Stochastic Induction of Decision Trees with Application to Learning Haar Tree

“decisiontree.m” is the primary function. The code is running on MNIST, Signmnist, FashionMNIST, Isolet, Satimage, covertype. The tree also is going to form from depth 5 to 12. ( if just one specific data set and depth is desired, the appropriate number for “ds” and “Depth” needs to be set. 
To calculate the training error, the trained tree will be pass to tree-predict. Later on, the test set will pass to the “tree-prediction” function to calculate the test error.
The result will be saved in an excel file in the given path.

The datasets are available in the following links:

https://www.kaggle.com/oddrationale/mnist-in-csv
https://www.kaggle.com/datamunge/sign-language-mnist
https://www.kaggle.com/zalando-research/fashionmnist
https://www.kaggle.com/gorangsolanki/isolet-dataset
https://datahub.io/machine-learning/satimage
https://archive.ics.uci.edu/ml/datasets/covertype



