Y.G. Bae (36) [Avatar] Offline
#1
At (2) PREDICT & COMPARE: Make a Prediction, Calculate Output Error and Delta,

layer_2_delta = (layer_2 - walk_stop[0:1]) 
              = -0.02129555 - 1
              = -1.02129555


Therefore, layer_2_delta should be -1.02 instead of 0.14.


At (3) LEARN: Backpropagate From layer_2 to layer_1,

layer_1_delta = layer_2_delta.dot(weights_1_2.T)
              = [-1.02129555] * [ 0.07763347, -0.16161097, 0.370439]
              = [-0.07928672, 0.16505257, -0.3783277]
layer_1_delta *= relu2deriv(layer_1)
              = [-0.07928672, 0.16505257, -0.3783277] * [0, 1, 0]
              = [0, 0.16505257, 0]

Hence, layer_1_delta for the 2nd column (row in the diagram) should be .17 instead of -.17.