Skip to content
By using Twitter’s services you agree to our Cookies Use. We and our partners operate globally and use cookies, including for analytics, personalisation, and ads.
  • Home Home Home, current page.
  • About

Saved searches

  • Remove
  • In this conversation
    Verified accountProtected Tweets @
Suggested users
  • Verified accountProtected Tweets @
  • Verified accountProtected Tweets @
  • Language: English
    • Bahasa Indonesia
    • Bahasa Melayu
    • Català
    • Čeština
    • Dansk
    • Deutsch
    • English UK
    • Español
    • Filipino
    • Français
    • Hrvatski
    • Italiano
    • Magyar
    • Nederlands
    • Norsk
    • Polski
    • Português
    • Română
    • Slovenčina
    • Suomi
    • Svenska
    • Tiếng Việt
    • Türkçe
    • Ελληνικά
    • Български език
    • Русский
    • Српски
    • Українська мова
    • עִבְרִית
    • العربية
    • فارسی
    • मराठी
    • हिन्दी
    • বাংলা
    • ગુજરાતી
    • தமிழ்
    • ಕನ್ನಡ
    • ภาษาไทย
    • 한국어
    • 日本語
    • 简体中文
    • 繁體中文
  • Have an account? Log in
    Have an account?
    · Forgot password?
    New to Twitter?
    Sign up
fchollet's profile
François Chollet
François Chollet
François Chollet
Verified account
@fchollet

Tweets

François CholletVerified account

@fchollet

Deep learning @google. Creator of Keras. Author of 'Deep Learning with Python'. Opinions are my own.

United States
fchollet.com
Joined August 2009

Tweets

  • © 2021 Twitter
  • About
  • Help Center
  • Terms
  • Privacy policy
  • Cookies
  • Ads info
Dismiss
Previous
Next

Go to a person's profile

Saved searches

  • Remove
  • In this conversation
    Verified accountProtected Tweets @
Suggested users
  • Verified accountProtected Tweets @
  • Verified accountProtected Tweets @

Promote this Tweet

Block

  • Tweet with a location

    You can add location information to your Tweets, such as your city or precise location, from the web and via third-party applications. You always have the option to delete your Tweet location history. Learn more

    Your lists

    Create a new list

    Under 100 characters, optional
    Privacy

    Copy link to Tweet

    Embed this Tweet

    Embed this Video

    Add this Tweet to your website by copying the code below. Learn more

    Add this video to your website by copying the code below. Learn more

    Hmm, there was a problem reaching the server.

    By embedding Twitter content in your website or app, you are agreeing to the Twitter Developer Agreement and Developer Policy.

    Preview

    Why you're seeing this ad

    Log in to Twitter

    · Forgot password?
    Don't have an account? Sign up »

    Sign up for Twitter

    Not on Twitter? Sign up, tune into the things you care about, and get updates as they happen.

    Sign up
    Have an account? Log in »

    Two-way (sending and receiving) short codes:

    Country Code For customers of
    United States 40404 (any)
    Canada 21212 (any)
    United Kingdom 86444 Vodafone, Orange, 3, O2
    Brazil 40404 Nextel, TIM
    Haiti 40404 Digicel, Voila
    Ireland 51210 Vodafone, O2
    India 53000 Bharti Airtel, Videocon, Reliance
    Indonesia 89887 AXIS, 3, Telkomsel, Indosat, XL Axiata
    Italy 4880804 Wind
    3424486444 Vodafone
    » See SMS short codes for other countries

    Confirmation

     

    Welcome home!

    This timeline is where you’ll spend most of your time, getting instant updates about what matters to you.

    Tweets not working for you?

    Hover over the profile pic and click the Following button to unfollow any account.

    Say a lot with a little

    When you see a Tweet you love, tap the heart — it lets the person who wrote it know you shared the love.

    Spread the word

    The fastest way to share someone else’s Tweet with your followers is with a Retweet. Tap the icon to send it instantly.

    Join the conversation

    Add your thoughts about any Tweet with a Reply. Find a topic you’re passionate about, and jump right in.

    Learn the latest

    Get instant insight into what people are talking about now.

    Get more of what you love

    Follow more accounts to get instant updates about topics you care about.

    Find what's happening

    See the latest conversations about any topic instantly.

    Never miss a Moment

    Catch up instantly on the best stories happening as they unfold.

    1. hardmaru‏ @hardmaru 8 Feb 2018

      hardmaru Retweeted Denny Britz

      Implementing fully connected nets, convnets, RNNs, backprop and SGD from scratch (using pure python, numpy, or even JS) and training these models on small datasets is a great way to learn how neural nets work. Invest time to gain valuable intuition before jumping onto frameworks.https://twitter.com/dennybritz/status/961829329985400839 …

      hardmaru added,

      Denny Britz @dennybritz
      A group of experienced researchers trying to fit a simple XOR function with a Deep Neural Net and a grid search over super complex activation functions and optimizers in PyTorch. No success. Of course, vanilla numpy w/o fancy tricks works just fine. 😂 https://discuss.pytorch.org/t/unable-to-learn-xor-representation-using-2-layers-of-multi-layered-perceptron-mlp/13287/18 …
      Show this thread
      8 replies 89 retweets 473 likes
      Show this thread
    2. François Chollet‏Verified account @fchollet 7 Mar 2018
      Replying to @hardmaru

      Implementing neural nets teaches you how to implement neural nets. It gives you an algorithmic understanding of how they work. It doesn't teach you what they do, or what they can or can't achieve. To learn that, you should apply them on a range of real problems (not XOR/MNIST)

      2 replies 10 retweets 54 likes
    3. François Chollet‏Verified account @fchollet 7 Mar 2018
      Replying to @fchollet @hardmaru

      Knowing the algorithm != Knowing how/why it works. My recommendation: do Kaggle competitions, and use more than neural nets. Use many different ML models. Make it visual. Plot your features.

      1 reply 21 retweets 85 likes
    4. François Chollet‏Verified account @fchollet 7 Mar 2018
      Replying to @fchollet @hardmaru

      It's perhaps clearer with a simpler example: implementing SVD teaches you how to implement SVD. Once you're done, you still don't know what it does. Using SVD on various datasets and plotting & using the results is what gives you intuition about SVD. Also, learning math

      2 replies 4 retweets 27 likes
      François Chollet‏Verified account @fchollet 7 Mar 2018
      Replying to @fchollet @hardmaru

      Grad students knew how to implement neural nets in C in 2000. And they didn't have good intuition about them. A high school student playing with NN frameworks in 2018 can develop stronger understanding of NNs in a matter of days -- just thanks to a better application context

      6:37 AM - 7 Mar 2018
      • 29 Retweets
      • 125 Likes
      • ankur kothari catalin nv DesignRevision ranktutorials. 📚💡 Daylen Yang HolyMotors Nouroz Rahman Ge Yang Wayde Gilliam
      7 replies 29 retweets 125 likes
        1. New conversation
        1. Kyle Kastner‏ @kastnerkyle 7 Mar 2018
          Replying to @fchollet @hardmaru

          Understanding is great, until you hit a bug. Going down to the bottom is a reminder of how many *implementation details* actually matter. Random seeds, tie breaking in max-pool, whether the gradient is correct (eps tolerance can still be broken when chained), others.

          1 reply 1 retweet 7 likes
        2. Kyle Kastner‏ @kastnerkyle 7 Mar 2018
          Replying to @kastnerkyle @fchollet @hardmaru

          Examples - SVD only deterministic up to sign (S is sqrt of eigenvalues), based on implementation, border filling in convolution, bias initialization, numerical stability in running mean/variance (http://www.dtic.mil/dtic/tr/fulltext/u2/a133112.pdf …), numerical stability of losses in prob/log space, etc.

          1 reply 0 retweets 4 likes
        3. Show replies
        1. Susan Groff‏ @SusanGroff1 7 Mar 2018
          Replying to @fchollet @hardmaru

          Would you say it's hands on learning?

          0 replies 0 retweets 0 likes
          Thanks. Twitter will use this to make your timeline better. Undo
          Undo
        1. Theofanis Karaletsos‏ @Tkaraletsos 7 Mar 2018
          Replying to @fchollet @hardmaru

          I'd disagree they did not have good intuitions when I look at old papers by Hinton and Schmidhuber. They had perfectly amazing intuitions, possibly better ones because they could not as easily make the pixels fly in GANs and had to use their brains for a bit before engineering.

          0 replies 0 retweets 2 likes
          Thanks. Twitter will use this to make your timeline better. Undo
          Undo
        4. New conversation
        1. François Chollet‏Verified account @fchollet 7 Mar 2018
          Replying to @fchollet @hardmaru

          I wrote my first neural nets (actually in C) as a student in 2009. Taught me more about C than about NNs. It took me a few more years until I started understanding what neural nets do and what makes them useful. The key was better tools + application to many real-world datasets

          2 replies 2 retweets 22 likes
        2. ericmjl@localhost:8888‏ @ericmjl 7 Mar 2018
          Replying to @fchollet @hardmaru

          I think it's a combination of both. @DavidDuvenaud and his colleagues taught me how to write neural nets from scratch using Python + autograd, and that gave me an intuition of the internals. Applying it to problems gave me an intuition of what it could do. I needed both!

          0 replies 0 retweets 4 likes
          3. End of conversation
        1. Giovanni Jurado‏ @giovanniJurado 7 Mar 2018
          Replying to @fchollet @hardmaru

          Which framework is that? I want to learn!!

          0 replies 0 retweets 0 likes
          Thanks. Twitter will use this to make your timeline better. Undo
          Undo
        1. Farhan Ahmad‏ @farhanhubble 7 Mar 2018
          Replying to @fchollet @hardmaru

          Abstraction is absolutely necessary to be able to experiment quickly & to reason about more complex things. Most of mathematics is also about creating abstractions. However all abstractions are leaky (https://www.joelonsoftware.com/2002/11/11/the-law-of-leaky-abstractions/ …) and sooner or later bite you in unmentionable places.

          0 replies 0 retweets 0 likes
          Thanks. Twitter will use this to make your timeline better. Undo
          Undo

      Loading seems to be taking a while.

      Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.

        Promoted Tweet

        false

        • © 2021 Twitter
        • About
        • Help Center
        • Terms
        • Privacy policy
        • Cookies
        • Ads info