logistic迴歸損失函數與梯度下降

邏輯迴歸中，假設函數hypothesis hθ(x) h θ ( x ) 爲： hθ(x)=g(θTx)=11+e−θTx=g(θTx+b) h θ ( x ) = g ( θ T x ) = 1 1 + e − θ T x = g ( θ T x + b ) 當 hθ(x)=g(θTx)≥0.5 h θ ( x ) = g ( θ T x ) ≥ 0.5 , 即 θTx≥0 θ T x ≥ 0

>>阅读原文<<