I saw examples of neural networks that can determine what is shown in the picture with a certain probability. What kind of neural networks or how to get the probability that the element belongs to a particular class?

  • For this, deep convolutional networks (CNN - Convolutional Neural Networks) are used. The answer to the question "how to get the probability that the element belongs to a certain class", if you answer it clearly and with examples it will take 15 to 20 printed pages. This is contrary to the ideology of this site - here they answer specific questions that have arisen during implementation / coding / design. You can start reading from here - MaxU
  • If I translated everything correctly, then the SoftMax transform on the last layer of stats.stackexchange.com/questions/265905/… will suit me - Pavel

0