This repository contains an implementation of a GPT-style language model using PyTorch. The model is designed to handle character-level tokenization and utilizes a Transformer architecture for natural language processing tasks. This specific implementation is focused on predicting the type of graph based on a given query.
Follow the steps below to run the project:
This script will extract the necessary data for training the model.
python data-extract.py
After extracting the data, run the training script to train the model.Please ensure that your comment down the pickle section in training.p Ensure that the batch-size is passed with the command (Ex: -batch_size 32)
python training.py
Finally, run the gpt.py file to use the trained model for generating responses. Ensure that the batch-size is passed with the command (Ex: -batch_size 32)
python gpt.py
Special thanks to Elliotcodes for helping me understand how to create a GPT-style model.