Less than 1 week –
Less than 10 hrs/week –
I am looking for a programmer familiar with scikit-learn to compare the training and testing time of decision trees and random forests with sparse data against the same metrics with data converted to dense.
The final analysis should show how testing and training times change by the following parameters: the density (percentage of non-zero feature values), the number of features, the number of training samples and finally the number of testing samples.
Please note that benchmarking should be done on ...