updated the app.py because of gpu error

app.py

@@ -1,6 +1,5 @@
 import gradio as gr
 from transformers import AutoTokenizer, AutoModelForCausalLM
-import torch
 from peft import PeftModel
 
 # Set the model name and parameters
@@ -9,8 +8,8 @@ max_seq_length = 2048  # Choose any! We auto support RoPE Scaling internally!
 dtype = None  # None for auto detection
 load_in_4bit = False  # Use False if we aren't using 4bit quantization
 
-# Check device availability (GPU or CPU)
-device = "cuda" if torch.cuda.is_available() else "cpu"
+# Set the device to CPU explicitly
+device = "cpu"
 
 # Load model and tokenizer
 
@@ -23,7 +22,7 @@ model = AutoModelForCausalLM.from_pretrained(base_model_name)
 adapter_model_name = "xiddiqui/News_Summarizer"  # Your model path on Hugging Face
 model = PeftModel.from_pretrained(model, adapter_model_name)
 
-# Move model to the appropriate device (GPU or CPU)
+# Move model to CPU (no need for GPU)
 model.to(device)
 
 # Define the summarization function
@@ -45,7 +44,7 @@ def generate_summary(input_text):
         return_tensors="pt", 
         truncation=True, 
         max_length=max_seq_length
-    ).to(device)  # Ensure computations are done on the same device as the model (CPU or GPU)
+    ).to(device)  # Ensure computations are done on the CPU
 
     # Generate summary
     summary_ids = model.generate(