gptmed 0.0.1__tar.gz → 0.3.0__tar.gz

{gptmed-0.0.1/gptmed.egg-info → gptmed-0.3.0}/PKG-INFO

@@ -1,31 +1,10 @@
 Metadata-Version: 2.4
 Name: gptmed
-Version: 0.0.1
+Version: 0.3.0
 Summary: A lightweight GPT-based language model framework for training custom question-answering models on any domain
 Author-email: Sanjog Sigdel <sigdelsanjog@gmail.com>
 Maintainer-email: Sanjog Sigdel <sigdelsanjog@gmail.com>
-License: MIT License
-        
-        Copyright (c) 2026 Your Name
-        
-        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.
-        
+License-Expression: MIT
 Project-URL: Homepage, https://github.com/sigdelsanjog/gptmed
 Project-URL: Documentation, https://github.com/sigdelsanjog/gptmed#readme
 Project-URL: Repository, https://github.com/sigdelsanjog/gptmed
@@ -37,7 +16,6 @@ Classifier: Intended Audience :: Science/Research
 Classifier: Intended Audience :: Education
 Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
 Classifier: Topic :: Software Development :: Libraries :: Python Modules
-Classifier: License :: OSI Approved :: MIT License
 Classifier: Programming Language :: Python :: 3
 Classifier: Programming Language :: Python :: 3.8
 Classifier: Programming Language :: Python :: 3.9
@@ -51,6 +29,7 @@ Requires-Dist: torch>=2.0.0
 Requires-Dist: sentencepiece>=0.1.99
 Requires-Dist: numpy>=1.24.0
 Requires-Dist: tqdm>=4.65.0
+Requires-Dist: pyyaml>=6.0
 Provides-Extra: dev
 Requires-Dist: pytest>=7.0.0; extra == "dev"
 Requires-Dist: black>=22.0.0; extra == "dev"
@@ -69,6 +48,10 @@ A lightweight GPT-based language model framework for training custom question-an
 [![Python 3.8+](https://img.shields.io/badge/python-3.8+-blue.svg)](https://www.python.org/downloads/)
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
 
+## 📖 [Complete User Manual](USER_MANUAL.md) | [Quick Start](#quick-start)
+
+> **New to GptMed?** Check out the [**step-by-step User Manual**](USER_MANUAL.md) for a complete guide on training your own model!
+
 ## Features
 
 - 🧠 **Custom GPT Architecture**: Lightweight transformer model for any Q&A domain
@@ -78,6 +61,27 @@ A lightweight GPT-based language model framework for training custom question-an
 - 📦 **Lightweight**: Small model size suitable for edge deployment
 - 🛠️ **Complete Toolkit**: Includes tokenizer training, model training, and inference utilities
 
+## Table of Contents
+
+- [Features](#features)
+- [Installation](#installation)
+- [Quick Start](#quick-start)
+- [Package Structure](#package-structure)
+  - [Core Modules](#core-modules)
+  - [Model Components](#model-components)
+  - [Training Components](#training-components)
+  - [Inference Components](#inference-components)
+  - [Data Processing](#data-processing)
+  - [Utilities](#utilities)
+- [Model Architecture](#model-architecture)
+- [Configuration](#configuration)
+- [Documentation](#documentation)
+- [Performance](#performance)
+- [Examples](#examples)
+- [Contributing](#contributing)
+- [License](#license)
+- [Support](#support)
+
 ## Installation
 
 ### From PyPI (Recommended)
@@ -204,27 +208,134 @@ config = TrainingConfig(
 )
 ```
 
-## Project Structure
+## Package Structure
+
+### Core Modules
+
+The `gptmed` package contains the following main modules:
+
+```
+gptmed/
+├── model/                  # Model architecture and configurations
+├── inference/              # Text generation and sampling
+├── training/               # Training loops and datasets
+├── tokenizer/              # Tokenizer training and data processing
+├── data/                   # Data parsers and formatters
+├── configs/                # Training configurations
+└── utils/                  # Utilities (checkpoints, logging)
+```
+
+### Model Components
+
+**`gptmed.model.architecture`** - GPT Transformer Implementation
+
+- `GPTTransformer` - Main model class
+- `TransformerBlock` - Individual transformer layers
+- `MultiHeadAttention` - Attention mechanism
+- `FeedForward` - Feed-forward networks
+- `RoPEPositionalEncoding` - Rotary position embeddings
+
+**`gptmed.model.configs`** - Model Configurations
+
+- `get_tiny_config()` - ~2M parameters (testing)
+- `get_small_config()` - ~10M parameters (recommended)
+- `get_medium_config()` - ~50M parameters (high quality)
+- `ModelConfig` - Custom configuration class
+
+### Training Components
+
+**`gptmed.training`** - Training Pipeline
+
+- `train.py` - Main training script (CLI: `gptmed-train`)
+- `Trainer` - Training loop with checkpointing
+- `TokenizedDataset` - PyTorch dataset for tokenized data
+- `create_dataloaders()` - DataLoader creation utilities
+
+**`gptmed.configs`** - Training Configurations
+
+- `TrainingConfig` - Training hyperparameters
+- `get_default_config()` - Default training settings
+- `get_quick_test_config()` - Fast testing configuration
+
+### Inference Components
+
+**`gptmed.inference`** - Text Generation
+
+- `TextGenerator` - Main generation class
+- `generator.py` - CLI command (CLI: `gptmed-generate`)
+- `sampling.py` - Sampling strategies (top-k, top-p, temperature)
+- `decoding_utils.py` - Decoding utilities
+- `GenerationConfig` - Generation parameters
+
+### Data Processing
+
+**`gptmed.tokenizer`** - Tokenizer Training & Data Processing
+
+- `train_tokenizer.py` - Train SentencePiece tokenizer
+- `tokenize_data.py` - Convert text to token sequences
+- SentencePiece BPE tokenizer support
+
+**`gptmed.data.parsers`** - Data Parsing & Formatting
+
+- `MedQuADParser` - XML Q&A parser (example)
+- `CausalTextFormatter` - Format Q&A pairs for training
+- `FormatConfig` - Formatting configuration
+
+### Utilities
+
+**`gptmed.utils`** - Helper Functions
+
+- `checkpoints.py` - Model checkpoint management
+- `logging.py` - Training metrics logging
+
+---
+
+## Detailed Project Structure
 
 ```
 gptmed/
 ├── model/
-│   ├── architecture/      # GPT transformer implementation
-│   └── configs/           # Model configurations
+│   ├── architecture/
+│   │   ├── gpt.py              # GPT transformer model
+│   │   ├── attention.py        # Multi-head attention
+│   │   ├── feedforward.py      # Feed-forward networks
+│   │   └── embeddings.py       # Token + positional embeddings
+│   └── configs/
+│       └── model_config.py     # Model size configurations
 ├── inference/
-│   ├── generator.py       # Text generation
-│   └── sampling.py        # Sampling strategies
+│   ├── generator.py            # Text generation (CLI command)
+│   ├── sampling.py             # Sampling strategies
+│   ├── decoding_utils.py       # Decoding utilities
+│   └── generation_config.py    # Generation parameters
 ├── training/
-│   ├── train.py          # Training script
-│   ├── trainer.py        # Training loop
-│   └── dataset.py        # Data loading
+│   ├── train.py                # Main training script (CLI command)
+│   ├── trainer.py              # Training loop
+│   ├── dataset.py              # PyTorch dataset
+│   └── utils.py                # Training utilities
 ├── tokenizer/
-│   └── train_tokenizer.py # SentencePiece tokenizer
+│   ├── train_tokenizer.py      # Train SentencePiece tokenizer
+│   └── tokenize_data.py        # Tokenize text data
+├── data/
+│   └── parsers/
+│       ├── medquad_parser.py   # Example XML parser
+│       └── text_formatter.py   # Q&A text formatter
 ├── configs/
-│   └── train_config.py   # Training configurations
+│   └── train_config.py         # Training configurations
 └── utils/
-    ├── checkpoints.py    # Model checkpointing
-    └── logging.py        # Training logging
+    ├── checkpoints.py          # Model checkpointing
+    └── logging.py              # Training logging
+```
+
+### Command-Line Interface
+
+The package provides two main CLI commands:
+
+```bash
+# Train a model
+gptmed-train --model-size small --num-epochs 10 --batch-size 16
+
+# Generate text
+gptmed-generate --prompt "Your question?" --max-length 100
 ```
 
 ## Requirements
@@ -237,14 +348,14 @@ gptmed/
 
 ## Documentation
 
-For detailed documentation, visit [GitHub Repository](https://github.com/yourusername/medllm).
+📚 **[Complete User Manual](USER_MANUAL.md)** - Step-by-step guide for training your own model
 
-### Key Guides
+### Quick Links
 
-- [Training Guide](docs/training.md)
-- [Inference Guide](docs/inference.md)
-- [Model Architecture](docs/architecture.md)
-- [API Reference](docs/api.md)
+- [User Manual](USER_MANUAL.md) - **Start here!** Complete training pipeline guide
+- [Architecture Guide](ARCHITECTURE_EXTENSION_GUIDE.md) - Understanding the model architecture
+- [Deployment Guide](DEPLOYMENT_GUIDE.md) - Publishing to PyPI
+- [Changelog](CHANGELOG.md) - Version history
 
 ## Performance
 
@@ -312,7 +423,8 @@ This project is licensed under the MIT License - see the [LICENSE](LICENSE) file
 
 ## Support
 
-- 📫 Issues: [GitHub Issues](https://github.com/sigdelsanjog/gptmed/issues)
+- � **[User Manual](USER_MANUAL.md)** - Complete step-by-step training guide
+- �📫 Issues: [GitHub Issues](https://github.com/sigdelsanjog/gptmed/issues)
 - 💬 Discussions: [GitHub Discussions](https://github.com/sigdelsanjog/gptmed/discussions)
 - 📧 Email: sanjog.sigdel@ku.edu.np
 

{gptmed-0.0.1 → gptmed-0.3.0}/README.md

@@ -6,6 +6,10 @@ A lightweight GPT-based language model framework for training custom question-an
 [![Python 3.8+](https://img.shields.io/badge/python-3.8+-blue.svg)](https://www.python.org/downloads/)
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
 
+## 📖 [Complete User Manual](USER_MANUAL.md) | [Quick Start](#quick-start)
+
+> **New to GptMed?** Check out the [**step-by-step User Manual**](USER_MANUAL.md) for a complete guide on training your own model!
+
 ## Features
 
 - 🧠 **Custom GPT Architecture**: Lightweight transformer model for any Q&A domain
@@ -15,6 +19,27 @@ A lightweight GPT-based language model framework for training custom question-an
 - 📦 **Lightweight**: Small model size suitable for edge deployment
 - 🛠️ **Complete Toolkit**: Includes tokenizer training, model training, and inference utilities
 
+## Table of Contents
+
+- [Features](#features)
+- [Installation](#installation)
+- [Quick Start](#quick-start)
+- [Package Structure](#package-structure)
+  - [Core Modules](#core-modules)
+  - [Model Components](#model-components)
+  - [Training Components](#training-components)
+  - [Inference Components](#inference-components)
+  - [Data Processing](#data-processing)
+  - [Utilities](#utilities)
+- [Model Architecture](#model-architecture)
+- [Configuration](#configuration)
+- [Documentation](#documentation)
+- [Performance](#performance)
+- [Examples](#examples)
+- [Contributing](#contributing)
+- [License](#license)
+- [Support](#support)
+
 ## Installation
 
 ### From PyPI (Recommended)
@@ -141,27 +166,134 @@ config = TrainingConfig(
 )
 ```
 
-## Project Structure
+## Package Structure
+
+### Core Modules
+
+The `gptmed` package contains the following main modules:
+
+```
+gptmed/
+├── model/                  # Model architecture and configurations
+├── inference/              # Text generation and sampling
+├── training/               # Training loops and datasets
+├── tokenizer/              # Tokenizer training and data processing
+├── data/                   # Data parsers and formatters
+├── configs/                # Training configurations
+└── utils/                  # Utilities (checkpoints, logging)
+```
+
+### Model Components
+
+**`gptmed.model.architecture`** - GPT Transformer Implementation
+
+- `GPTTransformer` - Main model class
+- `TransformerBlock` - Individual transformer layers
+- `MultiHeadAttention` - Attention mechanism
+- `FeedForward` - Feed-forward networks
+- `RoPEPositionalEncoding` - Rotary position embeddings
+
+**`gptmed.model.configs`** - Model Configurations
+
+- `get_tiny_config()` - ~2M parameters (testing)
+- `get_small_config()` - ~10M parameters (recommended)
+- `get_medium_config()` - ~50M parameters (high quality)
+- `ModelConfig` - Custom configuration class
+
+### Training Components
+
+**`gptmed.training`** - Training Pipeline
+
+- `train.py` - Main training script (CLI: `gptmed-train`)
+- `Trainer` - Training loop with checkpointing
+- `TokenizedDataset` - PyTorch dataset for tokenized data
+- `create_dataloaders()` - DataLoader creation utilities
+
+**`gptmed.configs`** - Training Configurations
+
+- `TrainingConfig` - Training hyperparameters
+- `get_default_config()` - Default training settings
+- `get_quick_test_config()` - Fast testing configuration
+
+### Inference Components
+
+**`gptmed.inference`** - Text Generation
+
+- `TextGenerator` - Main generation class
+- `generator.py` - CLI command (CLI: `gptmed-generate`)
+- `sampling.py` - Sampling strategies (top-k, top-p, temperature)
+- `decoding_utils.py` - Decoding utilities
+- `GenerationConfig` - Generation parameters
+
+### Data Processing
+
+**`gptmed.tokenizer`** - Tokenizer Training & Data Processing
+
+- `train_tokenizer.py` - Train SentencePiece tokenizer
+- `tokenize_data.py` - Convert text to token sequences
+- SentencePiece BPE tokenizer support
+
+**`gptmed.data.parsers`** - Data Parsing & Formatting
+
+- `MedQuADParser` - XML Q&A parser (example)
+- `CausalTextFormatter` - Format Q&A pairs for training
+- `FormatConfig` - Formatting configuration
+
+### Utilities
+
+**`gptmed.utils`** - Helper Functions
+
+- `checkpoints.py` - Model checkpoint management
+- `logging.py` - Training metrics logging
+
+---
+
+## Detailed Project Structure
 
 ```
 gptmed/
 ├── model/
-│   ├── architecture/      # GPT transformer implementation
-│   └── configs/           # Model configurations
+│   ├── architecture/
+│   │   ├── gpt.py              # GPT transformer model
+│   │   ├── attention.py        # Multi-head attention
+│   │   ├── feedforward.py      # Feed-forward networks
+│   │   └── embeddings.py       # Token + positional embeddings
+│   └── configs/
+│       └── model_config.py     # Model size configurations
 ├── inference/
-│   ├── generator.py       # Text generation
-│   └── sampling.py        # Sampling strategies
+│   ├── generator.py            # Text generation (CLI command)
+│   ├── sampling.py             # Sampling strategies
+│   ├── decoding_utils.py       # Decoding utilities
+│   └── generation_config.py    # Generation parameters
 ├── training/
-│   ├── train.py          # Training script
-│   ├── trainer.py        # Training loop
-│   └── dataset.py        # Data loading
+│   ├── train.py                # Main training script (CLI command)
+│   ├── trainer.py              # Training loop
+│   ├── dataset.py              # PyTorch dataset
+│   └── utils.py                # Training utilities
 ├── tokenizer/
-│   └── train_tokenizer.py # SentencePiece tokenizer
+│   ├── train_tokenizer.py      # Train SentencePiece tokenizer
+│   └── tokenize_data.py        # Tokenize text data
+├── data/
+│   └── parsers/
+│       ├── medquad_parser.py   # Example XML parser
+│       └── text_formatter.py   # Q&A text formatter
 ├── configs/
-│   └── train_config.py   # Training configurations
+│   └── train_config.py         # Training configurations
 └── utils/
-    ├── checkpoints.py    # Model checkpointing
-    └── logging.py        # Training logging
+    ├── checkpoints.py          # Model checkpointing
+    └── logging.py              # Training logging
+```
+
+### Command-Line Interface
+
+The package provides two main CLI commands:
+
+```bash
+# Train a model
+gptmed-train --model-size small --num-epochs 10 --batch-size 16
+
+# Generate text
+gptmed-generate --prompt "Your question?" --max-length 100
 ```
 
 ## Requirements
@@ -174,14 +306,14 @@ gptmed/
 
 ## Documentation
 
-For detailed documentation, visit [GitHub Repository](https://github.com/yourusername/medllm).
+📚 **[Complete User Manual](USER_MANUAL.md)** - Step-by-step guide for training your own model
 
-### Key Guides
+### Quick Links
 
-- [Training Guide](docs/training.md)
-- [Inference Guide](docs/inference.md)
-- [Model Architecture](docs/architecture.md)
-- [API Reference](docs/api.md)
+- [User Manual](USER_MANUAL.md) - **Start here!** Complete training pipeline guide
+- [Architecture Guide](ARCHITECTURE_EXTENSION_GUIDE.md) - Understanding the model architecture
+- [Deployment Guide](DEPLOYMENT_GUIDE.md) - Publishing to PyPI
+- [Changelog](CHANGELOG.md) - Version history
 
 ## Performance
 
@@ -249,7 +381,8 @@ This project is licensed under the MIT License - see the [LICENSE](LICENSE) file
 
 ## Support
 
-- 📫 Issues: [GitHub Issues](https://github.com/sigdelsanjog/gptmed/issues)
+- � **[User Manual](USER_MANUAL.md)** - Complete step-by-step training guide
+- �📫 Issues: [GitHub Issues](https://github.com/sigdelsanjog/gptmed/issues)
 - 💬 Discussions: [GitHub Discussions](https://github.com/sigdelsanjog/gptmed/discussions)
 - 📧 Email: sanjog.sigdel@ku.edu.np
 

gptmed-0.3.0/gptmed/__init__.py

@@ -0,0 +1,60 @@
+"""
+GptMed: A lightweight GPT-based language model framework
+
+A domain-agnostic framework for training custom question-answering models.
+Train your own GPT model on any Q&A dataset - medical, technical support,
+education, or any other domain.
+
+Quick Start:
+    >>> import gptmed
+    >>> 
+    >>> # 1. Create a config file
+    >>> gptmed.create_config('my_config.yaml')
+    >>> 
+    >>> # 2. Edit my_config.yaml with your settings
+    >>> 
+    >>> # 3. Train your model
+    >>> results = gptmed.train_from_config('my_config.yaml')
+    >>> 
+    >>> # 4. Generate answers
+    >>> answer = gptmed.generate(
+    ...     checkpoint=results['best_checkpoint'],
+    ...     tokenizer='tokenizer/my_tokenizer.model',
+    ...     prompt='Your question here?'
+    ... )
+
+Advanced Usage:
+    >>> from gptmed.model.architecture import GPTTransformer
+    >>> from gptmed.model.configs.model_config import get_small_config
+    >>> from gptmed.inference.generator import TextGenerator
+    >>> 
+    >>> config = get_small_config()
+    >>> model = GPTTransformer(config)
+"""
+
+__version__ = "0.3.0"
+__author__ = "Sanjog Sigdel"
+__email__ = "sigdelsanjog@gmail.com"
+
+# High-level API - Main user interface
+from gptmed.api import (
+    create_config,
+    train_from_config,
+    generate,
+)
+
+# Expose main components at package level for convenience
+from gptmed.model.architecture import GPTTransformer
+from gptmed.model.configs.model_config import ModelConfig, get_small_config, get_tiny_config
+
+__all__ = [
+    # Simple API
+    "create_config",
+    "train_from_config", 
+    "generate",
+    # Advanced API
+    "GPTTransformer",
+    "ModelConfig",
+    "get_small_config",
+    "get_tiny_config",
+]