Hi, I'm working on a bot to trade using neural networks. It is currently in development. If you want you can see the generated signals of laterization, long or short.

It is based on train images generated every minute by looking at the BTC / USD quote. The train is run weekly. Currently the train is over 40,000 images.

It is still in full development, if you are interested in taking a look at the signals this is the link:https://vafed.it/signals.php.

First time that I've heard a bot Neural signals.

So I'm also interested as to what it can bring 'new' to crypto trading. So I would also ask the OP to provide more details, although it is still in development phase, perhaps some initial background of the project itself.

Any details how to read those signals? Also based on the timestamp, it is not generated every minutes. It is every minutes for some signals and then gets long space before the next signal.

I honestly don't understand the signal from the link above.

What exactly do you mean by "Lateralization" do you mean this is the amount of your BTC?
And the records from your signals are the price movement of BTC where you lose and gains as up and low?

Btw, I'm still confused about how to read the bot / signal that you gave. How does it work and how do you read it? at least you provide a more detailed picture of the trade you make with your bot / signal so we understand more deeply.

Making a single page for the user's manual would be great because not every cryptocurrency users are familiar with the bots which is just familiar with the one who already used it before. More over for new people in cryptocurrency which just know about cryptocurrency less than one year.

Only participated in one such project based on blockchain, can't recall the name and can't even really dig it up in my history, but they found moderate success with similar others -- this was all before 2018 and 2019 though. After that, all these projects seemed to dry up on luck, and much like their blockchain fund counterparts, all ended up posting consistent losses.

It's almost as if they get worse with time, instead of better. I personally think it's always because the "market" is still too young and still unable to differentiate from genuine and authentic data. Some of these projects, at least, the one I worked with, had very capable academics behind them, willing to go quite deep into explaining how it worked.

They still, eventually, failed to produce consistently good results. How do you do different? Purely based on visuals? Charts do present differently visually, and those visuals miss a lot of white noise in between (temporary spikes for example).

Can you elaborate what is your background? This topic is very interesting to me and I'm curious what are requirements one has to meet to be able to start working with this type o coding. Do you have a lot of experience with programming and what architectures if so?

It seems like the first time I've ever heard of the neural network bot's name for trading in crypto. How this site actually gives you all the information you need to know how successful we can be if we work here If it is good and develops quickly then hopefully many will use it. We'll show you how to trade through this bot.

I try to answer the questions I have been asked.
I have been following the world of cryptocurrencies for about 3 years while trading for about 10 years. I have always been fascinated by bots to trade and in the past I have built several in the forex world.

As for cryptocurrencies, I developed a bot for about 3 months to arbitrage between 3 different exchanges for the XMR / USD pair (I mined Monero, when it was worth 10 USD).

I stopped developing the bot by arbitrage to focus on one for trading. I developed several, all of which were based on technical indicators. I didn't like them.

So I started studying neural networks with the Tensorflow framework.

They are currently in full development, it takes time and $$ money to develop.

The infrastructure you can see on the site is more complex than it appears.

These are the main details on which it is based:

Database: Mysql with a simple db

Frontend site: PHP

Backend: Python with Tensorflow, numpy, mathplotlib etc.

Exchange for my personal bot: Kraken with python API


Now the next step is to simulate the various models with the signals issued and to benchmark performance by including the fees of an exchange, as truthful as possible.

Next step will be to implement a Telegram bot and email notification of the signals.