DeepLabCut

@DeepLabCut

Markerless 3D Pose Estimation | Animal Tracking | Deep Learning Toolbox | Free & Open Source Code |

Vrijeme pridruživanja: veljača 2019.

Tweetovi

Blokirali ste korisnika/cu @DeepLabCut

Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @DeepLabCut

  1. Prikvačeni tweet
    14. stu 2019.

    Hi new followers, thanks for joining! DeepLabCut allows for efficient measurement of animal behaviors! Using deep learning, we provide a toolbox to generate custom neural networks for your scientific needs - free, open source:

    Poništi
  2. 2. velj

    : if you use the latest macOS Catania, anaconda is likely broken... check tips here:

    Poništi
  3. proslijedio/la je Tweet
    31. sij

    The full video annotated by the model. I took the video on an iPhone at 30fps. Seems to handle a dark brown/black horse quite well👇

    Prikaži ovu nit
    Poništi
  4. 31. sij

    2020 workshop materials are posted! 🎉👀 Check out our slides that cover important considerations from dataset curation, network creation to analysis:

    Poništi
  5. 31. sij

    Our DLC Hackathon application closes tomorrow (midnight anywhere on earth!) The focus will be on real-time applications: integration with software/hardware such as , , and others! Please chip in and have a say in the future of DLC!

    Poništi
  6. 29. sij

    CZI-DLC Fellow Jessy will be speaking about the latest updates to !

    Poništi
  7. proslijedio/la je Tweet
    14. sij

    💜 users, if you haven't already, we would love 2 things: (1) 2 min to fill out this form to let us know you're using DLC and give us feedback (2) consider giving a star ⭐️ on GitHub: Both metrics help us justify funding 4 DLC! 💜🙏

    Poništi
  8. proslijedio/la je Tweet
    22. pro 2019.

    “Break” day1: Slept for 15hrs and finally got to try out !!

    Poništi
  9. 22. sij

    >> commercial solutions! 🙌 awesome benchmarking, new post-DLC tools, plus best trailer video of all ...👇 (volume on!) 🤟

    Poništi
  10. proslijedio/la je Tweet
    22. stu 2019.

    : all the core functions of the Python package are documented in our user guide! 👇 a great place to start if you’re starting out with !

    Poništi
  11. proslijedio/la je Tweet
    14. sij

    🎉DLC 2.1.5 is up!! Minor but useful changes, including new GUI functions, faster dataset creation, & our 1st release with new CZI DLC Fellow, Jessy! Read more here: It's backwards compatible, so don't be afraid to upgrade! pip install deeplabcut==2.1.5 😎

    Prikaži ovu nit
    Poništi
  12. proslijedio/la je Tweet
    18. sij

    Thank you , and for hosting (for free!) this incredible workshop . Can't wait to track more videos! – mjesto: The Rowland Institute at Harvard

    Poništi
  13. proslijedio/la je Tweet
    18. sij

    It was a wonderful and intense few days at the workshop (second annual!). 💜 So amazed at all the progress students made. We are thankful to the group for supporting 🙌🏻, and to for the funding support!

    Poništi
  14. proslijedio/la je Tweet
    18. sij

    Heading back this chilly Boston morning from the wonderful workshop put on by . Very excited to add this powerful tool to our research!

    Poništi
  15. proslijedio/la je Tweet
    17. sij

    Me trying to use before the workshop...

    Prikaži ovu nit
    Poništi
  16. 18. sij

    And a little bump: inside your conda env or Docker run: pip install deeplabcut==2.1.5.1 :) includes a bug fix which only effected training via project GUI not run on MacOS (oh the joys of multi-OS support 😉)

    Prikaži ovu nit
    Poništi
  17. proslijedio/la je Tweet
    17. sij

    Update: I'm making progress! 🎉 Thank you & for hosting the workshop! By far one of the most informative and productive workshops I've attended

    Poništi
  18. proslijedio/la je Tweet
    17. sij

    Super useful training course on deep learning software for markerless video tracking. No more weeks of manual tracking! – mjesto: Rowland Institute

    Prikaži ovu nit
    Poništi
  19. proslijedio/la je Tweet
    17. sij

    Another : badly tracked videos can be added to the dataset and manually re-labelled to re-train . Data refinement is a painless and quick process and the new GUI is quite great! – mjesto: Rowland Institute

    Poništi
  20. 17. sij
    Poništi
  21. 14. sij

    💜 users, if you haven't already, we would love 2 things: (1) 2 min to fill out this form to let us know you're using DLC and give us feedback (2) consider giving a star ⭐️ on GitHub: Both metrics help us justify funding 4 DLC! 💜🙏

    Poništi

Čini se da učitavanje traje već neko vrijeme.

Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.

    Možda bi vam se svidjelo i ovo:

    ·