lab2 test

app.py

@@ -1,10 +1,17 @@
 import gradio as gr
 
 # Load model directly
-from transformers import AutoModel, AutoTokenizer
+# from transformers import AutoModel, AutoTokenizer
 
-model = AutoModel.from_pretrained("ID2223JR/gguf_model")
-tokenizer = AutoTokenizer.from_pretrained("ID2223JR/gguf_model")
+# model = AutoModel.from_pretrained("ID2223JR/gguf_model")
+# tokenizer = AutoTokenizer.from_pretrained("ID2223JR/gguf_model")
+
+from llama_cpp import Llama
+
+llm = Llama.from_pretrained(
+    repo_id="ID2223JR/gguf_model",
+    filename="GGUF_FILE",
+)
 
 # Data storage
 ingredients_list = []
@@ -38,12 +45,9 @@ 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)
+    response = llm.create_chat_completion(
+        messages=prompt,
+    )
     return response
 
 

requirements.txt

@@ -1,4 +1,5 @@
 gradio==5.1.0
 llama-cpp-python==0.2.24
 transformers==4.46.3
-torch==2.5.1
+torch==2.5.1
+hugingface-hub==0.25.2