Fabian Höring

@f_hoering

Software Engineer @ Criteo

Paris, France
fr.linkedin.com/in/fhoering
Vrijeme pridruživanja: svibanj 2017.
1 fotografija ili videozapis Fotografije i videozapisi

Tweetovi

Blokirali ste korisnika/cu @f_hoering

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

  1. proslijedio/la je Tweet
    27. sij

    We're in the middle of migrating one of our biggest apps (5000+ servers) to .net core. In yellow is CPU usage for .net framework + . In green is .net core 3.1 + aspnetcore. We've migrated other apps but it's the first time we see such a dramatic improvement

    19 replies 150 proslijeđenih tweetova 586 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  2. proslijedio/la je Tweet
    17. sij

    Put yourself in Berlin! The call for proposals to speak at the superb 2020 conference is open. Deadline to apply is 16 Feb. Conference in June. The call for submissions link:

    9 proslijeđenih tweetova 12 korisnika označava da im se sviđa
    Poništi
  3. 7. sij

    Blog post from on how they use pyspark with pex

    2 proslijeđena tweeta 7 korisnika označava da im se sviđa
    Poništi
  4. proslijedio/la je Tweet
    23. pro 2019.

    listed as most prolific French engineering blog in 💪Congrats & ! cc: and more

    8 proslijeđenih tweetova 28 korisnika označava da im se sviđa
    Poništi
  5. 20. stu 2019.

    Having access to a Hadoop cluster ? Want to run on it ? It has never been easier to launch a job and keep track of it. tf-yarn now integrates with your server. It tracks default metrics and all container logs. More infos on

    15 proslijeđenih tweetova 20 korisnika označava da im se sviđa
    Poništi
  6. proslijedio/la je Tweet
    3. lis 2019.

    and now opening our Criteo AI Lab first anniversary event! A full day of insighting panels about the future of AI

    17 proslijeđenih tweetova 18 korisnika označava da im se sviđa
    Poništi
  7. proslijedio/la je Tweet
    6. ruj 2019.

    My data are not big... there are huge 🤣 Obviously, the matrix is sparse. I tried the sklearn Logistic Regression in . It works! I trained my first models ... I will now push a couple of small PR in dask-glm to improve the support of sparse matrices.

    5 proslijeđenih tweetova 17 korisnika označava da im se sviđa
    Poništi
  8. proslijedio/la je Tweet
    4. ruj 2019.

    Added an example to Skein showing a simple web application that accepts a Python script and submits to run on YARN on behalf of a user. All of this plugs-in to Hadoop's security model (kerberos, etc...) transparently.

    1 reply 6 proslijeđenih tweetova 23 korisnika označavaju da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  9. 9. kol 2019.

    I made my first PR to I admit it is an easy one but it is also nice that 'pip install -e .' with mlflow master branch works out of the box to quickly test local changes.

    1 proslijeđeni tweet 14 korisnika označava da im se sviđa
    Poništi
  10. proslijedio/la je Tweet
    2. kol 2019.

    We made our first significant PR to ((link: ) …). The UI is up to 3X faster when it lists runs that contain timeseries metrics. Kudos to that made the PR and dbczumar that reviewed and amended it.

    1 reply 8 proslijeđenih tweetova 35 korisnika označava da im se sviđa
    Poništi
  11. proslijedio/la je Tweet
    27. srp 2019.

    That's quite astonishing. If I'm reading this right, PyTorch grew 194% in research use last year (based on arxiv mentions) and TensorFlow only grew 23%, meaning they're now neck and neck. In another year TensorFlow may just be a small minority player in research

    Prikaži ovu nit
    17 replies 178 proslijeđenih tweetova 610 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  12. proslijedio/la je Tweet
    26. srp 2019.

    The 42nd edition of took place in Paris, where we were born, and we couldn’t miss it! From top-notch keynotes to great papers and much more, get an overview of the awesome works we saw👏 Authors:

    8 proslijeđenih tweetova 14 korisnika označava da im se sviđa
    Poništi
  13. 23. srp 2019.

    Seems like one can directly setup a Hadoop3 cluster on Azure with HDInsight. Probably what they are going to do instead of migrating all their Hadoop jobs.

    As we look to scale our infrastructure for the next stage of LinkedIn’s growth, we’ve decided to begin a multi-year migration to the public cloud, specifically Azure. Learn more on our blog.
    1 proslijeđeni tweet
    Poništi
  14. proslijedio/la je Tweet
    11. srp 2019.

    Kickoff this afternoon: It is a go! This summer, my team will set up at a central service. Looking forward to our first pull requests!

    1 reply 6 proslijeđenih tweetova 17 korisnika označava da im se sviđa
    Poništi
  15. proslijedio/la je Tweet
    6. lip 2019.

    In our last article, we tell you more about a Python library we have built for training models on a YARN cluster! Thanks to and all writers who took part in the writing.

    15 proslijeđenih tweetova 30 korisnika označava da im se sviđa
    Poništi
  16. proslijedio/la je Tweet
    22. svi 2019.

    Prochain Tech Meetup à Paris : Data is the new oil, avec Elise ('13) et Marc ('07) sur les métiers de Data Scientist et Data Engineer. 📅 Mardi 4 Juin 📍

    1 reply 2 proslijeđena tweeta 5 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  17. 12. tra 2019.

    We released tf-yarn on pypi. If you want to run Tensorflow on Hadoop have a look at tf-yarn.

    tf-yarn is now on pypi tf-yarn is a library we develop at to run tensorflow on our hadoop cluster
    1 reply 3 proslijeđena tweeta 5 korisnika označava da im se sviđa
    Poništi
  18. proslijedio/la je Tweet
    19. velj 2019.

    Today, let's explore data using pyspark. Then train your first tensorflow DNN on our hadoop cluster.

    I’m consistently amazed by the wealth and breadth of talent we have here. Two days, 40 talks and workshops, all of very high quality. (Promise, I’ll get them to publish and talk at meetups !)
    1 proslijeđeni tweet 4 korisnika označavaju da im se sviđa
    Poništi
  19. proslijedio/la je Tweet
    9. sij 2019.

    📣To the experts out-there, we are offering a crash-course on combining lectures and exercises with ! 📅 20 seats available, apply before January 14th

    15 proslijeđenih tweetova 21 korisnik označava da mu se sviđa
    Poništi
  20. proslijedio/la je Tweet
    8. sij 2019.

    How to package code with Pex? Well, be prepared for two good news 🙌 ▶ managed to do so using ▶ he is sharing his experience on the matter directly with you ! 👉

    3 proslijeđena tweeta 10 korisnika označava da im se sviđa
    Poništi

@f_hoering još nije objavio nijedan Tweet.

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:

    ·