search

UnixJunkie / orf

OCaml Random Forests
Other
8 stars 1 forks source link

training a model is quite slow #16

Closed UnixJunkie closed 2 years ago

UnixJunkie commented 2 years ago

partially addressed by two PRs from @igarnier

UnixJunkie commented 2 years ago

Maybe we will come back to this in the future, using profiling on a high-dimensional and large dataset. E.g. I could find the target with the most molecules in ChEMBL and encode it using some unfolded counted FP.