sefakcmn00/tensorflow_machine_learning_simple-

Artificial Neural Network(ANN) Perceptron

Artificial Neural Network(ANN) Perceptron --> ARTIFICIAL NEURON

Input1 --->

Input2 ------------ -> DATA ----> Output

Bias ----->

Sigmoid Function = takes a value between 0 and 1 and is generally useful for our classification problems

Tanh(Hyperbolic Tangent) = Takes values ​​between -1 and 1, and with negative values ​​it provides a wider scope and is often used in zeroing operations.

ReLU(Rectified Linear Unit) = 0 and your end It takes value among others and is frequently encountered in the field of deep learning. The given operation takes the value of =0 if not, the value of -1 if not.

Linear Functions f(x)= takes x and can take infinite values.

Regression = Is there a relationship between the height of the children and the height of the fathers? Children's heights tend to be close to the mean in the total data set. Y=a*x + b

z=ag +bf(z)=estimatedValue(estimate of neuron) Quadratic Cost=sum(actualValue - approximate value)**2/ Cross Entropy Cost=( -1/n)sum(actualValueIn(estimatedValue)(1-actualValue)ln(1-exactValue)

Gradient Descent = the optimization function we use to find the minimum of a function We can use it to minimize the Cost function.