README

#To run the naive bayes implementation 

# Run in the top level directory
# This takes about 6 hours to run as it runs on the full review dataset

./naive_bayes_baseline.py

# Data preproceesing for word2vec
# To generate training and test data

./prepare_ip_text.py <training_text_fname> <training_labels_fname> <test_tax_fname> <test_labels_fname> <number_training> <number_test>

# To produce the word vector embeddings
# The input file is usually one of the training files generated in the previous step 
./word2vec_basic.py <ip_file> <op_file>

# To vectorize the review texts
./vectorize_reviews_summation.py <ip_fname> <word_vectors_fname> <op_filename> <embeddings_size>

# to run mluti-class logistic regression
./multiclass_logistic_regression.py <training_file> <est_file> <training_samples_per_category> <test_samples_per_category>