Comments (8)
@Syzygianinfern0: Urban Driver (like all other learned methods) is great at open-loop ego-forecasting, which is the primary focus in the evaluations in Marcel's earlier paper https://arxiv.org/abs/2302.07753
It is the same in our paper: the OLS (Open Loop Score) metric of Urban Driver is 76, compared to 38 for IDM (Table 2). See Section 4 for an explanation of why the CLS and OLS metrics are not aligned.
@LarryZhangy: we are still in the process of finalizing the contribution guide, please be patient and we will release it soon! I don't have a concrete suggestion for which paper is worth most to implement. Some interesting candidates could be PlanT (https://www.katrinrenz.de/plant/) or any of the nuPlan challenge submissions that placed 2nd to 4th (https://opendrivelab.com/AD23Challenge.html#nuplan_planning)
from tuplan_garage.
Thanks for your interest.
We plan to welcome community contributions to nuPlan garage for new results on the Val14 benchmark and the corresponding planner code. We want to make contributing to this project easy and transparent, so we will create a contribution guide after we have released our pre-trained models. The first release (for PDM-Open, PDM-Offset and PDM-Hybrid) is scheduled for this week.
from tuplan_garage.
Thanks for your response.
I'm very interested in contributing to this project and looking forward to seeing the contribution guidelines!
I wonder if there are any candidate papers for the Val14 benchmark ?
from tuplan_garage.
From https://arxiv.org/abs/2302.07753, it looks like Urban Driver has much better performance metrics as compared to IDM. But it is not that way in your paper. Is there a reason to it?
from tuplan_garage.
I want to contribute to this project, but i dont know which algorithm(from some paper) is worth to be Implemented.
from tuplan_garage.
Is there an estimated timeline on the visualization scripts?
from tuplan_garage.
@Syzygianinfern0 Adding the visualization scripts is not our highest priority at the moment. However, you can visualize our planners using the nuBoard from nuPlan’s devkit.
from tuplan_garage.
Update:
We just merged the initial version of our contribution guidelines (see here)
Hence, I am closing this issue.
Feel free to reopen if the guidelines do not answer all of your questions.
from tuplan_garage.
Related Issues (20)
- Error when trying to run pgp model with nuplan config: Metric target: "multimodal_trajectories" is not in model computed targets! HOT 4
- Questions about NuPlan to Use HOT 2
- Updated checkpoints for urban driver and PDM open missing HOT 7
- ROS bridge for real vehicle implementation HOT 5
- some questions about nuplan dataset HOT 5
- Callbacks'module problem before the start of training GC-PGP HOT 3
- Some question about "Val14" evaluation HOT 3
- Some errors feedback HOT 2
- AssertionError: Class to be of type <class 'pytorch_lightning.callbacks.base.Callback'>, but is <class 'omegaconf.dictconfig.DictConfig'>! HOT 2
- pdm_closed_planner trajectory's states have no dynamic information HOT 3
- How to accelerate simulation process HOT 1
- PDM-Hybrid that combined PDM-Closed with GC-PGP HOT 2
- Inconsistency between centerline and PDM-Closed trajectory HOT 4
- accelerate simulation by cuda HOT 3
- calculate inference time HOT 3
- time for training HOT 1
- Centerline and States encoding for PDM_Open and PDM_Offset models HOT 4
- Val14 Dataset HOT 1
- Visualization scripts release HOT 1
- L2 computation issue in open-loop settings HOT 1
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 tuplan_garage.