Metadata-Version: 2.4
Name: traini-ai
Version: 2.0.2
Summary: Traini AI SDK - Dog emotion analysis and human-dog communication platform
Home-page: https://github.com/Traini-Inc/traini-sdk
Author: Traini AI Team
Author-email: Traini AI Team <support@traini.ai>
License: MIT
Project-URL: Homepage, https://github.com/Traini-Inc/traini-sdk
Project-URL: Documentation, https://github.com/Traini-Inc/traini-sdk
Project-URL: Repository, https://github.com/Traini-Inc/traini-sdk
Project-URL: Bug Tracker, https://github.com/Traini-Inc/traini-sdk/issues
Keywords: traini,ai,dog,emotion,analysis,translation,audio,video,machine-learning,pet
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Developers
Classifier: Intended Audience :: Science/Research
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Classifier: Topic :: Multimedia :: Sound/Audio :: Analysis
Classifier: License :: OSI Approved :: MIT License
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
Requires-Python: >=3.8
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: numpy>=1.21.0
Requires-Dist: pillow>=9.0.0
Requires-Dist: requests>=2.27.0
Provides-Extra: dev
Requires-Dist: pytest>=7.0.0; extra == "dev"
Requires-Dist: pytest-cov>=3.0.0; extra == "dev"
Requires-Dist: black>=22.0.0; extra == "dev"
Requires-Dist: flake8>=4.0.0; extra == "dev"
Requires-Dist: mypy>=0.950; extra == "dev"
Provides-Extra: ml
Requires-Dist: torch>=1.12.0; extra == "ml"
Requires-Dist: torchvision>=0.13.0; extra == "ml"
Requires-Dist: opencv-python>=4.6.0; extra == "ml"
Provides-Extra: audio
Requires-Dist: librosa>=0.9.0; extra == "audio"
Requires-Dist: soundfile>=0.11.0; extra == "audio"
Requires-Dist: pydub>=0.25.0; extra == "audio"
Provides-Extra: all
Requires-Dist: torch>=1.12.0; extra == "all"
Requires-Dist: torchvision>=0.13.0; extra == "all"
Requires-Dist: opencv-python>=4.6.0; extra == "all"
Requires-Dist: librosa>=0.9.0; extra == "all"
Requires-Dist: soundfile>=0.11.0; extra == "all"
Requires-Dist: pydub>=0.25.0; extra == "all"
Dynamic: author
Dynamic: home-page
Dynamic: license-file
Dynamic: requires-python

# Traini AI - Dog Emotion Analysis SDK

![Python Version](https://img.shields.io/badge/python-3.8%2B-blue)
![License](https://img.shields.io/badge/license-MIT-green)
![Status](https://img.shields.io/badge/status-production-brightgreen)

Advanced AI-powered dog emotion detection and image analysis SDK combining deep learning emotion recognition with multi-model visual intelligence.

## 🌟 Features

- **🎯 High Accuracy**: 76-88% emotion detection accuracy using 4-model ensemble
- **🤖 Multi-Model Analysis**: Integrates GPT-4, Gemini 2.5, and Claude 3 for comprehensive visual understanding
- **📝 Dual Output**: Professional third-person descriptions + playful first-person narratives
- **⚡ Fast & Reliable**: Optimized inference pipeline with automatic image format detection
- **🔧 Easy to Use**: Simple API with just 3 lines of code

## 📦 Installation

```bash
pip install traini-ai
```

## 🚀 Quick Start

```python
from traini_ai import IntegratedImageAnalysisWithEmotion

# Initialize analyzer
analyzer = IntegratedImageAnalysisWithEmotion()

# Analyze dog image
result = analyzer.analyze_image('path/to/dog_image.jpg')

# Get results
print(result['third_person_description'])  # Professional analysis
print(result['first_person_description'])  # Fun dog perspective
```

## 📖 API Reference

### `IntegratedImageAnalysisWithEmotion`

Main class for dog image analysis.

#### Parameters

- `openai_api_key` (str, optional): OpenAI API key. Uses default if not provided.
- `google_api_key` (str, optional): Google API key. Uses default if not provided.
- `anthropic_api_key` (str, optional): Anthropic API key. Uses default if not provided.
- `use_ensemble` (bool, default=True): Whether to use 4-model ensemble for emotion detection.

#### Methods

##### `analyze_image(image_path: str) -> dict`

Analyzes a dog image and returns emotional and behavioral insights.

**Parameters:**
- `image_path` (str): Path to the dog image file. Supports JPEG, PNG, GIF, WebP.

**Returns:**
```python
{
    'third_person_description': str,  # Objective third-person analysis
    'first_person_description': str   # Playful first-person narrative
}
```

## 💡 Usage Examples

### Basic Usage

```python
from traini_ai import IntegratedImageAnalysisWithEmotion

analyzer = IntegratedImageAnalysisWithEmotion()
result = analyzer.analyze_image('happy_dog.jpg')

print("Third-Person Analysis:")
print(result['third_person_description'])
print("\nFirst-Person Narrative:")
print(result['first_person_description'])
```

### Custom API Keys

```python
analyzer = IntegratedImageAnalysisWithEmotion(
    openai_api_key='your-openai-key',
    google_api_key='your-google-key',
    anthropic_api_key='your-anthropic-key'
)
```

### Single Model Mode (Faster)

```python
# Use single model for faster inference
analyzer = IntegratedImageAnalysisWithEmotion(use_ensemble=False)
result = analyzer.analyze_image('dog.jpg')
```

## 📊 Performance

| Mode | Accuracy | Speed | Memory | Use Case |
|------|----------|-------|--------|----------|
| Ensemble (4 models) | 76-88% | ~3-5s | ~1.2GB | High accuracy needs |
| Single Model | 74-76% | ~1-2s | ~300MB | Real-time applications |

## 🎨 Output Examples

### Third-Person Description (Professional)

```
The image features a fluffy golden Pomeranian dog situated indoors on a
textured gray rug within a cozy living room setting. The dog's voluminous
fur and distinctively perked ears emphasize its breed traits, while its
dark, expressive eyes convey a sense of curiosity and attentiveness. The
dog is facing the camera with its head held high, displaying a slightly
open mouth that suggests a playful or relaxed demeanor...
```

### First-Person Description (Playful)

```
Omg, hi there! 😍 I'm just chillin' here on my favorite rug, feelin' all
cozy and stuff! My floofy fur is like, 10/10 fluffy for snuggles. 🐾 My
ears are perked 'cause I'm super curious about what you're up to! Maybe
you wanna throw me a toy or give me some treats? I promise I won't get
mad – just playful wags and happy barks! 🐶💖
```

## 🔬 Technology Stack

### Emotion Detection
- **AttentionResNet50**: Advanced attention mechanism
- **ResNet50**: Robust feature extraction
- **ResNet101**: Deep architecture for complex patterns
- **EfficientNet**: Optimized efficiency and accuracy

### Visual Analysis
- **GPT-4o-mini**: General visual understanding
- **Gemini 2.5 Flash Lite**: Fast image processing
- **Claude 3 Haiku**: Detailed emotion interpretation

## 🎯 Supported Emotions

The SDK can detect 13 different dog emotions:

- Happy
- Sad
- Angry
- Alert
- Relaxed
- Fear
- Anxiety
- Anticipation
- Appeasement
- Caution
- Confident
- Curiosity
- Sleepy

## 🖼️ Supported Image Formats

- JPEG/JPG
- PNG
- GIF
- WebP

## ⚙️ Requirements

- Python 3.8+
- PyTorch 1.9+
- OpenAI SDK
- Google Generative AI
- Anthropic SDK

## 📄 License

MIT License - see LICENSE file for details

## 🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

## 📞 Support

- **Documentation**: [https://docs.traini.ai](https://docs.traini.ai)
- **Issues**: [https://github.com/traini-ai/traini-sdk/issues](https://github.com/traini-ai/traini-sdk/issues)
- **Email**: support@traini.ai

## 🔄 Changelog

### v2.0.0 (2024-12-14)
- ✨ Added multi-model integration (GPT, Gemini, Claude)
- ✨ Implemented 4-model ensemble emotion detection
- ✨ Dual output: third-person + first-person descriptions
- 🔧 Automatic image format detection
- 📝 Simplified API with clean output
- 🚀 Production-ready with error handling

### v1.0.0 (2024-12-01)
- 🎉 Initial release
- 🤖 Single model emotion detection
- 📊 Basic image analysis

## 🙏 Acknowledgments

Built with advanced AI models from OpenAI, Google, and Anthropic.

---

**Made with ❤️ by the Traini AI Team**
