Update app.py

app.py

@@ -1,116 +1,103 @@
 from fastapi import FastAPI
 from fastapi.middleware.cors import CORSMiddleware
-from fastapi.staticfiles import StaticFiles
 from fastapi.responses import HTMLResponse
+from fastapi.staticfiles import StaticFiles
 import gradio as gr
 import numpy as np
 import faiss
 import logging
+import os
 import requests
 import json
-
-url = "https://api.fireworks.ai/inference/v1/chat/completions"
-payload = {
-  "model": "accounts/fireworks/models/deepseek-r1",
-  "max_tokens": 4096,
-  "top_p": 1,
-  "top_k": 40,
-  "presence_penalty": 0,
-  "frequency_penalty": 0,
-  "temperature": 0.6,
-  "messages": []
-}
-headers = {
-  "Accept": "application/json",
-  "Content-Type": "application/json",
-  "Authorization": "Bearer <API_KEY>"
-}
-requests.request("POST", url, headers=headers, data=json.dumps(payload))
-
-import os
-os.environ["TRANSFORMERS_CACHE"] = "/app/cache/transformers"
-os.environ["HF_HOME"] = "/app/cache/hf"
-os.environ["SENTENCE_TRANSFORMERS_HOME"] = "/app/cache/st"
-
 from sentence_transformers import SentenceTransformer
-model = SentenceTransformer("sentence-transformers/all-MiniLM-L6-v2")
-logging.basicConfig(level=logging.INFO)  # No file, just console
 
-from huggingface_hub import snapshot_download
-snapshot_download(
-    repo_id="sentence-transformers/all-MiniLM-L6-v2",
-    cache_dir="/app/cache/hf",
-)
+# === Environment variables (safe cache paths for HF Spaces) ===
+os.environ["TRANSFORMERS_CACHE"] = "/data/transformers"
+os.environ["HF_HOME"] = "/data/hf"
+os.environ["SENTENCE_TRANSFORMERS_HOME"] = "/data/st"
 
-# Initialize components
-app = FastAPI()
-model = SentenceTransformer('all-MiniLM-L6-v2')
+# === Load embedding model ===
+model = SentenceTransformer("sentence-transformers/all-MiniLM-L6-v2")
 index = faiss.IndexFlatL2(384)
 memory_text = []
 
-# Configure CORS
-app.add_middleware(
-    CORSMiddleware,
-    allow_origins=["*"],
-    allow_methods=["*"],
-    allow_headers=["*"],
-)
-
-# Autonomous Agent Function
+# === Fireworks API ===
+FIREWORKS_API_KEY = os.getenv("FIREWORKS_API_KEY")  # 🔐 Use env var instead of hardcoding
+FIREWORKS_URL = "https://api.fireworks.ai/inference/v1/chat/completions"
+
+def query_fireworks(prompt):
+    payload = {
+        "model": "accounts/fireworks/models/deepseek-r1",
+        "max_tokens": 4096,
+        "top_p": 1,
+        "top_k": 40,
+        "temperature": 0.6,
+        "messages": [{"role": "user", "content": prompt}],
+    }
+    headers = {
+        "Accept": "application/json",
+        "Content-Type": "application/json",
+        "Authorization": f"Bearer {FIREWORKS_API_KEY}"
+    }
+
+    response = requests.post(FIREWORKS_URL, headers=headers, data=json.dumps(payload))
+    result = response.json()
+    return result.get("choices", [{}])[0].get("message", {}).get("content", "⚠️ No response.")
+
+# === Autonomous Agent ===
 def autonomous_agent(input_text):
-    vec = model.encode([input_text])[0]  # Ensure shape is (384,)
+    vec = model.encode([input_text])[0]
     response = ""
 
     if index.ntotal > 0:
-        D, I = index.search(np.array([vec]), min(5, index.ntotal))  # ✅ missing closing paren fixed
-        matches = []
-
-        for idx, dist in zip(I[0], D[0]):
-            if idx != -1 and dist < 0.8:
-                matches.append(memory_text[idx])
-
+        D, I = index.search(np.array([vec]), min(5, index.ntotal))
+        matches = [memory_text[idx] for idx, dist in zip(I[0], D[0]) if idx != -1 and dist < 0.8]
         if matches:
-            response = "🧠 Related memories:\n- " + "\n- ".join(matches[:3])
-        else:
-            response = "🤖 No relevant memories found"
-    else:
-        response = "🤖 Memory is empty"
+            response += "🧠 Related memories:\n- " + "\n- ".join(matches[:3]) + "\n\n"
 
-    # Store new vector and memory
+    # Store current memory
     index.add(np.array([vec]))
     memory_text.append(input_text)
 
+    # 🔥 Query LLM (Fireworks)
+    llm_response = query_fireworks(input_text)
+    response += f"🤖 Response:\n{llm_response}"
     return response
 
-# Gradio Interface
+# === Gradio UI ===
 gradio_ui = gr.Interface(
     fn=autonomous_agent,
     inputs="text",
     outputs="text",
     title="Autonomous AI Agent",
-    description="Self-enhancing chatbot with vector memory",
-    flagging_mode="auto"
+    description="Self-enhancing chatbot with memory + Fireworks LLM",
+)
+
+# === FastAPI App ===
+app = FastAPI()
+
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=["*"], allow_methods=["*"], allow_headers=["*"],
 )
 
-# FastAPI Routes
+# === Root Web UI with embedded Gradio ===
 @app.get("/", response_class=HTMLResponse)
 async def root():
     return """
     <html>
-        <head>
-            <title>Autonomous AI Agent</title>
-        </head>
+        <head><title>Autonomous AI Agent</title></head>
         <body>
-            <h1>Autonomous AI Agent</h1>
-            <iframe src="/gradio" width="100%" height="500"></iframe>
+            <h2>Autonomous AI Agent</h2>
+            <iframe src="/gradio" width="100%" height="600"></iframe>
         </body>
     </html>
     """
 
-# Mount Gradio and static files
+# === Mount Gradio & static files ===
 app.mount("/gradio", gradio_ui.app)
 app.mount("/static", StaticFiles(directory="static"), name="static")
 
-# For Hugging Face Spaces
+# === For Hugging Face Spaces ===
 def get_app():
-    return app
+    return app