Depth Study Notes (1)

Logistic regression

Although logistic regression algorithm is literally a return to the word, but it is essentially a classification problem solving, the most intuitive expression which is a function Sigmod
${Sigmod(x)=\frac{1}{1+e^{\omega*x+b}}}$
Logstic loss function is:
${CostFuntion=-\frac{1}{m}\sum_{i=1}^{m}[y\log\hat{y}+(1-y)\log(1-\hat{y})]}$
A logistic regression unit may act as a node of the depth of the neural network, is also the basic constituent elements of the neural network.