Git Product home page Git Product logo

Comments (5)

lucabergamini avatar lucabergamini commented on June 4, 2024 1

Hi @johnwlambert; L5Kit is still an early release so hopefully a lot of what you suggest will be there soon :)

(1) Would it be possible to add the rendered notebooks with output in Github, rather than just the code?

Having the output in GitHub is a little bit tricky, especially if you want to keep them updated (even now jupyter makes some slight changes like the python version or other metadata). Maybe once the examples will get enough mature adding it won't be that cumbersome.

(2) Would you all be able to make a map notebook, to better understand the format, e.g. road vs. lane geometry, like this notebook?

We don't have that granularity yet, but given that those operations are performed in the code we'll do that if enough interest arises.

(3) To what precision is the aerial imagery aligned with the published map? It looks like Page Mill Road's road geometry doesn't align with the aerial map here:

I can't give you an exact number, but we're working to improve that. We're aware semantic and satellite don't match exactly.

(4) I ran into the following bug when installing requirements on Mac when I follow the setup instructions at pip install -r requirements.txt

numcodecs is a zarr dependency and contains, among other, the blosc compressor. As far as I know it's implemented in C++ so it may have some system dependencies. You can maybe try to update pip as the latest version may have a pre-compiled wheel for that

from l5kit.

johnwlambert avatar johnwlambert commented on June 4, 2024

Thanks for the quick reply Luca. I found I was able to push rendered IPython notebooks to Git without much trouble by just saving and committing the file after running all cells.

I found updating my pip installation didn't resolve those installation errors.

from l5kit.

lucabergamini avatar lucabergamini commented on June 4, 2024

I found I was able to push rendered IPython notebooks to Git without much trouble by just saving and committing the file after running all cells

yeah but the issue is not pushing them, is keeping them updated. The output cells are build artefacts, and as such it's not a good idea to upload them unless the notebook is very very stable. When things will get stable, we'll consider it.

I found updating my pip installation didn't resolve those installation errors.

you can try to install only zarr in a new env (pip install zarr). If that doesn't work, you might want to head to the zarr/numcodecs GitHub and ask there for support. zarr can run on mac, so I suspect it might be a system dependencies issue...

from l5kit.

pondruska avatar pondruska commented on June 4, 2024

Hi @johnwlambert, regarding the alignment in the figure:

The figure in the paper was aligned manually in adobe illustrator so there is some alignment gap we will fix in the next update. Thanks for pointing this out! This affects only the paper if you actually run the code rasterizes the two maps - aerial and semantic should be aligned very closely (< 1m error).

from l5kit.

johnwlambert avatar johnwlambert commented on June 4, 2024

That's good to know, thanks @pondruska.

from l5kit.

Related Issues (20)

Recommend Projects

  • React photo React

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

  • Vue.js photo Vue.js

    🖖 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. 📊📈🎉

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 ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.