app.py

import gradio as gr
from transformers import AutoTokenizer, AutoModelForSequenceClassification
import torch

# Load the model from the Hub or local directory
model_name = "mjpsm/recommendation-overview-classification-model"  # 🔁 Replace with your model path
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForSequenceClassification.from_pretrained(model_name)
id2label = model.config.id2label

# Inference function
def predict_tag(text):
    inputs = tokenizer(text, return_tensors="pt", truncation=True, padding=True)
    with torch.no_grad():
        outputs = model(**inputs)
    logits = outputs.logits
    predicted_class_id = torch.argmax(logits, dim=1).item()
    predicted_label = id2label[predicted_class_id]
    return predicted_label

# Gradio UI
demo = gr.Interface(
    fn=predict_tag,
    inputs=gr.Textbox(lines=4, placeholder="Enter student reflection..."),
    outputs="text",
    title="🧠 Recommendation Overview Classifier",
    description="Enter a student's reflection after a math game. The model will return a motivational recommendation tag.",
    examples=[
        "I got frustrated when I made a mistake but I didn’t give up.",
        "I asked my classmate for help and it finally made sense.",
        "It felt like budgeting in real life when I played that part of the game.",
        "Even though I was confused, I tried a new strategy and it worked.",
    ],
)

if __name__ == "__main__":
    demo.launch()