MCPcopy Index your code
hub / github.com/lawlite19/MachineLearning_Python / debugInitializeWeights

Function debugInitializeWeights

NeuralNetwok/NeuralNetwork.py:235–239  ·  view source on GitHub ↗
(fan_in,fan_out)

Source from the content-addressed store, hash-verified

233
234# 初始化调试的theta权重
235def debugInitializeWeights(fan_in,fan_out):
236 W = np.zeros((fan_out,fan_in+1))
237 x = np.arange(1,fan_out*(fan_in+1)+1)
238 W = np.sin(x).reshape(W.shape)/10
239 return W
240
241# 预测
242def predict(Theta1,Theta2,X):

Callers 1

checkGradientFunction · 0.85

Calls

no outgoing calls

Tested by

no test coverage detected