my_model/tabs/readme.md

Directory Overview: This directory contains all the atreamlit application pages:

################################################################################################################################

1. home.py

the home.py displays an introduction to the application with brief background and description of the application tools.

################################################################################################################################

2. results.py

The results.py module manages the interactive Streamlit demo for visualizing model evaluation results and analysis. It provides an interface for users to explore different aspects of model performance and evaluation samples.

Notes: Ensure the necessary dependencies are installed and properly configured. The run_demo function relies on the ResultDemonstrator class to generate plots and display results.

################################################################################################################################

3. run_inference.py

The run_inference.py is responsible for the running inference to test and use the fine-tuned models. It manages the user interface and interactions for a Streamlit-based Knowledge-Based Visual Question Answering (KBVQA) application. This module handles image uploads, displays sample images, and facilitates the question-answering process using the KBVQA model.

Notes:

• Ensure the necessary dependencies are installed and properly configured.
• The InferenceRunner class relies on the KBVQA model to generate answers to questions based on image analysis.

################################################################################################################################

4. model_arch.py

The model_arch.py displays the model architecture and accompanying abstract and design details for the Knowledge-Based Visual Question Answering (KB-VQA) model.

################################################################################################################################

5. dataset_analysis.py

The dataset_analysis.py module provides tools for analyzing and visualizing distributions of question types within given question datasets for Knowledge-Based Visual Question Answering (KBVQA). It supports operations such as data loading, categorization of questions, visualization, and exporting data to CSV files. This module leverages Streamlit for interactive visualization and Altair for plotting.

Notes: Ensure the necessary dependencies are installed and properly configured. The OKVQADatasetAnalyzer class leverages Altair for creating interactive visualizations and Streamlit for displaying these visualizations in a web app format. The run_dataset_analyzer function provides an overview of the dataset and utilizes the OKVQADatasetAnalyzer to visualize the data. This module has a dependency on the process_okvqa_dataset function from my_model.dataset.dataset_processor.