app.py

import gradio as gr
import numpy as np
import pandas as pd
import re
from tensorflow.keras.models import Sequential
from tensorflow.keras.models import load_model
from tensorflow.keras.layers import Dense
from transformers import BertTokenizer, TFBertModel
from sklearn.model_selection import train_test_split
from sklearn.metrics import confusion_matrix, classification_report
from nltk.corpus import stopwords
import tensorflow as tf
import nltk

# Download stopwords
nltk.download('stopwords')

# Load tokenizer and model
tokenizer = BertTokenizer.from_pretrained("bert-base-uncased")
bert_model = TFBertModel.from_pretrained("bert-base-uncased")


# Clean text
def remove_tags(txt):
    result = re.sub(r'<[^>]+>', '', txt)
    result = re.sub(r'https?://\S+', '', result)
    result = re.sub(r'[^a-zA-Z0-9\s]', ' ', result)
    return result.lower()

def remove_stop_words(txt):
    stop_words = set(stopwords.words('english'))
    return ' '.join([word for word in txt.split() if word not in stop_words])


# Load the trained model
classifier=load_model('movie_sentiment_model.h5')

# Single input BERT embeddings
def bert_embeddings(text, max_length=64):
    inputs = tokenizer(
        [text],
        return_tensors="tf",
        padding=True,
        truncation=True,
        max_length=max_length
    )
    outputs = bert_model(inputs['input_ids'], attention_mask=inputs['attention_mask'])
    cls_embeddings = outputs.last_hidden_state[:, 0, :]
    return cls_embeddings.numpy()

# Define Gradio function
def fn(test_review):
    review = remove_tags(test_review)
    review = remove_stop_words(review)
    cls_embeddings = bert_embeddings(review)
    prediction = classifier.predict(cls_embeddings)
    return "Positive" if prediction[0] > 0.5 else "Negative"

# Gradio Interface
description = "Give a review of a movie that you like (or hate, sarcasm ) and the model will let you know just how much your review truly reflects your emotions"
input_text = gr.Textbox(label="Enter Movie Review Text in English")
output_text = gr.Textbox(label="Output Text")

app = gr.Interface(
    fn=fn,
    inputs=input_text,
    outputs=output_text,
    title="Sentiment Analysis of Movie Reviews in English",
    description=description,
    allow_flagging="auto",
    flagging_dir='flagging_records'
)

app.launch(inline=False)