host

Dockerfile

@@ -0,0 +1,18 @@
+FROM python:3.9-slim
+
+WORKDIR /code
+
+COPY ./requirements.txt /code/requirements.txt
+RUN pip install --no-cache-dir --upgrade -r /code/requirements.txt
+
+# Create a non-root user for security
+RUN useradd -m -u 1000 user
+USER user
+ENV HOME=/home/user \
+    PATH=/home/user/.local/bin:$PATH
+
+# Copy scripts into the container
+COPY --chown=user . /code
+
+# Run Uvicorn on the port HF expects
+CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "7860"]

main.py

@@ -0,0 +1,35 @@
+from fastapi import FastAPI, HTTPException
+from pydantic import BaseModel
+from nlp_service import analyse_emergency
+
+app = FastAPI()
+
+class EmergencyRequest(BaseModel):
+    patient_id: str
+    message: str
+
+class EmergencyResponse(BaseModel):
+    patient_id: str
+    urgency: str
+    confidence: float
+    action: str
+    message: str
+
+@app.post("/api/emergency/analyse", response_model=EmergencyResponse)
+async def analyse(request: EmergencyRequest):
+    if not request.message.strip():
+        raise HTTPException(status_code=400, detail="Message cannot be empty")
+
+    result = analyse_emergency(request.message)
+
+    return EmergencyResponse(
+        patient_id=request.patient_id,
+        urgency=result["urgency"],
+        confidence=result["confidence"],
+        action=result["action"],
+        message=result["message"]
+    )
+
+@app.get("/")
+async def health_check():
+    return {"status": "BUTH NLP Service is running"}

nlp_service.py

@@ -0,0 +1,30 @@
+from transformers import pipeline
+
+# CHANGE THIS: Use your actual HF username and model name
+MODEL_ID = "firesolami/buth-nlp-model_lean" 
+
+# Load once when the server starts
+# It will download the weights from the Hub on the first run
+classifier = pipeline(
+    "text-classification",
+    model=MODEL_ID,
+    device=-1
+)
+
+ACTIONS = {
+    "Critical": {"action": "dispatch_ambulance", "message": "Ambulance has been dispatched."},
+    "High": {"action": "alert_admin_dashboard", "message": "Hospital staff have been alerted."},
+    "Moderate": {"action": "schedule_appointment", "message": "Your case has been noted. An appointment will be scheduled shortly."},
+    "Low": {"action": "provide_guidance", "message": "Your concern has been received. A staff member will provide guidance shortly."}
+}
+
+def analyse_emergency(text: str) -> dict:
+    result = classifier(text)[0]
+    label = result["label"]
+    confidence = round(result["score"] * 100, 1)
+    return {
+        "urgency": label,
+        "confidence": confidence,
+        "action": ACTIONS[label]["action"],
+        "message": ACTIONS[label]["message"]
+    }

requirements.txt

@@ -0,0 +1,5 @@
+fastapi
+uvicorn
+transformers
+torch
+pydantic