A full detail ppt about ho neural network work.

How do neural network work? Everything about neural network.

1. Let’s try to understand Neural Network with an example Of taking an image of panda They never saw a panda in their entity life. Let consider we hire some people to detect panda.

2. How they will Detect it?

3. Group 1 sunil raj malika Ear Eye Month

4. Group 2 sonu Mukesh Front Legs Back Legs

5. sunil raj malika EYE Ear Month Is this a panda head? Ramesh Group 1

6. Sonu Mukesh Front Legs Back legs Is this a panda Body? Swagat Group 2

7. Ramesh Swagat Raman Head Body So, it is panda, huu ..

8. sunil raj malika sonu Mukesh Ramesh Swagat Raman Group 1 Group 2

9. How they Work and get Trained ?

10. # Every person will return an output between 0 and 1 with some probability. 0.5 1 Not a Panda A bit look Like Panda A Panda

11. sunil raj malika sonu Mukesh Ramesh Swagat Raman 0.3 0.2 0.5 0.1 0.0 0.4 0.2 0.4 Then, it is not a panda. Y= EYE*0.3+EAR*0.2+MOUTH*0.5 Y= FRONT_LEGS*01+BACK_LEGS*0.0 Y= HEAD*0.4+BODY*0.2

12. Raman It is not a panda Supervise He knows the correct answer. No, it is a Panda. Raman you are Not working properly

13. Raman ok Supervise Get back and work properly

14. sunil raj malika sonu Mukesh Ramesh Swagat Raman 0.3 0.2 0.5 0.1 0.0 0.4 0.2 Guys, it is Actually a panda.

15. # Then all of them get some experience and learn from their failure. # Malika get excited as she predicted right answer and Learn more about to detect panda. Note: This steps continue until they do not achieve a great accuracy or able to detect panda.

16. Let’s try this once again after they get Train with a good accuracy.

17. sunil raj malika sonu Mukesh Ramesh Swagat Raman 0.8 0.7 1 0.9 0.6 0.9 0.85 1 Then, it a panda. Output layer Hidden layer Input layer Y= EYE*0.8+EAR*0.7+MOUTH*1 Y= FRONT_LEGS*0.9+BACK_LEGS*0.6 Y= HEAD*0.9+BODY*0.85

18. Raman It is a panda Supervise Yes, Raman it is a panda Now you reached to a good Accuracy and no need to train Further. You can join you job now

19. Now let’s look into reality.

20. Consider all the human in the given e xample as artificial neuron. They look like this: Hidden layer is more than any other. It is the most important layer.

21. What is inside a Artificial Neuron?

22. Noting just some mathematical function

23. An Artificial Neuron y   Y= +…+   w = Weight (score) x = Input b = bias Activation function Prediction function       y = output (0 to 1)

24. Programming Python Paradise Subscribe