hello everyone my name is Li-Qing I was
graduated from Massey University New Zealand major in computer science right
now I’m sitting at Cortex Labs I’m currently working on the community
operation and also a Junior Program today I’ll give you guys a brief
introduction to the Cortex Demo namely Silencer why it’s called Silencer
because it’s cool we’ll keep Silencer up to date and today
I’ll give you guys a tour of the first version of it we’re doing this as
industry tradition just like what ZCash did before their equihash came online
we kind of releasing a toy for you to play with if you just want to see the
demonstration of Silencer you can skip the first part and without further ado
let’s get started so first you would be able to find our
github through Cortex website then you can see many different
repositories and the ones where the word Cortex in it were developed by our team
but today we’re only interested in Cortex Silencer you can see that
Silencer was written in Python it stimulate the blockchain environment of
Cortex there are two files we want to first look at Node.py and the Block.py so Node.py is the backbone of Silencer it contains the hash and the
crypto libraries the transaction information and the user upload
functionality the second one is the Block.py as to what its name suggested
it contains information about blocks and the generation of the Genesis block and
new blocks and also the mining processes and it’s worth mention here this whole
new mechanism is the contribution that Cortex is making for the blockchain
which is AI smart contract with the AI small contract we’re able to handle very
complex information such as images audios videos and etc and let’s take a
look at three files CVM.py, Inference.py and the DefaultContract.py so CVM.py what is it
as what we stayed in our whitepaper CVM is the abbreviation for Cortex
Virtual Machine it provides a virtual machine environment for smart contract
to be executed on the GPU or FPGA and Silencer we’re only keeping the abstract
layer for the complete code we release them once cortex come online the second
one Inference.py it is in charge of smart contract inferences the models in
the data loader here for the execution of the inference calculation and lastly
the DefaultContract.py currently you only have one default contract for you to
play with and this one does a really simple job one transaction under one
account we’ve used the simplest example here if you have some knowledge about
the DApps on Ethereum you might find this structure familiar however what
make us special is this line of code it is executed on the GPU or the FPGA
instead of the CPU which is not achievable by modern blockchain and as
you can see here this smart contract is asking the CVM to make an inference and
all the parameters are visibly here then the transaction get printed out the
logic behind the smart contract is that if the full node have reached the
constants on death this is an image of a bird then gives you ten tokens you can
imagine this as a robot finds a bird on the street then you will receive ten
tokens as a reward so today I’m just giving you guys a brief introduction if
you want to see that code yourself you can find them on our github and if
you’re really interested you can also set up your own server running Silencer
our server hardware setup is also in our Silencer repository hmmm it’s just 8 1080TI nothing fancy not a big deal alright alright now let’s take a look at what
Silence already looks like open up this page you see the simulation
of the main chain and our only miner with address one has been mining for a
very very very long time and for each block that he mined he received five
tokens as a reward I wish I was doing this and here you can see we’ve added account system this is here just to prevent ddos attack but it’s actually a
blockchain everyone knows that right so now we can go through the register
process let’s pick a username do Silencer demo password repeat it I
think a typed it right register then we’ll see this panel each account has 10k of
tokens by default heres transfer and call for the execution of default contract
we’re interested in this one so let’s go in and click call as I mentioned
previously this is a Ethereum smart contract-like structure but different is
that the result of this smart contract is generated through the model inference
consensus first we can find a picture of a bird on the Internet
let’s search up bird and let’s see which one should we use a clear one I think
this one would do save it and go back to Silencer
let’s try to upload it and execute it now we wait for the full nodes to reach a
consensus and the miner will pack the transaction into the next block
hopefully this won’t take too long you can see here okay Dada Dada silence
have just approved our image and we got our 10 tokens reward
so this is the end of today’s demonstration in our next update we will
deploy Cortex Edge calculation to specific hardware allowing a robot to
accomplish a real world mission in order to receive some tokens sounds exciting
thank you for watching and I’ll see you next time bye bye you