humancompatibleai / overcooked-demo Goto Github PK
View Code? Open in Web Editor NEWWeb application where humans can play Overcooked with AI agents.
Web application where humans can play Overcooked with AI agents.
I've been trying to run a first build of the demo but I get this error:
=> => naming to docker.io/library/overcooked-demo_app 0.0s
Use 'docker scan' to run Snyk tests against images to find vulnerabilities and learn how to fix them
[+] Running 2/2
⠿ Network overcooked-demo_default Created 0.1s
⠿ Container overcooked-demo-app-1 Created 0.1s
Attaching to overcooked-demo-app-1
Error response from daemon: driver failed programming external connectivity on endpoint overcooked-demo-app-1 (4bcc12b01a8dfbf43e45318ecaaa79651e64a6aa7f483b3be00abb1738c42272): Error starting userland proxy: listen tcp4 0.0.0.0:80: bind: address already in use
The lsof -i :80
returns nothing.
I'm running in a dual boot with windows 10 and Ubuntu Focal 20.04
Client: Docker Engine - Community
Version: 20.10.9
API version: 1.41
Go version: go1.16.8
Git commit: c2ea9bc
Built: Mon Oct 4 16:08:29 2021
OS/Arch: linux/amd64
Context: default
Experimental: true
Server: Docker Engine - Community
Engine:
Version: 20.10.9
API version: 1.41 (minimum version 1.12)
Go version: go1.16.8
Git commit: 79ea9d3
Built: Mon Oct 4 16:06:37 2021
OS/Arch: linux/amd64
Experimental: false
containerd:
Version: 1.6.9
GitCommit: 1c90a442489720eec95342e1789ee8a5e1b9536f
runc:
Version: 1.1.4
GitCommit: v1.1.4-0-g5fd4c4d
docker-init:
Version: 0.19.0
GitCommit: de40ad0
I've tried restart the PC and the docker service, to no avail.
Thank you for your help.
Can help prevent mistakes caused by thinking one is playing with an agent but it's actually another.
Moved issue to overcooked_ai repo
Currently can load a agent for layout X into layout Y. Should update the agent to have a set of layouts for which it can be deployed, and only add agents to the drop down menus conditional on the layout selected.
Alternatively, just do as suggested in #4
Update layout names for web models, and clarify point about how to update agents
I tried to build this demo from Mac computer with macOS Catalina Version 10.15.7, and it failed. The message says:
Starting demo_app_1 ... done
Attaching to demo_app_1
app_1 | Traceback (most recent call last):
app_1 | File "app.py", line 12, in
app_1 | from flask import Flask, render_template, jsonify, request
app_1 | File "/usr/local/lib/python3.7/site-packages/flask/init.py", line 14, in
app_1 | from jinja2 import escape
app_1 | ImportError: cannot import name 'escape' from 'jinja2' (/usr/local/lib/python3.7/site-packages/jinja2/init.py)
demo_app_1 exited with code 1
Is it caused by OS or some package dependencies? Thanks!
I am trying to find a way to make human-human play possible as well, any helpful suggestions on how to make this happen?
Updating gh-pages requires to manually go through all changes imported to it to change "assets/xxx" mentions to "static/assets/xxx" mentions. We should find a way around this.
Hi,
I would like to train my BC model. How could I save trajectories in human-human mode?
Thanks
So, I have been attempting to make this repository work with the main overcooked_ai one and I have run into some issues. For reference, I was attempting to do the Overcooked-Demo-Legacy installation as I need and prefer the .js backend.
Given the previous errors ("no such file"), I figure it had something to do with a missing "cd overcooked-demo-legacy" rather than "cd overcooked-demo" only - so you should add this to your README file. Because once you go into that folder, then you can use 'npm install'. It then gives me this:
How should I proceed?
I tried following this guide: https://github.com/HumanCompatibleAI/overcooked-demo/tree/master/server/static/assets/agents/RllibSelfPlay_CrampedRoom to add my pretrained rllib agent and play with it, however that didn't work. Trying to lead the already given example doesn't work either as shown in attached picture. Is there an ray compatibility issue, as I saw when the tutorial was pushed into this repository, the corresponding ray version in human aware rl repository was lower, is that the issue?
Thank you for the codebase. I'm trying to use a pretrained rllib agent to test the demo / environment but I didn't seem to find a proper one in the repo. Are there these model checkpoints available?
Can replicate by using test_traj.json and scrolling back in the replay a couple of times
Hi, I'm trying to use the 'legacy' branch of this repo to visualize the agents trained using the code in the 'neurips2019' branch of 'human_aware_rl' (this)
However, when I'm following the installation instructions in README, I got some errors.
The errors were raised when I'm trying to run npm run build
.
/home/pc/overcooked-demo/static/js/demo/index.js:1
import $ from "jquery";
^
ParseError: 'import' and 'export' may appear only with 'sourceType: module'/home/pc/overcooked-demo/static/js/demo/replay-index.js:1
import $ from "jquery";
^
ParseError: 'import' and 'export' may appear only with 'sourceType: module'
Any idea how to solve this?
This would enable it to be also used also from the web demo
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.