+- Small Basic Forum (https://litdev.uk/mybb)
+-- Forum: Small Basic (https://litdev.uk/mybb/forumdisplay.php?fid=1)
+--- Forum: Discussion (https://litdev.uk/mybb/forumdisplay.php?fid=4)
+--- Thread: Creation of SB-Neuron. Ours. Branded.(v2) (/showthread.php?tid=133)
RE: Creation of SB-Neuron. Ours. Branded.(v2) - AbsoluteBeginner - 05-01-2025
This program of mine is getting very interesting.
But, I'm very sorry that I work so slowly.
My dream is to find out if I can manually install such a set of weights on the inputs of the SB-neuron that will allow one neuron to recognize the symbol “Cross”, even if this symbol has any size and position on the SB-display.
RE: Creation of SB-Neuron. Ours. Branded.(v2) - litdev - 05-01-2025
(05-01-2025, 10:13 AM)AbsoluteBeginner Wrote: My dream is to find out if I can manually install such a set of weights on the inputs of the SB-neuron that will allow one neuron to recognize the symbol “Cross”, even if this symbol has any size and position on the SB-display.
Clasically this would be done by the training with a very large training set and maybe pre-processing of the training data, so I think it is tough to do it manually, but interesting to try.
RE: Creation of SB-Neuron. Ours. Branded.(v2) - AbsoluteBeginner - 05-01-2025
Yes, exactly. It's interesting to try it.
RE: Creation of SB-Neuron. Ours. Branded.(v2) - AbsoluteBeginner - 05-09-2025
Hi all.
I am ready to publish a new version of the "NeuroLab" program.
But, I want to create a "User Manual" for this program.
Please advise me where would be the best place to place this "Manual"?
I don't like to place such a large text in the program code as comments. But if I create a separate text file, how will I publish this file along with the program code?
RE: Creation of SB-Neuron. Ours. Branded.(v2) - litdev - 05-09-2025
Hi, if you email it to me I will stick it in a pdf on this site.
admin@litdev.uk
RE: Creation of SB-Neuron. Ours. Branded.(v2) - AbsoluteBeginner - 05-09-2025
(05-09-2025, 05:00 PM)litdev Wrote: Hi, if you email it to me I will stick it in a pdf on this site.
admin@litdev.uk
Thank you. It will probably be good.
Now, it is interesting to know how many words there will be in this "Manual".
RE: Creation of SB-Neuron. Ours. Branded.(v2) - litdev - 05-09-2025
If you wanted to host it yourself something like github may be good.
RE: Creation of SB-Neuron. Ours. Branded.(v2) - AbsoluteBeginner - 05-10-2025
As the Mohican leader, Chingachgook (played in the movie by Gojko Mitic. His name appeared in the credits in German as "Georg Mitic".), liked to say: "The souls of my ancestors told me that I could save the text of the "User's Guide" by publishing it together with the code of the "NeuroLab" program, but as a separate file, which would have a version number one less than the program code file.".
I don't know why, but I liked this idea.
What do you say about this?
RE: Creation of SB-Neuron. Ours. Branded.(v2) - AbsoluteBeginner - 05-11-2025
While I was thinking about the text of the "User's Guide", I came up with the idea of adding another function to the program code.
This will delay the publication of the program a little, but it will be cooler.
RE: Creation of SB-Neuron. Ours. Branded.(v2) - AbsoluteBeginner - 05-12-2025
Hi all.
Finally I have published a new version of my program "NeuroLab". DWXL365.000-2
(short "User's Guide" - DWXL365.000-1 )
This version allows you to conduct some experiments and draw some conclusions.
It is already clear to me that a neural network capable of recognizing figures by features must have neurons in its first hidden layer that are trained to recognize not only the presence of features in an image, but also the location of the features and the distance between the features.
Only then can the second hidden layer of neurons be trained to detect the shape in the image using the data from the first layer.
At this point, I still hold the opinion that layers of neurons in a multilayer network should be trained one by one, rather than all layers at once. Otherwise, we train neurons, for example, of the second layer ON UNRELIABLE DATA, which the second layer receives from the still untrained neurons of the first layer.
Is this logical?..