Neural Networks
Refer to Introduction to Neural Networks
[Personal Code Example](https://github.com/lauradang/data-science-training/blob/master/neuron.py)
Neurons
General Process:
Takes input
Does some math
Outputs something
Weights (Aka Parameters)
Connects each neurons in one layer to the neurons in the next layer
Activation Function
Turns unbounded input into output that is bounded and thus, predictable
eg. Sigmoid function
Sigmoid Function
eg. Binary classification: The output would either be 0 or 1
What is happening in step 2 of process
Each input multiplied by some weight
→
All weighted inputs summed (basically dot product of input and weights) and bias is added
() + () +
Sum is passed into activation function
Neural Network
Neurons connected together
Hidden Layer
Layers between input and output layers
Last updated