This repo contains the code for QIA2023 Challenge. For more info, please refer to the report attached :)
When you run the code, please change the PATHS written in the code! Also note that the data is not public :(
Phase 1 consists of one single file run.ipynb Following the comments, you can see which part of the code refers to the training and testing. Running all cells will give you the csv file needed.
Phase 2 consists of two files. train.py and test.py After running train.py, you will get 4 json files, where each json file stores for the test set prediction by the best model of each dimension of MBTI. Then with these 4 files, run test.py to get the csv file needed.
Other .ipynbs (chatgpt.ipynb, lgbm.ipynb) include some of the other methods used for phase 2 prediction.