How to use datasets in Classification Learner App?
4 views (last 30 days)
Hello, I want to use the classification learner app. But before I do, I want to create multiple data objects, label them and then use all of them in the classification app. However, the data only has the same number of rows (not columns).
I wanted to put them all into a table, but I then realized when I wanted to label them with categorical letters, I would have to label every row of the table in order for the data to mean anything in the app. I don't want to label every row, I just want every 14 rows to be under one label. Is this by any chance, possible?
And would it be more efficient/easier to make every 14 rows its own dataset and import all the sets into the classification learner? From what I'm seeing, the learner wouldn't accept them all.
I have the MATLAB + Simulink Student Toolbox.
Sayan Saha on 11 May 2018
It seems that each data set actually corresponds to 14 rows. So you can make each 14 rows as one observation and label that observation. Here is an example. Consider the following is your original data :
Now assume each 2 rows will be labelled 'A' and 'B'. So the data will look like:
1 2 3 4 'A'
5 6 7 8 'B'