The required modules has been listed in the requirements.txt You can install them by pip install -r requirements.txt. Strongly recommend setup the environment in virtualenv

Steps:

1. Extract the data.tar.gz
2. Run merge.py to produce a CSV file instead of thousands of files. (Merged file provided, so this step is optional)
3. Run preprocessing.py, change the method you are going to apply (TFIDF or NN)
4. run run_tfidf.py or run_cnn.py

Result in Kaggle: Accuracy: 0.89709 Final Report is: ML_Project_Report.pdf