Updated simple_neural_network.py

[ved-et9]

->Created simple_neural_network.py from scratch

->It now contains Forward propagation as well as backward propagation
with training of the simple 2 layer neural network.

->Article Referred is a free resource avialable for evryone

Fixes #8785

• Add an algorithm?
• Fix a bug or typo in an existing algorithm?
• Documentation change?

Checklist:

• I have read CONTRIBUTING.md.
• This pull request is all my own work -- I have not plagiarized.
• I know that pull requests will not be merged if they fail the automated tests.
• This PR only changes one algorithm file. To ease review, please open separate PRs for separate algorithms.
• All new Python files are placed inside an existing directory.
• All filenames are in all lowercase characters with no spaces or dashes.
• All functions and variable names follow Python naming conventions.
• All function parameters and return values are annotated with Python type hints.
• All functions have doctests that pass the automated testing.
• [] All new algorithms include at least one URL that points to Wikipedia or another similar explanation.
• If this pull request resolves one or more open issues then the description above includes the issue number(s) with a closing keyword: "Fixes Source for simple_neural_network.py is member-only #8785 ".

[tianyizheng02]

This implementation is definitely a far improvement over our current implementation (and it runs much faster!), but you really shouldn't just copy the implementation from your source. However, I think it's fine if you significantly refactor it to make it more of your own work. @cclauss, thoughts?

One example of how you could rewrite and improve your replacement implementation is to write a class for the neural network: then initialize_network would just be the __init__ function and training would all fall under a NN object's methods. This would allow the user to create a NN and not have to pass the object into the training function.

[cclauss]

There is no software license associated with:

• https://machinelearningmastery.com/implement-backpropagation-algorithm-scratch-python/ or
• https://github.com/jbrownlee/Datasets
  and the author presumably is making or has made some revenue from its publication.

I think we should keep the current neural_network/simple_neural_network.py implementation in place. If we add this, it should be under a different filename.

I am not a fan of replacing other people's work with a copy of yet another person's work without some work to create value. If serious work is done to add type hints, doctests, etc. to the functions then this could considered a derivative work. However, given there is no license, I think we would need approval from @jbrownlee or @Jason2Brownlee before we land this in an MIT-licensed repo. https://github.com/TheAlgorithms/Python/blob/master/LICENSE.md

[Jason2Brownlee]

Sorry, I no longer own or operate machine Learning mastery - I cannot comment on licenses for the code. I recommend that you reach out to the company directly.

That being said, I wrote it as a port of my 2011 ruby version which is under creative commons here:
https://github.com/clever-algorithms/CleverAlgorithms

There are tons of algorithms there that may interest you for this project.