permission is hereby granted, free of charge, to any person obtaining a copy
        of this software and associated documentation files (the "Software"), to deal
        in the Software without restriction, including without limitation the rights
        to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
        copies of the Software, and to permit persons to whom the Software is
        furnished to do so, subject to the following conditions:
        
        The above copyright notice and this permission notice shall be included in all
        copies or substantial portions of the Software.
        
        THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
        IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
        FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
        AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
        LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
        OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
        SOFTWARE.
License-File: LICENSE
Keywords: artificial intelligence,deep learning,flow policy,robotic foundation model,transformers
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3.9
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Requires-Python: >=3.9
Requires-Dist: einops>=0.8.0
Requires-Dist: einx>=0.3.0
Requires-Dist: rotary-embedding-torch>=0.8.4
Requires-Dist: torch>=2.5
Requires-Dist: torchdiffeq
Requires-Dist: tqdm
Provides-Extra: examples
Provides-Extra: test
Requires-Dist: pytest; extra == 'test'
Description-Content-Type: text/markdown

<img src="./fig3.png" width="400px"></img>

## pi-zero-pytorch (wip)

Implementation of <a href="https://www.physicalintelligence.company/blog/pi0">π₀</a> the robotic foundation model architecture proposed by Physical Intelligence

Summary of this work would be that it is a simplified <a href="https://github.com/lucidrains/transfusion-pytorch">Transfusion</a> (Zhou et al.) with influence from <a href="https://arxiv.org/abs/2403.03206">Stable Diffusion 3</a> (Esser et al.), mainly the adoption of flow matching instead of diffusion for policy generation, as well as the separation of parameters (<a href="https://github.com/lucidrains/mmdit/blob/main/mmdit/mmdit_pytorch.py#L43">Joint Attention</a> from mmDIT). They build on top of a pretrained vision language model in the PaLI configuration with prefixed visual tokens from a ViT to Gemma 2B

## Install

```bash
$ pip install pi-zero-pytorch
```

## Usage

```python
import torch
from pi_zero_pytorch import π0

model = π0(
    dim = 512,
    dim_action_input = 6,
    dim_joint_state = 12,
    num_tokens = 20_000
)

vision = torch.randn(1, 1024, 512)
commands = torch.randint(0, 20_000, (1, 1024))
joint_state = torch.randn(1, 12)
actions = torch.randn(1, 32, 6)

loss, _ = model(vision, commands, joint_state, actions)
loss.backward()

# after much training

sampled_actions = model(vision, commands, joint_state, trajectory_length = 32) # (1, 32, 6)
```

## Citation

```bibtex
@misc{Black2024,
    author  = {Kevin Black, Noah Brown, Danny Driess, Adnan Esmail, Michael Equi, Chelsea Finn, Niccolo Fusai, Lachy Groom, Karol Hausman, Brian Ichter, Szymon Jakubczak, Tim Jones, Liyiming Ke, Sergey Levine, Adrian Li-Bell, Mohith Mothukuri, Suraj Nair, Karl Pertsch, Lucy Xiaoyang Shi, James Tanner, Quan Vuong, Anna Walling, Haohuan Wang, Ury Zhilinsky},
    url     = {https://www.physicalintelligence.company/download/pi0.pdf}
}
```

```bibtex
@inproceedings{Zhou2024ValueRL,
    title   = {Value Residual Learning For Alleviating Attention Concentration In Transformers},
    author  = {Zhanchao Zhou and Tianyi Wu and Zhiyun Jiang and Zhenzhong Lan},
    year    = {2024},
    url     = {https://api.semanticscholar.org/CorpusID:273532030}
}
```

```bibtex
@inproceedings{Yao2024FasterDiTTF,
    title   = {FasterDiT: Towards Faster Diffusion Transformers Training without Architecture Modification},
    author  = {Jingfeng Yao and Wang Cheng and Wenyu Liu and Xinggang Wang},
    year    = {2024},
    url     = {https://api.semanticscholar.org/CorpusID:273346237}
}
```
