nlp homework
This is the public homework repository for Natural Language Processing with Representation Learning DS-GA 1011.
For hw1, please run the tokenization frame in hw1.ipynb first. I delete the tokenization files since they are too large.
For hw2, please run the preprocessing in hw2.ipynb first. I also delete the tokenization files since they are too large.
save_model.py is used to save model. number_of_params.ipynb is used to calculate the number of params in each model. test.ipynb is used to get fine tune results.
code for cnn is in hw2_cnn.py; code for rnn is in hw2_rnn.py