Create app.py

app.py

@@ -0,0 +1,113 @@
+import gradio as gr
+import torch
+from transformers import AutoTokenizer, AutoModelForCausalLM
+
+# Load CodeGen model and tokenizer
+model_name = "Salesforce/codegen-2B-mono"  # Replace with the specific CodeGen model variant you need
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForCausalLM.from_pretrained(model_name)
+device = "cuda" if torch.cuda.is_available() else "cpu"
+model = model.to(device)
+
+def generate_response(input_text, max_length, temperature, top_p, top_k):
+    """
+    Generate response using the CodeGen model based on user input and selected parameters.
+    
+    Args:
+        input_text (str): The prompt or question for the model.
+        max_length (int): Maximum length of the generated text.
+        temperature (float): Sampling temperature for response creativity.
+        top_p (float): Nucleus sampling for generating top-p probable tokens.
+        top_k (int): Top-k sampling for generating top-k probable tokens.
+        
+    Returns:
+        str: Generated response from CodeGen.
+    """
+    try:
+        # Encode input and prepare input tensor
+        inputs = tokenizer(input_text, return_tensors="pt").to(device)
+        
+        # Generate text based on model output
+        outputs = model.generate(
+            inputs.input_ids,
+            max_length=max_length,
+            temperature=temperature,
+            top_p=top_p,
+            top_k=top_k,
+            do_sample=True,
+            num_return_sequences=1,
+            no_repeat_ngram_size=2
+        )
+        
+        # Decode and return the generated text
+        response = tokenizer.decode(outputs[0], skip_special_tokens=True)
+        return response
+    
+    except Exception as e:
+        return f"Error: {str(e)}"
+
+# Create Gradio interface
+with gr.Blocks() as codegen_app:
+    gr.Markdown("# CodeGen-powered Text Generation")
+    gr.Markdown("Generate high-quality, high-quantity output using the CodeGen model.")
+
+    # Input box for user prompt
+    with gr.Row():
+        input_text = gr.Textbox(
+            label="Input Text",
+            placeholder="Type your question or prompt here",
+            lines=3
+        )
+
+    # Sliders for customization
+    with gr.Row():
+        max_length = gr.Slider(
+            label="Max Length",
+            minimum=50,
+            maximum=1024,
+            step=10,
+            value=250,
+            interactive=True
+        )
+        temperature = gr.Slider(
+            label="Temperature",
+            minimum=0.1,
+            maximum=1.0,
+            step=0.1,
+            value=0.7,
+            interactive=True
+        )
+        top_p = gr.Slider(
+            label="Top-p (Nucleus Sampling)",
+            minimum=0.1,
+            maximum=1.0,
+            step=0.1,
+            value=0.9,
+            interactive=True
+        )
+        top_k = gr.Slider(
+            label="Top-k (Sampling Limit)",
+            minimum=0,
+            maximum=100,
+            step=5,
+            value=50,
+            interactive=True
+        )
+
+    # Output box
+    output_text = gr.Textbox(
+        label="Generated Response",
+        placeholder="The model's response will appear here",
+        lines=15
+    )
+
+    # Generate button to trigger response generation
+    generate_button = gr.Button("Generate Response")
+    generate_button.click(
+        fn=generate_response,
+        inputs=[input_text, max_length, temperature, top_p, top_k],
+        outputs=output_text
+    )
+
+# Launch the app
+codegen_app.launch()