datesim

This repository contains code for a tool named datesim, which processes images containing text (such as screenshots of conversations) and simulates conversations based on the extracted text using AI models.

Features

Image Processing: Upload images containing text (e.g., screenshots of WhatsApp conversations).
Text Extraction: Extract text from the uploaded images using Tesseract OCR (Optical Character Recognition).
Conversation Simulation: Simulate conversations based on the extracted text using AI models from OpenAI.
Database Integration: Store formatted conversation data in a SQLite database.

Prerequisites

Python 3.11
Tesseract OCR installed
OpenAI API key
Together API key
Dependencies listed in requirements.txt

Installation

Clone the repository:

git clone https://github.com/your-username/datesim.git

Install dependencies:
```
pip install -r requirements.txt
```
Set up environment variables:
- Create a .env file in the project directory.
- Add your OpenAI API key and Together API key in the .env file:
```
OPENAI_API_KEY=your-openai-api-key
TOGETHER_API_KEY=your-together-api-key
```
Update the path to the Tesseract executable (pytesseract.pytesseract.tesseract_cmd) if necessary.
Run the application:
```
streamlit run app.py
```

Usage

Launch the application.
Enter your username.
Upload images containing text.
Click "Process Images" to extract text and simulate conversations.
Then, click "Run Simulation" to simulate conversations.

Contributing