Usage

• Run the file named run_cnn.py to execute the program
• The second step is run the predict.py file
• All issues which without tag are stored in the data\issues_nolabel folder
• cnn_model.py build the cnn model, including configure the machine learning engine & set related parameters
• predict.py is use the training model to determine what the category a new issue belongs to
• issues_loader.py include the tool functions & open\read files & transform coding
• issues_spliter.py is focus on how to split different categories files to the relateds folder
• issues_group is use to group small files into a large file
• The training result stored in the folder named tensorboard