Skip to content

Latest commit

 

History

History
36 lines (18 loc) · 771 Bytes

README.md

File metadata and controls

36 lines (18 loc) · 771 Bytes

RoMA - LLM

The following instructions are for installing and executing the roma-llm case study on a linux/ubuntu machine with a GPU properly configured.

Setting up the environment

  1. Create a virtual environment using conda:

conda create -n roma-llm python=3.10

  1. Activate the environment:

conda activate roma-llm

  1. Install the requirements file

pip install -r requirements.txt

Executing the experiment

Pre-requisite - you should download the GLUE dataset to a folder named data.

1. Creating the dataset

python create_full_sentence_dataset.py

2. Running prediction on the dataset

python predict_sentence_dataset_full.py

3. Running RoMA analysis on the predictions

python analysis_full_prediction.py

Good Luck!