Skip to content
/ evalbot Public

This is an automated scoring system that leverages a trained BERT model to predict sentence similarity. By comparing student responses to reference answers using BERT's deep understanding of semantic meaning, EvalBot provides efficient and objective evaluation, moving beyond simple keyword matching to assess the conceptual accuracy of text.

0 stars 0 forks Branches Tags Activity
Star
Notifications You must be signed in to change notification settings

anu-gtb/evalbot

BranchesTags

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

17 Commits
.github/workflows
.github/workflows
 
 
src
src
 
 
.gitignore
.gitignore
 
 
README.md
README.md
 
 
dockerfile
dockerfile
 
 
model.py
model.py
 
 
requirements.txt
requirements.txt
 
 
scoring.py
scoring.py
 
 

Repository files navigation

src -> contains raw training and testing data files, preprocessed training and testing data files, separate files for data preprocessing, model fine-tuning and model evaluation

scoring.py -> Streamlit app/bot development

model.py -> preprocessing functions to be applied on the new data and prediction

dockerfile -> building docker image of streamlit app/bot

.github/workflows/deployment.yaml -> deployment of docker image to DockerHub through CI/CD pipeline

About

This is an automated scoring system that leverages a trained BERT model to predict sentence similarity. By comparing student responses to reference answers using BERT's deep understanding of semantic meaning, EvalBot provides efficient and objective evaluation, moving beyond simple keyword matching to assess the conceptual accuracy of text.

Topics

bot deep-learning transformers python3 pytorch embeddings artificial-intelligence sentence-similarity bert-model streamlit automated-scoring llms finetuning-llms

Resources

Readme
Activity

Stars

0 stars

Watchers

0 watching

Forks

0 forks
Report repository

Releases

No releases published

Packages

No packages published

Languages