Upload 2 files

app.py

@@ -1,5 +1,3 @@
-import torch
-from transformers import AutoModelForCausalLM, AutoTokenizer
 import openvino_genai
 import gradio as gr
 
@@ -19,9 +17,6 @@ pipe = openvino_genai.LLMPipeline(model=base_model_name, device=device, adapters
 print("Generate with LoRA adapter and alpha set to 0.75:")
 #print(pipe.generate(args.prompt, max_new_tokens=100, adapters=openvino_genai.AdapterConfig(adapter, 0.75)))
 
-# Load tokenizer
-tokenizer = AutoTokenizer.from_pretrained(base_model_name)
-tokenizer.pad_token = tokenizer.eos_token
 
 # Define prediction function
 def generate_response(prompt):

requirements.txt

@@ -1,5 +1,3 @@
-torch
 gradio
-transformers 
 huggingface_hub
 openvino-genai