Update app.py

app.py

@@ -10,19 +10,50 @@ from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
 
 subprocess.run('pip install flash-attn --no-build-isolation', env={'FLASH_ATTENTION_SKIP_CUDA_BUILD': "TRUE"}, shell=True)
 
-MODEL_ID = "nikravan/Marco_o1_q4"
-
+MODEL_ID = "nikravan/Marco-O1-q4"
 CHAT_TEMPLATE = "ChatML"
 MODEL_NAME = MODEL_ID.split("/")[-1]
 CONTEXT_LENGTH = 16000
 
-COLOR = "blue"  # تغییر رنگ به مشکی
-EMOJI = "🤖"
-DESCRIPTION = f"This is the {MODEL_NAME} model designed for testing thinking for general AI tasks."
+# Estableciendo valores directamente para las variables
+COLOR = "blue"  # Color predeterminado de la interfaz
+EMOJI = "🤖"  # Emoji predeterminado para el modelo
+DESCRIPTION = f"This is the {MODEL_NAME} model designed for testing thinking for general AI tasks."  # Descripción predeterminada
+
+latex_delimiters_set = [{
+        "left": "\\(",
+        "right": "\\)",
+        "display": False 
+    }, {
+        "left": "\\begin{equation}",
+        "right": "\\end{equation}",
+        "display": True 
+    }, {
+        "left": "\\begin{align}",
+        "right": "\\end{align}",
+        "display": True
+    }, {
+        "left": "\\begin{alignat}",
+        "right": "\\end{alignat}",
+        "display": True
+    }, {
+        "left": "\\begin{gather}",
+        "right": "\\end{gather}",
+        "display": True
+    }, {
+        "left": "\\begin{CD}",
+        "right": "\\end{CD}",
+        "display": True
+    }, {
+        "left": "\\[",
+        "right": "\\]",
+        "display": True
+    }]
 
 
 @spaces.GPU()
 def predict(message, history, system_prompt, temperature, max_new_tokens, top_k, repetition_penalty, top_p):
+    # Format history with a given chat template
     if CHAT_TEMPLATE == "Auto":
         stop_tokens = [tokenizer.eos_token_id]
         instruction = system_prompt + "\n\n"
@@ -43,6 +74,7 @@ def predict(message, history, system_prompt, temperature, max_new_tokens, top_k,
         instruction += f' {message} [/INST]'
     else:
         raise Exception("Incorrect chat template, select 'Auto', 'ChatML' or 'Mistral Instruct'")
+    print(instruction)
     
     streamer = TextIteratorStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True)
     enc = tokenizer(instruction, return_tensors="pt", padding=True, truncation=True)
@@ -70,11 +102,10 @@ def predict(message, history, system_prompt, temperature, max_new_tokens, top_k,
         outputs.append(new_token)
         if new_token in stop_tokens:
             break
-        result = "".join(outputs)
-        # تغییر قالب به Markdown و LaTeX
-        yield f"### $$ {result} $$"
+        yield "".join(outputs)
 
 
+# Load model
 device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
 quantization_config = BitsAndBytesConfig(
     load_in_4bit=True,
@@ -88,19 +119,27 @@ model = AutoModelForCausalLM.from_pretrained(
     attn_implementation="flash_attention_2",
 )
 
-gr.ChatInterface(
-    predict,
-    title=EMOJI + " " + MODEL_NAME,
-    description=DESCRIPTION,
-    additional_inputs_accordion=gr.Accordion(label="⚙️ Parameters", open=False),
-    additional_inputs=[
-        gr.Textbox("You are a code assistant.", label="System prompt"),
-        gr.Slider(0, 1, 0.3, label="Temperature"),
-        gr.Slider(128, 4096, 1024, label="Max new tokens"),
-        gr.Slider(1, 80, 40, label="Top K sampling"),
-        gr.Slider(0, 2, 1.1, label="Repetition penalty"),
-        gr.Slider(0, 1, 0.95, label="Top P sampling"),
-    ],
-    theme=gr.themes.Soft(primary_hue=COLOR),
-).queue().launch()
+# Create Gradio interface
+with gr.Blocks(theme=gr.themes.Soft(primary_hue=COLOR)) as demo:
+    chatbot = gr.Chatbot(label=EMOJI + " " + MODEL_NAME, latex=True)
+    system_prompt = gr.Textbox("You are a code assistant.", label="System prompt")
+    temperature = gr.Slider(0, 1, 0.3, label="Temperature")
+    max_new_tokens = gr.Slider(128, 4096, 1024, label="Max new tokens")
+    top_k = gr.Slider(1, 80, 40, label="Top K sampling")
+    repetition_penalty = gr.Slider(0, 2, 1.1, label="Repetition penalty")
+    top_p = gr.Slider(0, 1, 0.95, label="Top P sampling")
+    message = gr.Textbox(label="User Input")
+    submit = gr.Button("Submit")
+    
+    def respond(message, chatbot_history, system_prompt, temperature, max_new_tokens, top_k, repetition_penalty, top_p):
+        response = predict(message, chatbot_history, system_prompt, temperature, max_new_tokens, top_k, repetition_penalty, top_p)
+        chatbot_history.append((message, "".join(response)))
+        return chatbot_history
+    
+    submit.click(
+        respond, 
+        inputs=[message, chatbot, system_prompt, temperature, max_new_tokens, top_k, repetition_penalty, top_p],
+        outputs=chatbot
+    )
 
+demo.launch()