Update app.py

app.py

@@ -1,64 +1,85 @@
 import gradio as gr
-from huggingface_hub import InferenceClient
+import torch
+from transformers import AutoTokenizer, pipeline
+from sentence_transformers import SentenceTransformer
+import faiss
+import numpy as np
 
-"""
-For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-"""
-client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
+# Configuration
+class Config:
+    model_name = "TinyLlama/TinyLlama-1.1B-Chat-v1.0"
+    embedding_model = "all-MiniLM-L6-v2"
+    vector_dim = 384 
+    top_k = 3 
+    chunk_size = 256 
 
+# Vector Database
+class VectorDB:
+    def __init__(self):
+        self.index = faiss.IndexFlatL2(Config.vector_dim)
+        self.texts = []
+        self.embedding_model = SentenceTransformer(Config.embedding_model)
+    
+    def add_text(self, text: str):
+        embedding = self.embedding_model.encode([text])[0]
+        embedding = np.array([embedding], dtype=np.float32)
+        faiss.normalize_L2(embedding)
+        self.index.add(embedding)
+        self.texts.append(text)
+    
+    def search(self, query: str):
+        if self.index.ntotal == 0:
+            return []
+        query_embedding = self.embedding_model.encode([query])[0]
+        query_embedding = np.array([query_embedding], dtype=np.float32)
+        faiss.normalize_L2(query_embedding)
+        D, I = self.index.search(query_embedding, min(Config.top_k, self.index.ntotal))
+        return [self.texts[i] for i in I[0] if i < len(self.texts)]
 
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
+# Load Model
+class TinyChatModel:
+    def __init__(self):
+        self.tokenizer = AutoTokenizer.from_pretrained(Config.model_name)
+        self.pipe = pipeline("text-generation", model=Config.model_name, torch_dtype=torch.bfloat16, device_map="auto")
 
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
+    def generate_response(self, message: str, context: str = ""):
+        messages = [{"role": "user", "content": message}]
+        if context:
+            messages.insert(0, {"role": "system", "content": f"Context:\n{context}"})
+        prompt = self.tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
+        outputs = self.pipe(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
+        return outputs[0]["generated_text"].split("<|assistant|>")[-1].strip()
 
-    messages.append({"role": "user", "content": message})
+# Initialize
+vector_db = VectorDB()
+chat_model = TinyChatModel()
 
-    response = ""
+def chat_interface(user_input):
+    context = "\n".join(vector_db.search(user_input))
+    response = chat_model.generate_response(user_input, context)
+    vector_db.add_text(f"User: {user_input}\nAssistant: {response}")
+    return response
 
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
-
-        response += token
-        yield response
-
-
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
-demo = gr.ChatInterface(
-    respond,
-    additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
-    ],
-)
+def add_text_interface(text):
+    vector_db.add_text(text)
+    return "Text added to memory!"
 
+# Gradio UI
+demo = gr.Blocks()
+with demo:
+    gr.Markdown("# 🦙 TinyChat - AI Chatbot")
+    with gr.Row():
+        chatbot = gr.Chatbot()
+    with gr.Row():
+        user_input = gr.Textbox(label="Your Message")
+        send_btn = gr.Button("Send")
+    with gr.Row():
+        add_text_input = gr.Textbox(label="Add Knowledge to AI")
+        add_text_btn = gr.Button("Add Text")
+    
+    send_btn.click(chat_interface, inputs=user_input, outputs=chatbot)
+    add_text_btn.click(add_text_interface, inputs=add_text_input, outputs=gr.Textbox())
 
+# Launch
 if __name__ == "__main__":
     demo.launch()