GithubHelp home page GithubHelp logo

abdo-eldesokey / pncnn Goto Github PK

View Code? Open in Web Editor NEW
91.0 91.0 14.0 114.38 MB

The official PyTorch implementation for "Uncertainty-Aware CNNs for Depth Completion: Uncertainty from Beginning to End"

License: GNU General Public License v3.0

Python 100.00%

pncnn's People

Contributors

abdo-eldesokey avatar

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Watchers

 avatar  avatar

pncnn's Issues

About output images

I used the pretrained model on validation of KITTI depth dataset. I changed the parameter to save the output images but the images are all black. Is there some problems on saving?
Thank you.

About the uncertainty

Hi, I appreciate your work very much, but I have question on understanding. I want to know why the uncertainty of y can be estimated as the cov(y) in your paper on page 5 left column , formula (7) Thanks!

RuntimeError for model=f.CNN().to(device)

Hi, Thanks a lot for releasing the code. It is a very interesting work. I am trying to run it with the default parameters in workspace/kitti/pncnn, and got a following error at this line in main function model=f.CNN().to(device). May I ask for any suggestions? Many thanks.
image

How to evaluate your pretrained model?

Hey,

I simply want to run your pretrained model. How is that possible? What I have done:
I used your Kitti download script and modified the download path.
I run:

  1. "python main.py --evaluate workspace/kitti/pncnn/model_best.pth.tar"
  2. "python main.py --evaluate workspace/kitti/pncnn/checkpoint-9.pth.tar"
  3. "python main.py --evaluate workspace/kitti/pncnn/args.json" (with adjusting the path for the data)

all of them bring me the following message:
File "main.py", line 290
print('[Train] Epoch ({}) [{}/{}]: '.format(epoch, i+1, len(dataloader)), end='')
SyntaxError: invalid syntax

Why the error? Seems like it jumps in the training routine..? Could you pls give instructions how to run your code with your pretrained model? What exactly to choose as checkpoint path? Do I have to modify any file except the datadownloader (e.g. the args.json as well) ?

Thanks

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

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.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.