Depression Prediction with Gaussian Process
- Setup three main branches
documentation,technicalandprogram. Individual work should further branch from these branches. - Edit your work and push any work in progress into the respective branches. This is so that any rebase does not affect other's work.
- Push to master once you are complete.
- learner
- all codes are written in python here.
- mltools: python code adapted by UCI Machine Learning Class
- data: datas that used in the code. X is the features consisted of MFCC and Magnitude Spectrum; Y is the outcome, obtained using PHQ8 test.
- forests.py: code written to do prediction by random forests using development set after training the model using training set
- forests_crossValidation.py: code written to do 10-fold cross-validation test by random forests using the data provided in training set
- features_extractor/AudioFeaturesExtractor
- all codes are written in java here.
- it is developed using Eclipse (Version: Luna Service Release 2 (4.4.2))
- lib/: libraries included in the java project
- src/: the java codes written by me using template provided in CS2108 class.
- report
- all report-related files are placed here
If you encountered any errors with the .bib and .tex files, run make in the command line for Unix system. Else, compile the document in the following sequence:
pdflatex reportbibtex reportpdflatex report- Run the document in you favorite latex editor