app.py

import torch
from transformers import DistilBertTokenizer, DistilBertForSequenceClassification
import gradio as gr
import re
import nltk
from nltk.tokenize import word_tokenize
from nltk.corpus import stopwords
from nltk.stem import WordNetLemmatizer

# Download NLTK resources (if not already available)
nltk.download('punkt_tab')
nltk.download('stopwords')
nltk.download('wordnet')

# Preprocessing setup
stop_words = set(stopwords.words('english'))
lemmatizer = WordNetLemmatizer()

def preprocess_text(text):
    text = re.sub(r'[^A-Za-z\s]', '', text)
    text = re.sub(r'http\S+|www\S+|https\S+', '', text)
    text = re.sub(r'\s+', ' ', text).strip()
    text = text.lower()
    tokens = word_tokenize(text)
    tokens = [word for word in tokens if word not in stop_words]
    tokens = [lemmatizer.lemmatize(word) for word in tokens]
    return ' '.join(tokens)

# Load tokenizer and model
tokenizer = DistilBertTokenizer.from_pretrained("distilbert-base-uncased")
model = DistilBertForSequenceClassification.from_pretrained("distilbert-base-uncased", num_labels=3)
model.load_state_dict(torch.load("sent_model.pth", map_location=torch.device("cpu")))
model.eval()

# ✅ Correct label mapping based on LabelEncoder
idx2label = {0: "negative", 1: "neutral", 2: "positive"}

# Prediction function
def predict(text):
    clean_text = preprocess_text(text)
    inputs = tokenizer(clean_text, return_tensors="pt", truncation=True, padding=True, max_length=128)
    with torch.no_grad():
        outputs = model(**inputs)
        probs = torch.nn.functional.softmax(outputs.logits, dim=1)[0].numpy()

    return {idx2label[i]: float(round(probs[i], 4)) for i in range(3)}

# Gradio UI
interface = gr.Interface(
    fn=predict,
    inputs=gr.Textbox(lines=4, placeholder="Enter a YouTube comment..."),
    outputs=gr.Label(num_top_classes=3),
    title="🎯 YouTube Comment Sentiment Classifier",
    description="Predicts sentiment (positive, neutral, or negative) from YouTube comments using a custom DistilBERT model.",
    examples=[
        ["This video is absolutely amazing!"],
        ["Not bad, but could be better."],
        ["I hated this. Waste of time."]
    ]
)

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