Update app.py

app.py

@@ -4,14 +4,15 @@ from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline
 # Define the path where the model and tokenizer are saved
 save_directory = "RAG_model"
 
-
 # Load the model and tokenizer from the saved directory
-@st.cache(allow_output_mutation=True)
+@st.cache_resource
 def load_model():
     model = AutoModelForCausalLM.from_pretrained(
         save_directory,
-        torch_dtype="auto",  # Ensure automatic dtype selection
-        device_map="cpu"  # Explicitly set to CPU
+        torch_dtype=None,  # Ensure dtype is not set to a quantization dtype
+        device_map=None,   # Ensure no device_map for CPU
+        load_in_8bit=False,  # Ensure quantization is not used
+        load_in_4bit=False   # Ensure quantization is not used
     )
     tokenizer = AutoTokenizer.from_pretrained(save_directory)
     return model, tokenizer
@@ -51,4 +52,4 @@ if st.button("Generate"):
         st.write("Please enter a prompt to generate text.")
 
 # Add an example usage
-st.write("Example usage: Enter a prompt like 'What is Artificial Intelligence?' and click 'Generate'.")
+st.write("Example usage: Enter a prompt like 'What is Artificial Intelligence?' and click 'Generate'.")