MATLAB Answers

Loading multiple datasets to the classification learner app

6 views (last 30 days)
Guy Nir
Guy Nir on 16 Jun 2020
Commented: Guy Nir on 18 Aug 2020
I am using the classification learner app. My data is clusters of x,y,z points. My goal is to train the model to say whether the shape that these points make is what I defined as shape '0' or '1'. How do I load multiple, but separate datasets? That is, in each experiment I generate points that make up shape '0' or '1'. Now I want to take several experiments, and load to the classification learner. Obviously I need the learner to know that each experiment is by itself, so it can collect the shapes from each experiment.

  0 Comments

Sign in to comment.

Answers (1)

Aditya Patil
Aditya Patil on 18 Aug 2020
Combine the datasets together, either as a table or as a matrix, and load it into classification learner. More details on using classification learner app are available here.

  3 Comments

Guy Nir
Guy Nir on 18 Aug 2020
I need the learner to relate to each dataset separately. If you combine all points into a single dataset, they don’t mean anything.
Aditya Patil
Aditya Patil on 18 Aug 2020
In that case, you can train the model seperately for each dataset. Alternately, if each dataset is actually an entry, then reshape the dataset to be single row, and then combine the datasets. That way, the app will be able to recognise them all as separate entries.
Guy Nir
Guy Nir on 18 Aug 2020
Interesting! I’ll try the single-rows entry.
Thank you, Guy

Sign in to comment.

Products


Release

R2020a

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!