Spaces:

samuelolubukun
/

PII-Warden-API

Sleeping

App Files Files Community

samuelolubukun commited on Jun 17

Commit

9e2e91e

verified ·

1 Parent(s): aeb0a12

Upload 4 files

Browse files

Files changed (4) hide show

Dockerfile +34 -0
README.md +16 -5
app.py +152 -0
requirements.txt +5 -0

Dockerfile ADDED Viewed

	@@ -0,0 +1,34 @@

+# Use official lightweight Python image
+FROM python:3.10-slim
+# Create a secure, non-root user (required by Hugging Face Spaces)
+RUN useradd -m -u 1000 user
+# Set up environment variables
+ENV HOME=/home/user \
+    PATH=/home/user/.local/bin:$PATH \
+    PYTHONUNBUFFERED=1 \
+    HF_HOME=/home/user/.cache/huggingface
+# Set working directory
+WORKDIR $HOME/app
+# Copy requirements and install dependencies
+COPY --chown=user requirements.txt .
+RUN pip install --no-cache-dir --upgrade -r requirements.txt
+# Pre-cache the model during the image build process
+# This guarantees instant startup times when the Space boots up.
+RUN python -c "from transformers import pipeline; pipeline('token-classification', model='samuelolubukun/pii-ner-edge-optimized')"
+# Copy the rest of the application files
+COPY --chown=user . .
+# Switch to the non-root user
+USER user
+# Expose port 7860
+EXPOSE 7860
+# Start the application on port 7860
+CMD ["uvicorn", "app:app", "--host", "0.0.0.0", "--port", "7860"]

README.md CHANGED Viewed

@@ -1,10 +1,21 @@
 ---
-title: PII Warden API
-emoji: 🐨
-colorFrom: pink
-colorTo: blue
 sdk: docker
 pinned: false
 ---
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

 ---
+title: PII Warden AI
+emoji: 🛡️
+colorFrom: indigo
+colorTo: pink
 sdk: docker
+app_port: 7860
 pinned: false
 ---
+# PII Warden Cloud AI Space
+This is the Tier 2 Cloud AI Inference service for the PII Warden browser extension. It loads the fine-tuned `samuelolubukun/pii-ner-edge-optimized` DistilBERT NER model to detect unstructured PII entities (Names, Organizations, Locations).
+## How to use this Space with the Extension
+1. Copy the URL of this Hugging Face Space.
+   - Example: `https://<your-username>-pii-warden-ai.hf.space/analyze` (Note the `/analyze` path at the end!).
+2. Open the PII Warden browser extension popup.
+3. Paste the URL into the **"Cloud AI Endpoint"** field.
+4. The extension will automatically verify and connect, merging Cloud AI context detections with its local regex & checksum validator!

app.py ADDED Viewed

	@@ -0,0 +1,152 @@

+from fastapi import FastAPI, HTTPException
+from fastapi.middleware.cors import CORSMiddleware
+from fastapi.responses import HTMLResponse
+from pydantic import BaseModel
+from transformers import pipeline
+import uvicorn
+app = FastAPI(
+    title="PII Warden Cloud AI Endpoint",
+    description="Tier 2 Cloud AI Inference service for the PII Warden browser extension."
+)
+# Enable CORS (Cross-Origin Resource Sharing)
+# Critical so that browser extensions can send POST requests from any webpage.
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=["*"],
+    allow_credentials=True,
+    allow_methods=["*"],
+    allow_headers=["*"],
+)
+class AnalyzeRequest(BaseModel):
+    text: str
+# Load the Hugging Face Token Classification pipeline on startup
+print("Loading DistilBERT PII model into memory...")
+try:
+    nlp_pipeline = pipeline(
+        "token-classification",
+        model="samuelolubukun/pii-ner-edge-optimized",
+        aggregation_strategy="simple"
+    )
+    print("Model loaded successfully!")
+except Exception as e:
+    print(f"Error loading model: {e}")
+    nlp_pipeline = None
+@app.get("/", response_class=HTMLResponse)
+async def read_root():
+    return """
+    <!DOCTYPE html>
+    <html>
+    <head>
+        <title>PII Warden AI Endpoint</title>
+        <link href="https://fonts.googleapis.com/css2?family=Outfit:wght@400;600;700&display=swap" rel="stylesheet">
+        <style>
+            body {
+                font-family: 'Outfit', sans-serif;
+                background-color: #0b0f19;
+                color: #f3f4f6;
+                display: flex;
+                flex-direction: column;
+                justify-content: center;
+                align-items: center;
+                min-height: 100vh;
+                margin: 0;
+                background: radial-gradient(circle at top right, rgba(139, 92, 246, 0.15), transparent 60%),
+                            radial-gradient(circle at bottom left, rgba(236, 72, 153, 0.1), transparent 60%);
+            }
+            .card {
+                background: rgba(17, 24, 39, 0.75);
+                border: 1px solid rgba(255, 255, 255, 0.08);
+                border-radius: 16px;
+                padding: 32px;
+                max-width: 500px;
+                text-align: center;
+                backdrop-filter: blur(12px);
+                box-shadow: 0 8px 32px rgba(0, 0, 0, 0.5);
+            }
+            h1 {
+                font-size: 2rem;
+                margin-bottom: 8px;
+                background: linear-gradient(135deg, #ffffff 30%, #a78bfa 100%);
+                -webkit-background-clip: text;
+                -webkit-text-fill-color: transparent;
+            }
+            p {
+                color: #9ca3af;
+                font-size: 0.95rem;
+                line-height: 1.5;
+            }
+            .badge {
+                display: inline-block;
+                background: rgba(16, 185, 129, 0.15);
+                color: #10b981;
+                border: 1px solid rgba(16, 185, 129, 0.3);
+                padding: 6px 16px;
+                border-radius: 9999px;
+                font-size: 0.8rem;
+                font-weight: 600;
+                text-transform: uppercase;
+                letter-spacing: 0.05em;
+                margin-top: 16px;
+            }
+            .endpoint {
+                background: rgba(10, 10, 10, 0.4);
+                padding: 10px;
+                border-radius: 8px;
+                font-family: monospace;
+                font-size: 0.85rem;
+                color: #f472b6;
+                margin-top: 20px;
+                border: 1px solid rgba(255, 255, 255, 0.05);
+            }
+        </style>
+    </head>
+    <body>
+        <div class="card">
+            <h1>🛡️ PII Warden AI</h1>
+            <p>Your hosted client-side PII redactor cloud inference endpoint is live. Configure your browser extension to query the endpoint below for Tier 2 context analysis.</p>
+            <div class="endpoint">POST /analyze</div>
+            <div class="badge">Online & Active</div>
+        </div>
+    </body>
+    </html>
+    """
+@app.post("/analyze")
+async def analyze_text(request: AnalyzeRequest):
+    if nlp_pipeline is None:
+        raise HTTPException(status_code=503, detail="AI Model pipeline not initialized.")
+    try:
+        text = request.text
+        if not text.strip():
+            return []
+        predictions = nlp_pipeline(text)
+        formatted_results = []
+        for pred in predictions:
+            formatted_results.append({
+                "entity_group": pred["entity_group"],
+                "score": float(pred["score"]),
+                "word": pred["word"],
+                "start": int(pred["start"]),
+                "end": int(pred["end"])
+            })
+        return formatted_results
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=f"Inference error: {str(e)}")
+@app.get("/health")
+async def health_check():
+    return {"status": "healthy", "model_loaded": nlp_pipeline is not None}
+if __name__ == "__main__":
+    # Hugging Face Spaces require listening on port 7860
+    uvicorn.run("app:app", host="0.0.0.0", port=7860, reload=False)

requirements.txt ADDED Viewed

	@@ -0,0 +1,5 @@

+fastapi
+uvicorn
+pydantic
+transformers
+torch