My Codebase for Tensor Flow Neural Networks Install the requirements.txt and run the files in order to

• Train a model
• Test the Model
• Run Apply the Neural Network

input data > weight > hidden layer 1 ( activation function ) > weights > hiddent layer 2 > ... output layer

compare output to intended output > cost function (cross entropy) optimisation function (optimizer) > minimize cost (AdamOptimizer...SGP, AdaGrad)

backwards progation of weights for optimisation

feed forward + backprop = epoch --aka. cycle to lower cost function '''