Update app.py

app.py

@@ -1,26 +1,20 @@
 import gradio as gr
 import time
 from ctransformers import AutoModelForCausalLM  # Please ensure this import is correct
-from download_model import download_model
+from huggingface_hub import hf_hub_download
 
 PROMPT_TEMPLATE = (
-    "<s>" "[INST]"
-    "<<SYS>>"
-    """You are a dedicated public health assistant, trained to support community health workers (CHWs) in their essential role of enhancing community 	health. Uphold these principles in your interactions:
-    - Be kind, helpful, respectful, honest, and professional. Think step by step before answering each question. Think about whether this is the right answer, would others agree with it? Improve your answer as needed.
-    - Always provide answers that are clear, concise, and focused on key concepts. Highlight main points and avoid unnecessary repetition.
-    - Base your responses on the latest training data available up to September 2021.
-    - Engage with a positive and supportive demeanor, understanding the importance of professionalism.
-    - Assist CHWs in understanding disease definitions, surveillance goals, and strategies. Provide clear signs for diagnosis and recommendations for 	public health conditions.
-    - Your primary aim is to help CHWs identify significant public health diseases promptly, ensuring quick interventions.
-    - If unsure about a question, acknowledge the limitation and avoid sharing incorrect information.
-    """
-    "<</SYS>>" "[/INST]" "</s>"
+  
 )
 
 def load_llm():
-    llm = AutoModelForCausalLM.from_pretrained("Llama-2-7b-chat-q8-gguf",
-    model_type='llama',
+
+    # Set gpu_layers to the number of layers to offload to GPU. Set to 0 if no GPU acceleration is available on your system.
+    llm = AutoModelForCausalLM.from_pretrained(
+    "s3nh/PY007-TinyLlama-1.1B-Chat-v0.2-GGUF",
+    model_file="PY007-TinyLlama-1.1B-Chat-v0.2.Q4_K_M.gguf",
+    model_type="llama",
+    gpu_layers=0
     max_new_tokens = 1096,
     repetition_penalty = 1.13,
     temperature = 0.1
@@ -36,7 +30,7 @@ def llm_function(message, chat_history):
     output_texts = response
     return output_texts
 
-title = "Llama 7B GGUF Demo"
+title = "这里是小兮辞"
 
 examples = [
     'What is yellow fever.',