lab2 test

app.py

@@ -1,11 +1,12 @@
 import gradio as gr
 
-# Load model directly
-from transformers import AutoModel, AutoTokenizer
+from llama_cpp import Llama
+
+llm = Llama.from_pretrained(
+    repo_id="ID2223JR/gguf_model",
+    filename="GGUF_FILE",
+)
 
-# Load the LoRA model and tokenizer
-tokenizer = AutoTokenizer.from_pretrained("ID2223JR/lora_model")
-model = AutoModel.from_pretrained("ID2223JR/lora_model")
 
 # Data storage
 ingredients_list = []
@@ -39,13 +40,8 @@ def submit_to_model():
         ingredients_list
     )
 
-    # Tokenize and pass the prompt to the model
-    inputs = tokenizer(prompt, return_tensors="pt")
-    outputs = model.generate(**inputs, max_new_tokens=100)
-
-    # Decode the model output
-    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
-    return response
+    response = llm.create_chat_completion(messages=prompt)
+    return response.choices[0].text
 
 
 # App

requirements.txt

@@ -1,3 +1,2 @@
-transformers==4.46.3
 gradio==5.1.0
-torch==2.5.1
+llama_cpp==0.2.24