README.md

---
title: GeoScience Exam Marker
emoji: 💬
colorFrom: yellow
colorTo: purple
sdk: gradio
sdk_version: 5.0.1
app_file: app.py
pinned: false
license: mit
short_description: 'A chatbot to evaluate geoscience student assessment papers'
---


# Model Name
geobot_chat-v0

path to model:  ShebMichel/geobot_teacher-v0
## Overview
This bot is about assessing geo student assignment paper with eith a pass or fail mark.

## TESTING THE BOT
However, as of today Nov 7, 2024, to use it,

1- Load your data in json format or copy the below and save it as a json:

{
    "short_answer_questions": {
        "question1": {
            "question": "How do sedimentary rocks form?",
            "answer": "Sedimentary rocks form from the accumulation of sediments."
        },
        "question2": {
            "question": "How do metamorphic rocks form?",
            "answer": "Metamorphic rocks form under pressure and heat, altering existing rocks."
        },
        "question3": {
            "question": "What are the stages of the rock cycle?",
            "answer": "The rock cycle involves the transformation of rocks through melting, cooling, weathering, and erosion."
        }
    },
    "long_answer_questions": {
        "question1": {
            "question": "What is the purpose of down-hole logging?",
            "answer": "It is used to gather detailed information about rock layers and fluids."
        },
        "question2": {
            "question": "What types of logs are used in down-hole logging?",
            "answer": "Common logs include resistivity logs, sonic logs, and density logs."
        }
    }
}

2- If successfully loaded, you will see all the json data into the Processed Questions box

3- For the example shown above, you either click:
   - SHORT ANSWER QUESTIONS
   - LONG ANSWER QUESTIONS

   Once selected, the Processed Questions box will only show the questions for the selected options.
   Also, in the "Edit Selected Question" box, you will have the first question as input text, 

4 - Now click submit for the model to run.
    The result is printed on "Model Response" box.

5- click clear if you want to do it again

##



An example chatbot using [Gradio](https://gradio.app), [`huggingface_hub`](https://huggingface.co/docs/huggingface_hub/v0.22.2/en/index), and the [Hugging Face Inference API](https://huggingface.co/docs/api-inference/index).