Case 1: I would apply active learning to query a small chunk of samples gradually to label them and my model is being trained during this process. After a certain number of iterations, I have a training dataset with specific performance of the model.
Case 2: I re-train the model from scratch with the training dataset in case 1.
Question 1: do you think the performance of the model will be the same in both cases? why, please?
Question 2: do you think there is a difference to apply active learning to label the testing dataset after case 1 or apply it after case 2?
Unless I do not understand your question, but the best way to go about this is to label the entire dataset and use train_test_split in sklearn to set aside a portion of the dataset for testing.
I am not sure about your question, but do you mean unsupervised learning where you do not know the labels and you want your classifiers to cluster them?
I will encourage you to read:
Pattern Recognition and Machine Learning
Authors: Bishop, Christopher
https://www.springer.com/gb/book/9780387310732
You test and see...? Unless this is a school work and you should figure it out...?
- Badges
- Science topic