Create app.py

app.py

@@ -0,0 +1,45 @@
+import gradio as gr
+import torch
+from transformers import AutoModelForMultipleChoice, AutoTokenizer
+
+# Load the model and tokenizer
+model_id = "roberta-large-mnli"
+model = AutoModelForMultipleChoice.from_pretrained(model_id)
+tokenizer = AutoTokenizer.from_pretrained(model_id)
+
+# Define the preprocessing function
+def preprocess(sample):
+    question = sample["question"]
+    choices = [sample[choice] for choice in "ABCDE"]
+    inputs = [f"{question} {choice}" for choice in choices]
+    tokenized = tokenizer(inputs, truncation=True, padding=True, return_tensors="pt")
+    return {
+        "input_ids": tokenized["input_ids"],
+        "attention_mask": tokenized["attention_mask"]
+    }
+
+# Define the prediction function
+def predict(data):
+    inputs = torch.stack(data["input_ids"])
+    masks = torch.stack(data["attention_mask"])
+    with torch.no_grad():
+        logits = model(inputs, attention_mask=masks).logits
+    predicted_indices = torch.argmax(logits, dim=1)
+    answers = [chr(ord('A') + idx) for idx in predicted_indices]
+    return answers
+
+# Create the Gradio interface
+iface = gr.Interface(
+    fn=predict,
+    inputs=gr.inputs.Input(type="json"),
+    outputs=gr.outputs.Label(num_top_classes=1, label="Predicted Answer"),
+    live=True,
+    examples=[
+        {"question": "What is the capital of France?", "A": "Paris", "B": "London", "C": "Berlin", "D": "Madrid", "E": "Rome"}
+    ],
+    title="Multiple-Choice Question Answering",
+    description="Enter a question and answer choices (A to E) to get the predicted answer.",
+)
+
+# Run the interface
+iface.launch()