Update app.py

app.py

@@ -1,79 +1,69 @@
 import gradio as gr
 import torch
 from transformers import AutoTokenizer, AutoModelForCausalLM
-import spaces
 
-title = """# 🙋🏻‍♂️Welcome to 🌟Tonic's ☯️🧑‍💻Yi-Coder-9B-Chat Demo!"""
-description = """Yi-Coder-9B-Chat is a 9B parameter model fine-tuned for coding tasks. This demo showcases its ability to generate code based on your prompts. Yi-Coder is a series of open-source code language models that delivers state-of-the-art coding performance with fewer than 10 billion parameters. Excelling in long-context understanding with a maximum context length of 128K tokens. - Supporting 52 major programming languages:
-```bash
-  'java', 'markdown', 'python', 'php', 'javascript', 'c++', 'c#', 'c', 'typescript', 'html', 'go', 'java_server_pages', 'dart', 'objective-c', 'kotlin', 'tex', 'swift', 'ruby', 'sql', 'rust', 'css', 'yaml', 'matlab', 'lua', 'json', 'shell', 'visual_basic', 'scala', 'rmarkdown', 'pascal', 'fortran', 'haskell', 'assembly', 'perl', 'julia', 'cmake', 'groovy', 'ocaml', 'powershell', 'elixir', 'clojure', 'makefile', 'coffeescript', 'erlang', 'lisp', 'toml', 'batchfile', 'cobol', 'dockerfile', 'r', 'prolog', 'verilog'
-  ```
-### Join us : 
-🌟TeamTonic🌟 is always making cool demos! Join our active builder's 🛠️community 👻 [![Join us on Discord](https://img.shields.io/discord/1109943800132010065?label=Discord&logo=discord&style=flat-square)](https://discord.gg/qdfnvSPcqP) On 🤗Huggingface:[MultiTransformer](https://huggingface.co/MultiTransformer) On 🌐Github: [Tonic-AI](https://github.com/tonic-ai) & contribute to🌟 [Build Tonic](https://git.tonic-ai.com/contribute)🤗Big thanks to Yuvi Sharma and all the folks at huggingface for the community grant 🤗
-"""
+titulo = """# 🤖 Bienvenido al Chatbot con Yi-9B"""
 
-# Define the device and model path
-device = "cuda" if torch.cuda.is_available() else "cpu"
-model_path = "01-ai/Yi-Coder-9B-Chat"
+descripcion = """Este chatbot utiliza el modelo Yi de 9B parámetros para generar respuestas. 
+Puedes mantener una conversación fluida y realizar preguntas sobre diversos temas."""
 
-# Load the tokenizer and model
-tokenizer = AutoTokenizer.from_pretrained(model_path)
-model = AutoModelForCausalLM.from_pretrained(model_path, device_map="auto").eval()
+# Definir el dispositivo y la ruta del modelo
+dispositivo = "cuda" if torch.cuda.is_available() else "cpu"
+ruta_modelo = "01-ai/Yi-9B-Chat"
 
-@spaces.GPU(duration=130)
-def generate_code(system_prompt, user_prompt, max_length):
-    messages = [
-        {"role": "system", "content": system_prompt},
-        {"role": "user", "content": user_prompt}
+# Cargar el tokenizador y el modelo
+tokenizador = AutoTokenizer.from_pretrained(ruta_modelo)
+modelo = AutoModelForCausalLM.from_pretrained(ruta_modelo, device_map="auto").eval()
+
+def generar_respuesta(historial, usuario_input, max_longitud):
+    mensajes = [
+        {"role": "system", "content": "Eres un asistente útil y amigable. Proporciona respuestas claras y concisas."}
     ]
-    text = tokenizer.apply_chat_template(
-        messages,
+    
+    for entrada in historial:
+        mensajes.append({"role": "user", "content": entrada[0]})
+        mensajes.append({"role": "assistant", "content": entrada[1]})
+    
+    mensajes.append({"role": "user", "content": usuario_input})
+    
+    texto = tokenizador.apply_chat_template(
+        mensajes,
         tokenize=False,
         add_generation_prompt=True
     )
-    model_inputs = tokenizer([text], return_tensors="pt").to(device)
-
-    generated_ids = model.generate(
-        model_inputs.input_ids,
-        max_new_tokens=max_length,
-        eos_token_id=tokenizer.eos_token_id  
+    
+    entradas_modelo = tokenizador([texto], return_tensors="pt").to(dispositivo)
+    ids_generados = modelo.generate(
+        entradas_modelo.input_ids,
+        max_new_tokens=max_longitud,
+        eos_token_id=tokenizador.eos_token_id
     )
-    generated_ids = [
-        output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
+    
+    ids_generados = [
+        output_ids[len(input_ids):] for input_ids, output_ids in zip(entradas_modelo.input_ids, ids_generados)
     ]
+    
+    respuesta = tokenizador.batch_decode(ids_generados, skip_special_tokens=True)[0]
+    historial.append((usuario_input, respuesta))
+    return historial, ""
 
-    response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
-    return response
-
-def gradio_interface():
-    with gr.Blocks() as interface:
-        gr.Markdown(title)
-        gr.Markdown(description)
-
-        system_prompt_input = gr.Textbox(
-            label="☯️Yinstruction:",
-            value="You are a helpful coding assistant. Provide clear and concise code examples.",
-            lines=2
-        )
-        user_prompt_input = gr.Code(
-            label="🤔Coding Question",
-            value="Write a quick sort algorithm in Python.",
-            language="python",
-            lines=15
-        )
-        code_output = gr.Code(label="☯️Yi-Coder-7B", language='python', lines=20, interactive=True)
-        max_length_slider = gr.Slider(minimum=1, maximum=1800, value=650, label="Max Token Length")
-
-        generate_button = gr.Button("Generate Code")
-        generate_button.click(
-            generate_code,
-            inputs=[system_prompt_input, user_prompt_input, max_length_slider],
-            outputs=code_output
-        )
-
-    return interface
+def interfaz_gradio():
+    with gr.Blocks() as interfaz:
+        gr.Markdown(titulo)
+        gr.Markdown(descripcion)
+        
+        chatbot = gr.Chatbot(label="Historial de chat")
+        msg = gr.Textbox(label="Tu mensaje")
+        clear = gr.Button("Limpiar")
+        
+        max_longitud_slider = gr.Slider(minimum=1, maximum=1000, value=500, label="Longitud máxima de la respuesta")
+        
+        msg.submit(generar_respuesta, [chatbot, msg, max_longitud_slider], [chatbot, msg])
+        clear.click(lambda: None, None, chatbot, queue=False)
+    
+    return interfaz
 
 if __name__ == "__main__":
-    interface = gradio_interface()
-    interface.queue()
-    interface.launch()
+    interfaz = interfaz_gradio()
+    interfaz.queue()
+    interfaz.launch()