langvision 0.0.2__tar.gz → 0.1.0__tar.gz

{langvision-0.0.2 → langvision-0.1.0}/MANIFEST.in

@@ -1,4 +1,3 @@
-include LICENSE
 include README.md
 include requirements.txt
 recursive-include docs * 

langvision-0.1.0/PKG-INFO

@@ -0,0 +1,49 @@
+Metadata-Version: 2.1
+Name: langvision
+Version: 0.1.0
+Summary: Efficient LoRA Fine-Tuning for Vision LLMs with advanced CLI and model zoo
+Home-page: https://github.com/langtrain-ai/langtrain
+Author: Pritesh Raj
+Author-email: priteshraj10@gmail.com
+Keywords: vision,transformer,lora,fine-tuning,deep-learning,computer-vision
+Classifier: Development Status :: 4 - Beta
+Classifier: Intended Audience :: Developers
+Classifier: Intended Audience :: Science/Research
+Classifier: Operating System :: OS Independent
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.8
+Classifier: Programming Language :: Python :: 3.9
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
+Classifier: Topic :: Software Development :: Libraries :: Python Modules
+Requires-Python: >=3.8
+Requires-Dist: torch>=1.10.0
+Requires-Dist: torchvision>=0.11.0
+Requires-Dist: numpy>=1.21.0
+Requires-Dist: pillow>=8.3.0
+Requires-Dist: tqdm>=4.62.0
+Requires-Dist: pyyaml>=6.0
+Requires-Dist: tensorboard>=2.9.0
+Requires-Dist: wandb>=0.13.0
+Provides-Extra: dev
+Requires-Dist: pytest>=7.0.0; extra == "dev"
+Requires-Dist: pytest-cov>=4.0.0; extra == "dev"
+Requires-Dist: pytest-mock>=3.8.0; extra == "dev"
+Requires-Dist: black>=22.0.0; extra == "dev"
+Requires-Dist: isort>=5.10.0; extra == "dev"
+Requires-Dist: flake8>=5.0.0; extra == "dev"
+Requires-Dist: mypy>=1.0.0; extra == "dev"
+Requires-Dist: pre-commit>=2.20.0; extra == "dev"
+Requires-Dist: bandit>=1.7.0; extra == "dev"
+Provides-Extra: docs
+Requires-Dist: sphinx>=5.0.0; extra == "docs"
+Requires-Dist: sphinx-rtd-theme>=1.0.0; extra == "docs"
+Requires-Dist: myst-parser>=0.18.0; extra == "docs"
+Requires-Dist: sphinx-autodoc-typehints>=1.19.0; extra == "docs"
+Provides-Extra: examples
+Requires-Dist: jupyter>=1.0.0; extra == "examples"
+Requires-Dist: ipywidgets>=7.6.0; extra == "examples"
+Requires-Dist: tensorboard>=2.9.0; extra == "examples"
+Requires-Dist: wandb>=0.13.0; extra == "examples"

langvision-0.0.2/src/langvision.egg-info/PKG-INFO → langvision-0.1.0/README.md

@@ -1,46 +1,4 @@
-Metadata-Version: 2.4
-Name: langvision
-Version: 0.0.2
-Summary: A package for finetuning vision models.
-Author-email: Pritesh Raj <priteshraj10@gmail.com>
-License: MIT License
-        
-        Copyright (c) 2025 Plim
-        
-        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. 
-Project-URL: Homepage, https://github.com/langtrain-ai/langtrain
-Project-URL: Documentation, https://github.com/langtrain-ai/langtrain/tree/main/docs
-Project-URL: Source, https://github.com/langtrain-ai/langtrain
-Project-URL: Tracker, https://github.com/langtrain-ai/langtrain/issues
-Requires-Python: >=3.8
-Description-Content-Type: text/markdown
-License-File: LICENSE
-Requires-Dist: torch>=1.10
-Requires-Dist: numpy
-Requires-Dist: tqdm
-Requires-Dist: pyyaml
-Requires-Dist: scipy
-Requires-Dist: matplotlib
-Requires-Dist: pillow
-Dynamic: license-file
-
-# Langvision: Vision LLMs with Efficient LoRA Fine-Tuning
+# Langvision: Efficient LoRA Fine-Tuning for Vision LLMs
 
 <hr/>
 <p align="center">
@@ -60,11 +18,46 @@ Dynamic: license-file
   <a href="https://github.com/psf/black"><img src="https://img.shields.io/badge/code%20style-black-000000.svg" alt="Code style: black"></a>
 </p>
 
-<p align="center">
-  <b>Langvision provides modular components for vision models and LoRA-based fine-tuning.</b><br/>
-  <span style="font-size:1.1em"><i>Adapt and fine-tune vision models for a range of tasks.</i></span>
-</p>
-<hr/>
+---
+
+## 🚀 Installation
+
+> **Requirements:**  
+> - Python 3.8 or newer  
+> - [PyTorch](https://pytorch.org/get-started/locally/) (CPU or GPU)
+
+**Install the latest release from PyPI:**
+
+```bash
+pip install langvision
+```
+
+[![PyPI version](https://img.shields.io/pypi/v/langvision.svg?label=View%20on%20PyPI)](https://pypi.org/project/langvision/)
+
+**Optional installs:**
+
+- For development:
+  ```bash
+  pip install langvision[dev]
+  ```
+- For documentation:
+  ```bash
+  pip install langvision[docs]
+  ```
+- For examples and notebooks:
+  ```bash
+  pip install langvision[examples]
+  ```
+- For GPU (CUDA 11.3+):
+  ```bash
+  pip install langvision[gpu]
+  ```
+
+**Troubleshooting:**
+- For GPU support, ensure you have the correct CUDA version and install PyTorch as per [official instructions](https://pytorch.org/get-started/locally/).
+- If you encounter issues, see [Troubleshooting](docs/troubleshooting.md) or [open an issue](https://github.com/langtrain-ai/langtrain/issues).
+
+---
 
 ## Quick Links
 - [Documentation](docs/index.md)
@@ -90,41 +83,86 @@ Dynamic: license-file
 - [Examples & Use Cases](#examples--use-cases)
 - [Extending the Framework](#extending-the-framework)
 - [Contributing](#contributing)
-- [FAQ](#faq)
+- [License](#license)
 - [Citation](#citation)
 - [Acknowledgements](#acknowledgements)
-- [License](#license)
+- [Advanced LLM Concepts](#advanced-llm-concepts)
 
 ---
 
 ## Features
-- LoRA adapters for parameter-efficient fine-tuning
+- LoRA adapters for efficient fine-tuning
 - Modular Vision Transformer (ViT) backbone
-- Model zoo for open-source visual models
+- Model zoo for vision models
 - Configurable and extensible codebase
-- Checkpointing and resume support
+- Checkpointing and resume
 - Mixed precision and distributed training
-- Built-in metrics and visualization tools
-- CLI for fine-tuning and evaluation
-- Extensible callbacks (early stopping, logging, etc.)
+- Metrics and visualization tools
+- CLI for training and evaluation
+- Callback support (early stopping, logging, etc.)
 
 ---
 
 ## Showcase
 
-Langvision is a framework for building and fine-tuning vision models with LoRA support. It is suitable for tasks such as image classification, visual question answering, and custom vision applications.
+Langvision is intended for building and fine-tuning vision models with LoRA. It can be used for image classification, visual question answering, and other computer vision tasks.
 
 ---
 
 ## Getting Started
 
-Install with pip:
+Install:
 
 ```bash
 pip install langvision
 ```
 
-Minimal example:
+### CLI Usage
+
+After installation, you can use the comprehensive CLI:
+
+```bash
+# Show all available commands
+langvision --help
+
+# Training commands
+langvision train --help
+langvision finetune --help
+
+# Model management
+langvision evaluate --help
+langvision export --help
+langvision model-zoo --help
+langvision config --help
+```
+
+**Quick Examples:**
+
+```bash
+# Train a model
+langvision train --dataset cifar10 --epochs 5 --batch_size 32
+
+# Fine-tune with advanced features
+langvision finetune --dataset cifar100 --epochs 10 --lora_r 8 --rlhf
+
+# Evaluate a trained model
+langvision evaluate --checkpoint model.pth --dataset cifar10
+
+# Export to ONNX
+langvision export --checkpoint model.pth --format onnx --output model.onnx
+
+# Browse model zoo
+langvision model-zoo list
+langvision model-zoo download vit_base_patch16_224
+
+# Create configuration
+langvision config create --template advanced --output my_config.yaml
+
+# Check version
+langvision --version
+```
+
+Example usage:
 
 ```python
 import torch
@@ -149,28 +187,28 @@ with torch.no_grad():
     print('Output shape:', out.shape)
 ```
 
-For more details, see the [Documentation](docs/index.md) and [src/langvision/cli/finetune.py](src/langvision/cli/finetune.py).
+See the [Documentation](docs/index.md) and [src/langvision/cli/finetune.py](src/langvision/cli/finetune.py) for more details.
 
 ---
 
 ## Supported Python Versions
-- Python 3.8+
+- Python 3.8 or newer
 
 ---
 
 ## Why langvision?
 
-- Parameter-efficient fine-tuning with LoRA adapters
-- Modular ViT backbone for flexible model design
-- Unified interface for open-source vision models
-- Designed for both research and production
-- Efficient memory usage for large models
+- Fine-tuning with LoRA adapters
+- Modular ViT backbone design
+- Unified interface for vision models
+- Suitable for research and production
+- Efficient memory usage
 
 ---
 
 ## Architecture Overview
 
-Langvision uses a modular Vision Transformer backbone with LoRA adapters in attention and MLP layers. This allows adaptation of pre-trained models with fewer trainable parameters.
+Langvision uses a Vision Transformer backbone with LoRA adapters in attention and MLP layers. This enables adaptation of pre-trained models with fewer trainable parameters.
 
 ### Model Data Flow
 
@@ -269,7 +307,7 @@ lora:
 ---
 
 ## Documentation & Resources
-- [API Reference](docs/api/index.md)
+- [API Reference](docs/index.md)
 - [Tutorials and Examples](docs/tutorials/index.md)
 - [Research Papers](#research-papers)
 - [Best Practices Guide](docs/best_practices.md)
@@ -340,12 +378,14 @@ model.finetune(dataset, config_path="configs/custom_config.yaml")
 - For advanced usage, see `src/langvision/cli/finetune.py`.
 
 ## Contributing
-We welcome contributions. See the [Contributing Guide](CONTRIBUTING.md) for details.
+Contributions are welcome. See the [Contributing Guide](CONTRIBUTING.md) for details.
 
-## License & Citation
+## License
 
 This project is licensed under the MIT License. See [LICENSE](LICENSE) for details.
 
+## Citation
+
 If you use langvision in your research, please cite:
 
 ```bibtex
@@ -370,3 +410,19 @@ We thank the following projects and communities:
   <b>Made in India 🇮🇳 with ❤️ by the langtrain team</b><br/>
   <i>Star ⭐ this repo if you find it useful!</i>
 </p>
+
+## Advanced LLM Concepts
+
+This library is designed to be extensible with advanced concepts in large language models (LLMs) and explainable AI, including:
+
+- **RLHF**: Reinforcement Learning from Human Feedback
+- **CoT**: Chain-of-Thought
+- **CCoT**: Contrastive Chain-of-Thought
+- **GRPO**: Generalized Reinforcement Policy Optimization
+- **RLVR**: Reinforcement Learning with Value Ranking
+- **DPO**: Direct Preference Optimization
+- **PPO**: Proximal Policy Optimization
+- **LIME**: Local Interpretable Model-agnostic Explanations
+- **SHAP**: SHapley Additive exPlanations
+
+See the [`src/langvision/concepts/`](src/langvision/concepts/) directory for modularized explanations and stubs for each concept. These serve as a foundation for future development and integration into the library's training and evaluation workflows.

langvision-0.1.0/docs/_config.yml

@@ -0,0 +1,11 @@
+remote_theme: pages-themes/cayman@v0.2.0
+plugins:
+- jekyll-remote-theme
+
+title: Langvision
+description: Efficient Vision LLMs with LoRA Fine-Tuning
+show_downloads: true
+github:
+  repository_url: https://github.com/langtrain-ai/langvision
+  zip_url: https://github.com/langtrain-ai/langvision/zipball/main
+  tar_url: https://github.com/langtrain-ai/langvision/tarball/main

langvision-0.1.0/requirements.txt

@@ -0,0 +1,34 @@
+# Core dependencies
+torch>=1.10.0
+torchvision>=0.11.0
+numpy>=1.21.0
+tqdm>=4.62.0
+pyyaml>=6.0
+scipy>=1.7.0
+matplotlib>=3.5.0
+pillow>=8.3.0
+timm>=0.6.0
+transformers>=4.20.0
+
+# Development dependencies (install with: pip install -r requirements-dev.txt)
+# pytest>=7.0.0
+# pytest-cov>=4.0.0
+# pytest-mock>=3.8.0
+# black>=22.0.0
+# isort>=5.10.0
+# flake8>=5.0.0
+# mypy>=1.0.0
+# pre-commit>=2.20.0
+# bandit>=1.7.0
+
+# Documentation dependencies (install with: pip install -r requirements-docs.txt)
+# sphinx>=5.0.0
+# sphinx-rtd-theme>=1.0.0
+# myst-parser>=0.18.0
+# sphinx-autodoc-typehints>=1.19.0
+
+# Example dependencies (install with: pip install -r requirements-examples.txt)
+# jupyter>=1.0.0
+# ipywidgets>=7.6.0
+# tensorboard>=2.9.0
+# wandb>=0.13.0

langvision-0.1.0/setup.py

@@ -0,0 +1,70 @@
+from setuptools import setup, find_packages
+
+setup(
+    name="langvision",
+    version="0.1.0",
+    description="Efficient LoRA Fine-Tuning for Vision LLMs with advanced CLI and model zoo",
+    author="Pritesh Raj",
+    author_email="priteshraj10@gmail.com",
+    url="https://github.com/langtrain-ai/langtrain",
+    packages=find_packages(where="src"),
+    package_dir={"": "src"},
+    python_requires=">=3.8",
+    install_requires=[
+        "torch>=1.10.0",
+        "torchvision>=0.11.0",
+        "numpy>=1.21.0",
+        "pillow>=8.3.0",
+        "tqdm>=4.62.0",
+        "pyyaml>=6.0",
+        "tensorboard>=2.9.0",
+        "wandb>=0.13.0",
+    ],
+    extras_require={
+        "dev": [
+            "pytest>=7.0.0",
+            "pytest-cov>=4.0.0",
+            "pytest-mock>=3.8.0",
+            "black>=22.0.0",
+            "isort>=5.10.0",
+            "flake8>=5.0.0",
+            "mypy>=1.0.0",
+            "pre-commit>=2.20.0",
+            "bandit>=1.7.0",
+        ],
+        "docs": [
+            "sphinx>=5.0.0",
+            "sphinx-rtd-theme>=1.0.0",
+            "myst-parser>=0.18.0",
+            "sphinx-autodoc-typehints>=1.19.0",
+        ],
+        "examples": [
+            "jupyter>=1.0.0",
+            "ipywidgets>=7.6.0",
+            "tensorboard>=2.9.0",
+            "wandb>=0.13.0",
+        ],
+    },
+    entry_points={
+        "console_scripts": [
+            "langvision=langvision.cli:main",
+        ],
+    },
+    classifiers=[
+        "Development Status :: 4 - Beta",
+        "Intended Audience :: Developers",
+        "Intended Audience :: Science/Research",
+        "Operating System :: OS Independent",
+        "Programming Language :: Python :: 3",
+        "Programming Language :: Python :: 3.8",
+        "Programming Language :: Python :: 3.9",
+        "Programming Language :: Python :: 3.10",
+        "Programming Language :: Python :: 3.11",
+        "Programming Language :: Python :: 3.12",
+        "Topic :: Scientific/Engineering :: Artificial Intelligence",
+        "Topic :: Software Development :: Libraries :: Python Modules",
+    ],
+    keywords=["vision", "transformer", "lora", "fine-tuning", "deep-learning", "computer-vision"],
+    include_package_data=True,
+    zip_safe=False,
+)

langvision-0.1.0/src/langvision/__init__.py

@@ -0,0 +1,82 @@
+"""
+langvision - Modular Vision LLMs with Efficient LoRA Fine-Tuning
+
+A research-friendly framework for building and fine-tuning Vision Large Language Models
+with efficient Low-Rank Adaptation (LoRA) support.
+"""
+
+__version__ = "0.1.0"
+__author__ = "Pritesh Raj"
+__email__ = "priteshraj10@gmail.com"
+
+# Core imports for easy access
+from .models.vision_transformer import VisionTransformer
+from .models.lora import LoRALinear, LoRAConfig, AdaLoRALinear, QLoRALinear
+from .models.resnet import resnet18, resnet34, resnet50, resnet101, resnet152
+from .models.multimodal import VisionLanguageModel, create_multimodal_model, CLIPLoss
+from .utils.config import default_config
+from .training.trainer import Trainer
+from .training.advanced_trainer import AdvancedTrainer, TrainingConfig
+from .data.datasets import get_dataset
+from .data.enhanced_datasets import (
+    EnhancedImageDataset, MultimodalDataset, DatasetConfig, 
+    create_enhanced_dataloaders, SmartAugmentation
+)
+from .utils.metrics import (
+    MetricsTracker, ClassificationMetrics, ContrastiveMetrics, 
+    EvaluationSuite, PerformanceMetrics
+)
+from .callbacks.base import Callback, CallbackManager
+from .concepts import RLHF, CoT, CCoT, GRPO, RLVR, DPO, PPO, LIME, SHAP
+
+# Version info
+__all__ = [
+    "__version__",
+    "__author__",
+    "__email__",
+    # Core Models
+    "VisionTransformer",
+    "resnet18", "resnet34", "resnet50", "resnet101", "resnet152",
+    "VisionLanguageModel", "create_multimodal_model",
+    # LoRA Components
+    "LoRALinear", "LoRAConfig", "AdaLoRALinear", "QLoRALinear",
+    # Training
+    "Trainer", "AdvancedTrainer", "TrainingConfig",
+    # Data
+    "get_dataset", "EnhancedImageDataset", "MultimodalDataset", 
+    "DatasetConfig", "create_enhanced_dataloaders", "SmartAugmentation",
+    # Utilities
+    "default_config", "MetricsTracker", "ClassificationMetrics", 
+    "ContrastiveMetrics", "EvaluationSuite", "PerformanceMetrics",
+    # Callbacks
+    "Callback", "CallbackManager",
+    # Loss Functions
+    "CLIPLoss",
+    # Concepts
+    "RLHF", "CoT", "CCoT", "GRPO", "RLVR", "DPO", "PPO", "LIME", "SHAP",
+]
+
+# Optional imports for advanced usage
+try:
+    from .callbacks import EarlyStoppingCallback, LoggingCallback
+    from .utils.device import get_device, to_device
+    __all__.extend([
+        "EarlyStoppingCallback",
+        "LoggingCallback", 
+        "get_device",
+        "to_device"
+    ])
+except ImportError:
+    pass
+
+# Package metadata
+PACKAGE_METADATA = {
+    "name": "langvision",
+    "version": __version__,
+    "description": "Modular Vision LLMs with Efficient LoRA Fine-Tuning",
+    "author": __author__,
+    "email": __email__,
+    "url": "https://github.com/langtrain-ai/langtrain",
+    "license": "MIT",
+    "python_requires": ">=3.8",
+}