Profile picture
François Chollet
@fchollet
, 3 tweets, 1 min read Read on Twitter
All content from the first edition of the book is still relevant -- all code examples will run -- except for the appendix about installing software.

I'm working on a 2nd edition that will cover the latest techniques and APIs -- possible release at the end of 2019 or early 2020.
In general, the new APIs *extend* prior Keras workflows rather than upending them -- in particular they make Keras completely flexible and extensible for the needs of deep learning researchers. Prior high-level workflows are still relevant.
As a new user, you won't see a huge difference in the Keras experience in TensorFlow 2.0. But as a researcher, you'll get access to entirely new workflows and features to work more productively.
Missing some Tweet in this thread?
You can try to force a refresh.

Like this thread? Get email updates or save it to PDF!

Subscribe to François Chollet
Profile picture

Get real-time email alerts when new unrolls are available from this author!

This content may be removed anytime!

Twitter may remove this content at anytime, convert it as a PDF, save and print for later use!

Try unrolling a thread yourself!

how to unroll video

1) Follow Thread Reader App on Twitter so you can easily mention us!

2) Go to a Twitter thread (series of Tweets by the same owner) and mention us with a keyword "unroll" @threadreaderapp unroll

You can practice here first or read more on our help page!

More from @fchollet see all

Profile picture
François Chollet
@fchollet
Human morphology can climb Mount Everest and do Olympic ice-skating even though it was evolved for running in the savanna. That still doesn't mean it is a general locomotion system -- it is hyperspecialized.
However, it *was* selected for strong adaptability. As a result, it works in a far wider range of environments than the one it comes from.
Human cognition is the same: it was selected, in part, for adaptability & strong generalization. As such, despite its specialization, it achieves a *degree* of generality. It can solve an incredible range of problems.
Read 5 tweets
Profile picture
François Chollet
@fchollet
I think consciousness isn't a requirement for high intelligence, and the AI agents of the future won't be conscious by default. Inversely, creatures with relatively low intelligence may possess some degree of consciousness.
I see consciousness as a behavioral guidance mechanism similar to emotions. Emotions, like consciousness, aren't a requirement for intelligence, and strong AI won't have feelings by default. But both consciousness & emotions are quite useful for embodied creatures like ourselves.
I also see consciousness as both a qualitative and quantitative attribute: there is a threshold that a system has to cross in order to be conscious at all (0 to 1 -- a thermostat isn't conscious), and past that, systems may be more or less conscious.
Read 3 tweets
Profile picture
François Chollet
@fchollet
Historically, there have been two opposing views of cognition. One in which intelligence is a general ability to learn from arbitrary data, and one in which intelligence is the result of a myriad of special-purpose systems shaped by millions of years of evolution.
The first view derives from Enlightenment thinkers, themselves influenced by the ancients. This seems to be the position of most deep learning newbies today. The second view originates with Darwin, and was embraced by first-generation AI thinkers like Minsky, etc.
According to the first view, the mind could be seen as a "blank slate" (or a randomly initialized neural network) capable of making sense of any problem or environment, without requiring hard-wired priors.
Read 7 tweets

Related threads

Profile picture
Oscar Chavira
@Oscar_Chavira
LA TRAICIÓN DE HOLLYWOOD.
1.- La fábrica de sueños.
Hollywood, California 1916.

Adolph Zukor acaba de cerrar una alianza con otras empresas menores en el ramo del entretenimiento y crea la Paramount Pictures Corporation.

Es el embrión de Hollywood, lo que después seería llamado "La fábrica de sueños".
Read 116 tweets
Profile picture
Values, believing, independent thinking, teamwork
@Bruno_Brunelli
🔴 #Software is eating the world, and it’s only the beginning

#ArtificialIntelligence #AI #MachineLearning #ML #FutureOfWork #Robots #Technology #Disruption

1/2
🔴 #Software is eating the world, and it’s only the beginning

#ArtificialIntelligence #AI #MachineLearning #ML #FutureOfWork #Robots #Technology #Disruption

2/2
Emerson, Lake & Palmer - From The Beginning (Live)

Read 11 tweets
Profile picture
Yehuda Katz 🥨
@wycats
When you're writing software, sometimes you find that it's useful to extract some common functionality into an abstraction.
That can happen because:

- there's more than one consumer of a common chunk of code
- the ppl working on a chunk of code are very different from those who work on the code it's being extracted from
- the extracted code has a diff domain from the code it's being extracted from
Extraction works better when:

- the code takes well-described inputs and turns them into outputs (it's pure)
- the extracted code is changing much more slowly than the code it's being extracted from
- the ppl working on the code are willing to commit to a notion of stability
Read 22 tweets
Profile picture
Lindsay Brown
@Lidsville
Today the First Nations challenge to #SiteC in BC Supreme Court resumes, asking for an injunction to stop work on the dam until the rest of the Treaty 8 infringement case is decided in court. A critical case for BC. Follow the #SiteC & #SiteCInjunction hashtags this week! #bcpoli
Heading over to 800 Smithe now for 9:45 entry. All are welcome to be in the courtroom - please join us. Latecomers not admitted; proceedings start 10 am sharp. #bcpoli
This final week of the #SiteCinjunction hearing involves lawyers from the Province of BC (we have already heard from @sage_legal on behalf of West Moberly & Prophet River FNs, and BC Hydro's lawyers. #bcpoli #SiteC
Read 238 tweets
Profile picture
Chad Loder ❇️
@chadloder
A thread for the "#software is not political" crowd. Whether you like it or not, all software is political because technology AFFECTS PEOPLE. If you came to #tech thinking you'd escape having to think about people, #politics, and society - you were mistaken. I'll show you why...
Before we start, remember this as you read. Just because YOU think the answer to any of these issues is clear-cut, it doesn't mean that issue is not political.

There are people who take the OPPOSITE position with just as much conviction, and they think it's clear-cut too.
Also - as an engineer, a developer or designer - if you choose to ignore the political and societal implications of your #technology, YOU HAVE MADE A POLITICAL CHOICE.
Read 12 tweets
Profile picture
Ｔａｙｌｏｒ Ｒｉｃｈｅｙ
@TheTaylorRichey
0s and 1s commonly create speech that courts use -- that means precedence is set. #Code is speech -- they're languages made of bits! Point being, ISPs shouldn't be allowed to deny computer speech from being seen by others. #NetNeutrality

Evidence in thread:
#NetNeutrality Evidence case 1: Bradley v. State: @Facebook evidence used to ID assailants and other witnesses. Its compelling because translated bits were used to ID people. It proves third party intermediary bits translated into media are used by courts. scholar.google.com/scholar_case?c…
#SaveNetNeutrality Evidence case 2: Homeland Security v. Twitter: @DHSgov seized data that was digitized and sent to intermediary @Twitter. Twitter sued its an ongoing investigation. Regardless, It appears consumer's translated bits were treated as speech? clearinghouse.net/detail.php?id=…
Read 11 tweets

Trending hashtags

#turnaround #project #Doesnt #SuperBowl #Parliament #Sitecinjunction #CNS #finserv #UNDRIP #NothingToSeeHere #Question #THREEtimes #SPOiLER #Evil #Cryptography #EVIDENCE #KochBrothers #cannabispolicy #scale #technology #FalseFlag #oil #MuellerThursday #Can #Wisdom
Follow Us on Twitter!

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just three indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3.00/month or $30.00/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!