I have implemented 3 bootstrapping models for learning product details, then I did the comparison with more different performance measure and found out which model learned good, but I want to do something like optimization/ensemble (is these possible with the models result?) or please suggest some other simple process to conclude my work. Moreover, the work was performed in an unsupervised manner. So please help me how to do improvisation in my models results (like tp,tn,fp,fn or learned product details). Thanks in advance.
I do not understand your requirement precisely, but maybe Friedman test can applied to the comparison of the results.
Jopsef
As Jopsef mentioned, it is a bit difficult to follow what you are actually asking for. Still, if you do have some performance measure (does not matter if it is based on supervised or unsupervised methods), standard model selection techniques like using both a validation and test set can be useful for optimizing models with many hyperparameters, such as ensembles.
This coursera video discusses the strategy in general: https://www.coursera.org/learn/machine-learning/lecture/QGKbr/model-selection-and-train-validation-sets
This python library provides an easy-to-use parallel python implementation (on top of sklearn): http://dask-searchcv.readthedocs.io/en/latest/
thanks for your ideas. I will try these suggestions. once again thanks all
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Algorithms