This repository goes into exploring the transcripts of the Berkeley Restaurant Project (BeRP) for generating random but meaningful sentences.
- BeRP Corpus: The BeRP dataset comprises transcripts of conversations related to restaurant services in Berkeley, CA. This project utilizes the BeRP transcripts as the primary data source for building the n-gram model and generating sentences. These transcripts provide a rich and diverse collection of restaurant-related dialogue.
- n-Gram Language Modeling: The project implements an n-gram model modeling techniques to analyze the sequential word patterns within the transcripts, which predicts the next word in a sequence based on the preceding n-1 words. Experimenting with different n-gram sizes can offer varying levels of context dependence in the generated sentences.
lm_model.py
: This file houses the core functionality of the trained model, which utilizes the n-gram language modeling approach and BeRP transcripts.testing_ngram_lm.ipynb
: This Jupyter Notebook contains comprehensive unit tests to ensure the model's correctness across various scenarios, including edge cases.test_minitrainingprovided.py
: This Python file encompasses all the unit tests used to evaluate the model's performance.