deeplearing吴恩达assignment

编程作业中函数输入输出及相关项思维导图，主要包括initialize parameters deep；forward propagation；Cost function；Backward propagation module；update parameters等方面知识，希望对你有帮助！

helper function

• initialize parameters deep
  • W[L]，b[L].L is the layer of network
  • Input: layer dims
  • Output: list parameters("W[L]","b[l]")
• forward propagation
  • linear forward
    • Input: A w b
    • Z = np.dot(W,A) + b
    • Output: Z, cache(A,W,b)
  • linear_activation_forward
    • Input：A_prev,W,b,activation(sigmoid or relu)
    • Output: A,cache
  • L_model_forward
    • Input：X parameters
    • Output: AL, caches
• Cost function
  • compute_cost
    • Input: AL,Y
    • Output: cost
• Backward propagation module
  • linear_backward
    • Input: (dZ, cache)
      • 子主题
    • Output(dA_prev,dW,db)
  • linear_activation_backward
    • Input: dA, cache, activation
    • Output：dA_prev,dW,db
  • L_model_backward
    • Input: AL, Y, caches
    • Output: grads("dAL","dWL","dbL")
• update_parameters
  • Input: parameters, grads, learning_rate
  • Output： parameters