Create app.py

app.py

@@ -0,0 +1,53 @@
+import gradio as gr
+from transformers import pipeline
+
+def model_inference(model_name, task, input_data):
+    try:
+        # Load the model pipeline dynamically based on user selection
+        model_pipeline = pipeline(task, model=model_name)
+        # Perform the inference
+        result = model_pipeline(input_data, max_length=100)
+        # Handle different output formats
+        if isinstance(result, list):
+            return result[0]['generated_text'] if 'generated_text' in result[0] else str(result)
+        return result
+    except Exception as e:
+        # Return error message to the user interface
+        return f"An error occurred: {str(e)}"
+
+def setup_interface():
+    # Define the available models and tasks
+    models = {
+        "Text Generation": ["gpt2", "EleutherAI/gpt-neo-2.7B"],
+        "Text Classification": ["bert-base-uncased", "roberta-base"],
+        "Token Classification": ["dbmdz/bert-large-cased-finetuned-conll03-english"]
+    }
+
+    tasks = {
+        "Text Generation": "text-generation",
+        "Text Classification": "text-classification",
+        "Token Classification": "token-classification"
+    }
+
+    with gr.Blocks() as demo:
+        gr.Markdown("### Hugging Face Model Playground")
+        with gr.Row():
+            selected_task = gr.Dropdown(label="Select Task", choices=list(models.keys()), value="Text Generation")
+            model_name = gr.Dropdown(label="Select Model", choices=models[selected_task.value])
+        input_data = gr.Textbox(label="Input", placeholder="Type here...")
+        output = gr.Textbox(label="Output", placeholder="Results will appear here...")
+
+        # Update the model dropdown based on task selection
+        def update_models(task):
+            return gr.Dropdown.update(choices=models[task])
+
+        selected_task.change(fn=update_models, inputs=selected_task, outputs=model_name)
+
+        # Run model inference when input data changes
+        input_data.change(fn=model_inference, inputs=[model_name, selected_task, input_data], outputs=output)
+
+    return demo
+
+if __name__ == "__main__":
+    interface = setup_interface()
+    interface.launch()