Training program for SB-Neuron.

[z-s]

Hii
I don't know about ai so much but i have a suggestion.
Like the cross recognition can we make a ai to recognise ASCII ART.
https://www.asciiart.eu/text-to-ascii-art

[AbsoluteBeginner]

Hii
I don't know about ai so much but i have a suggestion.
Like the cross recognition can we make a ai to recognise ASCII ART.
https://www.asciiart.eu/text-to-ascii-art

Hello, Z-S. 

Do I understand you correctly that you want to create a SB neural network that will convert an ASCII ART IMAGE into text? 

[z-s]

Yes as first step and may could recognise pattern and images after it

[AbsoluteBeginner]

Yes as first step and may could recognise pattern and images after it

(translated by Google translator)

The information I have says that you have two ways to achieve your goal.

The first way is the classical one.
First you have to choose a classical neural network structure that can do the job you need.
Then you will need to prepare data to train this neural network.
Then the neural network must learn from this data.
If the neural network cannot learn, then you will be forced to change the structure of the neural network and try to train it.

If you are not a specialist in classical neural networks, then it will be difficult for you to do this kind of work correctly.
It is for this case that I created for myself SB-neurons, which I can consciously collect into a SB-neural network.
In this case, the structure of this SB neural network will be formed automatically, in accordance with the task that each SB neuron will perform.

Creating an SB neural network requires less work than preparing data for training a classical neural network and training it.
But, to create a SB neural network, the developer must clearly understand what work each SB neuron should perform.

[AbsoluteBeginner]

(translated by Google translator)

Hello everyone, dear friends. 

The discussion of the topic "Threads" gave excellent results.
Now it will be clear to anyone how to use "Threads" correctly.

I dream that we will achieve the same result in the topic "Training the SB-Neuron".
In this topic I don't have a definitive answer to two questions:

1. Why can we use different Activation Functions, but train a neuron only by changing the input weights? Perhaps the fact that we can use different Activation Functions is a hint that there is a way to train a neuron by changing the parameters of the Activation Function.
   
2. Which learning method is best: changing the input weights; changing the Activation Function parameters; or a combination of these two methods?
   

Why do I ask these questions?
Most likely, I am not the only fan of programming in Small Basic who is interested in being able to construct his own neural network by inserting a neuron with special settings into any place in the neural network.  
At the moment, we do not have this option. We have access to a neural network consisting of identical neurons. We can only change the number of neurons and layers. This network learns only by changing the weights on the inputs of neurons.

As a result, using such a neural network requires special data preparation. Such a neural network cannot solve many types of tasks. And the accuracy of the responses of such a neural network is not very high.

I invite people who are also interested in this topic to take part in the discussion.
If successful, we will all receive a very interesting tool for our research and entertainment.