Incongruity-SemEval-2020-Task-7

Final project for Natural Language Processing (CS 159) at Harvey Mudd College in Spring 2021. We analyzed incongruity in humor based on SemEval 2020 Task 7 and the Humicroedit database.

GloVe Vectors

Download GloVe vectors from the following link: http://nlp.stanford.edu/data/glove.6B.zip. Then put them in a folder called /glove.

Code to use GloVe vectors adapted from https://sites.google.com/g.hmc.edu/cs159spring2021/labs/lab-4-glove-vectors.

Duluth at SemEval 2020 Task 7

We ran the Duluth models using the code at https://github.com/dora-tang/SemEval-2020-Task-7. We used the following versions:

Python              3.7.10
torch               1.8.1
torchtext           0.9.1
numpy               1.20.1
pandas              1.2.4
transformers        4.5.1
allennlp            2.4.0
spacy               2.3.5
nltk                3.5
tensorboardX        2.2

We also made the following changes to the Duluth code. We replaced src/util.py line 31 with:

from torchtext.legacy import data

We also replaced src/models.py line 295:

outputs = self.transformer(inp, attention_mask=inp_mask, return_dict = False)