Malte Pietsch

@malte_pietsch

Co-Founder of deepset | NLP Engineer

Berlin, Germany
deepset.ai
Vrijeme pridruživanja: travanj 2010.
2 fotografije ili videozapisa Fotografije i videozapisi

Tweetovi

Blokirali ste korisnika/cu @malte_pietsch

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

  1. proslijedio/la je Tweet
    3. velj

    is more than just public code. It's a mindset of sharing, being transparent and collaborating across organizations. It's about building on the shoulders of other projects and advancing together the state of technology (1/N) , , ,

    1 reply 9 proslijeđenih tweetova 38 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  2. proslijedio/la je Tweet
    28. sij

    It's challenging to keep track of all the latest out there. What was again the difference between and ? What's the core idea behind ? Here's a little (not comprehensive) that we use for workshops

    1 reply 14 proslijeđenih tweetova 41 korisnik označava da mu se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  3. proslijedio/la je Tweet
    20. sij

    GitHub Repo Spotlight №3: Transfer Learning library for NLP called FARM:: With FARM you easily use BERT, XLNet, and others easily for any downstream NLP tasks. FARM is great for fast prototyping too.

    1 reply 3 proslijeđena tweeta 11 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  4. 22. sij

    Just migrated our basic inference API in from to . Really love how simple it is! Looking forward to a deeper integration. Great work ! FastAPI: Haystack:

    1 reply 9 korisnika označava da im se sviđa
    Poništi
  5. proslijedio/la je Tweet
    17. sij

    Are you doing in a non-english language? Try the multilingual XLM-R model! It gave us amazing results in German (for the SOTA chasers: yes, it's also outperforming previous results with BERT & Co). Blog:

    5 proslijeđenih tweetova 14 korisnika označava da im se sviđa
    Poništi
  6. proslijedio/la je Tweet
    15. sij

    v1.4: customizable mobile builds, Distributed Model Parallelism via experimental RPC API, Java Bindings, Chaining LRSchedulers Summary: Release Notes: Last release for Python 2 (bye bye!)

    1 reply 148 proslijeđenih tweetova 378 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  7. proslijedio/la je Tweet
    16. sij

    Introducing Reformer, an efficiency optimized architecture, based on the Transformer model for language understanding, that can handle context windows of up to 1 million words, all on a single accelerator with only 16GB of memory. Read all about it ↓

    14 replies 474 proslijeđena tweeta 1.353 korisnika označavaju da im se sviđa
    Poništi
  8. proslijedio/la je Tweet
    13. pro 2019.

    How big is the CO₂ bucket for 1.5°C? Well, the bucket is about to overflow in a few short years, unless we: 1. Turn off the tap (urgently) 2. Put a hole in the bottom to remove CO₂ (negative emissions)

    548 replies 12.329 proslijeđenih tweetova 15.801 korisnik označava da mu se sviđa
    Poništi
  9. proslijedio/la je Tweet
    15. pro 2019.

    As promised: here are the slides from Malte's talks in Warsaw! - Keynote at : - Talk at HumanTech: Reach out to us if you have some large polish text data set (> 10GB) and want to train a polish BERT or ALBERT.

    2 proslijeđena tweeta 4 korisnika označavaju da im se sviđa
    Poništi
  10. 4. pro 2019.

    Honoured to give a keynote at next week about ! Looking forward to the conference, the people and the city. Let me know, if you are also there and want to grab a coffee.

    Fascinating lectures🧠🤖, wonderful speakers 🗣- all during two days full of knowledge and inspiration‼️Available you want to know what we have prepared for you, read our schedule for the conference.📥📆
    1 proslijeđeni tweet 1 korisnik označava da mu se sviđa
    Poništi
  11. proslijedio/la je Tweet
    28. stu 2019.

    Excited to announce two releases with a common theme: Bringing to the industry! 1) : We have a new framework joining the family! Focus: all you need for at scale: indexing docs, retrievers, labeling ...!

    1 reply 4 proslijeđena tweeta 13 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  12. 18. stu 2019.

    It was really a pleasure to speak (again) at in Berlin and see so many new faces joining the community. As many of you asked me about slides and the framework. Here they are: Slides: FARM framework:

    3 proslijeđena tweeta 9 korisnika označava da im se sviđa
    Poništi
  13. proslijedio/la je Tweet
    5. stu 2019.

    We're releasing the 1.5billion parameter GPT-2 model as part of our staged release publication strategy. - GPT-2 output detection model: - Research from partners on potential malicious uses: - More details:

    60 replies 706 proslijeđenih tweetova 1.791 korisnik označava da mu se sviđa
    Poništi
  14. proslijedio/la je Tweet
    28. lis 2019.

    We just released 0.3.0 🎉 with two new language models: and are joining the .

    2 proslijeđena tweeta 9 korisnika označava da im se sviđa
    Poništi
  15. proslijedio/la je Tweet
    18. lis 2019.

    Great to see our opensource resources being used at the competition: - Winner Task 2.1 & 2.2: Used GermanBERT (UPB, ) - Winner Task 2.3: Used FARM & GermanBERT ( , )

    1 proslijeđeni tweet 10 korisnika označava da im se sviđa
    Poništi
  16. proslijedio/la je Tweet
    10. lis 2019.

    We just released 0.2.1 🎉. Another major step towards simpler for the industry! - Better parallelization of preprocessing (less RAM) - Multilabel classification - Simpler multitask learning - Upgrading to transformers 2.0

    1 proslijeđeni tweet 9 korisnika označava da im se sviđa
    Poništi
  17. proslijedio/la je Tweet
    8. lis 2019.

    This is a nice diagram by Zhengyan Zhang and that shows how many recent pretrained language models are connected. The GitHub repo contains a full list of relevant papers:

    6 replies 311 proslijeđenih tweetova 781 korisnik označava da mu se sviđa
    Poništi
  18. proslijedio/la je Tweet
    30. kol 2019.

    Hear Malte Pietsch talk about "The Imagenet Moment for NLP: Tips & Tricks for Effective Transfer Learning" at Berlin 2019!

    1 proslijeđeni tweet 1 korisnik označava da mu se sviđa
    Poništi
  19. proslijedio/la je Tweet
    13. ruj 2019.

    It's great to see the growing landscape of NLP transfer learning libraries: - pytorch-transformers by : - spacy-pytorch-transformers by : - FARM by

    5 replies 153 proslijeđena tweeta 507 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  20. proslijedio/la je Tweet
    19. kol 2019.

    We just released FARM 0.2.0 - making faster & easier: - 🚀 Significant speed up of preprocessing & training - 🙂 More user friendly processors for custom datasets - 🐞 Several bug fixes Check it out:

    3 proslijeđena tweeta 5 korisnika označava da im se sviđa
    Poništi

@malte_pietsch 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:

    ·