No tensorflow and/or pytorch, pure numpy and math equations using the MNIST dataset.
- Activation Function: ReLU and ReLU_deriv
- Last Activation Function: softmax
- Propagation: Foward and Backwards
- Encoding: one hot encoding
- Optimization: Gradient descent
Model had a 87% accuracy on Iteration 500
Iteration: 490
[2 7 7 ... 8 6 1] [2 7 9 ... 8 6 1]
0.8702439024390244