cai-group123-assignment

Model Card for Model ID

<!-- This model is Desinged as part of Conversation AI Asignment 2, we used IRFC Annual reports of 2023-24 and 2024-25 and extracted the data then used for model training and evaluation . -->

Comparative Financial QA System: RAG vs Fine-Tuning

Model Details

Objective: Develop and compare two systems for answering questions based on company financial statements (last two years):

Retrieval-Augmented Generation (RAG) Chatbot: Combines document retrieval and generative response. Fine-Tuned Language Model (FT) Chatbot: Directly fine-tunes a small open-source language model on financial Q&A.

Model Description

This model is Desinged as part of Conversation AI Asignment 2, we used IRFC Annual reports of 2023-24 and 2024-25 and extracted the data then used for model training and evaluation.

• Developed by: [Assignment Group 123]
• Funded by [optional]: [Group 123]
• Shared by [optional]: [Group 123]
• Model type: [RAG, Fine-Tune]
• Language(s) (NLP): [NLP]
• License: [More Information Needed]
• Finetuned from model [optional]: [gpt2]

Model Sources [optional]

• Repository: [https://huggingface.co/nvsngurram/cai-group123-assignment]
• Paper [optional]: [More Information Needed]
• Demo [optional]: [More Information Needed]

Uses

<!-- -->

Question and Answer based System for the learning of implementation of RAG and Fine-Tuning the model.

Direct Use

<!-- -->

Used to Analyse the pdf documents, let's say financial reports of a company can be analysed and give summary of it based on the user query in a short and compact manner.

[More Information Needed]

Downstream Use [optional]

<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->

[More Information Needed]

Out-of-Scope Use

<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->

responsive text generation

Bias, Risks, and Limitations

<!-- This section is meant to convey both technical and sociotechnical limitations. -->

[More Information Needed]

Recommendations

<!-- Using this model we can read any pdf files and ask questions out of it and get answers generated -->

Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.

How to Get Started with the Model

Use the code below to get started with the model. Step 1: clone the git repo into local Step 2: install the required libraries by running pip install -r requirement.txt Step 3: check if Annual reports are under data/annual_reports and QandA.txt in the data/raw/QandA.txt Step 4: run the data_extraction.py(python src/data_extraction.py) script to convert from .PDF to .txt Step 5: run the rag_ft_qa.py(python src/rag_ft_qa.py) script to pre-process, segmentation of data, tokenization, creation of model, pre-train the model then Implement of RAG techniques(cross encoder), Fine-tune the model with dataset prepared, re-ranking the results and extracting best of it, compared the results and represented in tabular format. Step 6: run the streamlit_cli.py(python streamlit_cli.py) script to run the application on streamlit

[More Information Needed]

Training Details

Training moto is to train the model with sample 10 questions and evaluate the performace after training the model.

Training Data

used Annual Report 2023-24.pdf and Annual Report 2024-25.pdf annual reports for raw data, then converted into .txt files and then segmented and trained with 400 chunck dataset.

[More Information Needed]

Training Procedure

<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->

Preprocessing [optional]

[More Information Needed]

Training Hyperparameters

• Training regime: [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->

Speeds, Sizes, Times [optional]

<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->

[More Information Needed]

Evaluation

<!-- This section describes the evaluation protocols and provides the results. -->

Testing Data, Factors & Metrics

Testing Data

<!-- This should link to a Dataset Card if possible. -->

[More Information Needed]

Factors

<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->

[More Information Needed]

Metrics

<!-- These are the evaluation metrics being used, ideally with a description of why. -->

[More Information Needed]

Results

Summary

Model Examination [optional]

<!-- Relevant interpretability work for the model goes here -->

[More Information Needed]

Environmental Impact

<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->

Carbon emissions can be estimated using the Machine Learning Impact calculator presented in Lacoste et al. (2019).

• Hardware Type: [More Information Needed]
• Hours used: [More Information Needed]
• Cloud Provider: [More Information Needed]
• Compute Region: [More Information Needed]
• Carbon Emitted: [More Information Needed]

Technical Specifications [optional]

Model Architecture and Objective

[More Information Needed]

Compute Infrastructure

[More Information Needed]

Hardware

[More Information Needed]

Software

[More Information Needed]

Citation [optional]

<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->

BibTeX:

[More Information Needed]

APA:

[More Information Needed]

Glossary [optional]

<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->

[More Information Needed]

More Information [optional]

[More Information Needed]

Model Card Authors [optional]

[More Information Needed]

Model Card Contact

[More Information Needed]