Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

The purpose of generating training data and why it's so slow? #11

Open
zhanghm1995 opened this issue Nov 19, 2023 · 2 comments
Open

The purpose of generating training data and why it's so slow? #11

zhanghm1995 opened this issue Nov 19, 2023 · 2 comments

Comments

@zhanghm1995
Copy link

Hi, thanks for opening this great work!

I wonder why the command to generate training data in step 1 is almost the same as the Evaluation process? And why the step for generating training data is so slow, and it will evaluate the metrics like the below for the nuScenes dataset?
image

Are there any ideas to accelerate the training data generation step?

@zhanghm1995 zhanghm1995 changed the title The purpose of generate training data and why it's so slow? The purpose of generating training data and why it's so slow? Nov 19, 2023
@AronDiSc
Copy link
Collaborator

Hi @zhanghm1995, thank you for your interest in our work.

Generating the data takes so long because we use a larger covisibility graph than the one used during inference to achieve slightly better pose estimates. This is specified through the --r3d3_graph_type=droid_slam parameter. Furthermore, the pre-computed cost volumes of the larger graph do not fit into (our) gpu-memory, thus we use the "memory efficient" approach (--r3d3_corr_impl=lowmem) as proposed by the authors of RAFT, which is a bit slower. To speed up the process you could also remove the two parameters and resort to the same covisibility graph used during inference.

The metrics are generated because Lidar information is available for NuScenes. We do not make use of those metrics during dataset generation. You could easily turn it off in the config files and use it for other datasets where you might not have Lidar available.

@zhanghm1995
Copy link
Author

Hi, @AronDiSc , thanks for your patient reply. It helps me a lot.

Since I'm not familiar with this framework, I want to know more details about this repo.

  1. If I remove the --r3d3_corr_impl=lowmem parameter and I have enough GPU memory to generate the data, could it cause performance degradation?

  2. If I don't want to train all training sets in the nuScenes dataset, I just try to overfit R3D3 in one scene, would it be reasonable to reproduce your results? I have already done this experiment in scene-1094, I got the Loss and metric like below, I'm not sure it's ok or not.
    image

  3. How to visualize the inference results like yours in your paper?
    image

  4. How to resume the checkpoint in the training stage.

Thanks in advance.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants