Update app.py

app.py

@@ -6,15 +6,12 @@ import torch
 import optimum
 from transformers import (AutoModelForCausalLM, AutoTokenizer, LlamaForCausalLM, LlamaTokenizer, GenerationConfig, pipeline,)
 from fastapi.middleware.cors import CORSMiddleware
-from pyngrok import ngrok
 
 os.environ['PYTORCH_CUDA_ALLOC_CONF'] = 'expandable_segments:True'
 torch.cuda.empty_cache()
 torch.cuda.set_per_process_memory_fraction(0.8)  # Adjust the fraction as needed
 
 app = FastAPI(root_path="/api/v1")
-ngrok_tunnel = ngrok.connect(7860)
-print(ngrok_tunnel.public_url)
   
 # Load the model and tokenizer
 model_name_or_path = "TheBloke/Wizard-Vicuna-7B-Uncensored-GPTQ"