This directory contains code for binary random forest classifier and experiments on the MNIST dataset.
You can train and test the binary random forest model on the MNIST dataset by the following command in this directory:
makeTBD. See the README for regression code for now because the implementation policy is almost the same as the regression code.
We have tried our binary random forest on the MNIST dataset which is a famouse dataset for an image classification task. We've created the input binary vector by applying thresholding to the image and flatten it.
The following figure is a summary of our binary random forest classifier on the MNIST dataset. The figure shows that our binary random forest is applicable to MNIST, however, the score is far from our expectation (we expected more than 90% within 10.0 sec inference time).
You can replicate our experiments by running bash runall.bash.
