Update app.py

app.py

@@ -1,54 +1,52 @@
+import os
 import gradio as gr
 from sentence_transformers import SentenceTransformer
 from memory_store import MemoryStore
-from graph_view import visualize_graph
+from graph_reasoner import GraphReasoner
+from graph_view import visualize_reasoned_graph
 
-# --- Initialize Core ---
-model = SentenceTransformer("sentence-transformers/all-MiniLM-L6-v2")
+# --- Initialize Environment ---
+os.environ["TRANSFORMERS_CACHE"] = "/home/user/.cache"
+model = SentenceTransformer("all-MiniLM-L6-v2")
 memory = MemoryStore()
+reasoner = GraphReasoner()
 
-# --- Chat Processing ---
-def process_input(user_input, chat_history):
-    user_input = user_input.strip()
-
-    # handle visualization
-    if user_input.lower() in ["visualize memory", "show memory", "see memory"]:
-        memories = getattr(memory, "memories", [])
-        if not memories:
-            chat_history.append([user_input, "No memories yet to visualize."])
-        else:
-            visualize_graph(memories)
-            chat_history.append([user_input, "🧠 Memory graph generated. Check the right panel."])
-        return chat_history
-
-    # normal memory flow
+# --- Core Logic ---
+def process_input(user_input, history):
+    if not user_input.strip():
+        return history + [("User input empty.", "Please type something meaningful.")]
+
+    # Encode input
     vector = model.encode(user_input)
+
+    # Store memory
     save_status = memory.add_memory(user_input, vector)
+
+    # Retrieve related memories
     related = memory.retrieve_relevant(vector)
-    recall = "\n".join([f"• {txt} (score: {round(score,4)})" for txt, score in related])
-    response = f"{save_status}\nHere’s what I recall most related:\n{recall}"
-    chat_history.append([user_input, response])
-    return chat_history
-
-# --- Graph Display Helper ---
-def render_graph():
-    visualize_graph(memory.memories)
-    return "🧠 Memory graph updated."
-
-# --- Interface ---
-with gr.Blocks(theme=gr.themes.Soft()) as demo:
-    gr.Markdown("## 🧠 Aventra OS — Memory Graph Interface")
-
-    chatbot = gr.Chatbot(height=400)
-    msg = gr.Textbox(label="Type your message here...")
-    with gr.Row():
-        send_btn = gr.Button("Send")
-        clear_btn = gr.Button("Clear Chat")
-        graph_btn = gr.Button("Visualize Memory")
-
-    send_btn.click(process_input, [msg, chatbot], [chatbot])
-    msg.submit(process_input, [msg, chatbot], [chatbot])
-    clear_btn.click(lambda: [], None, chatbot)
-    graph_btn.click(render_graph, None, None)
-
-demo.launch()
+    recall_text = "\n".join([f"• {r[0]} (score: {round(r[1], 4)})" for r in related])
+
+    # Build response
+    response = f"{save_status}\n\nHere’s what I recall that’s most relevant:\n{recall_text if related else 'No related context yet.'}"
+    history.append((user_input, response))
+    return history
+
+def visualize_memory():
+    relationships = reasoner.infer_relationships(memory.memories, model)
+    output_html = visualize_reasoned_graph(memory.memories, relationships)
+    return f"🧠 Cognitive Memory Graph updated: {output_html}"
+
+# --- UI Layout ---
+with gr.Blocks(title="Aventra OS — Memory Engine") as demo:
+    gr.Markdown("## 🧠 Aventra Memory Interface")
+    chatbot = gr.Chatbot(label="Aventra Conversation", height=400)
+    user_input = gr.Textbox(label="Message", placeholder="Type something like 'My name is Tirrek'...")
+    send_btn = gr.Button("Send")
+    graph_btn = gr.Button("Visualize Memory")
+    clear_btn = gr.Button("Clear Chat")
+
+    send_btn.click(process_input, inputs=[user_input, chatbot], outputs=[chatbot])
+    graph_btn.click(visualize_memory, inputs=None, outputs=None)
+    clear_btn.click(lambda: None, None, chatbot, queue=False)
+
+demo.launch(server_name="0.0.0.0", server_port=7860)