Delete frontend_pam.py

frontend_pam.py

@@ -1,108 +0,0 @@
-# filename: frontend_pam.py (UPDATED FOR HF INFERENCE API)
-
-import os
-import json
-import random
-import requests
-from datetime import datetime
-from typing import Dict, Any, Optional
-
-# --- Constants for Data Paths ---
-BASE_DIR = os.path.dirname(os.path.abspath(__file__))
-DATA_DIR = os.path.join(BASE_DIR, "data")
-
-APPOINTMENTS_FILE = os.path.join(DATA_DIR, "appointments.json")
-RESOURCES_FILE = os.path.join(DATA_DIR, "resources.json")
-FOLLOW_UP_FILE = os.path.join(DATA_DIR, "follow_up.json")
-PERMISSIONS_FILE = os.path.join(DATA_DIR, "permissions.json")
-
-# --- HuggingFace Inference API Setup ---
-HF_API_TOKEN = os.getenv("HF_READ_TOKEN")
-HF_HEADERS = {"Authorization": f"Bearer {HF_API_TOKEN}"}
-HF_ENDPOINTS = {
-    "intent": "https://api-inference.huggingface.co/models/typeform/distilbert-base-uncased-mnli",
-    "sentiment": "https://api-inference.huggingface.co/models/distilbert-base-uncased-finetuned-sst-2-english",
-    "summarizer": "https://api-inference.huggingface.co/models/google/flan-t5-large",
-    "chat": "https://api-inference.huggingface.co/models/tiiuae/falcon-7b-instruct"
-}
-
-# --- Load JSON Helper ---
-def load_json(filepath: str) -> Dict[str, Any]:
-    try:
-        with open(filepath, 'r') as f:
-            return json.load(f)
-    except FileNotFoundError:
-        print(f"CRITICAL: Data file not found at: {filepath}")
-        return {}
-    except json.JSONDecodeError as e:
-        print(f"CRITICAL: Failed to decode JSON from {filepath}: {e}")
-        return {}
-
-# --- Inference API Call Helper ---
-def hf_infer(task: str, payload: Any) -> Any:
-    url = HF_ENDPOINTS.get(task)
-    if not url:
-        return {"error": f"Invalid task: {task}"}
-    response = requests.post(url, headers=HF_HEADERS, json=payload)
-    if response.status_code != 200:
-        return {"error": f"HF API Error ({response.status_code})", "details": response.text}
-    return response.json()
-
-# --- Agent Initialization ---
-def load_frontend_agent() -> 'FrontendPAM':
-    print("Initializing Frontend PAM using HF Inference API...")
-    data = {
-        "APPOINTMENTS": load_json(APPOINTMENTS_FILE),
-        "RESOURCES": load_json(RESOURCES_FILE),
-        "FOLLOW_UP": load_json(FOLLOW_UP_FILE),
-        "PERMISSIONS": load_json(PERMISSIONS_FILE)
-    }
-    return FrontendPAM(data)
-
-# --- PAM Personality ---
-PAM_TONE = ("I am PAM — the cool older sister... [Full description abridged]")
-GREETINGS = ["babe", "love", "friend", "girl", "boo", "sweetheart"]
-
-# --- Agent Class ---
-class FrontendPAM:
-    def __init__(self, data: Dict[str, Dict]):
-        self.APPOINTMENTS = data.get("APPOINTMENTS", {})
-        self.PERMISSIONS = data.get("PERMISSIONS", {})
-        self.RESOURCES = data.get("RESOURCES", {})
-        self.FOLLOW_UP = data.get("FOLLOW_UP", {})
-        self.user_id = "user_001"
-
-    def respond(self, user_text: str, backend_brief: Optional[str] = None) -> Dict[str, Any]:
-        if not user_text.lower().startswith("hey pam"):
-            return {"reply": "Oops, you didn’t say 'Hey Pam'. I only respond to respectful greetings — you know how I am. 💅"}
-
-        text = user_text.replace("PAM", "you").replace("pam", "you")
-        detected_intent = hf_infer("intent", {"inputs": text})[0].get("label", "unknown")
-        sentiment_result = hf_infer("sentiment", {"inputs": text})[0]
-        backend_summary = backend_brief or "No backend input provided."
-        greeting = random.choice(GREETINGS)
-
-        for term, allowed in self.PERMISSIONS.items():
-            if term in text.lower() and not allowed:
-                return {"reply": f"That’s something I’m not allowed to help with directly, {greeting}. But I can connect you to a safe resource or provider if you’d like."}
-
-        if "appointment" in text.lower():
-            appt = self.APPOINTMENTS.get(self.user_id)
-            if appt:
-                return {"reply": f"Hey {greeting}, you've got a {appt['type']} scheduled..."}
-            else:
-                return {"reply": f"Hey {greeting}, I don’t have any appointments saved..."}
-
-        elif any(keyword in text.lower() for keyword in ["cramp", "discharge", "bleed", "smell", "spotting", "fatigue", "mood", "missed"]):
-            return {"reply": f"Okay {greeting}, I pulled a few resources I think will help..."}
-
-        else:
-            prompt = f"{PAM_TONE}\n... User said: {text}\nPAM:"
-            chat_output = hf_infer("chat", {"inputs": prompt})
-            reply = chat_output[0].get("generated_text", "Sorry love, I didn’t catch that. Try again?").split("PAM:")[-1].strip()
-            return {
-                "intent": detected_intent,
-                "sentiment": sentiment_result,
-                "summary": backend_summary,
-                "reply": reply
-            }