app.py

# ==================================================================
# app.py for Hugging Face Gradio Space
# ==================================================================

# STEP 1: Install necessary libraries (will be installed by Gradio Space)
# !pip install gradio huggingface_hub fasttext

import gradio as gr
import fasttext
import re
from huggingface_hub import hf_hub_download
import os

# ==================================================================
# STEP 2: Download the FastText model from Hugging Face Hub
# ==================================================================
# Replace 'Sheshank2609/Complaint_Classifier' with your actual repo_id
REPO_ID = "Sheshank2609/Complaint_Classifier"
MODEL_FILENAME = "complaint_classifier.ftz"

# Download the model file. It will be cached locally in the Space.
model_path = hf_hub_download(repo_id=REPO_ID, filename=MODEL_FILENAME)

# Load the FastText model once
model = fasttext.load_model(model_path)

# ==================================================================
# STEP 3: Preprocessing Functions (MUST be the same as training)
# ==================================================================
def clean_text(text):
    text = str(text).lower()
    text = re.sub(r'\s+', ' ', text).strip()
    return text

def normalize_slang(text):
    replacements = {
        "bakwass": "bakwas",
        "boht": "bahut",
        "nhi": "nahi",
        "nai": "nahi",
        "yaarrr": "yaar",
        "yawwrrr": "yaar",
        "pls": "please",
        "plz": "please"
    }
    for k, v in replacements.items():
        text = text.replace(k, v)
    return text

def boost_keywords(text):
    if any(word in text for word in ["khana", "khaana", "Jevan", "food", "mess", "roti", "sabzi", "rice", "milk"]):
        text += " mess food quality eating"

    if any(word in text for word in ["ganda","flush","toilet", "washroom", "dirty", "dust", "garbage", "smell"]):
        text += " cleanliness hygiene sanitation"

    # TECH
    if any(word in text for word in ["wifi", "internet", "net", "network", "server", "pcs", "system"]):
        text += " technical network issue"

    # INFRA
    if any(word in text for word in ["ac", "fan", "light", "door", "bench", "lock", "window"]):
        text += " infrastructure maintenance"

    # ACADEMICS
    if any(word in text for word in ["teacher", "lecture", "class", "test", "exam", "assignment"]):
        text += " academics study"

    # RAGGING
    if any(word in text for word in ["ragging", "bully", "harass", "senior"]):
        text += " ragging harassment"

    return text

# ==================================================================
# STEP 4: Prediction Function for Gradio
# ==================================================================
def predict_complaint_gradio(complaint_text):
    # Apply the same preprocessing as during training
    processed_text = clean_text(complaint_text)
    processed_text = normalize_slang(processed_text)
    processed_text = boost_keywords(processed_text)

    # Make prediction
    # FastText predict returns a tuple: ([label], [probability])
    predictions = model.predict(processed_text, k=1)
    
    # Extract label and confidence
    label = predictions[0][0].replace("__label__", "")
    confidence = predictions[1][0] * 100

    return f"Department: {label}", f"Confidence: {confidence:.2f}%"

# ==================================================================
# STEP 5: Create Gradio Interface
# ==================================================================

# Define example inputs
examples = [
    ["The projector in room 301 is not working. It's urgent for the class."],
    ["My room is full of dust and cobwebs. Please send someone to clean it."],
    ["The food in the mess is very spicy and unhealthy."],
    ["The wifi is constantly disconnecting in my hostel room."],
    ["The teacher didn't explain the concept properly in today's lecture."],
    ["Seniors are bothering first-year students in the common area."]
]

iface = gr.Interface(
    fn=predict_complaint_gradio,
    inputs=gr.Textbox(lines=5, placeholder="Enter your complaint here..."),
    outputs=[gr.Textbox(label="Predicted Department"), gr.Textbox(label="Confidence")],
    title="Student Complaint Classifier",
    description="Enter a student complaint, and the model will classify it into the most relevant department along with a confidence score.",
    examples=examples,
    allow_flagging="manual" # If you want to allow users to flag incorrect predictions
)

# ==================================================================
# STEP 6: Launch the Gradio app
# ==================================================================
if __name__ == "__main__":
    iface.launch()