How to Write a Neural Network In Python

Description
No external libraries are used; only Python's inbuilt libraries.

Using a screen-sharing method, Codacus takes you through ow to create a simple neural network using Python. Python is a programming language, popular due to its ability to provide increased productivity. What's great about this language is that the edit-test-debug cycle is really fast due to the lack of compilation step. To debug Python programmes is really easy because a bad input nor a bug will cause a segmentation fault. When an error is discovered by an interpreter, it raises an exception. The interpreter will print a stack trace when the program hasn't been caught by the exception. A source level debugger allows evaluation of arbitrary expressions, inspection of global and local variables, setting breakpoints, stepping through the code line by line, etc. The debugger has been written in Python itself, which is a testimony to Python's introspective power. Aside from this, usually the fastest way to debug a programme is to add to the source a few print statements: the quick edit-test-debug cycle makes this easy approach really effective.

For code, visit: thecodacus.com/neural-network-scratch-python-no-libraries/