Comments (5)
Hi @anahrendra , thanks for this note. You're right that the implementation and paper differ slightly, and I didn't catch this before! The code is correct, the paper contains a mistake here.
I think the effect of this should be a constant scaling of the reward function, since these two reward terms are added together at every timestep:
-Gabe
from walk-these-ways.
Hi! Thanks for your implementation. That makes sense.
However, I tried to run a training with your code (nothing is changed at all), but I could not obtain the results as what you have in the pretrained weight. A bit of summary is as follows:
- 4000 iterations, still able to walk, but cannot track given gaits well
- 45000 iterations, completely fail to even hold its position.
Could you by any chance directly train using your github code to verify if there is something missing? And how many iterations are required to produce a good result?
Thanks in advance for your help!
from walk-these-ways.
@anahrendra , I did some local testing and I suspect this is because of the larger gravity range in this repo's default config: https://github.com/Improbable-AI/walk-these-ways/blob/master/scripts/train.py#L49
In the paper, we used max gravity randomization 1.0
but I seem to have provided the config with more challenging max gravity randomization of 2.0
here. Training with this range might be unstable. Sorry about that!
Please try changing line 49 of train.py to:
Cfg.domain_rand.gravity_range = [-1.0, 1.0]
And let me know if that fixes the issue. On my machine, it converges after around 10k iterations.
By the way, the friction range in this codebase is also a bit wider than in the paper, and the max footswing height is higher. See appendix Table 6 https://arxiv.org/pdf/2212.03238.pdf for the values used in the paper. I may just update the repo in a bit to have all the original parameters
P.S. to debug with a bit faster convergence and verify that everything else is working properly, you can try turning off gravity randomization entirely ( Cfg.domain_rand.gravity_range = [-0.0, 0.0]
).
from walk-these-ways.
Hi!
Thanks a lot for your detailed support. I will try your fix as soon as possible and let you know about the results.
Have a nice weekend!
from walk-these-ways.
I have updated the default parameters in train.py
(728058d)
from walk-these-ways.
Related Issues (20)
- Generating your own actuator network training data HOT 1
- Is there a specific reason not to run controller on pi instead of Nano? HOT 1
- How to run a newly trained network on the Go1? HOT 3
- Output test.py HOT 1
- How to get joint torque values from StateEstimator? Maybe tau_est? HOT 6
- How to make it automatically run the repo's NN rather than the original on boot? HOT 3
- How to deploy this code on custom model? HOT 1
- How to create data to retrain the actuator net? HOT 1
- Is Kp = 20 too soft? HOT 11
- Velocity estimation HOT 4
- decimation of 4 seems to create some instability have you been successful with lower values like 2? HOT 9
- Can't find docker image after sudo ../installer/install_deployment_code.sh HOT 1
- When deployed, Go1 still moves despite neutral commands are given HOT 1
- When recording data using L2 button, generated data does not contain "tau_est" HOT 4
- Code not running on Untree A1 robot HOT 9
- I have confusion about foot clearance. HOT 3
- open /var/lib/docker/tmp/docker-import-221834566/repositories: no such file or directory HOT 2
- ./lcm_position: error while loading shared libraries: liblcm.so.1: cannot open shared object file: No such file or directory HOT 4
- ValueError invalid values HOT 1
- When we change some terrain in this situation why robots are start at sky? HOT 2
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
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.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from walk-these-ways.