1
$\begingroup$

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.

$\endgroup$
  • $\begingroup$ can you clarify your step 3? I do not understand what it is trying to say. $\endgroup$ – skim Dec 7 '18 at 14:46
1
$\begingroup$

In general these steps are correct, but are few clarifications would be good. For supervised learning, you can:

  1. pick model to use
  2. split data into training and test set
  3. train model, or determine weights by optimization of our error function (inputting in training data)
  4. use trained model to predict on test data
  5. utilize metrics of interest to evaluate our model (e.g. MSE for continuous outcomes, AUC for binary)

This however is a very elementary pipeline to train a supervised machine learning model. There are many other steps taken to ensure our model is as robust as possible (like cross validation, feature selection, hyperparameter optimization, model selection...). I would strongly recommend learning about these other steps and seeing how they integrate with the above pipeline.


For unsupervised learning, the process depends a lot on the task, but in general it involves optimizing some objective function using the data provided. This is usually done using all the data in one go (not split into training and test) unlike supervised learning. This can be seen more clearly with examples, take methods like PCA and k clustering; the steps taken to apply these methods very much depend on the method used.


Going further, I would recommend Andrew Ng's coursera course on machine learning. It is very accessible and a great introduction to this field, covering both unsupervised/supervised learning.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.