This is the benchmark of [VSM TF + Bernoulli Naive Bayes]
Competition Website: NCCU WSM - User Mood Prediction
Default Environment:
- python version: 2.7 (you can modify the code to fit the python 3.x)
- scikit-learn: http://scikit-learn.org/stable/install.html
Step 1. Change the file path & folder in 1042_WSM_benchmark.py
8 train_folder = './data/train/' # the folder containing the training data
9 test_folder = './data/test/' # folder containing the testing data
10 label_fname = './data/train_labels.csv' # the path of train_labels.csv
Step 2. Run the program:
python 1042_WSM_benchmark.py
It will create a solution file named submission_NB.csv.
Step 3. Submit the submission_NB.csv to the competition website.