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

for evaluation #10

Closed
Huskie377 opened this issue Nov 2, 2023 · 5 comments
Closed

for evaluation #10

Huskie377 opened this issue Nov 2, 2023 · 5 comments

Comments

@Huskie377
Copy link

Huskie377 commented Nov 2, 2023

i have problem about the below code
python evaluation/evaluate_depth.py --config <PATH_TO_MD4ALL>/config/eval_md4allDDa_50m_robotcar_test.yaml
how can i get the GT to evaluation? the Datasets->RobotCar in step 5 has shown the way ,but i fail to get the GT,it was complex.
I want the GT to run the evaluate_depth.py,could you share the link that i can download the GT? Thankyou !

@sgasperini
Copy link
Contributor

Hi @Huskie377,

Thank you for your interest in our work.

As indicated in #5, the ground truth is obtained by running the script data/robotcar/precompute_depth_gt.py. Can you be more specific about why this step goes wrong?

Thanks

@Huskie377
Copy link
Author

I added robotcar_dataset_sdk in the following directory so that precompute_depth_gt.py calls the package in robotcar_dataset_sdk. But when I run it, I have the following diagram error. Is that why I use windows?
屏幕截图 2023-11-02 163517
屏幕截图 2023-11-02 163314

@morbi25
Copy link
Contributor

morbi25 commented Nov 2, 2023

Hello @Huskie377,

Currently, I assume that this ModuleNotFoundError is related to a missing configuration of the PYTHONPATH when directly including the robotcar_dataset_sdk as a subfolder in your md4all-main project.

However, to simplify the incorporation of the robotcar_dataset_sdk, I would recommend using the pip package we created. This should have already been included with the installation of the requirements. Maybe something went wrong in this step. :)

To resolve your problem, you can try the following steps below:

  1. Check if the robotcar_dataset_sdk is correctly installed in your environment by checking whether 'robotcar_dataset_sdk' is contained in the pip package list executing:
    pip list
  2. If not, please install the pip package in your environment using the following command:
    pip install git+https://github.com/morbi25/robotcar-dataset-sdk
  3. If the installation succeeded, you can delete the robotcar_dataset_sdk folder from your project md4all-main again.

In general, to precompute the ground truth depth, you can use the following command:

python data/robotcar/precompute_depth_gt.py --dataroot <PATH_TO_ROBOTCAR_ROOT_DIR> --scenes 2014-12-09-13-21-02 2014-12-16-18-44-24 --mode val test

Please make sure to replace <PATH_TO_ROBOTCAR_ROOT_DIR> with the location of the RobotCar dataset on your local drive.

I hope these steps help you to resolve the problem; do not hesitate to reach out again if you have further questions! Have a nice day! :)

@sgasperini
Copy link
Contributor

Hi, I am closing this for inactivity.
If you are still experiencing issues after following the suggestions, feel free to reopen the issue and let us know.
Thanks

@Huskie377
Copy link
Author

Huskie377 commented Nov 15, 2023

I am so sorry to bother you again,when i use pip to install the package ,i solve the previous problems.

But when i run the code on my own env 'python data/robotcar/precompute_depth_gt.py --dataroot D:\md4all-main\robotcar --scenes 2014-12-16-18-44-24 --mode test' ,the error is
'OSError: Could not find scan files for given time range in directory D:\md4all-main\robotcar\2014-12-16-18-44-24\lms_front'
I have download 2014-12-16-18-44-24_lms_front01 02 03 04 05 06 and put them in the file structure as you show in the github.I have prepare the file structure like you show .Because I didn't have enough memory to download all the stereo left images, I only downloaded a portion of the left images.Is the cause of the above error due to the absence of the precompute_pose_gt.py run file? I really lack the knowledge to generate GT from this dataset. Hope to get your reply to solve this problem.
This is my file structure:
123

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

3 participants