Discover and read the best of Twitter Threads about #DOJO
Most recents (7)
On #biocompute news, $TSLA Tesla AI day gave some technical details on how they do their training and video labelling. It seems $NVDA Nvidia GPUs are the norm, with a 14,000 HPC heavily optimised on the software side. #pytorch #AVX2 #CUDA #SMT #smol #CUDNN 
None of this is for the purpose of #Bioinformatics applications, rather here it's for Full Self-Driving software, but the technical details show some choices for high-throughput #AI training that one could compare to the #ComputationalBiology #ComputeAcceleration world.
More at bit.ly/biocompute
@boblautenbach @Alan_L_Lovejoy @Ryu_Zoku Yes Chris Lattner is very interesting, been following his #LLVM and #SWIFT work.
1/
1/
@boblautenbach @Alan_L_Lovejoy @Ryu_Zoku Actually #ZokuTalk adopts the portion of SWIFT #EBNF for #Unicode as they seems have done good job on that, as well as parts of the EBNF for numerics such as "1_234.00" to expand privative data types entry.
\2
\2
@boblautenbach @Alan_L_Lovejoy @Ryu_Zoku Borrowing C & Assembly data types: 0b0011, 0xFF, keeping 16r0A, plus many others.
Actually the numeric formats are one of more complicated part of the #EBNF of #ZokuTalk and the only non-#MOBS (#Messages #Objects #Block #Syntax) part of the evolved syntax.
\3
Actually the numeric formats are one of more complicated part of the #EBNF of #ZokuTalk and the only non-#MOBS (#Messages #Objects #Block #Syntax) part of the evolved syntax.
\3
A short thread on accelerated computing. How $TSLA's #Dojo, $GOOG's #TPU #Cerebras's #WSE-2, #Graphcore's #IPU, and others compare to $NVDA 's #GPU. This is an extract of work we published a few weeks ago. 👇1/6
It is difficult to compare chips. Designing a chip is all about managing trade-offs between multiple dimensions. A chip has limited potential resources and the architect allocates them over these multiple dimensions. 2/6
With this framework in mind, one sees alternatives to #GPU have fundamentally different architectures, favoring the flow of data across the chip vs. the flow of data between the memory and the chip. 3/6
So instead of a working HPC, what we saw in #Dojo is a single node that’s is running a rudimentary NN on a single test bench. Designing a chip is easy. What’s hard is building the compiler, runtime scheduler in a HPC environment at scale. none of which @Tesla is anywhere close to
When asked, its brushed off as “No but we will”. This exapod doesn’t exist it’s a photoshopped image and won’t exist for many years. That’s why there’s no MLPerf benchmarks. There is only one exapod in existence & that’s @Google TPU version 4. In use today by @Waymo #Dojo
By the time this is working and ready in the presented form and specs there will already be a TPU V5. @Tesla will always be 3-4 years behind on this front. But this won’t stop the fanboys because ignorance is bliss.
I will be live tweeting @Tesla AI Day Here. Personally looking at the data flow, latency, low power and performance of #Dojo and how it compares to @Google #TPU version 4 and other HPC systems and how it directly relate to #AutonomousVehicles #FSDBeta
Well as usual, Tesla is fashionably late and we are almost 18 mins past the start time. And the beat goes on...
#AIDay starts with a demo of FSD with the driver griping the steering wheel with his left hand in what looks to be the streets of California. #DojovsTPU #Dojo @Tesla
1/3
Já retuitei isso, mas vale o ênfase:
O projeto #Dojo é um passo IMPORTANTÍSSIMO em direção à FSD.
Imagine uma neural net onde todos os carros rodando nas ruas conectam-se em tempo real.
O Dojo é o cérebro, processando quantidades imensas de dados e imagens em tempo real.
Já retuitei isso, mas vale o ênfase:
O projeto #Dojo é um passo IMPORTANTÍSSIMO em direção à FSD.
Imagine uma neural net onde todos os carros rodando nas ruas conectam-se em tempo real.
O Dojo é o cérebro, processando quantidades imensas de dados e imagens em tempo real.
2/3
Exemplo:
Imagina que o Tesla Ă sua frente identifica um buraco no asfalto.
O informação Ă© processada imediatamente, e distribuĂda para toda a frota, fazendo assim com que o seu carro desvie do buraco. 🤯
Exemplo:
Imagina que o Tesla Ă sua frente identifica um buraco no asfalto.
O informação Ă© processada imediatamente, e distribuĂda para toda a frota, fazendo assim com que o seu carro desvie do buraco. 🤯
1. #Alleged #Kidnapper Chases #Woman into #Karate #Dojo, Gets #Buttocks Kicked by #Charlotte, #NorthCarolaina #Karate Instructor, #RandallEphraim, who is #Today’sHero. vice.com/en_us/article/…
2. How did it happen? A large man followed a frantic woman inside #Ephraim’s #Charlotte, North Carolina #karate studio claiming he was there to pick her up. She repeatedly insisted that she did not know him and that he “tried to #kidnap her. nydailynews.com/news/crime/ny-…
3. The man refused to leave and tried to force himself further into the #Dojo. In #Ephraim's words: "I then went into action defending myself and got him out of the #Dojo...Once outside he attempted to attack again and was dealt with accordingly." nypost.com/2019/01/04/wou…
