Hugging Face's logo

Spaces:
yamanavijayavardhan
/
answer-grading-app
Running

App Files Community
answer-grading-app / README.md
yamanavijayavardhan's picture
yamanavijayavardhan
update_
666fb5d
preview code
|
raw
history blame contribute delete
1.21 kB
metadata
title: Answer Grading App
emoji: 📊
colorFrom: pink
colorTo: green
sdk: docker
pinned: false
license: apache-2.0
app_port: 7860

Answer Grading App

This is an AI-powered answer grading application that can grade student answers using multiple similarity metrics and LLM-based scoring.

Features

  • Support for PDF and CSV input formats
  • Multiple similarity metrics:
    • TF-IDF scoring
    • Semantic similarity
    • Word-level similarity
    • LLM-based scoring
  • Handwritten text recognition (HTR) support
  • Rate limiting and CORS protection
  • Comprehensive logging

Usage

  1. Upload your query file (text file with questions)
  2. Choose input format (PDF or CSV)
  3. Upload answer files (PDFs or CSV)
  4. Get graded results

API Endpoints

  • /: Main interface
  • /compute_answers: Process and generate answers
  • /compute_marks: Calculate marks for student answers
  • /check_logs: View application logs

Environment Variables

  • PORT: Server port (default: 7860)
  • HF_HOME: Hugging Face cache directory
  • GENSIM_DATA_DIR: Gensim data directory

Rate Limits

  • 200 requests per day
  • 50 requests per hour

Dependencies

See requirements.txt for full list of dependencies.