I am new to Tapkee and I can't find how to project new data not used while building a model.
The issue is that I have large data sets and I would like to estimate a dimensionality reduction model using a subset of my data, and apply the model afterwards to the complete data set.
I understand that this may not make sense for some of the methods, but it does for some others. What I would need to do is something similar to retrieving the projection matrix of a PCA in order to apply it to new data. The question is, how to do it in a generic way with Tapkee?
I think I should be able to do that using a ProjectingFunction, but I can't find an example doing that.
Hi,
I am new to Tapkee and I can't find how to project new data not used while building a model.
The issue is that I have large data sets and I would like to estimate a dimensionality reduction model using a subset of my data, and apply the model afterwards to the complete data set.
I understand that this may not make sense for some of the methods, but it does for some others. What I would need to do is something similar to retrieving the projection matrix of a PCA in order to apply it to new data. The question is, how to do it in a generic way with Tapkee?
I think I should be able to do that using a ProjectingFunction, but I can't find an example doing that.
Thanks in advance for your help.