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. François Chollet‏Verified account @fchollet 5 May 2020

      François Chollet Retweeted François Chollet

      This is exactly what happened. Now the UW model has been readjusted to predict 134k deaths total. The new model still suffers from the same flaw: it assumes something will change in May/June to cause the epidemic's trajectory to suddenly change direction and get under control.https://twitter.com/fchollet/status/1256297472407662593 …

      François Chollet added,

      François CholletVerified account @fchollet
      24 hour & 2.5k dead later and we're now at 65k confirmed deaths. The UW model is still predicting 72k total deaths *by the end of the epidemic*. Reality: we'll blow past it in 5 days. https://twitter.com/fchollet/status/1255937714244816896 …
      6 replies 22 retweets 123 likes
      Show this thread
    2. François Chollet‏Verified account @fchollet 5 May 2020

      Perhaps it will, perhaps it won't. But it's an odd thing for a model to assume, since our future response cannot be inferred from past data. You might as well make the reverse assumption: "gradual reopenings will cause the trajectory to change course, in the wrong direction"

      3 replies 3 retweets 25 likes
      Show this thread
    3. François Chollet‏Verified account @fchollet 5 May 2020

      That's what throws me off with these models: they don't appear to acknowledge that how the future turns out is a consequence of our actions. They start from the assumption that the curve will go flat, and then they fit a curve that fits past data plus this unjustified assumption.

      2 replies 3 retweets 36 likes
      Show this thread
      François Chollet‏Verified account @fchollet 5 May 2020

      A better model would be presented as a set of possible actions taken by the fed govt / state govts / the public, and outcomes conditioned on these actions. This would involve actually modeling transmission and economic/social activity, rather than fitting a curve.

      4:08 PM - 5 May 2020
      • 9 Retweets
      • 45 Likes
      • Sakthi Geek Prdctnomics - The economics of digital products Kbmajeed Oriol Abril Pla oliver rosario joao Michael Fadi Badine Carry You Home
      6 replies 9 retweets 45 likes
        1. François Chollet‏Verified account @fchollet 5 May 2020

          And don't get me started on the "cubic model" from the WH (cubic fit in Excel?) predicting 0 deaths in two weeks

          6 replies 4 retweets 44 likes
          Show this thread
          Thanks. Twitter will use this to make your timeline better. Undo
          Undo
        1. New conversation
        2. Thomas G. Dietterich‏ @tdietterich 5 May 2020
          Replying to @fchollet

          Isn't that what scenario-based modeling tries to do? You consider different scenarios for government/society behavior and then predict how the disease will respond. Someone must be doing this, yes?

          3 replies 0 retweets 1 like
        3. Simon Clavet‏ @profbof 5 May 2020
          Replying to @tdietterich @fchollet

          The pandemic behaves like an inverse self-fulfilling prophecy. If people think it will go up, it goes down. Now people seem to think it'll be ok.

          0 replies 0 retweets 1 like
        4. End of conversation
        1. E-quipper - grateful for the mask you wear  😷  💉 💉‏ @annwitbrock 5 May 2020
          Replying to @fchollet

          It's weird that people think that when barely plateauing, you can somehow get down the other side of the curve without it costing roughly as many lives as going up the curve. That somehow there isn't full on infection going on out there.

          0 replies 0 retweets 0 likes
          Thanks. Twitter will use this to make your timeline better. Undo
          Undo
        1. Biggus Discuss‏ @biggusdisscus 5 May 2020
          Replying to @fchollet

          That would only give a perception of work to more nerds just to come with an overfittet complex bullshit model with tons of assumptions in it resulting with similar predictive power or worse. But I agree this is pathetic.

          0 replies 0 retweets 0 likes
          Thanks. Twitter will use this to make your timeline better. Undo
          Undo
        1. Carry You Home‏ @quantadan 5 May 2020
          Replying to @fchollet

          That would be more difficult and would require people/modelers/policy makers to actually be competent.

          0 replies 0 retweets 0 likes
          Thanks. Twitter will use this to make your timeline better. Undo
          Undo
        1. Yüksel Günal‏ @yuksel_gunal 6 May 2020
          Replying to @fchollet

          Actually, there is one recently built by @cmmid_lshtm. One can see how school closures, social distancing, elderly shielding, self-isolation, and lockdowns may impact cases, hospitalizations and deaths https://cmmid.github.io/visualisations/covid-transmission-model …

          0 replies 0 retweets 1 like
          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