---
tags:
- legal
- bert
- classification
- kenya-constitution
- legal-ai
license: apache-2.0
language: en
library_name: transformers
datasets: kenya-constitution-2010
model_name: legalbert_finetuned
widget:
  - text: "What are the fundamental rights under the Kenyan Constitution?"
---

# 📜 LegalBERT Fine-Tuned for Kenya Constitution 2010

## 📝 Model Description  
This model is a fine-tuned **LegalBERT** for **legal text classification** based on the **Kenya Constitution 2010**.  
It is designed to classify constitutional provisions and assist in legal research by analyzing legal texts with high precision.

## 🏗️ Model Details  
- **Base Model**: [LegalBERT](https://huggingface.co/nlpaueb/legal-bert-base-uncased)  
- **Fine-Tuned For**: **Kenyan constitutional law classification**  
- **Architecture**: `BertForSequenceClassification`  
- **Number of Layers**: `6`  
- **Hidden Size**: `512`  
- **Attention Heads**: `8`  
- **Vocabulary Size**: `30,522`  
- **Max Position Embeddings**: `512`  
- **Dropout Rate**: `0.1`  
- **Transformers Version**: `4.48.3`  

## 🚀 How to Use This Model  
You can load and use the model easily with **Transformers**:

```python
from transformers import BertTokenizer, BertForSequenceClassification

model_name = "CoolCerebralTech/legalbert_finetuned"

# Load tokenizer
tokenizer = BertTokenizer.from_pretrained(model_name)

# Load model
model = BertForSequenceClassification.from_pretrained(model_name)

⚖️ Inference: Legal Text Classification
The model predicts legal categories based on input queries.

🔹 Example: Classifying a Legal Question
import torch

def predict_legal_query(query):
    """Runs inference on a legal query using LegalBERT."""
    inputs = tokenizer(query, return_tensors="pt", truncation=True, padding=True, max_length=512)

    with torch.no_grad():
        outputs = model(**inputs)
        logits = outputs.logits
        predicted_class = torch.argmax(logits, dim=1).item()  # Get highest scoring class

    return predicted_class

# 🔹 User Query
query = "What are the fundamental rights under the Kenyan Constitution?"
prediction = predict_legal_query(query)

# 🔹 Label Mapping (Modify as needed)
label_mapping = {0: "Civil Rights", 1: "Criminal Law", 2: "Property Law", 3: "Other"}
predicted_label = label_mapping.get(prediction, "Unknown")

print(f"📌 Query: {query}")
print(f"🔹 Predicted Legal Category: {predicted_label}")

✅ Sample Output
📌 Query: What are the fundamental rights under the Kenyan Constitution?
🔹 Predicted Legal Category: Civil Rights

📌 Intended Use
Legal Research & Analysis
Kenyan Constitutional Law Classification
Automated Legal Document Processing
⚠️ Limitations & Warnings
This model is fine-tuned on Kenyan constitutional law and may not generalize to other legal systems.
Always consult a qualified legal expert before making legal conclusions.