"New Neural" is an Open Source neural network engine project for the MetaTrader 5 platform. - page 85

New comment
 
Serj_Che:


We've already done everything for ourselves.

For some reason this "everything" is not visible as part of the terminal.

I guess not so much or not so well done.

there is nothing to show.

 
sergeev:

For some reason this"all" is not visible in the composition of the terminal.

I guess not so much or not so well done.

there is nothing to show.

First of all, I did not put dots over the e, it's your interpreter is wrong, read as is.

What can I show you?)

What I needed - all done.

Are you peeking at my terminal?)

 
Renat:

Anyone interested in participating directly?

We will provide funding and general management. Let's start a joint project in MQL5 Storage and get started.

Yes, we do. You can take part in it. Who will manage the project? I mean, who to contact?
 

I myself will not write, I do it clumsily, but a little command :D

So: stop speculating what network to take, who will be the boss, etc..

I have already said it is necessary to split up and start writing a finished product.

We need to decide who will write the code that creates the neuron, who will write code that creates a grid of neurons, someone to write the output, still need file operations, in short, each piece, then everything will be easy and fast.

What network to choose? Any network to start with!

 
The problem is that there is no general concept yet, no idea of the structural functioning of the final product.
 

It's not going to happen: great minds have been battling for a year and a half to create it!

do you want to continue this?

 
fellow:

It's not going to happen: great minds have been battling for a year and a half to create it!

do you want to continue this?

And no one fought, at first they slackened all the spears and quietly and peacefully fucked it up.
 
fellow:

I myself will not write, I do it clumsily, but a little command :D

So: stop speculating what network to take, who will be the boss, etc..

I have already said it is necessary to split up and start writing finished products.

We need to decide who will write the code that creates the neuron, who will write code that creates a grid of neurons, someone to write the output, still need file operations, in short, each piece, then everything will be easy and fast.

About what to choose a network? Any network for a start!

All masters of command. Even the methaquotes have tried to command, and even finance.

But when it comes to business - right away the brakes. Even the methaquots do not know what they want.

And if it comes to any problem, from the methaquots to wait three years to help, such as security policy does not allow or we have no time.

The budget has not yet been defined.

Here is the beginning (I did not vote for it).

P.S. Did the methaquotes make you commander-in-chief?)

 
fellow:

We need to decide who will write the code that creates the neuron, who will write code that creates a grid of neurons, someone to write the output, still need file operations, in short, each piece, then everything will be easy and fast.

Another hero.)

I was almost ready to be the organizer. I outlined the architecture and the workflow. But people were drawn in the direction of genetics and universal network architectures, so I gave up in the end.

Now I will be lured into this project only by the very tasty morsels, especially if there is an alglib.

 
Serj_Che:

What's so shameful about it?

Well, here's more proof - I didn't spit for nothing :) and in general - Rosenblatt's perseptron rules ))
1...787980818283848586878889909192...100
New comment
Reason: