pirag 0.1.0__tar.gz

pirag-0.1.0/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2025 jyje with studio.r4iny
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

pirag-0.1.0/PKG-INFO

@@ -0,0 +1,132 @@
+Metadata-Version: 2.4
+Name: pirag
+Version: 0.1.0
+Summary: CLI Projects of On-Premise RAG. You can use your own LLM and vector DB. Or just add remote LLM servers and vector DB.
+Author-email: semir4in <semir4in@gmail.com>, jyje <jyjeon@outlook.com>
+Project-URL: Homepage, https://github.com/jyje/pilot-onpremise-rag
+Project-URL: Repository, https://github.com/jyje/pilot-onpremise-rag
+Project-URL: Issue, https://github.com/jyje/pilot-onpremise-rag/issues
+Requires-Python: >=3.9
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: python-dotenv<1.2
+Requires-Dist: loguru<0.8
+Requires-Dist: pytest<8.4
+Requires-Dist: black<25.2
+Requires-Dist: ragas<0.3
+Requires-Dist: pymilvus<2.6
+Dynamic: license-file
+
+<div align="center">
+
+# pirag: pilot-onpremise-rag
+
+<!-- <img alt="RAG Logo" src="docs/rag-logo.jpg" width="450" style="object-fit: contain; max-width: 100%; aspect-ratio: 16 / 9;"> -->
+
+🌱 LLM+RAG CLI project operating in On-Premise environment
+
+[![Python](https://img.shields.io/badge/3.9+-3776AB?style=flat&logo=Python&logoColor=white&label=Python)](https://typer.tiangolo.com/)
+[![CLI](https://img.shields.io/badge/CLI-orange?style=flat&logo=iterm2&logoColor=white)](https://typer.tiangolo.com/)
+[![LLM](https://img.shields.io/badge/LLM-green?style=flat&logo=OpenAI&logoColor=white)](https://openai.com)
+[![LangChain](https://img.shields.io/badge/LangChain-blue?style=flat&logo=Langchain&logoColor=white)](https://langchain.com)
+[![Milvus](https://img.shields.io/badge/Milvus-red?style=flat&logo=Milvus&logoColor=white)](https://milvus.io/)
+[![MinIO](https://img.shields.io/badge/MinIO-red?style=flat&logo=MinIO&logoColor=white)](https://min.io/)
+<!-- [![Docker](https://img.shields.io/badge/Docker-blue?style=flat&logo=Docker&logoColor=white)](https://docker.com) -->
+
+</div>
+
+## 🚀 Introduction
+
+**pilot-onpremise-rag** is a CLI tool that implements a knowledge-based RAG (Retrieval-Augmented Generation) system with LLM. It provides powerful document retrieval and generation capabilities while ensuring data privacy.
+
+## 🔧 Setup
+
+### (Optional) Setup External Dependencies
+```bash
+git clone https://github.com/jyje/pilot-onpremise-rag
+cd pilot-onpremise-rag
+
+docker compose -f docker/compose.yaml up -d
+```
+
+### Install pirag
+```bash
+git clone https://github.com/jyje/pilot-onpremise-rag
+cd pilot-onpremise-rag
+
+pip install --upgrade -e ./app
+```
+
+## 📚 Usage
+
+### Basic Commands
+
+```
+# View help
+pirag --help
+
+# Train documents
+pirag train --source ./documents
+
+# Ask a question
+pirag ask "Give me a joke for Cat-holic."
+```
+
+## 🏗️ Project Structure
+
+```
+pilot-onpremise-rag/
+├── app/                        # Main application directory
+│   ├── main.py                 # CLI main entry point
+│   ├── setup.py                # Package setup configuration
+│   ├── pyproject.toml          # PEP 517/518 build configuration
+│   ├── requirements.txt        # Dependencies
+│   ├── logs/                   # Application logs
+│   └── rag/                    # RAG implementation
+│       ├── config.py           # Configuration settings
+│       ├── agent.py            # Agent implementation
+│       ├── ask/                # Query handling module
+│       ├── train/              # Document training module
+│       ├── test/               # Testing module
+│       └── doctor/             # Diagnostic tools
+├── VERSION                     # Project version
+├── docker/                     # Docker configuration
+├── assets/                     # Static assets (Files are not included)
+└── LICENSE                     # License information
+```
+
+## 🔄 How It Works
+
+1. **Document Training**: Process local documents and store in vector database
+2. **Search Engine**: Find document chunks related to user queries
+3. **Context Generation**: Create LLM prompts from retrieved documents
+4. **Response Generation**: Provide accurate responses via local LLM
+
+## 💡 Key Features
+
+- **Privacy Guaranteed**: All data and processing occurs locally
+- **Multiple Document Support**: Support for PDF, Markdown, TXT, DOCX, and other formats
+- **Custom LLM**: Compatible with various local LLM models
+- **Vector Database**: Vector DB integration for efficient document retrieval
+
+## 🧪 Performance Optimization
+
+| Configuration | Memory Usage | Response Speed | Suitable Use Cases |
+|--------------|-------------|---------------|-------------------|
+| Light Model | 4-6GB | Fast | Simple queries, low-spec hardware |
+| Medium Model | 8-12GB | Medium | General use, most queries |
+| Large Model | 16GB+ | Slow | Complex document analysis, expert answers |
+
+## 🔗 References
+
+- [LangChain Documentation](https://python.langchain.com/docs/get_started/introduction)
+- [LLM Optimization Techniques](https://huggingface.co/docs/optimum/index)
+- [RAG Paper](https://arxiv.org/abs/2005.11401)
+
+## Contributing
+
+Any contributions are welcome!
+
+### Current Maintainers
+- [Studio R4iny](https://github.com/studior4iny)
+    - [jyje](https://github.com/jyje), [semir4in](https://github.com/semir4in) (Same person)

pirag-0.1.0/README.md

@@ -0,0 +1,113 @@
+<div align="center">
+
+# pirag: pilot-onpremise-rag
+
+<!-- <img alt="RAG Logo" src="docs/rag-logo.jpg" width="450" style="object-fit: contain; max-width: 100%; aspect-ratio: 16 / 9;"> -->
+
+🌱 LLM+RAG CLI project operating in On-Premise environment
+
+[![Python](https://img.shields.io/badge/3.9+-3776AB?style=flat&logo=Python&logoColor=white&label=Python)](https://typer.tiangolo.com/)
+[![CLI](https://img.shields.io/badge/CLI-orange?style=flat&logo=iterm2&logoColor=white)](https://typer.tiangolo.com/)
+[![LLM](https://img.shields.io/badge/LLM-green?style=flat&logo=OpenAI&logoColor=white)](https://openai.com)
+[![LangChain](https://img.shields.io/badge/LangChain-blue?style=flat&logo=Langchain&logoColor=white)](https://langchain.com)
+[![Milvus](https://img.shields.io/badge/Milvus-red?style=flat&logo=Milvus&logoColor=white)](https://milvus.io/)
+[![MinIO](https://img.shields.io/badge/MinIO-red?style=flat&logo=MinIO&logoColor=white)](https://min.io/)
+<!-- [![Docker](https://img.shields.io/badge/Docker-blue?style=flat&logo=Docker&logoColor=white)](https://docker.com) -->
+
+</div>
+
+## 🚀 Introduction
+
+**pilot-onpremise-rag** is a CLI tool that implements a knowledge-based RAG (Retrieval-Augmented Generation) system with LLM. It provides powerful document retrieval and generation capabilities while ensuring data privacy.
+
+## 🔧 Setup
+
+### (Optional) Setup External Dependencies
+```bash
+git clone https://github.com/jyje/pilot-onpremise-rag
+cd pilot-onpremise-rag
+
+docker compose -f docker/compose.yaml up -d
+```
+
+### Install pirag
+```bash
+git clone https://github.com/jyje/pilot-onpremise-rag
+cd pilot-onpremise-rag
+
+pip install --upgrade -e ./app
+```
+
+## 📚 Usage
+
+### Basic Commands
+
+```
+# View help
+pirag --help
+
+# Train documents
+pirag train --source ./documents
+
+# Ask a question
+pirag ask "Give me a joke for Cat-holic."
+```
+
+## 🏗️ Project Structure
+
+```
+pilot-onpremise-rag/
+├── app/                        # Main application directory
+│   ├── main.py                 # CLI main entry point
+│   ├── setup.py                # Package setup configuration
+│   ├── pyproject.toml          # PEP 517/518 build configuration
+│   ├── requirements.txt        # Dependencies
+│   ├── logs/                   # Application logs
+│   └── rag/                    # RAG implementation
+│       ├── config.py           # Configuration settings
+│       ├── agent.py            # Agent implementation
+│       ├── ask/                # Query handling module
+│       ├── train/              # Document training module
+│       ├── test/               # Testing module
+│       └── doctor/             # Diagnostic tools
+├── VERSION                     # Project version
+├── docker/                     # Docker configuration
+├── assets/                     # Static assets (Files are not included)
+└── LICENSE                     # License information
+```
+
+## 🔄 How It Works
+
+1. **Document Training**: Process local documents and store in vector database
+2. **Search Engine**: Find document chunks related to user queries
+3. **Context Generation**: Create LLM prompts from retrieved documents
+4. **Response Generation**: Provide accurate responses via local LLM
+
+## 💡 Key Features
+
+- **Privacy Guaranteed**: All data and processing occurs locally
+- **Multiple Document Support**: Support for PDF, Markdown, TXT, DOCX, and other formats
+- **Custom LLM**: Compatible with various local LLM models
+- **Vector Database**: Vector DB integration for efficient document retrieval
+
+## 🧪 Performance Optimization
+
+| Configuration | Memory Usage | Response Speed | Suitable Use Cases |
+|--------------|-------------|---------------|-------------------|
+| Light Model | 4-6GB | Fast | Simple queries, low-spec hardware |
+| Medium Model | 8-12GB | Medium | General use, most queries |
+| Large Model | 16GB+ | Slow | Complex document analysis, expert answers |
+
+## 🔗 References
+
+- [LangChain Documentation](https://python.langchain.com/docs/get_started/introduction)
+- [LLM Optimization Techniques](https://huggingface.co/docs/optimum/index)
+- [RAG Paper](https://arxiv.org/abs/2005.11401)
+
+## Contributing
+
+Any contributions are welcome!
+
+### Current Maintainers
+- [Studio R4iny](https://github.com/studior4iny)
+    - [jyje](https://github.com/jyje), [semir4in](https://github.com/semir4in) (Same person)

pirag-0.1.0/app/main.py

@@ -0,0 +1,80 @@
+import argparse, os
+from dotenv import load_dotenv
+from loguru import logger
+
+load_dotenv(dotenv_path=os.environ.get('ENV_FILE', '.env'), override=True)
+
+from app.rag.config import top_parser, common_parser, setup_logger
+from app.rag.doctor import help as doctor_help, parser as doctor_parser, route as doctor_route
+from app.rag.train import help as train_help, parser as train_parser, route as train_route
+from app.rag.ask import help as ask_help, parser as ask_parser, route as ask_route
+from app.rag.test import help as test_help, parser as test_parser, route as test_route
+
+# Main parser
+parser = argparse.ArgumentParser(
+    formatter_class = argparse.ArgumentDefaultsHelpFormatter,
+    description = 'Pilot of On-Premise RAG',
+    parents = [top_parser],
+    add_help = False,
+)
+
+# Commands
+subparsers = parser.add_subparsers(
+    title = 'commands',
+    dest = 'command',
+)
+
+subparsers.add_parser(
+    'doctor',
+    help = doctor_help,
+    description = doctor_parser.description,
+    parents = [top_parser, common_parser, doctor_parser],
+    add_help = False,
+)
+
+subparsers.add_parser(
+    'train',
+    help = train_help,
+    description = train_parser.description,
+    parents = [top_parser, common_parser, train_parser],
+    add_help = False,
+)
+
+subparsers.add_parser(
+    'test',
+    help = test_help,
+    description = test_parser.description,
+    parents = [top_parser, common_parser, test_parser],
+    add_help = False,
+)
+
+subparsers.add_parser(
+    'ask',
+    help = ask_help,
+    description = ask_parser.description,
+    parents = [top_parser, common_parser, ask_parser],
+    add_help = False,
+)
+
+def main():
+    args = parser.parse_args()
+
+    command_message = f"with command: {args.command}" if args.command else ""
+    logger.info(f"RAG Started {command_message}")
+    
+    setup_logger(args.log_level)
+    logger.debug(f"Parsed arguments: {args}")
+
+    if args.command == 'doctor':
+        doctor_route(args)
+    elif args.command == 'ask':
+        ask_route(args)
+    elif args.command == 'train':
+        train_route(args)
+    elif args.command == 'test':
+        test_route(args)
+    else:
+        parser.print_help()
+
+if __name__ == '__main__':
+    main()

pirag-0.1.0/app/rag/agent.py

@@ -0,0 +1,64 @@
+from langchain_openai.llms import OpenAI
+from langchain_openai.embeddings import OpenAIEmbeddings
+from pymilvus import MilvusClient
+
+class Agent():
+    def __init__(
+        self,
+        llm_base_url: str,
+        llm_api_key: str,
+        llm_model: str,
+        embedding_base_url: str,
+        embedding_api_key: str,
+        embedding_model: str,
+        milvus_host: str,
+        milvus_database: str,
+        milvus_collection: str,
+    ):
+        self.llm_base_url = llm_base_url
+        self.llm_api_key = llm_api_key
+        self.llm_model = llm_model
+        self.embedding_base_url = embedding_base_url
+        self.embedding_api_key = embedding_api_key
+        self.embedding_model = embedding_model
+        self.milvus_host = milvus_host
+        self.milvus_database = milvus_database
+        self.milvus_collection = milvus_collection
+        
+        self._llm_client: OpenAI = None
+        self._embedding_client: OpenAIEmbeddings = None
+        self._milvus_client: MilvusClient = None
+
+    def retrieve_knowledge_base(self, query: str) -> str:
+        llm_client = self._get_llm_client()
+        embedding_client = self._get_embedding_client()
+        milvus_client = self._get_milvus_client()
+    
+        
+    def _get_llm_client(self) -> OpenAI:
+        if self._llm_client is None:
+            self._llm_client = OpenAI(
+                base_url=self.llm_base_url,
+                api_key=self.llm_api_key,
+                model=self.llm_model,
+            )
+        return self._llm_client
+
+
+    def _get_embedding_client(self) -> OpenAIEmbeddings:
+        if self._embedding_client is None:
+            self._embedding_client = OpenAIEmbeddings(
+                base_url=self.embedding_base_url,
+                api_key=self.embedding_api_key,
+                model=self.embedding_model,
+            )
+        return self._embedding_client
+
+    def _get_milvus_client(self) -> MilvusClient:
+        if self._milvus_client is None:
+            self._milvus_client = MilvusClient(
+                host=self.milvus_host,
+                database=self.milvus_database,
+                collection=self.milvus_collection,
+            )
+        return self._milvus_client

pirag-0.1.0/app/rag/config.py

@@ -0,0 +1,159 @@
+import argparse, os, sys
+from pathlib import Path
+from loguru import logger
+
+# Logger format constants
+TIME_FORMAT = "{time:YYYY-MM-DD HH:mm:ss.SSS!UTC}Z"
+FILE_FORMAT = f"{TIME_FORMAT} | {{level: <8}} | {{name}}:{{function}}:{{line}} - {{message}}"
+CONSOLE_FORMAT_FULL = f"<green>{TIME_FORMAT}</green> | <level>{{level: <8}}</level> | <cyan>{{name}}</cyan>:<cyan>{{function}}</cyan>:<cyan>{{line}}</cyan> - <level>{{message}}</level>\n"
+CONSOLE_FORMAT_SIMPLE = f"<green>{TIME_FORMAT}</green> | <level>{{level: <8}}</level> | <level>{{message}}</level>\n"
+
+# Initial logger setup before setup_logger()
+logger.remove()
+logger.add(
+    sink = sys.stderr,
+    level = "INFO", 
+    format = lambda record: CONSOLE_FORMAT_SIMPLE if record["level"].name == "INFO" else CONSOLE_FORMAT_FULL,
+    colorize = True
+)
+
+def setup_logger(log_level: str):
+    """Configure logger with specified level and outputs"""
+    
+    logger.remove()
+    
+    log_dir = Path("app/logs")
+    log_dir.mkdir(exist_ok=True, parents=True)
+    
+    # File handler
+    logger.add(
+        sink = log_dir / "app.log",
+        level = log_level,
+        rotation = "500 MB",
+        format = FILE_FORMAT,
+        serialize = False,
+        enqueue = True,
+        backtrace = True,
+        diagnose = True,
+        catch = True
+    )
+    
+    # Console handler
+    logger.add(
+        sink = sys.stderr,
+        level = log_level,
+        format = lambda record: CONSOLE_FORMAT_SIMPLE if record["level"].name == "INFO" else CONSOLE_FORMAT_FULL,
+        colorize = True
+    )
+
+
+class EnvDefault(argparse.Action):
+    """Custom argparse action that uses environment variables as defaults.
+
+    This action extends the standard argparse.Action to support reading default values
+    from environment variables. If the specified environment variable exists, its value
+    will be used as the default value for the argument.
+
+    For boolean flags (store_true/store_false), the environment variable is interpreted
+    as a boolean value where 'true', '1', 'yes', or 'on' (case-insensitive) are
+    considered True.
+
+    Args:
+        envvar (str): Name of the environment variable to use as default
+        required (bool, optional): Whether the argument is required. Defaults to True.
+            Note: If a default value is found in environment variables, required is set to False.
+        default (Any, optional): Default value if environment variable is not set. Defaults to None.
+        **kwargs: Additional arguments passed to argparse.Action
+
+    Example:
+        ```python
+        parser.add_argument(
+            '--log-level',
+            envvar='LOG_LEVEL',
+            help='Logging level',
+            default='INFO',
+            action=EnvDefault
+        )
+        ```
+
+    Note:
+        The help text is automatically updated to include the environment variable name.
+    """
+    def __init__(self, envvar, required=True, default=None, **kwargs):
+        if envvar and envvar in os.environ:
+            env_value = os.environ[envvar]
+            # Convert string environment variable to boolean
+            if kwargs.get('nargs') is None and kwargs.get('const') is not None:  # store_true/store_false case
+                default = env_value.lower() in ('true', '1', 'yes', 'on')
+            else:
+                default = env_value
+            logger.debug(f"Using {envvar}={default} from environment")
+        
+        if envvar:
+            kwargs["help"] += f" (envvar: {envvar})"
+        
+        if required and default:
+            required = False
+            
+        super(EnvDefault, self).__init__(default=default, required=required, **kwargs)
+        self.envvar = envvar
+
+    def __call__(self, parser, namespace, values, option_string=None):
+        setattr(namespace, self.dest, values if values is not None else self.default)
+
+
+# Top-level parser with common options
+top_parser = argparse.ArgumentParser(add_help=False)
+
+top_parser.add_argument(
+    '-h', '--help',
+    help = 'Show help message and exit',
+    default = argparse.SUPPRESS,
+    action = 'help',
+)
+
+top_parser.add_argument(
+    '--env-file',
+    envvar = 'ENV_FILE',
+    help = 'Path to environment file',
+    default = '.env',
+    type = str,
+    action = EnvDefault,
+)
+
+top_parser.add_argument(
+    '--log-level',
+    envvar = 'LOG_LEVEL',
+    help = 'Logging level',
+    default = 'INFO',
+    type = lambda x: x.upper(),
+    choices = ['DEBUG', 'INFO', 'WARNING', 'ERROR', 'CRITICAL'],
+    required = False,
+    action = EnvDefault,
+)
+
+top_parser.add_argument(
+    '--log-path',
+    envvar = 'LOG_PATH',
+    help = 'Path to log file',
+    default = '.logs',
+    type = str,
+    required = False,
+    action = EnvDefault,
+)
+
+top_parser.add_argument(
+    '--log-save',
+    envvar = 'LOG_SAVE',
+    help = 'Save log to file. If this flag is set, the log will be saved to the file specified in the `--log-path`.',
+    default = False,
+    const = True,
+    nargs = 0,
+    type = bool,
+    required = False,
+    action = EnvDefault,
+)
+
+common_parser = argparse.ArgumentParser(
+    add_help = False,
+)

pirag-0.1.0/app/requirements.txt

@@ -0,0 +1,11 @@
+# Core dependencies
+python-dotenv < 1.2
+loguru < 0.8
+
+# Development
+pytest < 8.4
+black < 25.2
+
+# RAG
+ragas < 0.3
+pymilvus < 2.6

pirag-0.1.0/app/setup.py

@@ -0,0 +1,28 @@
+from setuptools import setup, find_packages
+import os, sys
+import tomli
+
+# Load requirements
+with open(os.path.join(os.path.dirname(__file__), 'requirements.txt'), 'r') as f:
+    requirements = [line.strip() for line in f if line.strip() and not line.startswith('#')]
+
+# Get version from pyproject.toml
+with open(os.path.join(os.path.dirname(__file__), '../pyproject.toml'), 'rb') as f:
+    pyproject = tomli.load(f)
+    version = pyproject["project"]["version"]
+
+APP_NAME = "pirag"
+
+setup(
+    name = APP_NAME,
+    version = version,
+    packages = [".", "rag"],
+    package_dir = {"": ".", "rag": "rag"},
+    include_package_data = True,
+    install_requires = requirements,
+    entry_points = {
+        "console_scripts": [
+            f"{APP_NAME}=main:main",
+        ],
+    },
+)

pirag-0.1.0/pirag.egg-info/PKG-INFO

@@ -0,0 +1,132 @@
+Metadata-Version: 2.4
+Name: pirag
+Version: 0.1.0
+Summary: CLI Projects of On-Premise RAG. You can use your own LLM and vector DB. Or just add remote LLM servers and vector DB.
+Author-email: semir4in <semir4in@gmail.com>, jyje <jyjeon@outlook.com>
+Project-URL: Homepage, https://github.com/jyje/pilot-onpremise-rag
+Project-URL: Repository, https://github.com/jyje/pilot-onpremise-rag
+Project-URL: Issue, https://github.com/jyje/pilot-onpremise-rag/issues
+Requires-Python: >=3.9
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: python-dotenv<1.2
+Requires-Dist: loguru<0.8
+Requires-Dist: pytest<8.4
+Requires-Dist: black<25.2
+Requires-Dist: ragas<0.3
+Requires-Dist: pymilvus<2.6
+Dynamic: license-file
+
+<div align="center">
+
+# pirag: pilot-onpremise-rag
+
+<!-- <img alt="RAG Logo" src="docs/rag-logo.jpg" width="450" style="object-fit: contain; max-width: 100%; aspect-ratio: 16 / 9;"> -->
+
+🌱 LLM+RAG CLI project operating in On-Premise environment
+
+[![Python](https://img.shields.io/badge/3.9+-3776AB?style=flat&logo=Python&logoColor=white&label=Python)](https://typer.tiangolo.com/)
+[![CLI](https://img.shields.io/badge/CLI-orange?style=flat&logo=iterm2&logoColor=white)](https://typer.tiangolo.com/)
+[![LLM](https://img.shields.io/badge/LLM-green?style=flat&logo=OpenAI&logoColor=white)](https://openai.com)
+[![LangChain](https://img.shields.io/badge/LangChain-blue?style=flat&logo=Langchain&logoColor=white)](https://langchain.com)
+[![Milvus](https://img.shields.io/badge/Milvus-red?style=flat&logo=Milvus&logoColor=white)](https://milvus.io/)
+[![MinIO](https://img.shields.io/badge/MinIO-red?style=flat&logo=MinIO&logoColor=white)](https://min.io/)
+<!-- [![Docker](https://img.shields.io/badge/Docker-blue?style=flat&logo=Docker&logoColor=white)](https://docker.com) -->
+
+</div>
+
+## 🚀 Introduction
+
+**pilot-onpremise-rag** is a CLI tool that implements a knowledge-based RAG (Retrieval-Augmented Generation) system with LLM. It provides powerful document retrieval and generation capabilities while ensuring data privacy.
+
+## 🔧 Setup
+
+### (Optional) Setup External Dependencies
+```bash
+git clone https://github.com/jyje/pilot-onpremise-rag
+cd pilot-onpremise-rag
+
+docker compose -f docker/compose.yaml up -d
+```
+
+### Install pirag
+```bash
+git clone https://github.com/jyje/pilot-onpremise-rag
+cd pilot-onpremise-rag
+
+pip install --upgrade -e ./app
+```
+
+## 📚 Usage
+
+### Basic Commands
+
+```
+# View help
+pirag --help
+
+# Train documents
+pirag train --source ./documents
+
+# Ask a question
+pirag ask "Give me a joke for Cat-holic."
+```
+
+## 🏗️ Project Structure
+
+```
+pilot-onpremise-rag/
+├── app/                        # Main application directory
+│   ├── main.py                 # CLI main entry point
+│   ├── setup.py                # Package setup configuration
+│   ├── pyproject.toml          # PEP 517/518 build configuration
+│   ├── requirements.txt        # Dependencies
+│   ├── logs/                   # Application logs
+│   └── rag/                    # RAG implementation
+│       ├── config.py           # Configuration settings
+│       ├── agent.py            # Agent implementation
+│       ├── ask/                # Query handling module
+│       ├── train/              # Document training module
+│       ├── test/               # Testing module
+│       └── doctor/             # Diagnostic tools
+├── VERSION                     # Project version
+├── docker/                     # Docker configuration
+├── assets/                     # Static assets (Files are not included)
+└── LICENSE                     # License information
+```
+
+## 🔄 How It Works
+
+1. **Document Training**: Process local documents and store in vector database
+2. **Search Engine**: Find document chunks related to user queries
+3. **Context Generation**: Create LLM prompts from retrieved documents
+4. **Response Generation**: Provide accurate responses via local LLM
+
+## 💡 Key Features
+
+- **Privacy Guaranteed**: All data and processing occurs locally
+- **Multiple Document Support**: Support for PDF, Markdown, TXT, DOCX, and other formats
+- **Custom LLM**: Compatible with various local LLM models
+- **Vector Database**: Vector DB integration for efficient document retrieval
+
+## 🧪 Performance Optimization
+
+| Configuration | Memory Usage | Response Speed | Suitable Use Cases |
+|--------------|-------------|---------------|-------------------|
+| Light Model | 4-6GB | Fast | Simple queries, low-spec hardware |
+| Medium Model | 8-12GB | Medium | General use, most queries |
+| Large Model | 16GB+ | Slow | Complex document analysis, expert answers |
+
+## 🔗 References
+
+- [LangChain Documentation](https://python.langchain.com/docs/get_started/introduction)
+- [LLM Optimization Techniques](https://huggingface.co/docs/optimum/index)
+- [RAG Paper](https://arxiv.org/abs/2005.11401)
+
+## Contributing
+
+Any contributions are welcome!
+
+### Current Maintainers
+- [Studio R4iny](https://github.com/studior4iny)
+    - [jyje](https://github.com/jyje), [semir4in](https://github.com/semir4in) (Same person)

pirag-0.1.0/pirag.egg-info/SOURCES.txt

@@ -0,0 +1,14 @@
+LICENSE
+README.md
+pyproject.toml
+app/main.py
+app/requirements.txt
+app/setup.py
+app/rag/agent.py
+app/rag/config.py
+pirag.egg-info/PKG-INFO
+pirag.egg-info/SOURCES.txt
+pirag.egg-info/dependency_links.txt
+pirag.egg-info/entry_points.txt
+pirag.egg-info/requires.txt
+pirag.egg-info/top_level.txt

pirag-0.1.0/pirag.egg-info/dependency_links.txt

@@ -0,0 +1 @@
+

pirag-0.1.0/pirag.egg-info/entry_points.txt

@@ -0,0 +1,2 @@
+[console_scripts]
+pirag = app.main:main

pirag-0.1.0/pirag.egg-info/requires.txt

@@ -0,0 +1,6 @@
+python-dotenv<1.2
+loguru<0.8
+pytest<8.4
+black<25.2
+ragas<0.3
+pymilvus<2.6

pirag-0.1.0/pirag.egg-info/top_level.txt

@@ -0,0 +1 @@
+app

pirag-0.1.0/pyproject.toml

@@ -0,0 +1,39 @@
+[project]
+name = "pirag"
+version = "0.1.0"
+description = "CLI Projects of On-Premise RAG. You can use your own LLM and vector DB. Or just add remote LLM servers and vector DB."
+authors = [
+    { name="semir4in", email="semir4in@gmail.com" },
+    { name="jyje", email="jyjeon@outlook.com" }
+]
+readme = "README.md"
+requires-python = ">=3.9"
+dynamic = ["dependencies"]
+
+[project.urls]
+Homepage = "https://github.com/jyje/pilot-onpremise-rag"
+Repository = "https://github.com/jyje/pilot-onpremise-rag"
+Issue = "https://github.com/jyje/pilot-onpremise-rag/issues"
+
+[project.scripts]
+pirag = "app.main:main"
+
+[build-system]
+requires = [
+    "setuptools>=61.0",
+    "wheel<0.46",
+    "build>=1.0.3",
+    "pip>=23.0",
+    "tomli<2.1",
+]
+build-backend = "setuptools.build_meta"
+
+[tool.setuptools]
+packages = ["app", "app.rag"]
+
+[tool.setuptools.package-dir]
+app = "app"
+"app.rag" = "app/rag"
+
+[tool.setuptools.dynamic]
+dependencies = {file = ["app/requirements.txt"]}

pirag-0.1.0/setup.cfg

@@ -0,0 +1,4 @@
+[egg_info]
+tag_build = 
+tag_date = 0
+