Update app.py

app.py

@@ -1,17 +1,32 @@
-import gradio as gr
+# Install necessary libraries
+# !pip install transformers
 
-# Load the fine-tuned Llama 2 model
-model = "models/mlabonne/llama-2-7b-guanaco"
+from transformers import AutoTokenizer, pipeline
+import torch
 
-# Create a Gradio interface
-iface = gr.Interface(
-    fn=model,  # Use the fine-tuned Llama 2 model as the function
-    inputs="text",  # Input is text
-    outputs="text",  # Output is also text
-    live=True,  # Enable live updates without button click
-    theme="compact",  # Use a compact theme for the interface
-    description="Fine-tuned Llama 2: Enter a prompt to get a model-generated response."
+# Model and prompt details
+model_name = "mlabonne/llama-2-7b-guanaco"
+prompt = "What is a large language model?"
+
+# Load tokenizer and pipeline
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+text_generation_pipeline = pipeline(
+    "text-generation",
+    model=model_name,
+    torch_dtype=torch.float16,
+    device_map="auto",
+)
+
+# Generate text using the provided prompt
+sequences = text_generation_pipeline(
+    f'<s>[INST] {prompt} [/INST]',
+    do_sample=True,
+    top_k=10,
+    num_return_sequences=1,
+    eos_token_id=tokenizer.eos_token_id,
+    max_length=200,
 )
 
-# Launch the Gradio app
-iface.launch()
+# Print the generated text
+for seq in sequences:
+    print(f"Generated Text: {seq['generated_text']}")