- the project using two different dataset. The first is {ids.mat、SegPatches.mat}, the second is {class.mat、histogram.mat}
- first dataset, we must extract features from dataset and to build dictionary.
- second dataset, we have build the dictionary, directly using
- load_input.py load data, return labels, patches, count
- main.py using the first version dataset
- main_xubo.py using the second version dataset
- pLSA.py include calu two matrixs and selected the relevant words by pointing the number of concepts
- SVM.py, we using this file to cross validation by using SVM
- Tools.py, include some functions that sometimes need to use
- just run main.py or main_xubo.py