Coder Social home page Coder Social logo

robocook's People

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Watchers

 avatar  avatar  avatar  avatar

robocook's Issues

ModuleNotFoundError: No module named 'rosbag'

Hello, Thanks for your great work.
I followed the instructions to activate the environment. But when I ran the .sh file, e.g. "bash scripts/perception/perception.sh", there was an error "ModuleNotFoundError: No module named 'rosbag'". And I didn't see the rosbag in the environment.yml.

I tried some ways to install the rosbag and ros_numpy packages, but all my attempts failed. Could you please tell me what the correct way to install these two packages?

Question regarding the purpose of the `memory` input to the DynamicsPredictor

Hi Haochen,

Thank you for open-sourcing your work. I have a question regarding the memory input to the DynamicsPredictor. This memory is treated as the attribute of each particle. However, it seems that during training, this memory is always a zero tensor. Could you please explain the intended purpose and functionality of this memory input?

Thank you very much in advance!

Question about the Role of GNN-Based Dynamics Model during Inference

Thank you for open-sourcing this great work! I would like to ask you about the role of the GNN-based dynamics model outlined in your paper.

In the paper's introduction, you mentioned "we combine a PointNet-based tool classification module with learned dynamics models to determine the most appropriate tool to use at the current task stage", while in Section 6.1, it appears that both the Tool Predictor and Policy Network operate based on the defined target or sub-target point clouds, rather than on predictions from the dynamics model. Thus, I'm a little bit confused about the usage of this GNN-based dynamics model during deployment and wonder if I might have misunderstood the implementation details.

I appreciate your precious time and assistance in clarifying this!

Regarding the 3D models of the robot tools

Hello Haochen,
Firstly, thank you very much for contributing such interesting work to the community. I am working on a paper on robot learning in the kitchen domain, and I found the tools that you designed for the robot being very useful. We want to use a similar tool design as yours in our paper. However, I could only find the 3D model of 8 of the 15 tools of your paper. Additionally, the holder part of the rollers and pressing tools do not present in the stl files. I wonder if it is possible for you to also open source those 3D models.

Your help is much appreciated!

License

Hello, would you mind adding a license to this project?

Thank you!

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.