Updated app

app.py

@@ -11,18 +11,7 @@ model = AutoModelForCausalLM.from_pretrained(
     torch_dtype="auto",
 ).to("cuda" if torch.cuda.is_available() else "cpu")  # Check for GPU availability
 
-# Define the Gradio interface
-iface = gr.Interface(
-    fn=generate_code,
-    inputs=[gr.Textbox(lines=2, placeholder="Enter your Python code prompt")],
-    outputs="textbox",
-    title="Python Code Completion",
-    description="Generate code completions using a large language model.",
-)
-
 # Define the main function for code generation
-
-
 def generate_code(prompt):
     inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
     tokens = model.generate(
@@ -34,6 +23,15 @@ def generate_code(prompt):
     generated_code = tokenizer.decode(tokens[0], skip_special_tokens=True)
     return generated_code
 
+# Define the Gradio interface
+iface = gr.Interface(
+    fn=generate_code,
+    inputs=[gr.Textbox(lines=2, placeholder="Enter your Python code prompt")],
+    outputs="textbox",
+    title="Python Code Completion",
+    description="Generate code completions using a large language model.",
+)
+
 
 # Launch the Gradio app
 iface.launch()

requirements.txt

@@ -1,3 +1,5 @@
-torch
 gradio
+torch
 transformers
+einops
+accelerate