# HAPI SDK
A plug-and-play SDK for detecting and reducing hallucinations in Large Language Model outputs using semantic entropy analysis and trained classifiers.
## Features
- **Hallucination Detection**: Uses trained classifiers to detect hallucinations in LLM outputs
- **Semantic Entropy Analysis**: Advanced semantic analysis to identify uncertain or inconsistent outputs
- **Easy Integration**: Simple API that works with any Hugging Face model
- **Multiple Detection Methods**: Combines classifier-based and semantic entropy-based approaches
- **Real-time Analysis**: Generate and analyze outputs in real-time
## Installation
```bash
pip install HAPI-SDK
```
## Quick Start
```python
from dehallucinate_sdk import DeHallucinationClient
# Initialize the client
client = DeHallucinationClient(
model_id="meta-llama/Llama-2-7b-chat-hf",
license_key="your-license-key"
)
# Generate and analyze output
prompt = "Explain quantum entanglement in simple terms."
output, flagged_sentences = client.generate_output(prompt)
print("Generated Output:", output)
print("Flagged Sentences:", flagged_sentences)
```
## Available Methods
The `DeHallucinationClient` exposes the following main methods:
### 1. `generate_output(prompt, max_tokens=512)`
Generates a response and performs hallucination analysis in one step.
```python
output, flagged_sentences = client.generate_output(
"What is the capital of France?",
max_tokens=100
)
```
### 2. `semantic_entropy_check(prompt, num_generations=5)`
Analyzes semantic entropy to detect potential hallucinations.
```python
entropy_score, is_hallucinated = client.semantic_entropy_check(
"Tell me about the history of artificial intelligence"
)
```
### 3. `sentence_contains_hallucination(sentence, context="")`
Checks if a specific sentence contains hallucinations.
```python
is_hallucinated = client.sentence_contains_hallucination(
"The capital of France is Berlin.",
context="Geography facts"
)
```
### 4. `generate(prompt, max_tokens=512)`
Simple text generation without hallucination analysis.
```python
output = client.generate("Write a story about space exploration", max_tokens=200)
```
### 5. `generate_sentence(prompt)`
Generates a single sentence response.
```python
sentence = client.generate_sentence("Complete this: The best way to learn programming is")
```
## Supported Models
- Llama-2-7b-chat (meta-llama/Llama-2-7b-chat-hf)
- Falcon-40b (tiiuae/falcon-40b)
- Llama-2-7b (meta-llama/Llama-2-7b-hf)
- MPT-7b (mosaicml/mpt-7b)
- More models coming soon!
## Configuration
### Environment Variables
```bash
export OPENAI_API_KEY="your_openai_api_key"
export HUGGINGFACE_API_KEY="your_hf_token"
```
### Advanced Usage
```python
# Configure with custom parameters
client = DeHallucinationClient(
model_id="meta-llama/Llama-2-7b-chat-hf",
license_key="your-license-key",
device="cuda", # or "cpu"
temperature=0.7,
use_semantic_entropy=True,
confidence_threshold=0.8
)
# Batch processing
prompts = [
"What causes climate change?",
"How do vaccines work?",
"Explain machine learning basics"
]
results = []
for prompt in prompts:
output, flagged = client.generate_output(prompt)
results.append({"prompt": prompt, "output": output, "flagged": flagged})
```
## Error Handling
```python
try:
output, flagged = client.generate_output("Your prompt here")
except Exception as e:
print(f"Error during generation: {e}")
```
## Contributing
We welcome contributions! Please see our [GitHub repository](https://github.com/cyprienn967/HAPI_SDK) for more information.
## License
MIT License - see LICENSE file for details.
## Support
For questions, suggestions, or issues, please contact us or open an issue on GitHub.
Raw data
{
"_id": null,
"home_page": "https://github.com/cyprienn967/HAPI_SDK",
"name": "HAPI-SDK",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.7",
"maintainer_email": "True <cyprienseydoux@gmail.com>",
"keywords": "llm, hallucination, detection, ai, nlp, semantic-entropy, machine-learning",
"author": "True",
"author_email": "True <cyprienseydoux@gmail.com>",
"download_url": "https://files.pythonhosted.org/packages/21/ba/70324a18d7c961a152445f62eb1dc8595c62c8fb98a09feb54db5b0a296c/hapi_sdk-0.2.23.tar.gz",
"platform": null,
"description": "# HAPI SDK\n\nA plug-and-play SDK for detecting and reducing hallucinations in Large Language Model outputs using semantic entropy analysis and trained classifiers.\n\n## Features\n\n- **Hallucination Detection**: Uses trained classifiers to detect hallucinations in LLM outputs\n- **Semantic Entropy Analysis**: Advanced semantic analysis to identify uncertain or inconsistent outputs\n- **Easy Integration**: Simple API that works with any Hugging Face model\n- **Multiple Detection Methods**: Combines classifier-based and semantic entropy-based approaches\n- **Real-time Analysis**: Generate and analyze outputs in real-time\n\n## Installation\n\n```bash\npip install HAPI-SDK\n```\n\n## Quick Start\n\n```python\nfrom dehallucinate_sdk import DeHallucinationClient\n\n# Initialize the client\nclient = DeHallucinationClient(\n model_id=\"meta-llama/Llama-2-7b-chat-hf\",\n license_key=\"your-license-key\"\n)\n\n# Generate and analyze output\nprompt = \"Explain quantum entanglement in simple terms.\"\noutput, flagged_sentences = client.generate_output(prompt)\n\nprint(\"Generated Output:\", output)\nprint(\"Flagged Sentences:\", flagged_sentences)\n```\n\n## Available Methods\n\nThe `DeHallucinationClient` exposes the following main methods:\n\n### 1. `generate_output(prompt, max_tokens=512)`\nGenerates a response and performs hallucination analysis in one step.\n\n```python\noutput, flagged_sentences = client.generate_output(\n \"What is the capital of France?\", \n max_tokens=100\n)\n```\n\n### 2. `semantic_entropy_check(prompt, num_generations=5)`\nAnalyzes semantic entropy to detect potential hallucinations.\n\n```python\nentropy_score, is_hallucinated = client.semantic_entropy_check(\n \"Tell me about the history of artificial intelligence\"\n)\n```\n\n### 3. `sentence_contains_hallucination(sentence, context=\"\")`\nChecks if a specific sentence contains hallucinations.\n\n```python\nis_hallucinated = client.sentence_contains_hallucination(\n \"The capital of France is Berlin.\",\n context=\"Geography facts\"\n)\n```\n\n### 4. `generate(prompt, max_tokens=512)`\nSimple text generation without hallucination analysis.\n\n```python\noutput = client.generate(\"Write a story about space exploration\", max_tokens=200)\n```\n\n### 5. `generate_sentence(prompt)`\nGenerates a single sentence response.\n\n```python\nsentence = client.generate_sentence(\"Complete this: The best way to learn programming is\")\n```\n\n## Supported Models\n\n- Llama-2-7b-chat (meta-llama/Llama-2-7b-chat-hf)\n- Falcon-40b (tiiuae/falcon-40b) \n- Llama-2-7b (meta-llama/Llama-2-7b-hf)\n- MPT-7b (mosaicml/mpt-7b)\n- More models coming soon!\n\n## Configuration\n\n### Environment Variables\n\n```bash\nexport OPENAI_API_KEY=\"your_openai_api_key\"\nexport HUGGINGFACE_API_KEY=\"your_hf_token\"\n```\n\n### Advanced Usage\n\n```python\n# Configure with custom parameters\nclient = DeHallucinationClient(\n model_id=\"meta-llama/Llama-2-7b-chat-hf\",\n license_key=\"your-license-key\",\n device=\"cuda\", # or \"cpu\"\n temperature=0.7,\n use_semantic_entropy=True,\n confidence_threshold=0.8\n)\n\n# Batch processing\nprompts = [\n \"What causes climate change?\",\n \"How do vaccines work?\", \n \"Explain machine learning basics\"\n]\n\nresults = []\nfor prompt in prompts:\n output, flagged = client.generate_output(prompt)\n results.append({\"prompt\": prompt, \"output\": output, \"flagged\": flagged})\n```\n\n## Error Handling\n\n```python\ntry:\n output, flagged = client.generate_output(\"Your prompt here\")\nexcept Exception as e:\n print(f\"Error during generation: {e}\")\n```\n\n## Contributing\n\nWe welcome contributions! Please see our [GitHub repository](https://github.com/cyprienn967/HAPI_SDK) for more information.\n\n## License\n\nMIT License - see LICENSE file for details.\n\n## Support\n\nFor questions, suggestions, or issues, please contact us or open an issue on GitHub.\n\n\n",
"bugtrack_url": null,
"license": null,
"summary": "A plug-and-play SDK for de-hallucinating outputs from LLMs using semantic entropy and trained classifiers",
"version": "0.2.23",
"project_urls": {
"Bug Reports": "https://github.com/cyprienn967/HAPI_SDK/issues",
"Homepage": "https://github.com/cyprienn967/HAPI_SDK",
"Source": "https://github.com/cyprienn967/HAPI_SDK"
},
"split_keywords": [
"llm",
" hallucination",
" detection",
" ai",
" nlp",
" semantic-entropy",
" machine-learning"
],
"urls": [
{
"comment_text": null,
"digests": {
"blake2b_256": "905ed96b33c71fcf92767c549594a3ae4916c08ec08164c711df0a7b4abb7167",
"md5": "32fbc5121cc4d19ee292874d628f1156",
"sha256": "c80ffa95f511396ed15e4285c53f437d0e817ff1521567fbc211e8187420d271"
},
"downloads": -1,
"filename": "hapi_sdk-0.2.23-py3-none-any.whl",
"has_sig": false,
"md5_digest": "32fbc5121cc4d19ee292874d628f1156",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.7",
"size": 15958002,
"upload_time": "2025-07-12T02:44:15",
"upload_time_iso_8601": "2025-07-12T02:44:15.663148Z",
"url": "https://files.pythonhosted.org/packages/90/5e/d96b33c71fcf92767c549594a3ae4916c08ec08164c711df0a7b4abb7167/hapi_sdk-0.2.23-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": null,
"digests": {
"blake2b_256": "21ba70324a18d7c961a152445f62eb1dc8595c62c8fb98a09feb54db5b0a296c",
"md5": "bf6489daa4abf99aee8a4c3a641ac44f",
"sha256": "e3e640d7a65b222978152b6bf5ab235e22282b62a64982ce92ab2e34d333f8fb"
},
"downloads": -1,
"filename": "hapi_sdk-0.2.23.tar.gz",
"has_sig": false,
"md5_digest": "bf6489daa4abf99aee8a4c3a641ac44f",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.7",
"size": 15940752,
"upload_time": "2025-07-12T02:44:18",
"upload_time_iso_8601": "2025-07-12T02:44:18.656165Z",
"url": "https://files.pythonhosted.org/packages/21/ba/70324a18d7c961a152445f62eb1dc8595c62c8fb98a09feb54db5b0a296c/hapi_sdk-0.2.23.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2025-07-12 02:44:18",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "cyprienn967",
"github_project": "HAPI_SDK",
"github_not_found": true,
"lcname": "hapi-sdk"
}
JSON
