Loading multiple datasets to the classification learner app

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.

Answer 1

[Aditya Patil]

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.

[Guy Nir]

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]

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]

Interesting! I’ll try the single-rows entry.

Thank you, Guy