Metadata-Version: 2.1
Name: SchoginiAI
Version: 0.2.1
Summary: A sample AI toolkit by Schogini Systems with Retrieval-Augmented Generation (RAG).
Home-page: https://github.com/schogini/SchoginiAI
Author: Sreeprakash Neelakantan
Author-email: schogini@gmail.com
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Requires-Python: >=3.7
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: langchain>=0.3.14
Requires-Dist: langchain-community>=0.3.14
Requires-Dist: openai>=1.59.3
Requires-Dist: tiktoken>=0.8.0
Requires-Dist: faiss-cpu>=1.9.0.post1
Requires-Dist: langchain-openai
Requires-Dist: langchain-chroma
Requires-Dist: langchain-pinecone
Requires-Dist: langchain_experimental
Requires-Dist: chromadb
Requires-Dist: pinecone-client
Requires-Dist: pypdf

# SchoginiAI

SchoginiAI is a robust AI toolkit developed by **Schogini Systems**, providing advanced **Retrieval-Augmented Generation (RAG)** capabilities. It leverages the power of [LangChain](https://langchain.com/), [OpenAI](https://openai.com/), and multiple vector storage options like **FAISS**, **ChromaDB**, and **Pinecone**. This makes it a versatile tool for building AI-driven solutions for various applications.

---

## 🚀 Features

- **PDF & Text Processing**: Extracts knowledge from PDF files or raw text for vector storage.
- **Flexible Vector Stores**: Supports FAISS, ChromaDB, and Pinecone for vector management.
- **Advanced Text Chunking**: Includes configurable chunking strategies like `RecursiveCharacterTextSplitter` and `TokenTextSplitter`.
- **Environment-Based Configurations**: Manages API keys and settings securely using `.env` files.
- **RAG Capabilities**: Combines vector retrieval with OpenAI’s LLMs for generating relevant, context-aware responses.

---

## 🛠 Installation

### From PyPI

Install SchoginiAI directly from PyPI:

```bash
pip install SchoginiAI
```

### From Source

Clone the repository and install locally:

```bash
git clone https://github.com/yourusername/SchoginiAI.git
cd SchoginiAI
pip install .
```

Replace `yourusername` with your GitHub username.

---

## 🔧 Configuration

### Environment Variables

Create a `.env` file to securely store your API keys:

```dotenv
OPENAI_API_KEY=your_openai_api_key
PINECONE_API_KEY=your_pinecone_api_key
```

### Custom Settings

Edit the `settings.py` file to configure the following options:

- **Vector Store Type**: Choose between `faiss`, `chroma`, or `pinecone`.
- **Text Splitter Strategy**: Select chunking strategies like `RecursiveCharacterTextSplitter` or `TokenTextSplitter`.
- **Chunk Size and Overlap**: Control the size and overlap of text chunks.

Example `settings.py` snippet:

```python
vector_store_type = "chroma"  # Options: 'faiss', 'chroma', 'pinecone'
text_splitter_strategy = "RecursiveCharacterTextSplitter"
chunk_size = 1000
chunk_overlap = 200
```

---

## 📚 Usage

### Build Vector Store from Text

```python
from SchoginiAI import SchoginiAIRAG

rag_ai = SchoginiAIRAG()

sample_text = """
Schogini Systems is a pioneer in AI Chatbots.
We specialize in automation solutions for small businesses.
"""

rag_ai.build_vector_store(sample_text)
rag_ai.save_vector_store()
```

### Query the Vector Store

```python
from SchoginiAI import SchoginiAIRAG

rag_ai = SchoginiAIRAG()
rag_ai.load_vector_store()

response = rag_ai.ask_question("What services does Schogini Systems provide?")
print("Response:", response)
```

### Build Vector Store from PDF

```python
from SchoginiAI import SchoginiAIRAG

rag_ai = SchoginiAIRAG()
rag_ai.build_vector_store_from_pdf("path/to/your/pdf_file.pdf")
rag_ai.save_vector_store()
```

---

## 🐳 Docker Usage

### Build Docker Image

```bash
docker build --no-cache -t schoginiai .
```

### Run Docker Commands

#### Create Vector Store

- **Using FAISS**:
  ```bash
  docker run --rm -e OPENAI_API_KEY="your_openai_api_key" -e PINECONE_API_KEY="your_pinecone_api_key" schoginiai create
  ```

- **Using ChromaDB**:
  ```bash
  docker run --rm -e OPENAI_API_KEY="your_openai_api_key" -e PINECONE_API_KEY="your_pinecone_api_key" schoginiai create
  ```

#### Query Vector Store

- **Using FAISS**:
  ```bash
  docker run --rm -e OPENAI_API_KEY="your_openai_api_key" -e PINECONE_API_KEY="your_pinecone_api_key" schoginiai query
  ```

- **Using ChromaDB**:
  ```bash
  docker run --rm -e OPENAI_API_KEY="your_openai_api_key" -e PINECONE_API_KEY="your_pinecone_api_key" schoginiai query
  ```

Replace `"your_openai_api_key"` and `"your_pinecone_api_key"` with your actual keys.

---

## 🗃 Project Structure

```
SchoginiAI/
├── SchoginiAI/
│   ├── __init__.py
│   └── main.py
├── examples/
│   ├── usage_example.py
│   ├── knowledge_creation.py
│   └── .env
├── tests/
│   ├── test_faiss.py
│   ├── test_chroma.py
├── settings.py
├── README.md
├── requirements.txt
├── Dockerfile
├── doit.sh
└── setup.py
```

---

## 📦 Dependencies

SchoginiAI relies on the following Python packages:

- `langchain>=0.0.200`
- `openai>=0.28.1`
- `faiss-cpu>=1.7.6`
- `chromadb>=0.3.22`
- `python-dotenv>=0.21.0`

Install these dependencies using:

```bash
pip install -r requirements.txt
```

---

## 🛡 License

This project is licensed under the **MIT License**.

---

## 📚 Additional Resources

- [LangChain Documentation](https://langchain.com/docs/)
- [OpenAI API Documentation](https://beta.openai.com/docs/)
- [FAISS Documentation](https://github.com/facebookresearch/faiss)
- [ChromaDB Documentation](https://www.trychroma.com/docs/)
- [Python-Dotenv Documentation](https://pypi.org/project/python-dotenv/)

---

With this setup, SchoginiAI is ready to handle advanced AI-driven tasks with ease and flexibility. For more details, refer to the source code or open an issue in the repository.
