Update app.py

app.py

@@ -6,59 +6,62 @@ For more information on `huggingface_hub` Inference API support, please check th
 """
 client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
 
+from model import load_model, load_tokenizer
+from utils import clean_output, get_shap_values
+import torch
+import shap
+import matplotlib.pyplot as plt
 
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
+tokenizer = load_tokenizer()
+model = load_model()
 
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
+def gradio_generate(context, num_questions, max_length):
+    input_prompt = f"generate question: {context.strip()}"
+    inputs = tokenizer(input_prompt, return_tensors="pt", truncation=True, padding="longest").to(model.device)
+    outputs = model.generate(
+        input_ids=inputs["input_ids"],
+        attention_mask=inputs["attention_mask"],
+        max_length=max_length,
+        num_return_sequences=num_questions,
+        do_sample=True,
+        top_p=0.95,
+        temperature=1.0
+    )
+    decoded = tokenizer.batch_decode(outputs, skip_special_tokens=True)
+    questions = clean_output(decoded)
+    return "\n".join([f"{i+1}. {q}" for i, q in enumerate(questions)])
 
-    messages.append({"role": "user", "content": message})
+def gradio_shap(context):
+    input_prompt = f"generate question: {context.strip()}"
+    try:
+        shap_values, tokens = get_shap_values(tokenizer, model, input_prompt)
+        fig, ax = plt.subplots(figsize=(10, 2))
+        shap.plots.text(shap.Explanation(values=shap_values, data=tokens), display=False)
+        plt.tight_layout()
+        return fig
+    except Exception as e:
+        return f"SHAP explanation failed: {e}"
 
-    response = ""
-
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
-
-        response += token
-        yield response
-
-
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
-demo = gr.ChatInterface(
-    respond,
-    additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
-    ],
-)
+with gr.Blocks() as demo:
+    gr.Markdown("# 🧠 C3QG – Context-Controlled Question Generation with FLAN-T5")
+    context = gr.Textbox(label="📄 Paste your context paragraph here:", lines=6)
+    num_questions = gr.Slider(1, 5, value=3, label="Number of Questions")
+    max_length = gr.Slider(32, 128, value=64, label="Max Output Tokens")
+    generate_btn = gr.Button("🔄 Generate Questions")
+    questions_output = gr.Textbox(label="🎯 Generated Questions")
+    shap_btn = gr.Button("Show SHAP Explanation")
+    shap_output = gr.Plot(label="SHAP Token Importance")
 
+    generate_btn.click(
+        gradio_generate,
+        inputs=[context, num_questions, max_length],
+        outputs=questions_output
+    )
+    shap_btn.click(
+        gradio_shap,
+        inputs=context,
+        outputs=shap_output
+    )
 
 if __name__ == "__main__":
     demo.launch()