Ben Falk
Ben Falk
Attempting to use a `url` tag in markdown doesn't actually render the URL or static file into HTML, it just displays whatever was entered, including the url tag. Am I...
So that it passes the estimator test. First thing to do is to make the model pickleable.
#218 adds dependency of eigen. need to update dockerfile when that gets merged. `sudo apt install libeigen3-dev`
The following is a list of parameters that [sklearn random forest](https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.RandomForestClassifier.html) has that we do not. - [ ] criterion (gini/entropy) - [ ] min_samples_leaf: The minimum number of samples...
Currently, releasing on pypi/docker is a chore and needs resources we don't always have available (OSX for instance). We can have Travis push the the release for us whenever we...
**Describe the bug** Travis fails intermittently in C++ tests **To Reproduce** ??? ```test: ./fpTests/RerF/../../../src/forestTypes/basicForests/rerf/../labeledData.h:37: T fp::labeledData::midVal(const fp::labeledData&) [with T = double]: Assertion `dataElement != (dataElement+otherData.dataElement)/2.0' failed. Makefile:87: recipe for target...
**Is your feature request related to a problem? Please describe.** we have binning as an option for RerF.py but not for sklearn rerfClassifier **Describe the solution you'd like** add them
Different combinations of `max_features` and `feature_combinations` are resulting in segfaults when called from the sklearn class (rerfClassifier). Needs to be reproduced on different systems but known failures in OSX. @MrAE...
for instance: `self.estimators_` contains a list of trees in sklearn However, we store multiple trees in each bin, with multiple bins representing the forest. also, trees are interleaved within the...