I am new for machine learning and I am tried to understand basic steps to get final modal of Logistic Regression.
I know Logistic Regression is supervisory learning technique. Therefore we want to give training data to training the modal. According to my understanding, I can take steps to get the final modal as follows.
Step 1 - Make an algorithm. For Logistic Regression is y = sigmoid(W x + B) function. Get zero or some value for W and B.
Step 2 - Give sample known training data and find W and B values. x(1), x(2)...x(m) inputs and get y(1), y(2)...y(m) outputs. Gradient Descent use for find W and B that minimizes the cost function.
Step 3 - Then apply W and B values which I found.y^ = sigmoid(W x + B). And then again apply sample known training data to get the predicting of y^.
Step 4 - Get the final modal to test unknown data.
are these steps right? Please give me basic fundamental steps to understand supervisory learning technique. I would like to know un-supervisory learning technique steps also.